Power regulation method, system, device and medium for vehicle hybrid energy storage device

Abstract

The application provides a power regulation method, system, device and medium of a vehicle hybrid energy storage device. The method comprises: acquiring the speed and acceleration of the vehicle at the current time, and predicting the load compensation value of the hybrid energy storage device at the future time; acquiring the state variable of the hybrid energy storage device at the current time and the current change amount of the fuel cell at the previous time, and obtaining the residual compensation value of the hybrid energy storage device at the future time; inputting the preset reference value and the load compensation value and the residual compensation value into the model in the model predictive control algorithm, and calculating the current change amount of the fuel cell of the vehicle in the preset control interval, and regulating the power distribution of the hybrid energy storage device in the control interval. The application can realize effective power distribution, thereby greatly improving the service life of the fuel cell.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fuel cells, in particular to a power regulation method, system, device and medium for a vehicle hybrid energy storage device. BACKGROUND

[0002] Under the background of carbon neutrality, the global automobile industry urgently needs to explore new energy fuels to reduce dependence on traditional petrochemical fuels. Hydrogen, as a new energy, has the advantages of being clean, non-polluting, widely available, and high in combustion energy. Fuel cells (FC) have become the main form of hydrogen energy application due to their high energy conversion efficiency and clean operation process, and have received widespread attention. Although fuel cells can be used as the main energy source for fuel cell vehicles, due to the internal electrochemical reaction, their dynamic response is relatively slow. Therefore, fuel cell vehicles usually introduce electric energy storage devices (such as lithium ion batteries, supercapacitors, lithium ion capacitors, etc.) with faster dynamic response characteristics to form a hybrid energy storage device.

[0003] The hybrid energy storage device still faces many problems in the power distribution between the fuel cell and the electric energy storage device. The existing technology generally only considers single aspects such as system voltage stability or FC current slope. However, there are still many aspects that have not been covered and should be considered simultaneously. For example, the State Of Charge (SOC) of the electric energy storage device should be within a reasonable range within an operating cycle and should remain unchanged after an operating cycle. The FC current should be as small as possible to avoid expensive hydrogen consumption. In addition, rapid changes in FC current will shorten the life of the FC, so the current change rate of the FC should be as small as possible. Therefore, it is necessary to provide a power regulation method, system, device and medium for a vehicle hybrid energy storage device. SUMMARY

[0004] In view of the above shortcomings of the prior art, the purpose of the present application is to provide a power regulation method, system, device and medium for a vehicle hybrid energy storage device to improve the problem that the hybrid energy storage device of the vehicle does not consider all factors affecting its current time operation in the prior art, resulting in the inability to reasonably distribute power to the hybrid energy storage device within the control interval.

[0005] To achieve the above object and other related objects, the present application provides a power regulation method for a vehicle hybrid energy storage device, the hybrid energy storage device of a hybrid electric vehicle comprising a fuel cell and an electric energy storage device, the electric energy storage device comprising one or more of a lithium ion battery, a supercapacitor, and a lithium ion capacitor, the method comprising the following processes:

[0006] obtaining the speed and acceleration of the vehicle at the current time, and predicting the load compensation value of the hybrid energy storage device at the future time;

[0007] The state variables of the hybrid energy storage device at the current moment and the current change of the fuel cell given at the previous moment are obtained to obtain the residual compensation value of the hybrid energy storage device at future moments; wherein, the state variables are the current value of the fuel cell and the state of charge of the energy storage device.

[0008] The preset reference value, load compensation value, and residual compensation value are input into the model in the model predictive control algorithm to obtain the change in fuel cell current of the vehicle within the preset control range, and to regulate the power distribution of the hybrid energy storage device within the control range; wherein, the preset reference value is the preset charge state of the energy storage device and the preset current value of the fuel cell.

[0009] In one embodiment of the present invention, obtaining the vehicle's speed and acceleration at the current moment and predicting the load compensation value of the hybrid energy storage device at a future moment includes:

[0010] The vehicle's current speed and acceleration are input into a pre-trained load compensation model to predict the vehicle's speed and acceleration at future times;

[0011] The load compensation value is obtained based on the velocity and acceleration at the future time.

[0012] In one embodiment of the present invention, obtaining the state variables of the hybrid energy storage device at the current moment and the current change of the fuel cell given at the previous moment, and obtaining the residual compensation value of the hybrid energy storage device at a future moment, includes:

[0013] The current state variable and the current change of the fuel cell given in the previous time step are input into the pre-trained residual compensation model to obtain the offline residual.

[0014] Based on the actual and predicted values ​​of the state variables at the current moment, the online residual is obtained, and the online residual is weighted and summed with the offline residual to obtain the residual compensation value of the hybrid energy storage device; wherein, the predicted value is obtained by solving the model in the model predictive control algorithm at the previous moment.

