Technical field

The object of the present invention is a control method for controlling a fluid plant comprising a central device, a distribution network in fluid connection with one or more diffuse utilities that are characterized by a variable load over time.

The proposed invention falls within the scope of the adjustment of centralized fluid plants.

Prior art

Adjusting prior art centralized fluid plants is based on sensors installed in the central device that measure an averaged value associated with all the utilities. For example, the return temperature sensor from the utilities indicates how much energy the generator needs to reach the delivery set- point value.

The pump (or fan) is often adjusted with the use a constant pressure jump of the pump (or fan) and enables the flowrate to be modulated in function of the number of active utilities (and thus of the loss of load on the hydraulic system).

One prior art example of heating, ventilation and air conditioning systems (“HVAC”) that have a need that depends on outer temperatures consists in numerous applications of managing the delivery temperature of the generator with a climatic compensation curve (known as “climatic curve”). When the outer temperatures are cold (e.g. -5°C) the delivery temperature is accordingly high (e.g. 80°C) to compensate for the high heat loss from the building. On the other hand, when the outer temperatures are lower (e.g. 15°C), the delivery temperature is lower (e.g. 60°C).

Adjusting the pump (or the fan) and the generator with an averaged value in the central heating unit and not with the more critical utility, as occurs in the prior art, is limiting because not all the utilities require the same temperature and head needs, thus, with utilities that are variable over time, the ideal operation conditions vary over time.

In a first example of this type of adjustment of known type, applied for example to the sector of refrigerator utilities, a first utility 1 is always active and requires an operating function that is hardly constraining (required temperature of 7°C and a low head 5 mCE), on the other hand, a second utility 2 is more constraining (required temperature of 2°C and 10 mCE head) but operates only for a few hours a day. If we do not look at the operating status of the utilities, the centralized system must always work at 2°C and 10 mCE, on the other hand if we control the system with the disadvantaged utility we can for some hours of the day use operating conditions that are less strict and improve the total consumption energy efficiency: 7°C and 5 mCE.

In a second example of adjustment of known type, there is a single utility with the required power that varies over time and does not have the same operating constraints over time. For example, the heating terminals of a given environment have great need in the morning that decreases progressively over the day. In addition, the need on a Monday, after two days of shut-down plants, is much greater than on other working days of the week.

Using a centralized climatic curve without looking at the utilities, it is very difficult to match the set-point temperature of the boiler with the actual temperature required by the utility.

A further example of the management of the central device with a critical utility can consist of the case of heat storage (e.g. an ice tank), in which the requested temperature is a function of the loading level of the utility (e.g. ice level). As a result, the model of the utility gives us information on the need of the utility, which will be met by the adjusted central device. In the prior art, as the storage model is missing, the necessary information on the need to be met will be missing. Lastly, a further advantage of looking at the utility and not the centralized conditions consists of the possibility of optimizing the anticipated switch-on or switch-off in function of the actual need of the utility and not in function of an averaged value of all the utilities. In the prior art, the need of the critical utility is “hidden” inside the averaged value so that in the central device it will be necessary to consider a safety margin to avoid misunderstandings and manage the absence of exact information.

Object of the invention

The technical task on which the present invention is based is to propose a control method for controlling a fluid plant comprising a central device, a distribution network in fluid connection with one or more diffuse utilities that overcomes the aforesaid drawbacks of the prior art.

In particular, the object of the present invention is to provide a control method for controlling a fluid plant that enables the energy consumption to be reduced and the service provided by the various utilities to be improved.

A further object of the present invention is to provide a control method for controlling a fluid plant that permits monitoring of the variables and parameters, adjusting and optimisation of the energy that is simple and efficient.

In a first aspect of the invention, the aforesaid objects are achieved by a control method for controlling a fluid plant comprising a central device, a distribution network in fluid connection with one or more diffuse utilities that are characterized by a variable load over time; the method comprising:

- a step of configuration of a plurality of inertia acceptability parameters of each utility present in the fluid plant;

- a first step of characterization of a model configured to identify the variables and estimate the parameters of the model necessary to predict a use profile in a reduced temporal horizon by using a numeric prediction model (MNP); - a second step of characterization of the inertia of the fluid plant;

- a prediction step for predicting the needs of the heat transfer fluid, performed by means of the numeric prediction model (MNP) defined in the first characterization step, said prediction step being configured to estimate an optimum use profile (PUO);

- a step of adjusting said fluid plant, comprising the substeps of: o processing the need of the central device assuming the absence of inertia of the fluid plant; o comparison between the results of the inertia characterization step with the inertia acceptability parameters defined in the configuration step; o if the inertia of the fluid plant is significant:

- construction (108) of a model of the fluid plant that comprises the inertia of the fluid plant;

- processing the need of the central device by taking account of the inertia of the fluid plant and of the future variation of the need;

- adjusting the central device.

