Artificial Intelligence-based lifetime analytic calculations for intelligent devices of building infrastructure systems

The invention relates to the field of technical building infrastructure, in particular to building infrastructure systems such as light systems, building infrastructure management systems, and components including building infrastructure devices of such systems. A method for detecting anomalous events for a building infrastructure device, a corresponding program, a building infrastructure device, and a building infrastructure system are proposed. The invention concerns in particular use of data from intelligent building infrastructure devices in building infrastructure management systems for increasing availability of the building infrastructure system, e.g. by an improved building infrastructure control extending lifetime of the intelligent building infrastructure devices.

The technical building infrastructure, sometimes referred under the term building management system (BMS, also building automation system BAS), includes a computer-based control system installed in a building that controls and monitors the building's mechanical and electrical equipment including heating, ventilation, lighting, power systems, fire alarm systems, and security systems. A BMS consists of software and hardware components, wherein the software is usually configured in a hierarchical manner. The systems linked to a BMS typically represent a large proportion of the energy consumption of the building, in particular when lighting is included. BMS systems are a critical component to managing energy demand. Moreover, the technical building infrastructure includes systems that typically arrange a large number of devices over an extensive area in and around the building, which require regular maintenance, identification and replacement of defective equipment, and add further significant cost to maintaining the building infrastructure.

For the example of a light system, the devices may include luminaires, lighting modules for emitting light, driver devices for driving the lighting modules, presence detectors, switches, further control gear, light system management servers, linked via dedicated communication equipment, e.g. a DALI®, ZigBee® based control network, and emergency power supply including batteries for storing electric energy. Many of the components of the light system include intelligent devices that, e.g., include control circuits based on microcontrollers, ASICs, FPGAs. Intelligent devices may include light driver devices, sensor devices or light control devices.

Current driver devices forming a key element of modern light systems often provide basic operational data, collected by the driver device, in particular an LED driver device, and transmitted via a DALI® interface to a central light system management server. The operational data may be utilized in several ways, for example, the provided operational data may be used for tracking the power consumption of the individual devices in the entire light system, for monitoring a current LED current and a currently set dimming level of the LED driver device, as well as reporting failure modes. Thus, the operational data may be used for tracking the current status of the light system. Although there is already some information available in the operational data provided by the plurality of LED driver devices included in the light system, the available operational data may enable monitoring a current status of the light system and its power consumption, but lacks the capability to predict current and future states of the light system reliably with sufficient detail.

In short, the current use of status information provided by intelligent devices situated in the field after configuration of the building infrastructure system is not processed with regard to future control of the building infrastructure system, e.g. with regard to indication of a remaining lifetime of the building infrastructure devices or managing the building infrastructure system in a manner to increase system availability, and to reduce maintenance efforts of the building infrastructure system.

In a first aspect, the method according to claim i achieves the aforementioned object. The program according to a second aspect, the building infrastructure device according to third aspect, and the building infrastructure system according to a fourth aspect provide further advantageous solutions to the problem.

The features of the dependent claims define further advantageous embodiments.

The first aspect concerns the method for detecting an anomalous event in a building infrastructure device of a building infrastructure system. The method comprises, in a training phase, obtaining parameter data for at least one physical parameter of the building infrastructure device during operation of the building infrastructure device in the building infrastructure system during a predefined time period; generating a usage profile for the building infrastructure device based on the obtained parameter data, wherein the usage profile includes a characteristic parameter range for the at least one physical parameter determined based on the obtained parameter data, or training at least one artificial intelligence-based model for detecting at least one anomalous event in the building infrastructure device based on the obtained parameter data. The method further comprises, in a monitoring phase, obtaining current parameter data for the at least one physical parameter of the building infrastructure device; determining whether an anomalous event in the building infrastructure device has occurred by comparing the obtained current parameter data with the usage profile of the building infrastructure device, or determining whether the anomalous event in the building infrastructure device has occurred based on the obtained current parameter data, and the trained at least one artificial intelligence-based model; and generating and outputting a signal based on the determined anomalous event.

The method performs the training phase, corresponding to a usage profile generating phase, in the installed and configured building infrastructure system in the field, at the site and in the environment of the specific application.

The building infrastructure system may include or correspond to a light system. Given this case, the building infrastructure device may be a presence-detecting device, an ambient light sensor device or a light driver device, in particular an LED driver device. The building infrastructure device may be any intelligent component of the building infrastructure system. An intelligent component corresponds to an electronic device that includes a control circuit, e. g„ a programmable control circuit including integrated circuits (IC) such as a microcontroller, an ASIC and an FPGA.

In one specific embodiment, the method obtains, in a control circuit or a sensor circuit in the building infrastructure device, the parameter data and the current parameter data by measuring the at least one physical parameter of the building infrastructure device or by estimating based on measurements of the at least one physical parameter of the building infrastructure device, and generates the usage profile individually associated with the building infrastructure device.

An anomalous event is an event, in which at least one specific physical parameter of the building infrastructure device has a parameter value that deviates from the range of parameter values normally detected or expected due to the physical and in particular electrical design of the building infrastructure device. The physical parameter may have an uncertain or unexpected value. Occurrence of an anomalous event in the driver device includes that the building infrastructure device, at least for a limited time, is in an anomalous operational state. Occurrence of an anomalous event may also include that at at least one electrical or mechanical interface of the driver device with its environment, an unusual or unexpected value for a measurable physical characteristic is detected or measured.

Implementing the claimed method enables to use processes of predictive maintenance for the building infrastructure system based on events determined by evaluating physical parameters of the building infrastructure devices in the building infrastructure system. Performing predictive maintenance has the advantage of reducing downtime of the building infrastructure system, which in turn increases availability of the building infrastructure system and decreases overall operating cost of the building infrastructure system.

The method uses the available basic control infrastructure of the building infrastructure system for collecting parameter data on individual building infrastructure devices in their operational environment and determining individual usage profiles for the driver devices in their actual environment based on the collected parameter data. The determined usage profiles enable to derive from the perspective of the individual building infrastructure device, whether current parameter data for the physical parameter indicates problems for the individual building infrastructure device or the system, or indicates that there might arise problems in the future. For example, a conspicuous cluster of overvoltage spikes on a mains supply line representing an increased stress level for the elements at a mains supply interface of the building infrastructure device, may reduce expected lifespan of the electric circuit elements. An operator may take suitable measures based on the knowledge provided by the signal generated by the method, for example taking additional measures to suppress overvoltage spikes, envisaging an early replacement of the building infrastructure device, preferably with a type of building infrastructure device having a high resilience to overvoltage spikes on the mains supply interface.

