Field of the Invention

The present invention disclosure relates to monitoring of an object by sensors, and more

particularly to a method and computer device for providing a visualization of sensor measurements. Background of the Invention

Numerous facilities in various industries require detection of anomaly at an early stage, which include hydroelectric power plants, atomic reactors, aircrafts, medical devices, manufacturing devices for

semiconductors etc.

Once an anomaly is identified, an alarm may be raised. Then, an expert needs to examine the trends of measurements across e.g. 40-50 sensors in order to interpret and explain the anomaly, which is a time consuming and laborious task.

European patent application No. 10813558.3 describes a method and system for early detection of anomaly of a plant or a facility and is incorporated herein in its entirety by reference. The method comprises anomaly detection, evaluation of level of effect of each signal, construction of determination conditions, and display of e.g. numbers of abnormal sensors corresponding to the anomaly. However, useful information associated with the anomaly can be missing from the displayed content.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an improved method and system for visualization of sensor measurements.

According to an embodiment of a first aspect of the present invention, there is provided a method for providing a visualization of measurements of a

plurality of sensors, comprising:

- visualizing a measurement of a first sensor by a first node, visual attributes of the first node being determined based on the measurement of the first sensor;

- visualizing a measurement of a second sensor by a second node, visual attributes of the second node being determined based on the measurement of the second sensor;

- wherein the first node and the second node being in a same graph.

The visual attributes of a node may include any one of the following: a color of the node, a size of the node, a pattern of the node, or any combination thereof.

In an embodiment, the method further comprises presenting a line connection between the first node and the second node if the two sensors give correlated outputs. The two sensors may be algorithmically determined to give correlated outputs if, for instance, increase in value of first sensor output is strongly correlated with increase in value of second sensor output and vice versa.

At least one attribute of the line connection between the two sensors gives visual indication of any one of the following:

- an increasing correlation;

- a decreasing correlation; or - a maintained correlation.

- Attributes of the line connection can be: type of line (straight line or curve), line weight, color, animation (such as blinking line) or the like .

Optionally, the color of each node is determined by the following: if the measurement of the

corresponding sensor indicates no anomaly, the color is green; if the measurement of the corresponding sensor indicates a first level of anomaly, the color is yellow; if the measurement of the corresponding sensor indicates a second level of anomaly, the color is red, and the second level being higher than the first level.

Optionally, the size of each node being

proportional to a level of anomaly indicated by the measurement of the corresponding sensor.

Optionally, the graph' s spatial visualization can be customized in multiple ways such as:

- Homomorphic (nodes organized similar to how the sensors are physically placed on the object) with optional overlay of a diagram depicting the ob ect .

- Clustered (nodes of a similar type are clustered together)

Optionally, the visualization may be periodically saved as a graph, enabling a playback of the

visualization later.

According to embodiments of a second aspect of the present invention disclosure, there is provided a computer device, comprising:

a processor,

a memory storing computer-readable instructions that, when executed by the processor, instruct the processor to perform the aforementioned method. According to embodiments of a third aspect of the present invention, there is provided a system,

comprising: a plurality of sensors; and the computer device aforementioned, the computer device is

configured to receive measurements from the plurality of sensors and provide a visualization of the

measurements .

According to embodiments of a fourth aspect of the present invention disclosure, there is provided a non- transitory computer readable medium storing computer- readable instructions that, when executed by a

processor of a computer device, instruct the processor to perform the aforementioned method.

Inventors of the present invention disclosure find it might be advantageous if sensor measurements are presented to a user (e.g. an expert dealing with anomalies) in a more informative manner comparing to European patent application No. 10813558.3. The latter only indicates the reference number (s) of the

"abnormal" sensor (s) to the user, which may not be really helpful if the algorithm is not fully reliable to give a fully-automatic diagnosis.

With the aid of the method, computer device, system and non-transitory computer readable medium, sensor measurements of a plurality of sensors can be summarized and visualized in one graph that conveys the issue, provides an overview of impact of the issue, and even visually suggests where to look for root cause and possible solution, saving hours or even days of the expert' s time spent on anomaly interpretation and explanation. Accordingly, maintenance action can be defined and executed quickly, supporting a fast turnaround time to fix the anomaly. The proposed solution can be particularly helpful, by means of providing informative representation of sensor measurements, where a fully-automatic diagnosis of anomalies is not available or not reliable.

