A heat pump system based on robust model prediction and a control method thereof

Abstract

The application discloses a kind of based on robust model prediction heat pump system and its control method, heat pump system includes heat pump cabin heating loop and heat pump cabin refrigeration loop, superimposes nominal system control law and additional feedback control law, as the control signal of heat pump system, control compressor speed, first electronic expansion valve opening, second electronic expansion valve opening and fan air volume;Nominal system control law is solved by constructing nominal system space state prediction model in nominal system model prediction controller, establishes heat pump system performance index function and state quantity constraint and control quantity constraint;Additional feedback controller input k time actual system state quantity, k+1 nominal system state quantity, additional feedback controller is stabilized at equilibrium point by Lyapunov function, and then additional feedback control law is solved.The application makes the heat management of electric vehicle more effective and energy saving.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of new energy vehicle thermal management, and particularly relates to a heat pump system based on robust model prediction and a control method thereof. BACKGROUND

[0002] With the development of pure electric vehicles, passengers have higher requirements for the vehicle's cruising range, and the decrease of the cruising range of pure electric vehicles in low-temperature environments is a problem that needs to be solved. Although heat pump technology can improve heating efficiency, it also has problems such as low heating efficiency in low-temperature environments and frequent defrosting. Therefore, it is necessary to develop an efficient vehicle thermal management control method to improve the heating efficiency in the cabin and reduce energy consumption, so that the thermal management of electric vehicles is more effective and energy-saving. SUMMARY

[0003] Therefore, the application provides a heat pump system based on robust model prediction and a control method thereof.

[0004] The application achieves the above technical purpose through the following technical means.

[0005] A control method of a heat pump system based on robust model prediction comprises the following steps:

[0006] (1) In the nominal system model predictive controller, a nominal system space state prediction model is constructed:

[0007]

[0008] According to the nominal system space state prediction model, a heat pump system performance index function is established:

[0009]

[0010] At the same time, the state quantity constraint and the control quantity constraint of the nominal system are determined as:

[0011]

[0012]

[0013] Wherein: N p is a prediction time domain, N c is a control time domain, Q is a weight of a state quantity, R is a weight of a control quantity, x ref (k+1) is a target reference state quantity at k+1 time, is a nominal system state quantity at k time, is a nominal system control quantity, A is a state transition matrix, B is a control matrix, is a nominal system control law at k time, indicates a nominal system state quantity, denotes the nominal system control variable, denotes the Pontryagin set subtraction, Z is the robust positive invariant set of the system, χ is the state variable constraint set, and U is the control variable constraint set;

[0014] Solving the above constrained nonlinear programming problem, x ref (k+1) is the nominal system state variable at time k+1 in the nominal system space state prediction model At the same time, according to the nominal system state variable x The nominal system control law at time k is output from the nominal system space state prediction model

[0015] (2) The additional feedback controller inputs the actual system state variable x The additional feedback controller is stabilized at the equilibrium point by means of Lyapunov function, and the additional feedback control law u feedback (k) is solved.

[0016] (3) Superimposing the nominal system control law u and the additional feedback control law u feedback (k) as the control signal of the heat pump system, the compressor speed, the first electronic expansion valve opening degree, the second electronic expansion valve opening degree and the fan air volume are controlled.

[0017] Further, the matrices A and B are solved by the heat pump system space state model, and the heat pump system space state model is:

[0018] x(k+1)=Ax(k)+Bu(k);

[0019] According to the simulation data of the heat pump system space state model, the unknown parameters in the matrices A and B are identified, and the input variables are the compressor speed, the first electronic expansion valve opening degree, the second electronic expansion valve opening degree and the fan air volume in turn, and the output variables are the cabin outlet temperature, the high pressure of the circuit, the low pressure of the circuit and the temperature of the circuit in turn;

[0020] Based on the given working condition cycle data of the heat pump system space state model simulation experiment, the least square method is used to identify the parameters of the heat pump system space state model.

[0021] Further, the actual system space state prediction model obtained from the heat pump system space state model is:

[0022] x(k+1)=Ax(k)+Bu(k)+w;x∈χ,u∈U

[0023] wherein: w is an unknown disturbance, and |w|≤w bound , w boundThe upper limit of the interference.

[0024] Further, a Gaussian radial basis neural network is used to predict unknown disturbances, which are:

[0025] w = omega *T (z) + xi

[0026] wherein omega * is an optimal weight matrix, z is an input, xi is a model error, and |xi| <= xi0, ||omega * || <= omega0.