By collecting parameter data from intelligent elements of the building infrastructure system, e.g. the light devices, different types of sensor devices and control devices, and combining the collected parameter data with the operational data in the lighting management system, the artificial intelligence based-model is trained to find anomalies, e.g. the anomalous events, in the data. The trained artificial intelligence-based model enables to rely upon known patterns in the data that were generated during the training phase in the operational building infrastructure system. The artificial intelligence-based model provides a realistic, e.g. for the specific application scenario in the field, explanation to the current data, and enables to identify a potential solution to an identified issue. The potential solution is output in the signal, which may inform a technician to correct a mains supply characteristic, which leads to a failure pattern due to mains characteristics, for example. The output signal may control a chatbot, for example, to generate and forward a message to an operator that includes information on a pattern found by using the artificial intelligence-based model, and a recommendation how to react on the anomalous event in order to improve an expected lifetime of elements of the building infrastructure system. The provided information in the signal thereby enables to increase operational lifetime of the devices of the of the building infrastructure system, to intelligently schedule maintenance of the building infrastructure system, and thereby reduces downtime and running cost of the building infrastructure system.

The method according to an embodiment includes, in the training phase, obtaining management data for operation of the building infrastructure system during the predefined time period; and training the at least one artificial intelligence-based model for detecting at least one anomalous event in the building infrastructure device based on the obtained parameter data and the obtained management data. In the monitoring phase, the method comprises obtaining current management data for operation of the building infrastructure system; and determining whether the anomalous event in the building infrastructure device has occurred based on the obtained current parameter data, the current management data and the trained at least one artificial intelligence-based model.

Management data is data from the building infrastructure management system level concerning operation of the building infrastructure system, which may, for example include time series of data, dimming characteristics during operation over time periods, and which does not relate to measured parameters, e.g. physical parameters, internally to the building infrastructure device.

Combining the parameter data acquired from the individual building infrastructure devices with the management data from the building infrastructure system, which relates to the actual operation of the system in its particular environment enables to train the artificial intelligence based-model to determine patterns in the current parameter data and management data, that enable to deduce early indications of the anomalous events.

According to an embodiment, the trained at least one artificial intelligence-based model includes a stored pattern including values of the parameter data for the at least one physical parameter of the building infrastructure device during operation of the building infrastructure device in the building infrastructure system during the predefined time period, and values for the management data for operation of the building infrastructure system the during the predefined time period, wherein the stored pattern is associated with determining the occurrence of the anomalous event. Stored patterns of parameter data, management data, and associated anomalous events, possibly also with control information for the signal for controlling output information and action recommendations enables to evaluate the current state of the system and possible future evolvements of the current state by applying techniques of pattern recognition and machine learning.

The method according to one embodiment includes generating a message to an operator based on the output signal, wherein the message includes at least one of information of the determined anomalous event, and a recommended action in response to the determined anomalous event.

The method enhances the understanding of the operator of the building infrastructure system in its specific application scenario by respective output information and provides targeted recommendations for improving availability of the system and reduces running cost, thereby maximizing return of investment for the specific installation of technical building infrastructure in the field.

According to an embodiment, the method includes computing an expected remaining lifetime of the building infrastructure device based on the obtained parameter data and the obtained management data.

The expected lifetime of a building infrastructure device, e.g. a light driver device, is dependent on its temperature during operation. This temperature is usually determined by the ambient temperature and also significantly influenced by a load current provided by the light driver device. The expected lifetime of the building infrastructure device for particular combinations of load currents and ambient temperatures may be listed in lifetime tables specific the particular type of building infrastructure device. A datasheet of the building infrastructure device may include such lifetime tables. For example, a lifetime table of a conventional LED light driver may reveal a lifetime of 50,000 h for the device running at an ambient temperature of e.g. 50°C corresponding to a temperature T _c of 8o°C of the light driver while outputting a load current of e.g. 400 mA at a 100% dimlevel to a LED module. However, since the LED light driver will not operate continuously at a dimlevel of 100%, the actual lifetime of the LED light driver might exceed the value shown in the lifetime table. Thus, the lifetime table often fails in showing realistic lifetime expectations of the LED light driver in their actual installation environment.

An electronic circuit of the building infrastructure device may comprise one or more electrolytic capacitors. Typically, the expected lifetime of the electronic circuit depends on the temperature of the electrolytic capacitors during operation. The electrolytic capacitors is regularly the initial point of failure due to temperature strain in the electronic circuit.

A further limiting factor on the expected lifetime of building infrastructure devices is a number of write/erase cycles of flash memory of a microcontroller of the building infrastructure device, which is a typical component of intelligent building infrastructure devices. For example, the flash memory of the building infrastructure devices may have a lifetime of 10k write/erase cycles.

Often, the specification of building infrastructure devices requires that a minimum lifetime of the gear operated at the most demanding operating point in the building infrastructure system needs to be equal to or exceed a minimum duration. This requirement leads to increased electric component cost, as more resilient and thus use of more expensive electric components ensures to reach the required minimum lifetime. Nevertheless, the real building infrastructure device is often operated for long periods at less demanding conditions. Thus, cheaper electric components would have been sufficient to reach the specified target lifetime.

The embodiment achieves the advantage that a remaining expected lifetime of the building infrastructure device is determined more accurately based on the actual use of the building infrastructure devices in the particular environment, e.g. based on parameter data of each individual building infrastructure device and operating conditions based on the actual infrastructure management data. For example, both the temperature of the building infrastructure device and the memory usage building infrastructure device can be considered.

The recommended action may include adapting at least one operational parameter of the building infrastructure device based on the determined anomalous event in the output signal for increasing the expected remaining lifetime of the building infrastructure device.

Thus, by adapting the operational parameters of the building infrastructure devices individually based on their individual expected remaining lifetime, the expected remaining lifetime of individual devices may be adjusted to each other, thereby achieving a simultaneous end of expected life for the building infrastructure devices of one area, e.g. one warehouse facility, and therefore reduce maintenance cost.