One or more steps of the method can be implemented by means of a computer.

In a second aspect of the invention, the aforesaid aims are achieved by a control system for controlling a fluid plant comprising:

- a central device; a distribution network in fluid connection with one or more diffuse utilities that are characterized by a variable load over time;

- a processing unit comprising:

• a module of configuration of the fluid plant, configured to configure a plurality of inertia acceptability parameters of each utility present in the fluid plant;

• a first characterization module configured to identify the variables and estimate the model parameters that are necessary to predict a use profile in a reduced temporal horizon, by using a numeric prediction model (MNP); • a second characterization module configured to determine the inertia of the fluid plant;

• a prediction module configured make a prediction the needs of the heat transfer fluid using the numeric prediction model (MNP) defined by the first characterization module, said prediction module being configured to estimate an optimum use profile (PUO);

• an adjusting module for adjusting the fluid plant configured to: o process the need of the central device, assuming the absence of inertia of the fluid plant; o compare the inertia of the fluid plant, determined by the second characterization module, with the inertia acceptability parameters defined by the configuration module; o if the inertia of the system is significant:

- construct a model of the fluid plant that comprises the inertia of the fluid plant; and

- process the need of the central device in the actual case, by taking account of the inertia of the system and of the future variation of the need;

• an adjusting module for adjusting the central device configured to send an adjusting signal of the central device.

In a third aspect of the invention, the aforesaid objects are achieved by a computer programme that actuates one or more of the steps of the method.

Brief description of the drawings

Further features and advantages of the present invention will become more apparent from the illustrative and thus non-limiting description of a preferred, but non-exclusive, embodiment of a method and a system for controlling a fluid plant comprising a central device, a distribution network in fluid connection with one or more diffuse utilities that are characterized by a variable load over time, as illustrated in the accompanying drawings, in which: figure 1 illustrates schematically the block diagram of the control system for controlling a fluid plant, in accordance with the present invention; figure 2 illustrates a first embodiment of a fluid plant, according to the present invention; figure 3 illustrates a second embodiment of a fluid plant with a closed loop circuit, according to the present invention.

Detailed description of preferred embodiments of the invention In the following description, fluid plant means medium or large dimension circuits that exploit a technical fluid for conveying heat. Examples of technical fluids: technical water, chilled water, glycol water, diathermic oil, steam, air, compressed air.

A central device means a central heating unit consisting of at least one generator (for example, a boiler, a refrigerator unit, a cooling tower or heat exchanger, batteries), at least one distribution pump or fan and a system for adjusting the generator (for example set-point temperature) and/or adjusting the pump or fans (for example, the number of revolutions). Distribution network means a network that distributes a fluid from one point to several utilities. Via the carrier fluid, the distribution network takes the heat to the various utilities (in heating mode) or extracts heat from the various utilities (in cooling or heat recovery mode).

By diffuse utilities, we mean at least two distinct points of use of the service generated by the central device. The diffuse utilities have an adjustment at the utility level intended to reach the local adjustment set- point value. Some examples of adjustment of the diffuse utilities comprise reaching the environmental set-point conditions, reaching the set-point temperature conditions of the storage, the ice level, or the emission power or the temperature value of the utility side set-point.

In a first aspect, the present invention describes a control method for controlling a fluid plant 1 comprising a central device 10, a distribution network 20 in fluid connection with one or more diffuse utilities 30 that are characterized by a variable load over time.

The central device 10 can, for example, consist of at least one heating generator or of at least one refrigerator unit and/or of both. Further, the central device 10 also comprises at least one distribution pump or fan, an adjustment system for adjusting said at least one generator 10a, 10b (for example, adjusting the set-point value of the delivery temperature) and/or adjusting the pump or fan.

According to a first non-limiting embodiment, as illustrated in figure 2, the distribution network 20 can have a tree topology.

According to the embodiment of figure 3, the distribution network 20 consists of a closed loop.

The control method 100 for controlling a fluid plant 1 according to the present invention is shown by the flow diagram of figure 1.

The method 100 comprises initially a step 101 of configuration of a plurality of inertia acceptability parameters of each utility 30 present in the fluid plant 1.

The configuration step can comprise defining a plurality of inertia acceptability parameters and defining other parameters associated with the fluid plant and with the control system. Some parameter embodiments comprise the time step, the distinction of comfort and/or saving periods, weights for the terms of the cost function described below, etc.

From the configuration step 101, a characterization step 111 follows that in turn comprises two substeps 102 and 103.

