Patent Application
20130144444
The invention relates to the field of command and control of an HVAC (Heating, Ventilation and Air-Conditioning) system or machine. The invention relates more particularly to a method making it possible to determine parameters for regulating such an HVAC system and for optimizing the energy consumption of this system.
The solutions currently implemented for regulating an HVAC system are not optimal in terms of the performance, reliability and energy efficiency obtained. This is usually due to the fact that the user of HVAC systems has little knowledge in the fields of energy regulation or optimization. Specifically, the adjustment of the parameters of the existing regulation loops of the HVAC systems is often carried out by habit, which may cause regulation difficulties within the systems (considerable excesses, pumping phenomena, instabilities etc), these difficulties resulting in energy losses and risks of breakage of the badly regulated equipment. Similarly, the operating points are also determined by habit and not on the basis of a perfect control of the process: for example, the high floating pressure of the heat pumps is usually determined by a constant difference between the condensation temperature and the outside temperature.
Tools exist making it possible to achieve good regulation, such as for example the implementation of advanced algorithms such as predictive control, or energy optimization functions. However, such tools are very difficult to use for a user who is not an expert in these fields.
One object of the present invention is to provide a method for determining parameters for regulating an HVAC system improving the command and control of the HVAC system, making it possible notably to reduce the energy consumption of the HVAC system and to increase the operating reliability of the system thus regulated.
For this, the present invention provides a method for determining parameters for regulating an HVAC system comprising thermal elements designed to be regulated by regulation loops, characterized in that it comprises at least the following steps:
This method makes it possible to determine regulation parameters based on which it is possible to easily program, without requiring expertise on the part of the user in the automation, regulation, optimization or diagnostics fields, the programmable controller or controllers of an HVAC system.
This method makes it possible to manage two aspects: the regulation itself of the system, but also the optimization of the energy consumption, for example the electricity consumption, of the HVAC system.
Thus, the more effective command and control obtained by virtue of the method makes it possible to solve the problems associated with the regulation difficulties, with the energy inefficiencies and if necessary with the installation of diagnostic functions.
This method finally allows a user who is not an expert in automation to easily use a regulation of an HVAC system.
During this method, a simulation is carried out in steady state of the thermal and electrical behaviour of the modelled HVAC system. Such a steady state corresponds to a stable state over time of the HVAC system, that is to say a state in which the variations of the signals of the system, notably the output physical magnitudes and the control signals, are substantially zero over time (that is to say, for a signal x of the system, such that dx/dt=0). In the context of the present document, a “steady state” therefore does not signify a state that is “permanently” present over time, but rather a stable state.
The dynamic state of the modelled HVAC system corresponds to an unstable state over time of the HVAC system caused by varying the value of a signal on at least one input of the system, that is to say by applying a transitory signal to this input of the system. In the context of the present document, a “transitory” signal corresponds to a signal that can vary over time. The dynamic state is therefore a temporary state following a modification on at least one input of the system.
The simulation in dynamic state of the thermal and electrical behaviour of the modelled HVAC system can therefore be carried out by varying values of input signals of the thermal elements of the modelled HVAC system, the computation of the values of the parameters of the regulation loops being carried out based on parameters of transfer functions of the thermal elements corresponding to the ratios between the values of the output physical magnitudes of the thermal elements and the values of input signals of the thermal elements.
The regulation loops may preferably be of the predictive control type. Thus, the method can use techniques of advanced control of the HVAC system, for example a PFC (Predictive Functional Command) control.
The simulation in dynamic state of the thermal and electrical behaviour of the modelled HVAC system can be carried out by successively applying, on the inputs of each thermal element, signal variations of a step type.
The simulation in dynamic state of the thermal and electrical behaviour of the modelled HVAC system can be carried out for values of the set points for which the energy consumption of the modelled HVAC system is lowest.
The invention also relates to a method for regulating an HVAC system comprising thermal elements, comprising at least the steps of:
The thermal and electrical modelling of the HVAC system may comprise at least the steps of:
The definition of the regulation loops may comprise the steps of description, in the software, of:
The method may also comprise, prior to use of the method for determining parameters for regulating the HVAC system, a step of defining ranges of possible values of control signals designed to be delivered by the regulators. Thus, the user can specify operating fields of the system.
The simulation in steady state of the thermal behaviour of the modelled HVAC system may also determine values of control signals designed to be applied at the input of the thermal elements.
The definition of the regulation loops may also comprise a definition of ranges of possible values of the set points of the regulation loops. Thus, the user can specify the operating point or points of the system for which the simulations are carried out.