The building infrastructure device maybe a driver device, in particular a light driver device, a presence-detecting sensor device, an ambient light sensor device, or a building infrastructure control device. The cited examples of infrastructure devices usually include electronic circuitry with intelligent components, e.g. microcontrollers and memories, and often also comprise internal sensor circuits for determining temperatures, voltages and electric currents. Thus, these building infrastructure devices benefit particularly from the embodiments of the presented method. The method includes obtaining, in a control circuit or a sensor circuit of the building infrastructure device, the parameter data and the current parameter data by measuring the at least one physical parameter of the building infrastructure device or by estimating based on measurements of the at least one physical parameter of the building infrastructure device, and generating the usage profile individually associated with the building infrastructure device.

The method according to one embodiment comprises, in the monitoring phase, increasing a counter value of a counter of a predefined failure event when determining that the anomalous event has occurred, and comparing the counter value with a threshold value, and outputting the signal based on the determined anomalous event in case the counter value is equal to or exceeds the threshold value.

This embodiment enables to monitor anomalous events in the building infrastructure device, which in itself do not qualify as a highly critical failure, which is to be reported and acted upon immediately in order to mitigate the consequences for the building infrastructure system. The counted anomalous events may by their number of occurrence, or their frequency or rate of occurrence, indicate that a state of health of the building infrastructure device is decreasing over time.

According to a further embodiment, the method comprises, in the monitoring phase, increasing a counter value of a counter of the predefined failure event when determining that the anomalous event has occurred, and evaluating a rate of increase of the counter value, and outputting the signal based on the determined anomalous event in case the evaluated rate of increase of the counter value increases.

An increasing rate of anomalous events may indicate that a breakdown of the building infrastructure device may be imminent, and preventive replacement of the driver device may be advisable in order to avoid an at least partial shutting down of the building infrastructure system.

The method according to embodiment may, in the monitoring phase, comprise, in the step of comparing the obtained current parameter data with the usage profile of the building infrastructure device, comparing the obtained current parameter data with a characteristic parameter range included in the usage profile, and determining that the anomalous event occurred when the current parameter data is outside the characteristic parameter range.

Comparing whether the measured parameter value is within an expected range of parameter values as included in the usage profile of the building infrastructure device, makes for a computationally simple and simultaneously reliable detection of outliers for physical parameters during operation of the building infrastructure device. Implementing the comparison requires only limited processing resources, for example in a microprocessor already present in the building infrastructure device. Most building infrastructure device include integrated circuits, for example for controlling a switch of a switched power supply circuit, often used for purposes such as power factor correction (PFC) or DC/DC voltage conversion in the building infrastructure device.

According to an embodiment, the method comprises determining whether the anomalous event in the building infrastructure device has occurred by evaluating the obtained current parameter data and obtained current parameter data of at least one previous processing cycle, and the usage profile of the building infrastructure device with respect to trends in the current parameter data, current parameter data of the at least one previous processing cycle and the usage profile.

Analyzing the obtained current parameter data of a current processing cycle in the monitoring phase together with stored current parameter data of at least one previous processing cycle may yield further information on the building infrastructure device and its current state, as well as enable to predict future states of the building infrastructure device. For example, the obtained parameter values may steadily approach a limit of the parameter range included in the usage profile, without yet being outside the parameter range. Thus, determining a trend based on the current parameter data and the past instances of parameter data may enable to predict whether the parameter will violate the limits of the parameter range of the usage profile in the near future provided the determined trend continue. This prediction may involve a computationally efficient extrapolation of based on the determined trend for the physical parameter over current parameter data over at least two processing cycles. A variation over time of the current parameter data with regard to the parameter ranges included in the usage profile of the building infrastructure device may yield information on future values to expected in the current parameter data, and therefore of future system states of the building infrastructure system as a whole, and the individual building infrastructure device in particular.

The method may include, in the monitoring phase, updating the usage profile including the characteristic parameter range based on the current parameter data. This is in particular advantageous for adapting the usage profile, in particular the parameter ranges in the usage profile, to changes in environmental conditions or to aging processes of technical equipment, for example in the mains supply infrastructure or the building infrastructure device.

The at least one physical parameter may include at least one of an average dimming level for a predetermined time interval, a characteristic dimming level curve for the predetermined time interval, an average power output for the predetermined time interval, a mains supply voltage provided to the building infrastructure device, a mains supply current provided to the building infrastructure device, a mains frequency of the mains supply, an average power consumption of the building infrastructure device, a load voltage provided by the building infrastructure device, a load current provided by the building infrastructure device, a power loss of the building infrastructure device, an energy conversion efficiency of the building infrastructure device, an average rectifier half-bridge frequency, an internal DC bus voltage ripple value, a temperature of a control circuit of the building infrastructure device, a temperature on a printed circuit board of the building infrastructure device, a temperature within a housing of the building infrastructure device.

Known building infrastructure devices collect operational data with a limited extent. Contrary thereto, the proposed method is adapted to monitor and to analyze a wide range of physical parameters characterizing on the one hand external interfaces of the building infrastructure device with the building infrastructure system and, on the other hand, internal parameters of the building infrastructure device. Analyzing the cited physical parameters may reveal if there exist certain trends over time in the obtained parameter data, which may pose a threat to the building infrastructure device, which may indicate future problems, for example when anomalous events defined and detected based on these physical parameters occur with a certain predefined frequency. Measuring the cited physical parameters, and combining measuring the parameter values with a simple analysis of the measured values using the usage profile during the monitoring phase enables to implement an in-depth analysis while simultaneously the required processing resources are limited, and may even be provided by control circuits already present in the building infrastructure device. This enables an efficient implementation of the method, which is economically advantageous.

According to an embodiment, the method comprises, in the monitoring phase at least one of adjusting a least one setting of the building infrastructure device based on the output signal, storing data on the anomalous event based on the output signal in a data storage, and outputting, based on the output signal, an alert to a building management server or an operator.

The signal that is generated based on the detection of an anomalous event includes information on the driver device in its current application, which enables to take a range of actions based on the signal. For example, adapting settings of the driver device becomes possible, in order to improve its capability to cope with the detected anomalous event. Adapting settings of the driver device might be automatically initiated by the building management system server based on the signal output by the driver device.

Alternatively or additionally, data on the anomalous event may be stored in a log file of the building infrastructure device, either stored in the building infrastructure device, or stored associated with the building infrastructure device. This provides historical parameter data, which enables an off-line analysis of the environment of the building infrastructure device, or enables accumulating data on highly relevant events in the usage history of the building infrastructure device.