[0027] A heat pump system based on robust model prediction, comprising a compressor heater, a first on-off valve, a first electronic expansion valve, a second on-off valve, a third on-off valve, an air conditioner evaporator, an outdoor heat exchanger, a second electronic expansion valve, a third electronic expansion valve, a water side heat exchanger, a high-pressure coolant heater, a liquid accumulator and a power battery; the suction port of the compressor is in communication with the outlet of the liquid accumulator, the exhaust port of the compressor is in communication with the first port of the heater, the second port of the heater is connected to the first port of the first electronic expansion valve at one end and to the second port of the second on-off valve at the other end through a pipeline node C; the second port of the first electronic expansion valve is connected to the first port of the air conditioner evaporator at one end and to the first port of the third on-off valve at the other end through a pipeline node A; the second port of the air conditioner evaporator is connected to the first port of the second electronic expansion valve at one end and to the second port of the third electronic expansion valve at the other end through a pipeline node F; the second port of the second electronic expansion valve is in communication with the first port of the outdoor heat exchanger; the second port of the outdoor heat exchanger is in communication with the first port of the second on-off valve at one end and with the first port of the first on-off valve at the other end through a pipeline node B; the second port of the first on-off valve is in communication with the inlet of the liquid accumulator through a pipeline node D and is connected to the pipeline node E through the pipeline node D, and the pipeline node E is also in communication with the second port of the third on-off valve and the second port of the high-pressure coolant heater; the water side heat exchanger and the power battery are connected between the first port of the third electronic expansion valve and the first port of the high-pressure coolant heater.

[0028] In the above technical solution, the heat pump system further comprises a blower and a fan, the blower provides the required air flow for the heat exchange between the refrigerant and air of the outdoor heat exchanger, and the fan provides the required air flow for the heat exchange between the refrigerant and air of the air conditioner evaporator and the heater.

[0029] In the above technical solution, the heat pump system further comprises a loop high-pressure section pressure sensor and a loop low-pressure section pressure sensor, the loop high-pressure section pressure sensor is located between the heater and the pipeline node C, and the loop low-pressure section pressure sensor is located between the liquid accumulator and the pipeline node D.

[0030] The working mode of the heat pump system includes a heat pump cabin heating mode in a low-temperature environment and a heat pump cabin refrigeration mode in a high-temperature environment.

[0031] The present application has the following advantages:

[0032] (1) The control method of the present application superimposes a nominal system control law and an additional feedback control law as a control signal of the heat pump system to control the compressor speed, the first electronic expansion valve opening degree, the second electronic expansion valve opening degree and the fan air volume; that is, the model predictive control and the robust control are combined, the overshoot is smaller and the control response is more rapid, and the control method can solve the target limited optimization problem of a complex control system with multiple inputs and outputs.

[0033] (2) The heat pump system of the present application realizes the heat pump refrigeration mode and the cabin heating mode by using the robust model predictive control, the nominal system control law is obtained by constructing a nominal system space state prediction model in a nominal system model predictive controller, establishing a heat pump system performance index function and state quantity constraints and control quantity constraints for solving, thereby improving the heating efficiency in the driver's cabin, reducing the energy consumption, and making the thermal management of the electric vehicle more effective and energy-saving. BRIEF DESCRIPTION OF DRAWINGS

[0034] The features and advantages of the present application can be better understood by referring to the following detailed description in conjunction with the accompanying drawings, in which the same reference notations are used to represent the same elements throughout the several views of the vehicle drawings, and wherein:

[0035] Figure 1 The heat pump system structure diagram of the present application is shown in the figure;

[0036] Figure 2 is Figure 1 The communication connection diagram of the control module and the actuators of the heat pump system is shown in the figure;

[0037] Figure 3 is Figure 2 The schematic internal structure diagram of the control module is shown in the figure;

[0038] Figure 4 The cabin heating mode flow chart is shown in the figure;

[0039] Figure 5 The cabin cooling mode flow chart is shown in the figure;

[0040] Figure 6 The robust model predictive heat pump system control flow chart is shown in the figure;

[0041] In the figure: 101 - compressor, 102 - heater, 103 - first on-off valve, 104 - first electronic expansion valve, 105 - second on-off valve, 106 - third on-off valve, 107 - air conditioning evaporator, 108 - outdoor heat exchanger, 109 - blower, 110 - second electronic expansion valve, 111 - third electronic expansion valve, 112 - Chiller, 113 - high-pressure coolant heater, 114 - reservoir, 115 - fan, 116 - power battery, 117 - circuit high-pressure section pressure sensor, 118 - circuit low-pressure section pressure sensor, 1011 - compressor suction port, 1012 - compressor discharge port, 1021 - heater first port, 1022 - heater second port, 1031 - first on-off valve first port, 1032 - first on-off valve second port, 1041 - first electronic expansion valve first port, 1042 - first electronic expansion valve second port, 1051 - second on-off valve first port, 1052 - second on-off valve second port, 1061 - third on-off valve first port, 1062 - third on-off valve second port, 1071 - air conditioning evaporator first port, 1072 - air conditioning evaporator second port, 1081 - outdoor heat exchanger first port, 1082 - outdoor heat exchanger second port, 1101 - second electronic expansion valve first port, 1102 - second electronic expansion valve second port, 1111 - third electronic expansion valve first port, 1112 - third electronic expansion valve second port, 1121 - Chiller first port, 1122 - Chiller second port, 1131 - high-pressure coolant heater first port, 1132 - high-pressure coolant heater second port, 1141 - reservoir outlet, 1142 - reservoir inlet, 1161 - power battery first port, 1162 - power battery second port, 100 - heat pump system, 8000 - control module, 8001 - bus, 8002 - input interface, 8003 - memory, 8004 - processor, 8005 - output interface, 8101 - first interface, 8102 - second interface, 8103 - third interface, 8104 - fourth interface, 8200 - connection. DETAILED DESCRIPTION