These and other features, embodiments and advantages of the present invention disclosure are described in the accompanying claims, abstract and the following detailed description of non-limiting embodiments depicted in the accompanying drawings, in which description reference numerals are used which refer to corresponding reference numerals that are depicted in the drawings .

Similar reference numerals in different figures denote the same or similar objects. Objects and other features depicted in the figures and/or described in this specification, abstract and/or claims may be combined in different ways by a person skilled in the art .

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 illustrates a system according to an embodiment of the present invention disclosure.

Fig. 2 illustrates a flow chart of a method for providing visualization of measurements of multiple sensors, according to an embodiment of the present invention disclosure.

Fig. 3 illustrates a block diagram of a computer device according to an embodiment of the present invention disclosure.

Fig. 4a-4c illustrates graphs generated by visualizing measurements of a plurality of sensors.

DETAILED DESCRIPTION OF THE DEPICTED EMBODIMENTS In oil and gas industry, production-critical equipment such as rotating equipment are equipped with a large number of sensors. Those sensors generate measurements regularly which are evaluated by a computer to detect anomaly. A reliable early alarm of anomaly (that can give advance warning of developing problems) is generally preferred. On the other hand, if an alarm for anomaly is only raised when a sensor measurement exceeds a preset, numeric, critical threshold, there is very little time to plan

maintenance action and equipment downtime is not optimized .

Figure 1 illustrates a system 1 according to an embodiment of the invention. The system 1 includes an object 10, eleven sensors 202-222 monitoring the object 10, a computer device 30 receiving sensor measurements from the sensors via wired/wireless connection 50, and provide a visualization of the sensor measurements to a user 40 e.g. via a display thereof. The user 40 is expected to, when an alarm is raised for a detected anomaly, understand what went wrong and come up with a plan for appropriate maintenance actions to minimize the loss caused by the anomaly.

The object 10 can be a facility, equipment or any other object or collection of objects aforementioned which might need to be monitored by a plurality of sensors. The object 10 might be a group of different equipment, devices or components, and the sensors 202- 222 can be associated with different equipment, devices or components.

Fig. 2 illustrates a flow chart of a method 2 for providing a visualization of measurements of a

plurality of sensors. Hereinafter, reference will be made to the example in Fig. 1 when describing this method .

The method 2 includes steps S22, S24 and S26, which might be performed regularly, e.g. every 20 minutes, every hour.

In the illustrated example, in step S22, a

measurement of a first sensor (e.g. sensor 202) is visualized by a first node, a color of the first node and a size of the first node being determined based on the measurement of the first sensor. The sensor 202 might be a thermometer detecting a temperature at an outlet of a container A in the object 10.

In step S24, a measurement of visualizing a measurement of a second sensor (e.g. sensor 204) by a second node, a color of the second node and a size of the second node being determined based on the

measurement of the second sensor. The sensor 204 might be another thermometer detecting a temperature at an inlet of a container B in the object 10, the containers A and B might be in liquid communication. Liquid such as heated oil flows from the container A to the container B via a valve. In this example, it is assumed that both containers A and B are also

pressurized and therefore a pressure in each might be also detected by pressure sensors.

In step S26, a line connection, e.g. a line, is presented between the first node and the second node if the first sensor and the second sensor are correlated.

Those skilled in the art would appreciate that although only two sensors, i.e. a first sensor and a second sensor are mentioned in the flow chart in Fig. 2, the method 1 can be easily adapted to a case where a visualization is needed for more sensors such as that illustrated in Fig. 1. In that case, step 22 or step 24 might be repeated for each additional sensor, and step 26 might be repeated for each pair of sensors. Eventually, a visualization of measurements of the 11 sensors in Fig. 1 might be formed as a graph including 11 nodes, a line connection between each pair of nodes representing a pair of correlated sensors .

Each of steps S22-S26 might need an input, which can be generated as following:

Input for step S22

A measurement of sensor 202 is evaluated to have an input for step S22. To be specific, a method or system described in US patent No. 5548597 to Kayama et al. or European patent application No. EP2477086 Al might be used to generate that input, both are hereby

incorporated into the present application with the entirety by reference. The goal of the evaluation is to identify if the measurement (e.g. a temperature of 200°C) is abnormal, and if so what is the level of the anomaly. Of course, the exact measurement value per se, i.e. 200°C, can also be useful for steps S22.