The method may, in the training phase, comprise obtaining initial parameter data, wherein the initial parameter data includes predefined parameter ranges for the at least one physical parameter determined based on a physical and/or electrical structure of the building infrastructure device and/or the building management system, and generating the usage profile for the building infrastructure device based on the obtained parameter data and the obtained initial parameter data.

The initial parameter data provides a suitable starting point for the profile generating phase of the method, wherein basic assumptions for the national parameter data may be readily available from the design process of the building infrastructure device.

According to an embodiment, the method comprises executing the training phase in case of determining a structural change of the building infrastructure system.

In particular, automatically initiating a rerun of the profile generating phase enables to react to changes in the building infrastructure system and to adapt the usage profile is of the individual building infrastructure device to the amended system structure by repeating the steps of the profile generating phase. This reduces the number of unnecessarily reported anomalous events in the monitoring phase after the changes to the building infrastructure system. Repeating the profile generating phase provides updated usage profiles and therefore enables keep track with the implications of the amendments during the succeeding monitoring phase. Operator inventions are thereby minimized.

According to an embodiment, an edge gateway device, a light management server or a building management server of the building infrastructure system may execute the steps of the method at least in part.

The versatile structure of the method enables its implementation in current building infrastructure device with only limited processing resources, for example a microcontroller or ASIC of an AC/DC converter circuit of the building infrastructure device. However, at least some method steps may be performed on intermediate layers of the building infrastructure system, for example by an edge gateway device, or alternatively by a central server, for example a light management server or building management system server. The implementations may differ with respect to the required processing resources and memory capacity for executing the individual processing steps of the method on the one hand, and on the other hand with regard to the required transmission bandwidth for transmitting the obtained parameter data, the obtained current parameter data, the usage profile data, or the signals over a communication network linking the components of the building infrastructure system.

According to an embodiment, the method comprises executing the training phase for a plurality of building infrastructure devices of the building management system; storing the generated usage profile for the building infrastructure devices of the plurality of building infrastructure devices; and, in the monitoring phase, obtaining the current parameter data for the at least one physical parameter of each building infrastructure devices of the plurality of building infrastructure devices; determining for each building infrastructure devices whether an anomalous event has occurred by comparing the obtained current parameter data with the stored usage profile for each building infrastructure devices of the plurality of building infrastructure devices; generating and outputting by each building infrastructure devices, the signal to the building infrastructure system in case the anomalous event is determined; wherein the method further comprises determining a current state of the building management system based on the signals output by the plurality of building infrastructure devices.

Performing the method for the plurality of building infrastructure devices installed in the building infrastructure system enables to derive information on the current state and to predict future states of the entire building infrastructure system, based on the combined information provided by the plurality of building infrastructure devices.

According to the second aspect, the computer program comprises instructions, which when the computer program is executed by a computer or signal processor, cause the computer or digital signal processor to cariy out the method of the first aspect.

A building infrastructure device according to the third aspect

A building infrastructure device, in particular a driver device or a light driver device, comprising a converter circuit configured to output a load current, and a control circuit configured to perform the method according to the first aspect or one of its embodiments.

The building infrastructure device according to an embodiment includes a sensor circuit for sensing the at least one physical parameter, wherein the at least one physical parameter is an electrical parameter of at least one electrical component of the converter circuit, wherein the at least one electrical component of the converter circuit comprises or corresponds to at least one critical component with respect to a lifetime of the converter circuit.

A building infrastructure system according to the fourth aspect comprises a plurality of building infrastructure devices according to the third aspect, and a building management server. The building management server is configured to execute the method according to the first aspect.

The plurality of building infrastructure devices may each comprise a converter circuit configured to output a load current and a communication circuit for connecting via a communication network to a building management server.

The computer program according to the second aspect, the driver device according to the third aspect and the building infrastructure system according to the fourth aspect achieve corresponding advantages as the method according to the first aspect.

The description of embodiments refers to the enclosed figures, in which

Fig. 1 is a simplified flowchart of the profile generating phase and the monitoring phase of the method for detecting anomalous events in a driver device of a building infrastructure system according to an embodiment; Fig. 2 shows a schematic view of major blocks of a light driver device for driving a lighting module according to an embodiment;

Fig. 3 illustrates the hierarchical structure of a building management system including a light system with a plurality of light driver devices according to an application.

Fig. 4A displays an example for a converter circuit in a driver device;

Fig. 4B displays an example for change of toN-values in the converter circuit of fig. 4A with time;

Fig. 4C displays an example for determining nominal ranges for toN-values in the converter circuit of fig. 4A; and

Fig. 5 provides an example for determining nominal ranges for temperature values; and

Fig. 6 is a simplified flowchart of the profile generating phase and the monitoring phase of a method for detecting anomalous events in a driver device of a building infrastructure system according to an embodiment.

In the figures, corresponding elements have the same reference signs. The proportions and dimensions of the elements shown in the figures do not represent the emergency luminaire and the emergency sign unit to scale, but are merely chosen to describe the structure and function of the emergency luminaire.

Fig. 1 is a simplified flowchart of the profile generating phase and the monitoring phase of the method for detecting anomalous events in a building infrastructure device of a building infrastructure system according to an embodiment.

The building infrastructure device of the building infrastructure system may be a light driver device 1. Alternatively, the building infrastructure device maybe, for example, a light control device or a sensor device in the light system as one example of a particular building infrastructure system.

The method starts with performing the training phase. The training phase in fig. corresponds to a usage profile generating phase. The method performs training phase after installation and configuration of the building infrastructure system in the field, in particular on the site of a customer and in the specific environment of the customer.

In step Si, the method obtains parameter data for at least one physical parameter of the light driver device 1 during operation of the light driver device 1 in the building infrastructure system. The step Si is performed for a predefined time period.

The predefined time period (timeframe) may extend over timespan of one day, one week or even several weeks or months. Preferably, the length of the predetermined time period is selected in order to cover plural usage cycles for buildings such as day or a week.

The at least one physical parameter may include at least one of

- an average power output for a time period provided by the light driver device i,

- a mains supply voltage provided to the light driver device i,

- a mains supply current provided to the light driver device 1,

- a mains frequency of the mains supply, and

- an average power consumption of the light driver device 1.