[0042] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. For example, electronic expansion valves and single-way valves can be replaced by other reasonable types of valves. Rather, these embodiments are provided to provide a more thorough and complete understanding of the disclosure of the present invention. Various specific embodiments of the invention will now be described with reference to the accompanying drawings, which form part of this specification. It should be understood that ordinal numbers such as "first" and "second" used in this application are used merely for distinction and identification and have no other meaning, and do not indicate a specific order or association unless specifically specified. For example, the term "first three-way valve" does not imply the existence of a "second three-way valve," and the term "second electronic expansion valve" does not imply the existence of a "first electronic expansion valve."

[0043] Figure 1 This is a system diagram of a heat pump system 100 according to an embodiment of this application, illustrating the components of the heat pump system 100 and their connections. Figure 1As shown, the heat pump system 100 includes a compressor 101, a heater 102, a first on-off valve 103, a first electronic expansion valve 104, a second on-off valve 105, a third on-off valve 106, an air conditioning evaporator 107, an outdoor heat exchanger 108, a blower 109, a second electronic expansion valve 110, a third electronic expansion valve 111, a chiller 112, a high-pressure coolant heater 113, a reservoir 114, a fan 115, a power battery 116, a high-pressure circuit pressure sensor 117, a low-pressure circuit pressure sensor 118, and connecting pipes between each of the components represented by lines. The selection and role of each component of the heat pump system 100 are described as follows. The compressor 101 is selected as a scroll or other type of electric compressor, which functions to evaporate and compress the refrigerant into superheated steam and drive its flow in the refrigerant circulation system. The first on-off valve 103, the second on-off valve 105, and the third on-off valve 106 can be solenoid or electric on-off valves, which control the opening and closing of the valves. The first electronic expansion valve 104, the second electronic expansion valve 110, and the third electronic expansion valve 111 can be solenoid or electric expansion valves, which achieve temperature accuracy of superheat or subcooling by controlling the valve hole opening degree. The air conditioning evaporator 107 and the outdoor heat exchanger 108 are air-side heat exchangers, which provide heat exchange between air and refrigerant. The first blower 109 can be an electric blower of different types, which provides the required air flow for the outdoor heat exchanger 108 to provide heat exchange between refrigerant and air. The fan 115 can be a fan of different types, which not only provides the required air flow for the air conditioning evaporator 107 to provide heat exchange between refrigerant and air, but also provides the required air flow for the heater 102 to provide heat exchange between refrigerant and air. The chiller 112 is a water-side heat exchanger, which provides heat exchange of coolant. The high-pressure coolant heater 113 is a device for heating the coolant in a low-temperature state. The reservoir 114 separates the liquid refrigerant and gaseous refrigerant in the refrigerant circulation. The high-pressure circuit pressure sensor 117 is used to measure the refrigerant pressure in the high-pressure section of the refrigeration circuit. The low-pressure circuit pressure sensor 118 is used to measure the refrigerant pressure in the low-pressure section of the refrigeration circuit.

[0044] The connecting pipes between the components of the heat pump system 100 are described as follows. In the circuit, the suction port 1011 of the compressor is in communication with the outlet port 1141 of the liquid accumulator; the first port 1021 of the heater is in communication with the discharge port 1012 of the compressor; the second port 1022 of the heater is connected to one end of the first electronic expansion valve first port 1041 and the other end of the second switch valve second port 1052 through the pipe node C; the second port 1042 of the first electronic expansion valve is connected to one end of the air conditioner evaporator first port 1071 and the other end of the third switch valve first port 1061 through the pipe node A; the second port 1072 of the air conditioner evaporator is connected to one end of the second electronic expansion valve first port 1101 and the other end of the third electronic expansion valve second port 1112 through the pipe node F; the second port 1102 of the second electronic expansion valve is in communication with the outdoor heat exchanger first port 1081; the second port 1082 of the outdoor heat exchanger is in communication with one end of the second switch valve first port 1051 and the other end of the first switch valve first port 1031 through the pipe node B; the second port 1032 of the first switch valve is in communication with the liquid accumulator inlet port 1142 through the pipe node D, connected to the pipe node E and the third switch valve second port 1062 through the pipe node D, and in communication with the pipe node E and the high-pressure cooling liquid heater second port 1132 through the pipe node D; the first port 1111 of the third electronic expansion valve is in communication with the Chiller second port 1122; the first port 1161 of the power battery is in communication with the Chiller first port 1121; the second port 1162 of the power battery is in communication with the high-pressure cooling liquid heater first port 1131. The air blower 109 provides the required air flow for the heat exchange between the refrigerant and the air of the outdoor heat exchanger 108; the fan 115 provides the required air flow for the heat exchange between the refrigerant and the air of the air conditioner evaporator 107 and the heater 102; the circuit high-pressure section pressure sensor 117 is located between the heater 102 and the pipe node C; and the circuit low-pressure section pressure sensor 118 is located between the liquid accumulator 114 and the pipe node D. The heat pump system of the present application provides two working modes of heat pump air conditioning heating and refrigeration through a simple topology, thereby meeting different heat management requirements.