Alternatively, value ranges can be set for each parameter and status monitored by a sensor can be decided by comparing its measurement against the value ranges. Take sensor 202 as an example, value ranges can be preset as below:

Normal: [180, 250)

First level of anomaly (caution) : [250, 350)

Second level of anomaly (dangerous) : 350 or above, wherein the second level of anomaly being more severe than the first level of anomaly.

This (simple threshold) is just one way in which anomaly can be quantified. Alternate method of anomaly detection and quantification can also be applied, such as those that take the system as a whole to determine how abnormal the measured state is compared to normal.

Input for step S24

Input for step S24 can be obtained in a way as described for step 22. Similarly, value ranges can also be preset for other sensors .

Those skilled in the art would appreciate that more levels of status can be defined in addition to the aforementioned "normal", "first level of anomaly" and "second level of anomaly", which can be suitably represented by more colors in steps S22 and S24.

Input for step S26

Step S26 requires an indication of whether two sensors are correlated, which can be decided using the method described in the US patent No. 5548597.

Alternatively, a Normalized Mutual Information (NMI) between two nodes can be used, defined by the below equation (1) :

MI=H (X) +H (Y)—H (X, Y) (1)

where H (X) is the entropy of X, defined in

equation (2 ) :

H(X)=-∑p(x)log(p(x) ) (2)

MI can have values in the ranges of [0, ^∞ ) . NMI aims at normalizing these values and constraining them between [0,1] . This might be a desirable property, and NMI may be defined as below:

A line connection between two nodes X and Y may be presented if NMI(X,Y)>0, meaning the two sensors represented by the two nodes are more or less

correlated .

Steps S22-S26 will now be described with more details by referring to Fig. 1 and further referring to Figs. 4a-4c. Fig. 4a-4c illustrates graphs generated by visualizing measurements of a plurality of sensors, and they are provided with sequential timestamps 47, 47a and 47b, in this case, the visualization is done every 20 minutes.

In the graphs in Figs. 4a-4c, for simplicity, not all sensors in Fig. 1 are shown. In practice, people might choose to cover all sensors in this one graph or only a part of those which might be really crucial.

Referring to Fig. 4a, in the graph generated by visualization at 13:20:00, September 06, 2014, each sensor has a node marked using the numeral reference of the sensor. A color of each node is determined based on the measurement of the specific sensor. In the drawings, we use patterns to indicate different colors, for example, we use blank nodes like"(*∞)" indicate green, " (J)' to indicate yellow and to indicate red .

As mentioned above, the color code can be used to give a user a visualized and qualitative idea of the sensor measurement. In this example, "traffic light" colors are used, where green means normal or no anomaly, yellow means a first (lower) level of anomaly, and red means a second (higher) level of anomaly.

Different colors can be used in practice.

Examining the graph in Fig. 4a, a user may directly read the following information relating to the sensor measurements :

1. None of sensors 202-212 has detected an anomaly;

2. There is a correlation between 4 pairs of sensors : a. Sensor 202 and sensor 204, the

correlation indicated by the presented line connection 402;

b. Sensor 202 and sensor 206, the

correlation indicated by the line connection 404;

c. Sensor 204 and sensor 210, the

correlation indicated by the line connection 406; and

d. Sensor 210 and sensor 212, the

correlation indicated by the line connection 408.

3. Sensor 208 currently has no correlation with any other sensor.

As aforementioned, a size of each node can also be determined based on the measurement of the sensor. In an example, for each sensor and its associate node in the graph, the size of the node is proportional to a level of anomaly indicated by the measurement of the sensor. To emphasize the visualization of anomaly, for each node, the size (e.g. diameter of the round node) can be determined in such a way: when the measurement indicates no anomaly, the size change proportionally to the change of the measurement but slowly; when the measurement indicates a first level of anomaly, the size of the node changes proportionally to the change of the measurement but at a higher change rate, when the measurement indicates a second level of anomaly, the size of the node changes proportionally to the change of the measurement but at the highest change rate. In this way, the size of a node is defined as a monotonic function of the measurement value got by the sensor. By repeating a process described in S22 or S24 for all other sensors, the color and size of each node at 13:20:00 can be determined in order to provide the illustrated graph.