The method may also comprise, prior to the use of the method for determining parameters for regulating the HVAC system, a step of defining ranges of possible values of the physical magnitudes.
When the controller of the HVAC system is a programmable controller, the programming of the controller can be carried out at least by the use of the following steps:
When the controller of the HVAC system is a parameterizable controller, the programming of the controller may comprise at least one step of entering determined values of the set points and computed values of the parameters of the regulation loops into the controller.
The programming of the controller of the HVAC system may be carried out based on data in which the values of the set points and the values of the parameters of the regulation loops computed during the method for determining parameters for regulating the HVAC system are encrypted.
The invention also relates to a device for determining parameters for regulating an HVAC system comprising thermal elements designed to be regulated by regulation loops, comprising means for using a method for determining parameters for regulating an HVAC system as described above.
The present invention will be better understood on reading the description of exemplary embodiments given purely as an indication and in no way limiting by making reference to the appended drawings in which:
Identical, similar or equivalent portions of the various figures described below bear the same reference numbers so as to make it easier to switch from one figure to the other.
Various portions shown in the figures are not necessarily shown according to a uniform scale, in order to make the figures easier to read.
The various possibilities (variants and embodiments) must be understood to be not exclusive of one another and may be combined together.
Reference is made first of all to
The HVAC system 10 represented in
Now will be described the regulation and the energy optimization that will be carried out for each of the elements of the HVAC system 10 by the regulation and energy optimization method, with reference to
In this figure, a thermal element 20, for example a condenser, of the HVAC system 10 is shown. The thermal element 20 receives on an input 22 a control signal u designed to drive the thermal element 20. An output physical magnitude y of the element 20, for example a pressure or a temperature of the refrigerant output from the element 20, is measured on an output 24 of the element 20 and will be used for the regulation of the thermal element 20. The relationship between the physical magnitude y and the control signal u corresponds to a first transfer function H1(s) of the thermal element 20.
For each of the thermal elements of the HVAC system 10, the value of the output physical magnitude y depends on the value of the control signal u applied at the input of the element 20, but also depends on the values of other parameters, called external physical magnitudes influencing the energy performance of the thermal element, and therefore of the HVAC system. Such external physical magnitudes are for example an outside temperature, a temperature of a fluid entering the machine, a flow rate, a load ratio, a delay, etc. In the example of
The regulation of the thermal element 20 is carried out by a regulation loop 26 determining the value of the control signal u designed to be applied on the input 22 of the thermal element 20 based on the value of the physical magnitude y that is applied on a first input 28 of the regulation loop 26, of a regulation set point y* applied on a second input 30 of the regulation loop 26, of measurements of the external physical magnitudes Gext1, Gext2 and Gext3, but also of certain internal physical magnitudes of the thermal element 20 influencing the performance of the regulation of the thermal element 20.
The value of the regulation set point y* is determined by a set point generator 32 which makes it possible, based on the external physical magnitudes Gext1, Gext2 and Gext3 but also by at least one portion of the internal physical magnitudes of the thermal element 20 (for example a compressor control) and applied on the inputs 34 of the set point generator 32, to determine the best set point to be applied as an input of the regulation loop 26 in order to minimize the energy consumption of the thermal element 20. For example, for the optimization of the high pressure set point, the physical magnitudes to be taken into account may be the temperature of the outside air, the control of the compressors, and the low pressure. In the example of
The regulation loop 26 can use various types of regulation, for example a regulation of PID type. However, the regulation loop will advantageously carry out a regulation of PFC (Predictive Functional Control) type, which makes it possible, relative to a PID regulation, to use a dynamic model of the regulation process inside the controller (control part 14 of the HVAC system 10) and in real time producing the regulation loop 26 in order to anticipate the future behaviour of the thermal element 20. The use of a regulation of PFC type is for example described in the work of J. Richalet et al.: “La commande prédictive. Mise en œuvre et applications industrielles” [Predictive control. Use and industrial applications], Editions Eyrolles.
The steps of the method of regulation and of energy optimization of the HVAC system 10 that is used, making it possible to achieve the regulation of the thermal elements of the HVAC system 10 previously described and to parameterize the various regulation algorithms used will be described with reference to
First, the user describes, for example in a software used as a description tool, the operative part 12 of the HVAC system 10. This description relates both to the architecture of the HVAC system 10, that is to say to the links between the various thermal elements 20 of the HVAC system 10, and to the intrinsic features of the thermal elements 20 and the nominal operating fields of the system 10 (step 102).