[0045] Figure 2 is a control module and Figure 1 is a schematic diagram of the communication connection of the heat pump system in each actuator. As Figure 2As shown, the control module 8000 determines the operating status of each actuator in the heat pump system 100. The first interface 8101, second interface 8102, third interface 8103, and fourth interface 8104 of the control module 8000's output interface 8005 are respectively connected to the compressor 101, fan 115, first electronic expansion valve 104, and second electronic expansion valve 110 for communication. The control module 8000 controls the refrigerant flow rate by controlling the compressor 101's speed; the control module 8000 controls the fan airflow by controlling the fan 115's speed; and the control module 8000 regulates the refrigerant pressure and temperature by controlling the opening of the first electronic expansion valve 104 and the second electronic expansion valve 110.

[0046] Figure 3 yes Figure 2 The diagram shows a schematic internal structure of the control module. Figure 3 As shown, the control module 8000 of the heat pump system 100 includes a bus 8001, an input interface 8002, a memory 8003, a processor 8004, and an output interface 8005. Specifically, the memory 8003 stores programs, instructions, and data, while the processor 8004 reads programs, instructions, and data from the memory 8003 and can write data to the memory 8003. By executing the program and instruction read from the memory 8003, the processor 8004 exchanges signals through the input interface 8002 and the output interface 8004. Figure 3 As shown, the control module 8000's input interface 8002 receives operating requests and other operating parameters from the heat pump system 100 via connection 8200. The first interface 8101, second interface 8102, third interface 8103, and fourth interface 8104 of the output interface 8005 are respectively connected to the compressor 101, fan 115, first electronic expansion valve 104, and second electronic expansion valve 110 for communication. Through the program and instructions in the actuator 8003, the processor 8004 controls the operation of the heat pump system 100. Specifically, the control module 8000 can receive operating requests from the heat pump system 100 or signals from other components via the input interface 8002, and send control signals to each controlled component via the output interface 8005, thereby enabling the heat pump system 100 to operate in a specified working mode and switch between different modes.

[0047] Figure 4 and Figure 5 yes Figure 1 The system diagram of the heat pump system 100 shown illustrates the fluid flow state of the heat pump system 100 under different operating modes. Bold solid arrows indicate the direction and path of refrigerant flow, while solid lines indicate no fluid flow. Details are as follows. Figure 4 and Figure 5 The working mode shown.

[0048] Figure 4 yes Figure 1 The diagram shows the system diagram of the heat pump system 100 in a low-temperature environment, operating in a cabin heating mode. In a low-temperature environment, upon receiving a cabin heating command (or automatically generated by the control module 8000), the heat pump system 100 can transfer heat to the cabin via heat pump heating. Figure 4 As shown, high-temperature, high-pressure refrigerant flows from the compressor discharge port 1011 into the first port 1021 of the heater. As the high-temperature refrigerant passes through the heater 102, it releases heat through the air blown out by the fan 115, heating the cabin. At the second port 1022 of the heater, it reverts to low-temperature refrigerant and flows towards the first electronic expansion valve 104. Under the pressure reduction and accumulation effect of the first electronic expansion valve 104, a low-temperature, low-pressure liquid mist mixture is formed and flows into the first port 1071 of the air conditioner evaporator, where it is then evaporated by the air conditioner. The refrigerant flows from the second port 1072 of the refrigerant exchanger through the second electronic expansion valve 110 to the first port 1081 of the outdoor heat exchanger. At this time, the outdoor heat exchanger 108 acts as an evaporator. Through the blower 109, the outdoor heat exchanger 108 absorbs a large amount of heat from the ambient air, causing the refrigerant to become gaseous. This gaseous refrigerant then flows out from the second port 1082 of the outdoor heat exchanger, passing through pipe node B, the first switching valve 103, and pipe node D before flowing to the inlet 1142 of the receiver 114. The receiver 114 separates the liquid and gaseous refrigerants. The compressor suction port 1011 draws in the gaseous refrigerant from the receiver outlet 1141, beginning the next refrigerant cycle.