[0015] In one embodiment of the present invention, the predicted value of the state variable is obtained through a state-space equation, which is: x k+1 =Ax k +B u u k +B d d k +C1x ep +C2x ec Where A is the state coefficient matrix, B u For the input coefficient matrix, Bd is a disturbance coefficient matrix, C1 is an offline residual compensation factor, C2 is an online residual compensation factor, x k is a state variable at time k, u k is input data at time k, d k is a load disturbance at time k, x ep is an offline residual at time k, x ec is an online residual at time k, x k+1 is a state variable at time k+1.

[0016] In an embodiment of the present application, the model in the model predictive control algorithm into which the preset reference value and the load compensation value and the residual compensation value are input is used to obtain the current variation of the fuel cell in the preset control interval, and the power distribution of the hybrid energy storage device in the control interval is regulated, comprising:

[0017] The model in the model predictive control algorithm into which the preset reference value and the load compensation value and the residual compensation value are input is used to obtain the current variation of the fuel cell in the control interval according to the minimization of the cost function;

[0018] The current variation of the fuel cell in the control interval is input into a preset proportional-integral controller, the duty cycle of the boost converter in the control interval is predicted, and the state variable of the hybrid energy storage device in the control interval is regulated based on the duty cycle in the control interval, and the power distribution of the hybrid energy storage device is controlled; wherein the boost converter is connected in parallel with the fuel cell and the electrical energy storage device.

[0019] In an embodiment of the present application, before the speed and acceleration of the vehicle at the current time are obtained, a cost function of the hybrid energy storage device is constructed according to the preset constraint conditions of the fuel cell and the electrical energy storage device and the operating requirements of the hybrid energy storage device, and a model in the model predictive control algorithm is obtained.

[0020] In an embodiment of the present application, the load compensation model and the residual compensation model are offline trained Gaussian process regression models.

[0021] In an embodiment of the present application, a power regulation system for a hybrid energy storage device of a vehicle is also provided, the hybrid energy storage device of a hybrid vehicle comprising a fuel cell and an electrical energy storage device, the electrical energy storage device comprising one or more of a lithium ion battery, a supercapacitor, and a lithium ion capacitor, and the system comprising:

[0022] A load compensation acquisition module is configured to acquire the speed and acceleration of the vehicle at the current time, and predict the load compensation value of the hybrid energy storage device at the future time;

[0023] a residual error compensation obtaining module, configured to obtain a state variable of the hybrid energy storage device at a current time and a current variation of the fuel cell at a previous time, to obtain a residual error compensation value of the hybrid energy storage device at the current time, wherein the state variable is a current value of the fuel cell and a state of charge of the electric energy storage device;

[0024] a control module, configured to input a preset reference value, a load compensation value and the residual error compensation value into a model in a model predictive control algorithm, to obtain a current variation of the fuel cell of the vehicle in a preset control interval, and to regulate power distribution of the hybrid energy storage device in the control interval, wherein the preset reference value is a preset state of charge of the electric energy storage device and a preset current value of the fuel cell.

[0025] In an embodiment of the present application, a power control device of a hybrid energy storage device of a vehicle is also provided, comprising a processor and a memory coupled to the processor, and the memory stores program instructions, when the program instructions stored in the memory are executed by the processor, the method described in any of the above embodiments is implemented.

[0026] In an embodiment of the present application, a computer readable storage medium is also provided, comprising a program, when the program is run on a computer, the method described in any of the above embodiments is executed.

[0027] In summary, in the present application, by obtaining the speed and acceleration of the vehicle, the load compensation value can be obtained. And based on the current value of the fuel cell, and the state of charge of the electric energy storage device, and the current variation of the fuel cell, the residual error compensation value is obtained. The residual error compensation value and the load compensation value, and the preset reference value are input into the model in the model predictive control algorithm, to obtain the current variation of the fuel cell of the vehicle in the preset control interval. And the power of the hybrid energy storage device in the control interval can be regulated through the current variation of the fuel cell. The final power distribution is more reasonable, and the service life of the fuel cell is improved. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0029] Figure 1 A semi-active power system topology diagram of a fuel cell hybrid electric vehicle in the present application is shown;

[0030] Figure 2 A schematic diagram of a control scheme in the present application is shown.

[0031] Figure 3 Fig. 1 shows a flow chart of a power regulating method of a vehicle hybrid energy storage device according to an embodiment of the present application;

[0032] Figure 4 Fig. 2 shows a flow chart of step S10 according to an embodiment of the present application;

[0033] Figure 5 Fig. 3 shows a flow chart of step S20 according to an embodiment of the present application;

[0034] Figure 6 Fig. 4 shows a flow chart of step S30 according to an embodiment of the present application;

[0035] Figure 7 Fig. 5 shows a schematic diagram of a power regulating system of a vehicle hybrid energy storage device according to an embodiment of the present application.