Additionally or alternatively, the at least one physical parameter may comprise at least one of

- a load voltage provided by the light driver device i,

-a load current provided by the light driver device i,

-a power loss of the light driver device i,

- an energy conversion efficiency of the light driver device 1,

- an average rectifier half-bridge frequency in the light driver device i,

- an internal DC bus voltage ripple value in the light driver device i,

- a temperature of a control circuit 3 of the light driver device 1,

- a temperature on a printed circuit board of the light driver device 1, and - a temperature within a housing of the light driver device 1.

Some examples for physical parameters and methods for obtaining them will be discussed with reference the light driver device i of fig. 2.

In step S2 following to step Si, the method generates a usage profile for the light driver device i based on the obtained parameter data, which was acquired in step Si. The generated usage profile includes a characteristic parameter range for the at least one physical parameter, wherein step S2 determines the characteristic parameter range based on the obtained parameter data.

The generated usage profile is subsequently stored.

The method proceeds then to a monitoring phase, which essentially comprises steps S3, S4 and Ss.In the monitoring phase, the method cyclically performs steps S3-S4-S3 or S3-S4-S5. The monitoring phase bases on the usage profile associated with light driver device 1 and its application in the building infrastructure system, which was generated in the profile generating phase.

In step S3, the method obtains current parameter data for the at least one physical parameter of the light driver device 1.

In step S4, the method proceeds by determining whether an anomalous event occurred in the light driver device 1. In particular, in step S4, the obtained current parameter data is compared with the usage profile of the lighting driver device 1.

In case the method determines in step S4, that the method does not detect the anomalous event based on the current parameter data and the usage profile, the method returns to step S3.

In case the method determines in step S4, that the anomalous event occurred, the method proceeds to step S5. In step S5, the method generates and outputs a signal to the building infrastructure system based on the determined anomalous event from step S4.

The method illustrated in fig. 1 may be expanded to include further steps defining further advantageous embodiments.

In the monitoring phase, the method may add a process for updating the usage profile. The process for updating the usage profile may start with determining whether the current parameter data enables updating or recommends updating of the stored usage profile of the light driver device 1. If determining, that the usage profile of the light driver device i, the method uses the obtained current parameter data of the light driver device i and generates an updated usage profile for the light driver device i based on the stored usage profile and the current parameter data.

The process of updating the usage profile of the light driver device i may run in parallel to the monitoring process of steps S3, S4, S5. Alternatively or additionally, the process of updating the usage may be initiated automatically at regular time intervals, for example times of a reduced processing load for the data processing equipment of the building infrastructure system executing the method.

Alternatively or additionally, the process of updating the usage profile may be initiated manually by an operator or triggered by specific events. For example, in case an operator determines by an inspection that the current parameter data leading to an alert communicating an anomalous event at the light driver device 1 presents no sufficient risk to the light driver circuit 1 that merits an alert, the operator may initiate process of updating the usage profile.

Alternatively or additionally, the process of updating the usage profile may be initiated manually or automatically, in case a structural change in the building infrastructure system occurred. For example, when adding additional devices or removing devices, or amending the power grid of the building providing the mains supply 7, executing the process of updating the usage profile may be particularly advantageous.

In the profile generating phase, the method may perform the process for generating the usage profile in step S2 based on the obtained parameter data from step Si, and further taking initial parameter data into account. In this particular embodiment, the method starts with a step of obtaining initial parameter data, wherein the initial parameter data includes predefined parameter ranges for the at least one parameter determined based on a physical and/ or electrical structure of the light driver device 1 and/or the building management system.

The initial parameter data may include parameter data that is generated by simulating or analysing the circuit diagrams of the electric circuits of light driver device 1, vaiying input signals to and output signals provided by the light driver device 1.

The step S2 of generating the usage profile for the light driver device 1 then generates the usage profile data based on the obtained parameter data, which is obtained in the step Si and in combination with the obtained initial parameter data. Fig. 2 shows a schematic view of major blocks of a light driver device 1 for driving a lighting module according to an embodiment.

The light driver i is a particular example for a building infrastructure device (building management device). The light driver device 1 may form part of a luminaire. Nevertheless, the invention is not restricted to a light driver device 1, the driver device may alternatively be part of a dimming device for a window by driving an electrically operated blind, a window locking device, or a roll shutter. The driver device may drive an actuator, such as a door opener or a window opener.

Alternatively, the driver device may form part of a heating system, a venting and/or cooling device, such as an air conditioner, an air ventilator, a fan. The driver device may not only be part of a stationary device, alternatively a mobile device, e.g. a cleaning device, such as a cleaning robot, a window cleaner; a central control unit for controlling one or more building management devices; a user interface for controlling one or more building management devices, such as a switch (e.g. light switch), a touch panel, input device etc.; and any combination of the aforementioned devices.

In addition or alternatively, the building management device may comprise or correspond to any other building management device known in the art.

The light driver device 1 of fig. 2 comprises an AC/DC converter circuit 2, the control circuit 3, and a communication circuit 4.

The light driver device 1 may be an LED driver.

The light driver device 1 arranges its subassemblies AC/DC converter circuit 2, control circuit 3, and communication circuit 4 on one or more printed circuit boards (PCB). The subassemblies include essentially electronic circuits including a plurality of active and passive electronic components linked via the electric connections arranged on surfaces and within the PCB. The at least one PCB is located within a housing providing mechanical protection for the subassemblies of the light driver device 1.

The light driver device 1 has a plurality of interfaces.

A mains supply interface 6 of the light driver device 1 connects the light driver device 1 to a mains supply 7 of the building providing an AC mains supply. The light driver device 1, in particular the AC/DC converter circuit 2, generates a DC load current ILED and outputs the generated DC load current ILED via an LED interface 8 of the light driver device 1 to the lighting module 9.

The AC/DC converter circuit 2 may include electric circuitiy, which is configured to perform as a rectifier circuit, e.g. implemented as a bridge or half-bridge rectifier, to perform power factor correction (PFC), to generate a DC bus voltage based on the AC mains voltage input to the mains supply interface 6, to perform DC/DC-conversion to convert the DC bus voltage to at least one DC output voltage ULED for supplying the lighting module 9 via the LED interface 8.

The AC/DC converter circuit 2 may include at least one switched mode power supply circuit (SMPS) including a switch controlled by the control circuit 3 with a control signal 11.

The AC/DC converter circuit 2 may include a low voltage power supply circuits for generating supply voltages for the electric circuits included in the light driver device 1, e.g. the control circuit 3 and the communication circuit 4.