[0049] Figure 5 yes Figure 1 The diagram shows a heat pump system in cabin cooling mode under high-temperature conditions. When the heat pump system 100 detects a high ambient temperature, it cools the cabin. Figure 5As shown, in this mode, the high-temperature and high-pressure refrigerant flowing out of the compressor exhaust port 1012 flows into the heater first port 1021, and after passing through the heater 102, the refrigerant changes from gas to liquid. The liquid high-pressure refrigerant flows out of the heater second port 1022 and flows to the outdoor heat exchanger second port 1082 after passing through the second switch valve 105. At this time, the outdoor heat exchanger 108 is used for heat dissipation, and through the blower 109, the outdoor heat exchanger 108 releases a large amount of heat to the ambient air, so that the refrigerant becomes liquid and flows out of the outdoor heat exchanger first port 1081. The refrigerant flows through the second electronic expansion valve 110, and under the action of the pressure reduction and accumulation of the second electronic expansion valve 110, a low-temperature and low-pressure liquid mist mixture is formed and flows to the air conditioner evaporator second port 1072 through the pipeline node F. At this time, the refrigerant absorbs heat from the air blown by the fan 115 and performs refrigeration. Then, the refrigerant flows through the third switch valve to the reservoir inlet 1142 through the pipeline node A, the pipeline node E and the pipeline node D. After passing through the reservoir 114, the liquid refrigerant and the gaseous refrigerant are separated. The compressor suction port 1011 sucks in the gaseous refrigerant from the reservoir outlet 1141, and the next refrigerant cycle begins to work.

[0050] A heat pump system controller based on robust model prediction, comprising: a heat pump system space state model, a nominal system model predictive controller and an additional feedback controller; the heat pump system space state model is used as the prediction equation of the nominal system model predictive controller to predict the system state quantity in the prediction time domain in order to solve the optimal control quantity; the nominal system model predictive controller is used to solve the first part of the control law, i.e. the nominal system control law, which takes the target reference state quantity as the input, converts the optimal control problem into a nonlinear programming problem according to the prediction equation, and solves the control parameters: the additional feedback controller is used to drive the actual system state quantity to approach the state trajectory of the nominal system, so that the actual system approaches the nominal system infinitely.

[0051] The heat pump system controller based on robust model prediction superimposes the control law of the above-mentioned nominal system model predictive controller and the control law of the additional feedback controller as the final heat pump system control signal, which is sent to the heat pump system. The heat pump system feeds back the current state quantity to the additional feedback controller to form a closed-loop control.

[0052] A construction method of a heat pump system based on robust model prediction control, comprising the following steps:

[0053] Step (1): based on the perception of the vehicle's external environment and its own parameters, a heat pump system space state model is established:

[0054] x(k+1)=Ax(k)+Bu(k)

[0055] Wherein, x(k+1) represents the state quantity at k+1 time, x(k) represents the state quantity at k time, u(k) represents the input quantity at k time, A is a state transition matrix, and B is a control matrix.

[0056] The application models a heat pump system as a controlled object based on a state space equation, takes the state space equation of the heat pump system as a black box model, that is, takes the state transition matrix A and the control matrix B as unknown parameters, and takes the output matrix C as a unit matrix; after selecting the input quantity and the output quantity of the state space equation and the state quantity, the parameters of the matrix A and B are obtained through model simulation data driving and parameter identification. For the heat pump system of the application, a state space equation with 4 states and 4 inputs is selected as the system model, wherein the cabin air outlet temperature, the loop high pressure, the loop low pressure and the loop temperature are taken as the state quantity, the compressor speed, the first electronic expansion valve opening degree, the second electronic expansion valve opening degree and the fan air volume are taken as the input quantity, and the output matrix is a unit matrix, directly outputting the state quantity (that is, the output quantity) of the system; therefore, the expression of the heat pump system space state model is as follows:

[0057]

[0058] Wherein: x1(k), x2(k), x3(k) and x4(k) are respectively the cabin air outlet temperature, the loop high pressure, the loop low pressure and the loop temperature at k time, u1(k), u2(k), u3(k) and u4(k) are respectively the compressor speed, the first electronic expansion valve opening degree, the second electronic expansion valve opening degree and the fan air volume at k time, and x1(k+1), x2(k+1), x3(k+1) and x4(k+1) are respectively the cabin air outlet temperature, the loop high pressure, the loop low pressure and the loop temperature at k+1 time.