A first step of characterization 102 of a numeric model configured to identify the variables and estimate the parameters (involved in the process) of the model that are necessary to predict a use profile in a reduced temporal horizon by using a numeric prediction model (MNP); The object of step 102 is to identify which information is necessary to predict the use profile (e.g. power, temperature, flowrate or any other operating indicator like valve position or pump revolutions) in a use profile in a reduced temporal horizon and to construct a calculation instrument to make the prediction.

The numeric prediction models MNP of the variables of interest used by the method 100 can be of the grey box or conceptual type, based on knowledge of the physics of the system (white part of the method).

The grey box model means a model consisting of two parts, an explicit first part, that is based on knowledge of the physics of the fluid plant 1 , and a second “black” part that exploits the historic data for estimating the trend. The first characterization step 102 comprises a first substep that consists of selecting a numeric model of the response of an interest variable in function of the variables and of the parameters by numeric prediction models MNP of the variables of interest of grey box or conceptual type, based on a knowledge of the physics of the system and comprises a second substep that consists of identifying the parameters Pi of each numeric prediction model MNP by regressive methods without supervision of training and testing on historic data (i.e. black part of the method).

In the second characterization step 103, the inertia of the fluid plant 1 is determined.

The sequence of execution of the substeps 102 and 103 is performed parallel as illustrated in the diagram of figure 1 , i.e. by first performing the characterization substep 103 and then the substep 102.

From the two characterization substeps 102 and 103 we move to a prediction step 104 of the needs of the heat transfer fluid.

The prediction step 104 is performed by the numeric prediction model MNP defined in the first characterization step 102 (which could also be performed after the step 103).

The prediction step 104 is performed to estimate an optimum use profile PUO.

From the prediction step 104, the step 105 of adjusting the fluid plant 1 follows. As illustrated in figure 1 , the step 105 of adjusting the fluid plant 1 comprises the substeps 106, 107, 108, 109 and 110.

The processing substep 106 comprises calculating the need of the central device 10 assuming the absence of inertia of the fluid plant 1 (i.e. ideal case).

In the comparison substep 107, the results of the substep 103 of characterizing the inertia are compared with the inertia acceptability parameters defined in the initial configuration step 101. if the inertia of the system is significant, the method 100 proceeds from the step 107 to the steps 108 and 109.

In particular, in the step 108, a model is constructed that comprises the inertia of the fluid plant 1 and in the step 109 the need of the central device 10 is processed, by taking account of the inertia of the fluid plant 1 and the future variation of the need.

Lastly, the method 100 concludes with a step 110 of adjusting the central device 10.

If in the comparison of the step 107 it is ascertained that the inertia of the fluid plant 1 is not significant, the adjusting step 110 follows directly.

The first characterization step 102 comprises a processing step for processing a quality indicator of characterization by a step of simulation of the behaviour of the fluid plant 1 in a given period of time using only the numeric prediction model MNP with the identified parameters Pi and a step of comparison of the results of the simulation with real data, calculating a measurement of the difference between the reality and the simulation.

The prediction step 104 comprises processing said optimum use profile PUO to resolve an optimum problem using a method known as a Model Predictive Control (MPC), starting from a model and an FC cost function.

In particular, the cost function FC consists of a first term that forces the vicinity between the objective variable and the set-point value thereof and of a second term that forces the operation with the objective of minimizing energy consumption.

The first and second term have different weights configured to check how much importance to give to the service provided with respect to the reduction of energy consumption and which can be adjusted and balanced periodically or according to settings defined in the configuration step 101. For example, in order to balance the two needs at various moments of the day.

The first term is a distance between the objective variable and the set- point value thereof in a differentiable metric.

For example, the first term is the squared difference between the objective variable and the set-point thereof. If the service provided was comfort in terms of maintenance of a certain temperature it could be the squared difference between the ambient temperature and the desired temperature. The second term is a standardized measurement of energy quantities to be checked, also in a differentiable metric.

For example, the second term could be the squared absolute value of the energy.

A particular case of cost function FC could be provided by the linear combination of these two terms with coefficients « and b if comfort is preferred and on the other hand if a reduction of consumption is preferred. The numeric prediction model MPN comprises parameters and/or variables measured during the step of use of the MPN model that can be, for example, parameters describing the current status (for example, temperature, flowrate, power, level or any operating indicator like valve position or pump revolutions) or parameters of future operation in a reduced temporal horizon (for example, on the basis of the environment set-point - time schedule -, process set-point, activation times) and parameters and/or variables outside the fluid plant in a reduced temporal horizon (for example, weather forecast - e.g. temperature and outside humidity -, production plan). In other words, the step of reading the data corresponding to the variables occurs during the use of the prediction model.

In other words, the numeric prediction model MPN uses current status variables (e.g. objective variable and current set-point value thereof) and set-point values in the future (e.g., following activation of a specific utility in a set period of time in the future, for example in X hours, change of the set-point value in X hours from 16°C to 20°C).