The control circuit 3 may receive basic sensor signals 12 from the AC/DC converter circuit 2. These sensor signals 12 may in particular include sensor data currently measured in the lighting driver 1, e.g. presence of the AC mains supply voltage at the mains supply input 6.

The lighting module 9 may include one or typically a plurality of individual LEDs emitting light.

The light driver 1 may be an emergency light driver. The emergency light driver monitors presence or absence of the mains supply voltage at the mains supply interface 6. In case of detecting failure of the mains supply voltage based on the sensor signal 12, the control circuit 3 controls switching of the AC/DC converter circuit 2 to an alternate electric power source, e. g. to a DC energy storage such as a batteiy, in order to ensure a continuing load current ILED for a predetermined time with a predetermined current value at the LED interface 8 based on the alternate power source acting as an emergency power supply.

The control circuit 3 may receive external control commands in a driver control signal 13 for the light driver device 1 via the communication circuit 4 connected to a communication network. The driver control signal 13 may include the usage profile data generated externally to the light driver device 1. The control circuit 3 may generate and output a driver output signal 14 from the light driver device 1 to other devices in a light system 30 via the communication circuit 4 connected to the communication network. The driver output signal 14 may include the usage profile data generated internally to the light driver device 1, for example by the control circuit 3.

The driver output signal 14 may further include the parameter data and the current parameter data obtained by the light driver device 1, for example by the control circuit 3 and the sensor circuit 2.1.

The driver output signal 14 may further include the signal to the building infrastructure system in case the anomalous event is determined.

Fig. 2 displays the communication circuit 4 configured for wired communication via a communication interface 10 of the light driver device 1 to a communication bus 5 connecting other devices of a light infrastructure system with a light driver device 1.

Alternatively, or even additionally, the communication circuit 4 maybe adapted to perform wireless communication via one or more antennas with other devices of the light infrastructure system.

The communication circuit 4 and the communication interface 10 may communicate with the other devices of the light infrastructure system based on at least one of the lighting control standards DSI®, DALI®, DALI-2®, D4i® and KNX enabling a digital control of the light infrastructure system using a wired light control system.

The communication circuit 4 and the communication interface 10 may communicate with the other devices of the light infrastructure system based on at least one of the lighting control standards DALI+® enabling a digital control of the light infrastructure system using a wireless light control system. Additionally or alternatively, the wireless light control system may base on a wireless communication protocol defined by a wireless communication standard, e.g. ZigBee®, matter, Bluetooth Mesh®, and Bluetooth LE®.

A standard for communication circuit 4 and the communication interface 10 may base on the technologies defined in the series of technical standards known under IEC 62386.

The other devices connected to the communication interface 10 via the communication bus 5 may include system server 32, ON/OFF switches, dimmers, and presence detecting sensors. DALI+ devices communicate using existing DALI commands, but transmit and receive these commands over a wireless and/ or IP-based medium rather than the dedicated pair of wires used by DALI-2 and D4L

The control circuit 3 may be implemented using an integrated circuit (IC), e.g. a microcontroller, a microprocessor or an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA) or any combination of these elements. The control circuit 3 may form an integral part of the AC/DC converter circuit 2.

The control circuit 3 may include a data storage providing data storage capacity in form of a memory 3.1.

The memoiy 3.1 may store program data, including program data implementing at least parts of the method for detecting anomalous events in the light driver device 1. The memoiy 3.1 may also store application data when executing an application, in particular the program implementing at least parts of the method for detecting anomalous events in the light driver device 1.

The memoiy 3.1 may in particular store usage profile data.

Additionally or alternatively, the control circuit 3 and the memory 3.1 may generate, manage and store one or more event counters, which count the occurrence of anomalous events based on the current parameter data.

Additionally or alternatively, the memory 3.1 may store parameter data and/or current parameter data.

The AC/DC converter circuit 2 may include at least one sensor circuit 2.1.

The sensor circuit 2.1 is configured to obtain parameter data during the profile generating phase and current parameter data during the monitoring phase on at least one physical parameter of the light driver device 1. The parameter data may in particular be obtained by measuring the at least one physical parameter, or by computing or estimating the parameter data based on one or measurements of the at least one physical parameter.

The at least one physical parameter may include at least one electrical parameter characterizing the operation of the light driver device 1, and in particular the AC/DC converter circuit 2 in the building infrastructure system. Examples for the electrical parameter are - an average power output for a time period provided by the light driver device i via the LED interface 8,

- a load voltage provided by the light driver device lat the LED interface 8,

-a load current ILED provided by the light driver device 1 via the LED interface 8,

- a mains supply voltage provided to the light driver device i at the mains supply interface 6,

- a mains supply current provided to the light driver device 1 at the mains supply interface 6,

- a mains frequency of the mains supply at the mains supply interface 6,

- an average power consumption of the light driver device 1,

-a power loss of the light driver device i,

- an energy conversion efficiency of the light driver device 1,

- an average rectifier half-bridge frequency in the light driver device i, and

- an internal DC bus voltage ripple value in the light driver device i.

The electrical parameter, in particular including a voltage, a current or a frequency may be available in the control circuit 3, e.g., the control circuit 3 configured to control the AC/DC converter circuit 2, or may be measured by the sensor circuit 2.1. The sensor circuit 2.1 may include a shunt resistor and/or measuring bridge to provide suitable measurement signals, which may be processed in the control circuit 3 further in order to generate the parameter data.

The power loss of the light driver device 1 may be computed by subtracting an electrical output power of the light driver device 1 from the electrical input power of the light driver device 1.

The energy conversion efficiency of the light driver device 1 may be computed by dividing the electrical output power of the light driver device 1 by the electrical input power of the light driver device 1. The at least one physical parameter may include at least one operation parameter of the light driver device 1. The operation parameter may include one or more parameters for setting an operation state of the light driver device 1. The one or more operation parameters may correspond to control parameters for controlling operation of the light driver device 1. In the example of the light driver device 1 and the AC/DC converter circuit 2 providing a LED current ILED at the LED interface 8 to the lighting module 9, the at least one operation parameter may include a dimming level at which the lighting means 9 is controlled to operate. The diming level may have a percentage value ranging from 0% to 100%. A value of 0% for the dimming level corresponds to the LED module 9 emitting no light, and a value of 100% for the dimming level corresponds to the lighting means emitting a maximum of light. The at least one operation parameter may be known in the control circuit 3 controlling operation of the AC/DC converter circuit 2. Additionally or alternatively, the at least one operation parameter is received, e.g. via the communication circuit 4 and provided to the control circuit 3 in the control signal 13.