[0059] The application takes the heat pump state space equation as a black box model, needs to identify the unknown parameters in the matrix A and B according to model simulation data, the input quantity is the compressor speed, the first electronic expansion valve opening degree, the second electronic expansion valve opening degree and the fan air volume in sequence, and the output quantity is the cabin air outlet temperature, the loop high pressure, the loop low pressure and the loop temperature in sequence; based on the given working condition cycle data of the model simulation experiment, the least square method Levenberg-Marquardt (LM) algorithm is used to identify the parameters of the state space equation. The Levenberg-Marquardt (LM) algorithm is one of gradient optimization iterative solution methods. The LM algorithm uses the second order derivative of the objective function, and dynamically adjusts the convergence direction by using the direction vector, so as to increase the convergence performance and ensure good convergence speed. The variable iteration formula of the LM algorithm is as follows:

[0060] x(a+1)=x(a)-[J T(x(a))J(x(a)) + η(a) diag(J T (x(a))J(x(a)) -1 J T (x(a))r(x(a))

[0061] where a is the iteration step number, r(x) is the residual vector, J(x) is the Jacobian matrix, and:

[0062] r(x) = [r1(x), r2(x),..., r n (x)

[0063]

[0064] where q is the dimension of the optimization variable, and n is the number of data sample points.

[0065] The main steps of the LM algorithm are as follows:

[0066] 1) Let a = 0, and give the initial value x0, iteration accuracy ε0, and initial length η0;

[0067] 2) Calculate r(x(a)) and J(x(a));

[0068] 3) Calculate x(a+1) from the LM algorithm variable iteration formula, and calculate the objective function value;

[0069] 4) Determine whether the objective function value meets the accuracy ε0 requirement. If it does, the optimal solution x * = x(a+1), and the iteration ends, otherwise, continue to step 5);

[0070] 5) Calculate the difference between the current objective function value and the objective function value of the previous step. If the difference is positive, let η(a+1) = 10η(a), otherwise, η(a+1) = 0.1η(a);

[0071] 6) Let a = a + 1, and return to step 2).

[0072] Step (2): On the basis of the model in step (1), consider the external disturbance and establish the constraint of the nominal system.

[0073] Assume that the unknown disturbance caused by the external system is w, and |w|≤w bound , w bound is the upper limit of the disturbance, and the state space equation of the system is reconstructed as follows:

[0074] x(k+1) = Ax(k) + Bu(k) + w

[0075] A Gaussian radial basis function (RBF) neural network is used to predict unknown disturbances. This method has strong nonlinear fitting capabilities, simple learning rules, and is easy to implement on a computer. The neural network-based nonlinear observer can be constructed as follows:

[0076]

[0077]

[0078] in, This is the estimated value, and L is the observer gain. The estimated disturbance is constructed as follows:

[0079]

[0080] in, It is the input vector; ω T It is the estimated weight vector; It is an activation function, and satisfies It can be constructed as:

[0081]

[0082] Where: μ i With δ as the center, and δ as the width.

[0083] Based on the general approximation property of neural networks, the perturbation can be approximated as:

[0084] w=ω *T (z)+ξ

[0085] Where, ω * It is the optimal weight matrix; z = [x(k)] T ξ is the input; ξ is the model error, satisfying |ξ|≤ξ0. Since the perturbation is bounded, the optimal weights are also bounded, and satisfy ||ω|≤ξ0. * ||≤ω0.

[0086] Based on the above formula, if the target is perturbed, a neural network-based nonlinear observer can be constructed as follows:

[0087] x(k+i)=Ax(k+i-1)+Bu(k+i-1)+ω *T (z)+ξ

[0088] y(k+i)=Cx(k+i)

[0089] The state-space model expression of a nonlinear observer heat pump based on a neural network is as follows:

[0090]

[0091]

[0092] The actual system is defined as the system considering unmodeled dynamics and bounded disturbances, and its space state prediction model is as follows:

[0093] x(k+1) = Ax(k) + Bu(k) + w; x e χ, u e U

[0094] wherein χ is the state quantity constraint set, and U is the control quantity constraint set.

[0095] The nominal system is defined as the system without considering unmodeled dynamics and bounded disturbances, and its space state prediction model is as follows:

[0096]

[0097] is the control quantity of the nominal system at time k, is the state quantity of the nominal system at time k.

[0098] The actual system state quantity x(k) is the real state quantity of the heat pump system at time k, including the cabin air outlet temperature, the high pressure of the circuit, the low pressure of the circuit and the heat exchanger temperature, which are obtained through the signal parameters of the temperature sensor located at the cabin air outlet, the pressure sensor of the high pressure section of the circuit, the pressure sensor of the low pressure section of the circuit and the temperature sensor at the heat exchanger. The nominal system state quantity is the state quantity at time k output by the nominal system model prediction controller, including the cabin air outlet temperature, the high pressure of the circuit, the low pressure of the circuit and the heat exchanger temperature. Z is defined as the robust positive invariant set of the system, so that the state quantity constraint and the control quantity constraint of the nominal system are rewritten as:

[0099]

[0100]

[0101] wherein: represents the Minkowski set subtraction, represents the nominal system state quantity, represents the nominal system control quantity.

[0102] Step (3): The control law of the heat pump system controller is composed of the nominal system control law and the additional feedback control law:

[0103]

[0104] is the control law of the nominal system at time k, including the speed control signal of the compressor, the opening control signal of the first electronic expansion valve, the opening control signal of the second electronic expansion valve and the speed control signal of the fan.