[0036] Element No. Explanation

[0037] 100, a power regulating system of a vehicle hybrid energy storage device; 110, a load compensation obtaining module; 120, a residual compensation obtaining module; 130, a power control module. DETAILED DESCRIPTION

[0038] The present application is described herein with reference to specific embodiments thereof which are illustrated in the accompanying drawings. These embodiments are described in detail so that this application will be thorough and complete, and fully convey the scope of the application to those skilled in the art. Other advantages and novel features of the application will become apparent from the following detailed description, from the novel configurations, and from the organized, structured and thorough disclosure, when considered together with the drawings and the associated detailed description. Various embodiments can be more clearly understood by referring to the following description together with the accompanying drawings, in which: the use of the same reference numerals in different figures serves to

[0039] Reference will now be made to Figures 1 to 7It should be understood that the structure, proportion, size and the like shown in the drawings of the specification are only used to cooperate with the disclosed content, to be understood and read by those skilled in the art, and are not used to limit the conditions for implementing the application, so they do not have technical significance. Any modification of the structure, change of the proportion relationship or adjustment of the size, without affecting the effect and purpose that can be achieved by the application, should still fall within the scope of the disclosed technology.

[0040] When the embodiments give a numerical range, it should be understood that, unless otherwise stated by the application, each numerical range and any numerical value between the two endpoints can be selected. Unless otherwise defined, all technical and scientific terms used in the application and the understanding of the prior art by those skilled in the art and the description of the application can also use any method, equipment and material of the prior art similar or equivalent to the method, equipment and material described in the embodiments of the application to realize the application.

[0041] Please refer to Figures 1 to 3 The application provides a power regulation method of a vehicle hybrid energy storage device. A fuel cell is used as the main power supply, directly connected in series with a boost converter to supply power to a load, and an electric energy storage device is directly connected to the load to serve as an energy buffer and provide fast response. The application uses a model in a model predictive control algorithm to predict the power of the hybrid energy storage device. Based on the model in a linear model predictive control (MPC) algorithm considering load disturbance error and residual error, GPRC-MPC (MPC scheme with GP regression compensation) is introduced to predict and compensate the load disturbance error and residual error related to the linear MPC model. Thus, the final power distribution is more reasonable, and the service life of the fuel cell is improved.

[0042] Please refer to Figures 1 to 3 In an embodiment of the application, the power regulation method of the vehicle hybrid energy storage device includes the following processes:

[0043] S10, obtain the speed and acceleration of the vehicle at the current time, and predict the load compensation value of the hybrid energy storage device at the future time.

[0044] During the driving of the vehicle, the speed v(k) and the acceleration a(k) of the vehicle at the current time (i.e., the k time) are obtained, and the speed v(k) and the acceleration a(k) at the current time are input to the trained load disturbance model, the load disturbance model outputs a series of speed prediction values, and based on the speed prediction values, the corresponding load disturbance is calculated, and the load disturbance is taken as the load compensation value. It should be noted that the vehicle described in the present application includes but is not limited to a car, a bus and a truck.

[0045] Please refer to Figure 1 、 Figure 2 and Figure 3 , further, in an embodiment of the present application, in step S10, the speed and acceleration of the vehicle at the current time are obtained, and the load compensation value of the hybrid energy storage device at the future time is predicted, including:

[0046] S11, input the speed and acceleration of the vehicle at the current time to the pre-trained load compensation model, and predict the speed and acceleration of the vehicle at the future time;

[0047] S12, based on the speed and acceleration at the future time, obtain the load compensation value.

[0048] The load compensation model is used to provide load compensation for the hybrid energy storage device, and the current change prediction value of the fuel cell at the future time is predicted to make it more accurate. Specifically, the load compensation model is a Gaussian process regression model trained offline using v(k), a(k) and v(k+1) as training data. If the current time is k time, after inputting the speed v(k) and the acceleration a(k) of the vehicle at the k time to the trained load compensation model, the load compensation model will output the speed prediction value of N-1 time after the k time, and a series of speed prediction values v(k+1), v(k+2),..., v(k+N-1) are obtained. Wherein, N is the prediction step. The sampling time interval of adjacent speed prediction values is T s , according to the adjacent speed prediction value and the sampling time, the acceleration a(k+1), a(k+2),..., a(k+N-1) corresponding to the k+1 time, the k+2 time and the k+N-1 time can be predicted. When the speed prediction is realized by using the method proposed in the present application, generally, the predicted speed curve is very close to the actual speed curve.