Additionally or alternatively, the at least one physical parameter may comprise at least one of a temperature parameter, e.g.

- a temperature of a control circuit 3 of the light driver device 1,

- a temperature on the printed circuit board of the light driver device 1, and

- a temperature within a housing of the light driver device 1.

The temperature may be measured already by the control circuit 3 configured to control the AC/DC converter circuit 2, or maybe measured by the sensor circuit 2.1.

Fig. 3 illustrates the hierarchical structure of a building management system including or corresponding to a light system 30 with a plurality of light driver devices 1 in the light system 30 according to an embodiment.

The building management system comprises a building management server 31.

The building management server 31 is connected with a light management system server 32. The light management server 32 forms part of the light system 30, which furthermore includes a plurality of driver devices 1. The driver devices 1 may the structure of the LED driver device discussed with reference to figure 2 above. The light system 30 arranges a number of driver devices 1 directly connected with a light management system server 32. Moreover, the light system 30 includes a plurality of other devices 33, which also connect directly to the light management server 32.

The other devices 33 may comprise, but are not limited to, a luminaire, a dimming device for a shading a window, such as an electrical blind, a sensor device, e. g. a movement detector or a presence detector, a security camera (CCTV), a smoke detector, an ambient light sensor, a humidity sensor, a temperature sensor, an audio sensor, e.g. microphone. The other devices 33 may include a sprinkler device, an alarm device, a door locking device, window locking device, or a roll shutter.

The other devices 33 may comprise an information output device, e.g. a display, a security sign, a loudspeaker.

The other devices 33 may comprise an actuator, e.g. a door opener, a window opener.

The other devices 33 may include a heating, venting and/ or cooling device, for example an air conditioner, an air ventilator, a fan, a heating device or a humidifier.

The other devices 33 may further comprise a cleaning device, such as an autonomous cleaning device, a window cleaner; a control module for controlling one or more of the building management devices, a user interface for controlling one or more building management devices, such as a switch, e.g. an ON/OFF-switch, a dimming module, a touch panel, a numeric input device.

The other devices 33 may include any combination of the aforementioned examples of other devices 33.

Plural driver devices 1 are connected via an edge gateway device 34 to the light management system server 32. There may be additional other devices 33 not shown in figure 3, which are connected with the edge gateway device 34.

The building management server 31, the light management system server 32, the edge gateway device 34, the driver devices 1, and the other devices 33 may be linked via one or more communication networks, which may include wired and wireless communication networks based on same or on different communication standards. Fig. 4A displays an example for a converter circuit 2’ in a building infrastructure devices. The converter circuit 2’ may, e.g. implement a PFC circuit or a DC/DC-converter circuit of a light driver device 1, in particular a LED driver device.

The converter circuit 2’ is a generally known example for switched mode power supply (SMPS) for generating an output current IOUT using a controlled switch SW, which may be implemented using a field effect transistor (FET, in particular a MOSFET).

The depicted converter circuit 2’ is a SMPS circuit in Buck circuit topology.

The output current IOUT may be the load current ILED for driving a load, e.g. for driving the lighting module. The input voltage VIN may be a bus voltage V _B us provided from mains supply via mains supply interface 6 of the light driver device 1.

At a specific operating point, the converter circuit 2’ provides the output current IOUT at an output voltage VOUT = 400 V from an input voltage VIN = 200 V. A switching frequency of the switch SW is fsw = 200 kHz. Thus, a duty cycle of a switch control signal corresponds to VOUT / VIN = 0.5, and a time toN during which the switch SW is closed, equals toN = 2.5 ps.

Monitoring the parameter time toN of the switch SW may be performed using an embodiment of the disclosed method, e.g. as discussed with reference to fig. i.Fig. 4B displays as simplified example for a gradual change of toN-values in the converter circuit 2’ of fig. 4A over a prolonged period of time.

Fig. 4B illustrates the change of toN-values over some years. The depicted curve 41 for toN exceeds a first threshold 42, at a time ti and reaches and then exceeds a second threshold 43 at a time t ₂ . Exceeding the first warning threshold 42 may result in a warning that the light driver device 1 may come approach an expected end of its estimated lifetime. Exceeding the second warning threshold may result in a failure alert that the light driver device 1 may not work according to specifications due reaching the expected end of its estimated lifetime.

Fig. 4C displays an example for determining nominal ranges for toN-values of the converter circuit 2’ of fig. 4A.

A sudden change of the value of toN indicates a problem in the converter circuit 2’. For example, an accelerated aging process of electrical components, e.g. increased conduction losses of the diode D of the converter circuit 2’ may result in the time toN increasing, subsequently exceeding the first and even the second threshold 43. Exceeding the first and second thresholds results in outputting the respective warning signal and even switching of the lighting driver device 1 and outputting a failure alert message. Fig. 4C displays a sudden increase in toN starting at a time t ₃ and resulting in significantly steeper linear increase of the toN-curve 44 after the time t ₄ elapsed. The increased rate of change of toN curve 44 after the time t ₄ when compared to the rate of change before the time t ₃ indicates accelerated aging processes of the electronic circuitry of the converter circuit 2’. The remaining expected lifetime of the converter circuit 2’ will decrease rapidly after time t ₄ .

An initial value 45 for toN may be measured after production, or after installation of the light driver device 1 at the site.

A parameter range for toN including a lower limit value 45 for toN and an upper limit value 46 for toN may be set based on the initial value 45 for toN. Optionally, as shown in fig. 4C, the parameter range for toN including a lower limit value 45 for toN and an upper limit value 46 for toN may be dynamically adapted, e.g. at regular intervals during operational life of the light driver device 1. This avoids erroneously detecting anomalous events and outputting faulty warning messages or failure alerts.

An update of the parameter range for toN maybe performed weekly, monthly or at other predetermined intervals and avoids erroneously detecting anomalous events and reporting thereon due to aging processes of electronic components of the converter circuit 2’. Fig. 5 provides an example for determining nominal ranges for temperature values.