[0105] u feedback (k) is the additional feedback control law, which contains the speed control compensation signal of the compressor, the opening control compensation signal of the first electronic expansion valve, the opening control compensation signal of the second electronic expansion valve and the speed control compensation signal of the fan output by the additional feedback controller.

[0106] Through the additional feedback control law u feedback (k) makes the actual system state quantity x(k) consistent with the state quantity of the nominal system. of the nominal system.

[0107] Step (4): the design method of the nominal system model predictive controller includes:

[0108] The nominal system model predictive controller includes: the nominal system space state prediction model, the heat pump system performance index function, the state quantity constraint and the control quantity constraint of the nominal system.

[0109] The heat pump system performance index function contains two items, the first item is the difference between the target reference state quantity at k+1 time in the prediction time domain and the nominal system state quantity at k time in the nominal system space state prediction model in the prediction time domain ; the second item is the difference between the nominal system control law at k+1 time in the control time domain and the nominal system control law at k time in the nominal system space state prediction model, and finally the two items are added to write the heat pump system performance index function:

[0110]

[0111] Wherein: N p is the prediction time domain, N c is the control time domain, Q is the weight of the state quantity, R is the weight of the control quantity, x ref (k+1) is the target reference state quantity at k+1 time, that is, the state quantity that the heat pump system is expected to reach at k+1 time, including the target temperature of the cabin air outlet, the target high pressure of the circuit, the target low pressure of the circuit and the target temperature of the outdoor heat exchanger.

[0112] Solving the constrained nonlinear programming problem, taking the target reference state quantity x ref (k+1) at k+1 time as the nominal system state quantity at k+1 time of the nominal system space state prediction model, and according to the nominal system state quantity at k time, outputting the nominal system control law at k time by the nominal system space state prediction model. Wherein is the initial state quantity of the nominal system at k=0 time.

[0113] Step (5): the additional feedback controller is constructed by using a sliding mode control method, and a sliding surface is designed as σ=s T e(k), wherein s is a sliding surface coefficient, and the definition is is a deviation between an actual system state quantity and a nominal system state quantity at the k moment.

[0114] A Lyapunov function is defined as:

[0115]

[0116] In order to make the additional feedback controller stable at the equilibrium point σ=0, according to the Lyapunov stability principle, it is obtained that: Let wherein ε>0 is a sliding surface parameter.

[0117] Derivation is performed on σ=s T e(k), and it is obtained that:

[0118]

[0119] wherein ρ is a soft factor, and the following is obtained: Substituting the above formula, it is obtained that:

[0120]

[0121] Thus, it is obtained that:

[0122]

[0123] That is, u feedback (k)=u(k).

[0124] The u is applied to the heat pump system to form a closed loop control.

[0125] The embodiments are preferred embodiments of the present application, but the present application is not limited to the above embodiments, and any obvious improvement, replacement or modification made by those skilled in the art without departing from the essential content of the present application shall fall within the protection scope of the present application.

Claims

1. A control method of a heat pump system based on robust model prediction, characterized in that: (1) in a nominal system model predictive controller, a nominal system space state prediction model is constructed: a heat pump system performance index function is established according to the nominal system space state prediction model: the state quantity constraint and the control quantity constraint of the nominal system are determined at the same time: the matrix A and the matrix B are solved through a heat pump system space state model, the heat pump system space state model is: x (k+1) =Ax (k) +Bu (k) ; unknown parameters in the matrix A and the matrix B are identified according to simulation data of the heat pump system space state model, input quantities are compressor speed, first electronic expansion valve opening degree, second electronic expansion valve opening degree and fan air volume in turn, and output quantities are cabin air outlet temperature, loop high pressure, loop low pressure and loop temperature in turn; parameter identification is performed on the heat pump system space state model by using a least square method based on given working condition cycle data of simulation experiment of the heat pump system space state model; an actual system space state prediction model obtained from the heat pump system space state model is: x (k+1) =Ax (k) +Bu (k) +w; x∈χ,u∈U unknown disturbances are predicted by using a Gaussian radial basis neural network, and the unknown disturbances are: ​ ​ ​ wherein: N p is a prediction horizon, N c is a control horizon, Q is a weight of a state quantity, R is a weight of a control quantity, x ref (k+1) is a target reference state quantity at k+1 time, is a nominal system state quantity at k time, is a nominal system control quantity, A is a state transition matrix, B is a control matrix, is a nominal system control law at k time, denotes a nominal system state quantity, denotes a nominal system control quantity, denotes a Pontryagin set subtraction, Z is a robust positive invariant set of the system, χ is a state quantity constraint set, U is a control quantity constraint set; the target reference state quantity at k+1 time includes a cabin air outlet target temperature, a loop target high pressure, a loop target low pressure, and an outdoor heat exchanger target temperature, the nominal system control law at k time contains a speed control signal of a compressor, an opening control signal of a first electronic expansion valve, an opening control signal of a second electronic expansion valve, and a speed control signal of a fan, and the nominal system state quantity includes a cabin air outlet temperature, a loop high pressure, a loop low pressure, and a heat exchanger temperature. Solving the above constrained nonlinear programming problem, x ref (k+1) as the nominal system state quantity at time k+1 in the nominal system space state prediction model According to the nominal system state quantity at time k Output the nominal system control law at time k from the nominal system space state prediction model (2) the actual system state quantity x(k) at time k and the nominal system state quantity x(k+1) at time k+1 of the additional feedback controller input k The additional feedback controller is stabilized at the equilibrium point by Lyapunov function, and then the additional feedback control law u is solved feedback (k); the actual system state quantity includes cabin air outlet temperature, loop high pressure, loop low pressure and heat exchanger temperature, and the additional feedback control law includes a speed control compensation signal of the compressor, an opening control compensation signal of the first electronic expansion valve, an opening control compensation signal of the second electronic expansion valve and a speed control compensation signal of the fan; (3) the nominal system control law and the additional feedback control law u feedback (k) superimposing, as a control signal of the heat pump system, the compressor rotation speed, the first electronic expansion valve opening degree, the second electronic expansion valve opening degree, and the fan air volume.