The user therefore describes, initially, the architecture of the HVAC system 10 by choosing, in the software, generic components representing each of the thermal elements 20 of the system 10 (heat exchangers, valves, compressors, pumps, fans, etc), then by creating the links between the thermal elements (by linking them by pipes, cables, etc). Each of the generic components present in the software corresponds to a mathematical modelling of the thermal and electrical behaviour of a thermal element of the HVAC system. Thus, the description produced by the user therefore forms, in the software, a mathematical modelling of the general architecture of the HVAC system 10. An example of a description made by the user is shown in
Once the general architecture of the system has been described, the user then describes individually the features of each of the thermal elements of this architecture of the HVAC system 10.
For this, the description software may comprise a library in which the features of existing elements are stored. The work of the user then consists in selecting a reference in the library. This selection may be assisted by filtering functions: the user may for example enter information on the element (for example for an exchanger: plates, tubes and grilles, nominal power, name of the manufacturer, etc), allowing him to have access either directly to the element or to a short list of potential elements.
If the element is not present in the library, or if the software does not comprise such a library, the user may describe the component completely by entering a list of features corresponding to this element (for example, for a plate exchanger: counter-current or co-current flow, the number of plates, the spacing between plates, the corrugation angle, the width and height of each plate, the material of the plates, the empty weight of the exchanger, describe whether the exchanger is thermally insulated, etc). The features to be entered depend on the type of element (pump, condenser, compressor, etc).
It is also possible to import into the description software a file comprising the features of the elements to be described, for example in the form of a table containing several points, that is to say one or more tables of values giving the values of the output parameters (pressure, fluid flow rate, etc) of the elements of the HVAC system depending on the input parameter values of these elements (input temperature, power, water flow rate, etc). Such a file may be a library file of DLL type. A link is made between the inputs/outputs of the imported model and the inputs/outputs of the element modelled in the software.
In a variant, this description of the operative part of the HVAC system 10 can be made by the user by answering a list of predefined questions in the description software, each of the questions relating for example to one or more parameters of one of the thermal elements of the HVAC system 10. Such a description of the HVAC system can be made when the HVAC system forms an assembly of finite machines, that is to say an HVAC system the architecture of which is predefined in the description software and not modified by the user.
Thus, the information that the user enters is easy to obtain and corresponds to data relating to the elements of the HVAC system 10 that the user is able to understand: for example, if the user is a refrigerationist, these data are exchanger references, exchanger mechanical features (number of plates, space between plates, length and width of the plates, etc), or else data tables supplied by the manufacturers (for example, for compressors, these data tables may comprise the values of the low pressure, of the high pressure, of the aspiration temperature, of the flow rate of the refrigerant, etc).
The user then describes, in the description software, the command and control of the system, that is to say the regulation loops of the system making it possible to regulate the various thermal elements of the HVAC system (step 104). Thus, for each regulation loop to be implemented on the system, corresponding to the regulation loop 26 described above with reference to
For this, the user again takes up the system modelling previously carried out in the description software and first of all sets the sensors to the physical magnitudes that he wishes to measure and that are intended to be regulated. Thus, in the example of
Pre-actuators are then positioned on the actuators of the HVAC system 10 to be driven, that is to say on the actuators of the compressors 52 and 58, of the condensers 54 and 60, and the actuators of the expansion valves 56 and 62. These pre-actuators are designed to receive at the input the values of the control signals delivered by the regulation loops. The user then links the sensors 66, 68 and the pre-actuators to the regulation boxes 70 which will be typed according to the nature of the output physical magnitude to be regulated (for example: the box for speed regulation, for overheating regulation, for pressure regulation, for temperature regulation, for regulation of the pressure difference, etc), but also depending on the desired type of regulation, notably PID or predictive control. Finally, the user defines, in the regulation boxes 70, the desired operating fields, that is to say the ranges of possible values of the set points of the loops (
In a manner similar to the description of the operative part of the HVAC system described above, the description of the command and control of the HVAC system 10 can be carried out by the user by answering a list of predefined questions in the description software, these questions in this instance relating to the parameters for command and control of the HVAC system 10 to be carried out.
Based on this complete description (thermal elements+regulation loops) of the HVAC system 10 made by the user, by means of the description software, simulations of the modelled system will be used by computation modules of the software in order to obtain the values of the parameters of the regulation loops from which it will be possible to program the controller or controllers of the control part 14 of the HVAC system 10 in order to ensure the command and control of the HVAC system 10, and of the parameters of the set point generator 32, comprising notably the values of the optimum set points intended to be applied as an input of the various regulation loops and making it possible to optimize the energy consumption of the HVAC system 10 by minimizing it.