Combining of plural different physical parameters enables drawing conclusions on other physical parameters of the light driver device 1, management parameters of the building infrastructure system, and even environmental parameters of the building infrastructure, e.g. an ambient temperature in the building. Fig. 5 illustrates calculating the ambient temperature based on the measured temperature of the light driver device.

After switching on the light driver device 1 in the field, an initial temperature, e.g. TINTT = 23 °C is measured. Measuring the initial temperature TINIT immediately after switching on provides a measurement, which includes not yet a temperature increase resulting from the light driver device 1 itself due to circuit losses in the electronic circuitiy. This requires that the light driver device 1 had been switched off before switching it on, that was sufficiently long for the light driver device 1 to cool down to the ambient temperature in the environment of the light driver device 1. After a predetermined minimum time has elapsed after switching the light driver device 1 on, temperature of the light driver device i is measured again. Fig. 5 depicts a measured value of T = 70 °C for the temperature of the light driver device 1.

The predetermined minimum time is determined such that the light driver devices 1 has reached a stable temperature (constant temperature). Assuming the ambient temperature being constant, the increase in temperature due to the light driver device 1 itself can be computed as the difference between the measured initial temperature TINTT and the measured temperature of the light driver device 1 after a predetermined minimum time has elapsed.

The temperature curve 52 corresponds to the measured temperature after switching the light driver device 1 on.

Based on the determined value for the increase in temperature due to the losses from operating the light driver device 1, for any measured value of the temperature of the light driver device 1, a corresponding value for the ambient temperature may be estimated or calculated.

In the simplified example illustrated in fig. 5, the increase in temperature due to the losses from operating the light driver device 1 amounts to AT = 70 °C - 23°C = 47 °C.

Fig. 5 shows a temperature curve 54 for the temperature measured by a temperature sensor circuit of the light driver device 1. For any temperature value of the temperature curve 54, a corresponding ambient temperature value of the ambient temperature curve 53 may be determined by estimation or calculation as discussed before.Fig. 6 is a simplified flowchart of the training phase and the monitoring phase of a method for detecting anomalous events in a driver device of a building infrastructure system according to an embodiment.

The flowchart illustrates that the methods of the embodiments include a training phase or usage profile generating phase and a monitoring phase that are performed after installing and configuring the building infrastructure system on a site, e.g. in the operation environment where the building infrastructure system will operate throughout its operational lifetime.

It should be noted, that the building infrastructure system is a stationary system that is differs from a mobile lighting system installed in a vehicle. The environment of the building infrastructure system is characterized by slow changes with time of the day, week, and season, and by usage cycles of human persons in or around a building. The method starts with the training phase.

In a first step S61, the method includes obtaining parameter data from intelligent components of the building infrastructure system. In particular, step S61 includes obtaining parameter data for at least one physical parameter of the building infrastructure device during operation of the building infrastructure device in the building infrastructure system during a predefined time period.

Step S6i may include obtaining management data for operation of the building infrastructure system during the predefined time period.

Step S62 comprises training at least one artificial intelligence-based model for detecting at least one anomalous event in the building infrastructure device based on the obtained parameter data from step S61.

Step S62 may include training the at least one artificial intelligence-based model for detecting at least one anomalous event in the building infrastructure device based on the obtained parameter data and the obtained management data

The trained at least one artificial intelligence-based model (trained model) is then stored in a memory. In particular, the trained model is recorded in a memory of a building management server 31 or a light management system server 32.

The monitoring phase follows to the training phase. The monitoring phase essentially comprises the operational lifetime of the building infrastructure system. The method may regularly reenter the training phase for retraining the model(s).

The monitoring phase includes steps S63, S64 and S65. The monitoring phase includes detecting occurrence of anomalous events based on current parameter data and the stored at least one artificial intelligence-based model learned in the training phase.

In step S63, the method obtains current parameter data for the at least one physical parameter of the building infrastructure device.

Step S63 may include obtaining current management data for operation of the building infrastructure system. In step S64, the method proceeds with determining whether an anomalous event in the building infrastructure device has occurred based on the obtained current parameter data of step S63, and the trained at least one artificial intelligence-based model from the training phase.

Step S64 may include determining whether the anomalous event in the building infrastructure device has occurred based on the obtained current parameter data, the current management data and the trained at least one artificial intelligence-based model generated and stored in the training phase.

Step S64 may comprise at least two distinct processes executed sequentially.

In step S64.1, the method applies the trained at least one artificial intelligence-based model from the training phase on the obtained current parameter data from step S63.

In step S64.2, the method determines based on applying the at least one artificial intelligencebased model from the training phase on the obtained current parameter data whether anomalous event in the building infrastructure device has occurred based on the application of the trained model on the current parameter data.

After step S64, the method may re-enter the training phase, in particular using the new obtained parameter data from the monitoring phase.

In step S65 the method generating and outputting a signal based on the determined anomalous event, in case step S64, in particular step S64.2, did indeed determine that an anomalous event has occurred in the building infrastructure system.

Step S65 may comprise generating a message to an operator based on the output signal, wherein the message includes at least one of information of the determined anomalous event, and a recommended action in response to the determined anomalous event.

Step S65 may include computing an expected remaining lifetime of the building infrastructure device based on the obtained parameter data and the obtained management data.

The recommended action may include adapting at least one operational parameter of the building infrastructure device based on the determined anomalous event in the output signal for increasing the expected remaining lifetime of the building infrastructure device. For example, the building management server 31, in particular the light management server 32, may adapt the load current ILED generated and output by individual light driver devices 1 of a plurality of light driver devices 1 that illuminate a storage facility in a manner to align predicted remaining lifetimes of the light driver devices 1 of the plurality of light driver devices 1. Thereby, the costly replacement of light driver devices 1 installed under the ceiling of the storage facility requiring working platforms may be done in one single working cycle. This simple example illustrates the benefits of the embodiments of the method in terms of running cost of the storage facility due to enabling to use internally estimated physical parameters of the building infrastructure devices in combination with operational management parameters for learning usage profiles or artificial intelligence models to determine anomalous events and base management of the building infrastructure system thereon.

All steps which are performed by the various entities described in the present disclosure as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. In the claims as well as in the description the word “comprising” does not exclude the presence of other elements or steps.

Using the indefinite article “a” or “an” in combination with an entity does not exclude a plurality of entities A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that different dependent claims recite certain measures, steps, and features of the method, driver device or building infrastructure system does not exclude that a combination of these measures, steps, and features cannot combined in an advantageous implementation within scope of the claims.