2. The control method according to claim 1, characterized by, ​ ​ ​ ​ 3. The control method according to claim 2, characterized by, ​ ​ where: w is an unknown disturbance, and |w| < w bound , w bound is an upper bound on the disturbance.

4. The control method according to claim 3, characterized by, ​ w = ω *T (z) + ξ where: ω * is the optimal weight matrix, z is the input, ξ is the model error, and |ξ|≤ξ0, ||ω * ||≤ω0.

5. A heat pump system implementing the control method according to any one of claims 1 to 4, characterized in that, The heat pump system comprises a compressor (101), a heater (102), a first switch valve (103), a first electronic expansion valve (104), a second switch valve (105), a third switch valve (106), an air conditioner evaporator (107), an outdoor heat exchanger (108), a second electronic expansion valve (110), a third electronic expansion valve (111), a water-side heat exchanger (112), a high-pressure coolant heater (113), a liquid accumulator (114), and a power battery (116); the suction port of the compressor (101) is communicated with the outlet of the liquid accumulator (114), the exhaust port of the compressor (101) is communicated with the first port of the heater (102), the second port of the heater (102) is communicated with the second port of the second switch valve (105) through a pipeline node C, one end of the first port of the first electronic expansion valve (104) is communicated with the pipeline node C, and the other end of the first port of the first electronic expansion valve (104) is communicated with the second port of the third switch valve (106); one end of the second port of the first electronic expansion valve (104) is connected with the first port of the air conditioner evaporator (107), and the other end of the second port of the first electronic expansion valve (104) is connected with the first port of the third switch valve (106); one end of the second port of the air conditioner evaporator (107) is connected with the first port of the second electronic expansion valve (110), and the other end of the second port of the air conditioner evaporator (107) is connected with the second port of the third electronic expansion valve (111); the second port of the second electronic expansion valve (110) is communicated with the first port of the outdoor heat exchanger (108); the second port of the outdoor heat exchanger (108) is communicated with the first port of the second switch valve (105) through a pipeline node B, and the second port of the outdoor heat exchanger (108) is communicated with the first port of the first switch valve (103) through the pipeline node B; the second port of the first switch valve (103) is communicated with the inlet of the liquid accumulator (114) through a pipeline node D, and the second port of the first switch valve (103) is connected with a pipeline node E through the pipeline node D; the pipeline node E is further communicated with the second port of the third switch valve (106) and the second port of the high-pressure coolant heater (113); the water-side heat exchanger (112) and the power battery (116) are connected between the first port of the third electronic expansion valve (111) and the first port of the high-pressure coolant heater (113).

6. The heat pump system of claim 5, wherein, The heat pump system further comprises a blower (109) and a fan (115), the blower (109) provides the required air flow for the heat exchange between the refrigerant and air of the outdoor heat exchanger (108), and the fan (115) provides the required air flow for the heat exchange between the refrigerant and air of the air conditioner evaporator (107) and the heater (102).

7. The heat pump system of claim 5, wherein, The heat pump system further comprises a high-pressure circuit pressure sensor (117) and a low-pressure circuit pressure sensor (118), the high-pressure circuit pressure sensor (117) is located between the heater (102) and the pipeline node C, and the low-pressure circuit pressure sensor (118) is located between the liquid accumulator (114) and the pipeline node D.

8. The heat pump system of claim 5, wherein, The working modes of the heat pump system include a heat pump cabin heating mode in a low-temperature environment and a heat pump cabin refrigeration mode in a high-temperature environment.

Images