The adjusting step 105 further comprises the step of defining, given the profiles of the delivery temperature or flowrate for each utility (for example terminal) 30a, 30b, 30c, 30d, a critical profile that has to be reached at each step at the level of the central device.

Advantageously, the construction step 108, performed only in the case of significant inertia of the plant 1 , comprises the step of determining one or more of the following parameters of the fluid plant 1 : o heat dissipation DT between the central device and the various utilities; o inertia of the fluid plant 1 ; o sudden variations in the operating conditions in function of the capacity of the generator and/or of the pump or fan; o minimum and maximum operating limit on the control variable.

In particular, the sudden variation substep of the operating conditions is determined by defining a limit value for the variation of the control variable, for example the temperature, which depends on the capacity of the generator, for example, on the maximum power of the generator, on the dissipation and applying smoothing to the use profile so as to distribute the sudden variation in a set period of time preceding the variation.

If an increase of the central device 10 need is not feasible taking account of the inertia of the system, it is necessary to anticipate the increase to meet the need of the utility 30.

In other words, if we know that at 8 in the morning we have programmed a temperature jump, for example of 40°C in temperature, from 40°C to 80°C, but we have a jump limit of 20°C every X minutes, the method according to the invention will start to heat in advance, moving gradually from 40°C to 60°C, from 8 am less 2X minutes at 8 am minus X minutes and subsequently going to 80°C (from 8 am less X minutes to 8 am).

In a second aspect, the present invention discloses a control system for controlling a fluid plant 1 comprising a central device 10, a distribution network 20 in fluid connection with diffuse utilities 30 that are characterized by a variable load over time.

By diffuse utilities 30, we mean at least two distinct points of use of the service generated by the central device 10.

The system according to the present invention further comprises a processing unit.

The processing unit is in data communication with the central device 10 and with the utilities 30 of the distribution network 20.

The processing unit comprises a configuration module for configuring the fluid plant 1 , configured to configure parameters associated with the fluid plant 1 and with the control system, a first characterization module configured to identify the variables and estimate the model parameters that are necessary to predict a use profile in a use profile in a reduced temporal horizon, by using a numeric prediction model MNP, a second characterization module configured to determine the inertia of the circuit of the fluid plant 1 , a prediction module configured to predict the needs of the heat transfer fluid using the numeric prediction model MNP defined by the first characterization module, configured to estimate an optimum use profile PUO, an adjusting module of the fluid plant and an adjusting module for adjusting the central device 10 configured to send an adjusting signal to the central device 10.

In particular, the adjusting module for adjusting the fluid plant 1 is configured to: o process the need of the central device 10, assuming the absence of inertia of the fluid plant 1 ; o compare the inertia of the fluid plant 1 , determined by the second characterization module, with the inertia acceptability parameters defined by the configuration module; o if the inertia of the system is significant:

- construct a model of the fluid plant that comprises the inertia of the fluid plant 1 ; and

- process the need of the central device 10 in the actual case, by taking account of the inertia of the system and of the future variation of the need. In general, it should be noted that in the present context and in the subsequent claims, the processing unit is considered to be divided into distinct functional modules (memory modules or operating modules) for the sole purpose of describing the functionalities thereof clearly and completely.

This processing unit can consist of a single electronic device, appropriately programmed to perform the functionalities described, and the different modules can correspond to hardware entities and/or routine software that are part of the programmed device.

Alternatively, or in addition, these functions can be performed by a plurality of electronic devices over which the aforesaid functional modules can be distributed.

The processing unit can moreover rely on one or more processors to execute the instructions contained in the memory modules.

Preferably, the fluid plant 1 comprises a circuit of medium and large dimensions that exploits a technical fluid between at least technical water, chilled water, glycol water, diathermic oil, technical steam, air, compressed air.

Optionally, the central device 10 comprises a central heating unit consisting of at least: a generator, at least one distribution pump, a system for adjusting the generator (for example the set point of the delivery temperature) and/or adjusting the pump.

In a third aspect, the present invention relates to a programme for a computer that actuates one or more steps of the method and a method according any one of claims 1 and 10, characterized in that it is actuated by a calculator.

As a person skilled in the art can easily understand, the invention allows the drawbacks to be overcome that are highlighted above with reference to the prior art.

In particular, the present invention enables energy consumption to be reduced and the service to be improved that is provided by the various utilities of the centralized fluid plants. It is clear that the specific features are described in relation to different embodiments of the invention with an exemplary and non-limiting intent. Obviously a person skilled in the art can make further modifications and variants to the present invention, in order to satisfy contingent and specific needs. For example, the technical features described in relation to an embodiment of the invention can be extrapolated therefrom and applied to other embodiments of the invention. Such modifications and variations are moreover embraced within the scope of the invention as defined by the following claims.