Fig. 4b illustrates a visualization of measurements of the sensors at 13:40:00, 20 minutes after that of Fig. 4a.

Comparing Fig. 4b to Fig. 4a, the following

differences might be observed:

- Node 204a and node 204 both correspond to sensor 204, and node 204a is yellow and is bigger than node 204, which means at 13:40:00, sensor 204 is indicating an anomaly of a first level.

- A weight of line connection 406a is higher than line connection 406. Here, we use a weight of a line connection to indicate to what extent two sensors are correlated, and the weight changes proportionally to the correlation, e.g. the calculated NMI value. Thus, it can be seen that in the past 20 minutes, the correlation between sensors 204 and 210 has increased.

Fig. 4c illustrates a visualization of measurements of the sensors at 14:00:00, 20 minutes after that of Fig. 4b.

Comparing Figs. 4c and 4b, the following might be observed :

- Node 204b is red is even bigger than node 204a, indicating that the measurement of sensor 204 now indicates an anomaly of second level, which is more serious than 20 minutes ago.

- A correlation between sensor 202 and sensor 204b is still there, but has decreased comparing to Fig. 4b. When describing the change between Fig. 4b and Fig. 4a, we use a weight of a line connection to demonstrate to what extent two sensors are correlated. Fig. 4c is showing an alternative, where a change of correlation between two sensors are demonstrated by a pattern of a line connection. I.e., we use dotted line 402b to demonstrate that a

correlation has decreased. In general, a changed correlation might be a good indicator of anomaly at one or both sensors .

- Sensor 210 detects a first level of anomaly and dot 210b is now yellow.

- A correlation between sensors 204 and 210 is maintained comparing to Fig. 4b, even if both sensors demonstrates abnormal measurements.

This maintained correlation between sensors which both detects anomaly can be demonstrated by dotted and dashed line 406b. In this case, the user needs to further investigate whether both sensors are really detecting anomaly, or only one is .

In an example, graphs like those in Figs. 4a-4c are stored periodically, enabling a playback of the

visualization later. When playing such visualizations, a user is able to go through all the graphs generated in several days within several minutes, which is time efficient .

The method for providing visualization of

measurements of a plurality of sensors according to certain embodiments of the invention has been described as above. It shall be noted that although the steps are shown in a sequential way, they are not required to be executed in this specific order, nor required to be executed in series at all. Alternatively, they can be simultaneous, in a different serial order.

Fig. 3 illustrates a block diagram of a computer device, e.g. the computer device 30 in Fig. 1. It includes a processor 302, a memory 304 storing

computer-readable instructions that, when executed by the processor 302, instruct the processor 302 to perform the aforementioned method 2. The computer device 30 may further comprise a display 306,

configured to present the visualization to the user 40. Processor 50 is configured to receive measurements from sensors (e.g. those shown in Fig. 1) for executing the method 2.

The method, system and/or any other subject matter according to present invention disclosure are well adapted to attain the ends and advantages mentioned as well as those that are inherent therein.

The particular embodiments disclosed above are illustrative only, as the present invention disclosure may be modified, combined and/or practiced in different but equivalent manners apparent to those skilled in the art having the benefit of the teachings herein.

Furthermore, no limitations are intended to the details of construction or design herein shown, other than as described in the claims below.

It is therefore evident that the particular illustrative embodiments disclosed above may be altered, combined and/or modified and all such

variations are considered within the scope of the present invention as defined in the accompanying claims .

While any methods, systems and/or products embodying the invention are described in terms of "comprising," "containing," or "including" various described features and/or steps, they can also "consist essentially of" or "consist of" the various described features and steps. All numbers and ranges disclosed above may vary by some amount . Whenever a numerical range with a lower limit and an upper limit is disclosed, any number and any included range falling within the range is specifically disclosed. In particular, every range of values (of the form, "from about a to about b," or, equivalently, "from approximately a to b," or,

equivalently, "from approximately a-b") disclosed herein is to be understood to set forth every number and range encompassed within the broader range of values .

Also, the terms in the claims have their plain, ordinary meaning unless otherwise explicitly and clearly defined by the patentee.

Moreover, the indefinite articles "a" or "an", as used in the claims, are defined herein to mean one or more than one of the element that it introduces .

If there is any conflict in the usages of a word or term in this specification and one or more patent or other documents that may be cited herein by reference, the definitions that are consistent with this

specification should be adopted.