Specifically, for each regulation loop, the user can describe set points of fixed values. However, these fixed values are not usually optimal in terms of energy consumption of the HVAC system, notably because of the internal and external physical magnitudes that may vary and influence the energy performance of the system. In order to optimize the energy consumption of the system, it is therefore preferable to vary the values of the set points depending on these internal and external physical magnitudes which may influence the operation of the system: this is the role fulfilled by the set point generator 32 shown in
For this, a simulation in steady state of the HVAC system 10 will be carried out (step 106). This simulation in steady state is carried out on the basis of the modelling obtained by the description of the system previously carried out by means of the description software. The set point values of the regulation loops defined by the regulation boxes 70 (corresponding to the set point applied to the input 30 of the regulation loop 26), and the values of the internal and external physical magnitudes are considered to be input variables for this simulation of the HVAC system in steady state.
In order to carry out this simulation in steady state, the user may also enter, as input data, constraints on the values of the set points and on the controls of the actuators (for example, for a motor, that the speed must be between 30 Hz and 50 Hz). The user may also enter maximum and/or minimum values that one or more internal and external physical magnitudes can take.
Once all these data have been entered into the software, a computation module of the software will simulate the behaviour of the HVAC system in steady state, that is to say when the outputs, in this instance the values of the control signals intended to be applied at the input of the actuators, are stable over time (that is to say, for a control signal u, such that du/dt=0). When the HVAC system is operating in steady state, for each regulation loop, the set point value applied at the input of the regulation loop is substantially equal to the value of the measurement that is applied at the input of the regulation loop (the error computed by the regulation loop being in this case substantially zero).
The variables obtained at the output of this simulation in steady state correspond to the control signals intended to be sent to the input of the actuators of the HVAC system 10. Since these values are stable (steady state), the computation module can then compute, based on the values of these signals and on the electrical and thermal models of the elements of the HVAC system 10, the electrical power absorbed by the HVAC system 10 in steady state.
This simulation in steady state is therefore carried out by scanning the possible values of the internal and external physical magnitude. If there are n internal and external physical magnitudes, the set of values that these physical magnitudes can take G1, . . . , Gn is called D, i.e. the domain of these physical magnitudes, such that: D=(G1, G2, . . . , Gn).
The computation module of the software samples the domain D as a finite set of points forming a set Dd (the discrete domain D). For each point of Dd, the module determines the values of the optimum set points to be applied making it possible to obtain a minimal energy consumption for the system. This gives a function foptimum which associates, with each point of Dd, a value such that:
foptimum: Dd={(G1, . . . , Gn)}->{(set point 1,set point 2, . . . )}
The value associated with each point of Dd is therefore determined so that this value is that for which the energy consumption of the system is lowest.
Then, each optimum set point value is evaluated in the form of a mathematical function which will be for example, advantageously, a polynomial in G1, . . . , Gn. The coefficients of the polynomial can then be determined by a least square method on the basis of the two sets Dd and foptimum(Dd). This gives the values of the coefficients of each polynomial:
set point 1=P1(G1,G2,G3, . . . );
set point 2=P2(G1,G2,G3, . . . );etc.
These values of the coefficients therefore form a list of parameters for the set point generator 32, which will be used for the programming of the controller or controllers of the control part 14 of the HVAC system 10.
This simulation in steady state therefore makes it possible to determine, as a function of the values of the internal and external physical magnitudes, the optimal values of the set points to be applied to the regulation loops in order to minimize the energy consumption of the HVAC system 10 while observing the constraints imposed by the user.
A simulation of the system in dynamic state is then applied in order to determine the parameters of the regulation loops 26 of the HVAC system 10 (step 108). These parameters of the regulation loops are obtained by determining the transfer functions of each of the thermal elements, corresponding to the ratios between the values of the output physical magnitudes y and the values of the control signals u and of the physical magnitudes that are external and internal to the thermal element 20 influencing the energy performance and the regulation performance of the thermal element 20.
To perform this simulation in dynamic state, each element 20 is assumed to be for example a system of the 1st order with delay, that is to say a system each of the transfer functions of which is of the type:
where
y: output physical magnitude
u: input signal (control signal or internal or external physical magnitude)
K: gain
θ: delay
τ: time constant.
Although a model of the 1st order represents a good compromise between the accuracy of the results obtained and the complexity of the computations used, it is totally possible for the model used to be of an order higher than the 1st order.
Such a simulation in dynamic state makes it possible to evaluate, as a function of the internal or external physical magnitudes and of the controls of the actuators (for example the outside temperature, the temperature of the water of the pipe return, the water flow rate in the pumps, the rotation speed of a fan, etc), the values of the output physical magnitudes that it is desired to control (overheating temperature, high pressure, low pressure, flow rate etc), that will subsequently be used to determine the parameters of the regulation loops.
For this, a computation module of the software varies values of signals applied on the inputs 22 and 23 (designed to receive the control signals and the physical magnitudes) of the thermal elements 20 of the modelled HVAC system, that is to say applies transitory signals, such as control steps, on these inputs 22 and 23. These control steps are applied around the operating point desired by the user, this operating point corresponding to the set points and to the physical magnitudes external to the system: outside temperature, flow rates, etc.
The software carries out a sampling of the signals obtained at the output of the simulation and then determines, for each response obtained (that is to say for each control step applied), values of the variables K, θ and τ of each transfer function. Finally, based on all the values obtained, and by applying for example a method of least squares, the computation module then determines the parameters K, θ and τ to be applied for each regulation loop.
In a particular embodiment, it is possible to use the optimum set points obtained by virtue of the simulation in steady state as input data for carrying out the simulation of the system in dynamic state. Thus, this simulation in dynamic state is carried out at the particular operating point desired by the user and for which the energy consumption of the HVAC system is minimal.
The regulation and energy-optimization parameters obtained by using the simulations in steady state and in dynamic state are then used to program the controller or controllers of the HVAC system (step 110).
A functional module of the software can use these parameters to generate functional blocks, that is to say algorithms encoded in a programming language, designed to be imported directly or by means of a programming software into the controller or controllers of the control part 14 of the HVAC system 10.
The functional blocks thus obtained based on the simulations in dynamic state and in steady state are algorithms in the form of computer codes making it possible to carry out the regulation and energy optimization of the various elements of the HVAC system 10.
When the controllers of the control part 14 are programmable controllers, the language of these codes then corresponds to the programming language of the controller or controllers of the control part 14 of the HVAC system 10, and corresponds for example to the C language or to the “structured text” language. The code of each functional block is then exported in the form of a file, for example encrypted, to an information-storage means (server, hard disk, USB key, CD-ROM). The file or files thus exported can then be imported into a software for programming the controllers and thus be used by the person in charge of programming the controllers (step 110). The programming may notably consist in defining the interactions between the functional block or blocks relative to the energy optimization, and with the functional blocks relative to the regulation of the HVAC system 10. The outputs of the functional block or blocks relating to the energy optimization, on which the optimum set points are delivered, will notably be linked to the inputs of the functional blocks relating to the regulation, each regulation functional block being able to receive as an input the optimum set point relating to the element designed to be regulated by this regulation functional block. The user then downloads the control program generated on the basis of the programming previously carried out into the memory of the programmable controller or controllers
In a variant embodiment, it is possible for the controller or controllers not to be programmable controllers, but parameterizable controllers. In this case, the programming is not carried out by the user, the latter in this case transferring the list of parameters to the controller, these data being transferred directly (directly or via a data medium) into the parameterizable controller or controllers without being encoded in the form of algorithms. On the other hand, it is possible for the parameters to be transferred in an encrypted manner so as, for example, to be able to lock certain functionalities as a function of an offer level chosen by the user.
It is also possible for the regulation and optimization parameters to be copied by the user directly into the controller or controllers or into a software for programming the controllers. Here again, these parameters can be supplied to the user in an encrypted form for reasons of locking functionalities and/or in order to prevent errors when the parameters are copied into the controller or controllers, for example by introducing error codes mixed with the parameters.
In the example described above, the simulation in steady state is carried out prior to the simulation in dynamic state. However, it is also possible for the simulation in dynamic state of the system to be carried out prior to or simultaneously with the simulation in steady state of the HVAC system.
The algorithms used by the computation module will be a function notably of the nature of the regulation loops (PID, predictive control, etc). Moreover, the mathematical models relating to the thermal and electrical behaviour of the elements of the HVAC system used by the software to produce the simulations in dynamic state and in steady state can be known mathematical models, for example described in the thesis of P. Schalbart entitled “Modélisation du fonctionnement en régime dynamique d'une machine frigorifique bi-étagée à turbo-compresseurs—Application à sa regulation” [Modelling the operation in dynamic state of a two-stage cooling machine with turbocompressors—application to its regulation], Ecole doctorale MEGA, 2006.
| Number | Date | Country | Kind |
|---|---|---|---|
| 1056751 | Aug 2010 | FR | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP11/63218 | 8/1/2011 | WO | 00 | 2/11/2013 |