[0049] Since the power consumed by the vehicle during driving is regarded as a dynamic load, according to the vehicle dynamics, the instantaneous wheel force of the vehicle is shown in formula (1):

[0050] F wheel (k)=(m+m r )a(k)+mgC r +ρ air Cd A f v(k) 2 / 2 (1)

[0051] wherein F wheel (k) is the instantaneous wheel force at time k, m is the mass of the vehicle, m r is the equivalent mass of the rotating parts, a(k) is the acceleration of the vehicle at time k, g is the acceleration of gravity, C r is the coefficient of rolling resistance, p air is the air density, C d is the coefficient of aerodynamic drag, A f is the front area of the vehicle, and v(k) is the speed of the vehicle at time k. According to the above instantaneous wheel force, the load power is scaled down by k s times, and the load simulation power at time k can be obtained, as shown in equation (2):

[0052]

[0053] wherein P load (k) is the simulation power of the load at time k, k s is the power scaling factor, η dif is the efficiency of the differential in the load, η em is the efficiency of the motor in the load, η inv is the efficiency of the inverter in the load, and sgn is the sign function. It should be noted that the specific value of the power scaling factor can be adaptively selected by those skilled in the art according to actual needs, and is not limited herein. The simulation current of the vehicle at time k is shown in equation (3):

[0054]

[0055] wherein i load (k) is the load current value of the vehicle at time k, v sc (k) is the voltage value of the fuel cell at time k, which can be measured by a voltmeter. It can be understood that, for the convenience of calculation, P load (k) and i load (k) are regarded as the load power and the load current at time k in the calculation and description process of the present application. The load current value i load(k) the load disturbance d(k) at the current time k (i.e. the load compensation value at the current time). According to the above formula (3), the load current values at the future N-1 time points can be predicted, and thus the load disturbances d(k+1), d(k+2),..., d(k+N-1) at the N-1 time points after the time k can be obtained. It should be noted that if the speed prediction is not implemented, only the speed v(k) at the current time k and the load disturbance d(k) can be obtained, and the load disturbances at the future N-1 time points cannot be predicted.

[0056] It can be understood that the present application only takes the current time k as an example to illustrate the calculation process of the load disturbance, and the calculation process of the load disturbance at other time points is similar to that at the time k, which is not described herein. Since the load disturbance at the current time k is the load current value at the time, the load disturbance can be calculated according to the formula (3) and can also be directly measured by an instrument. Specifically, in another embodiment of the present application, the load disturbance is obtained by measuring with an ammeter. The ammeter is connected in series with the load, and thus the obtained load disturbance is more accurate, so that the hybrid energy storage device at the future time can be more accurately controlled.

[0057] S20, obtaining the state variable of the hybrid energy storage device at the current time and the current change amount of the fuel cell at the previous time, and obtaining the residual compensation value of the hybrid energy storage device at the future time; wherein the state variable is the current value of the fuel cell and the state of charge of the electric energy storage device.

[0058] It should be noted that the type of the electric energy storage device includes but is not limited to lithium ion batteries, supercapacitors, lithium ion capacitors and other energy storage devices with better power characteristics than fuel cells. The type of the electric energy storage device can be one or more, which is not limited herein. The present application takes the supercapacitor as an example to describe the related action principle of the supercapacitor, and the action principle of other types of electric energy storage devices is similar to that of the supercapacitor, which is not described herein.

[0059] Please refer to Figure 1 , Figure 2 and Figure 5 , specifically, in an embodiment of the present application, the step S20 includes the following process:

[0060] S21, inputting the state variable at the current time and the current change amount of the fuel cell at the previous time into the pre-trained residual compensation model to obtain an offline residual;

[0061] S22, obtaining an online residual based on the actual value and the predicted value of the state variable at the current time, and obtaining the residual compensation value of the hybrid energy storage device by weighted sum of the online residual and the offline residual; wherein the predicted value is obtained by solving the model in the model predictive control algorithm at the previous time.

[0062] The state variables, i.e. the current value of the fuel cell and the state of charge of the super capacitor at the current time k, and the current variation of the fuel cell given at the previous time k-1, are jointly input into the trained residual compensation model to obtain the offline residual x ep (k). In addition, by solving the model in the model predictive control algorithm at the previous time, the predicted value of the state variable of the model output in the model predictive control algorithm can be obtained. The actual value and the predicted value of the state variable at the current time k are subtracted to obtain the online residual x ec (k) at the current time. Wherein, the actual value of the state variable can be obtained by actually measuring when the hybrid energy storage device is running online, for example, the current value of the fuel cell can be measured by a current sensor, and the state of charge of the super capacitor can be obtained by a SOC estimation algorithm based on Kalman filter. The predicted value of the state variable can be seen from the following scheme. It can be understood that the offline residual includes the offline residual of the fuel cell current value, and the offline residual of the state of charge of the super capacitor, and for the same reason, the online residual includes the online residual of the fuel cell current value, and the online residual of the state of charge of the super capacitor.

[0063] S30, input the preset reference value and the load compensation value and the residual compensation value into the model in the model predictive control algorithm, to obtain the current variation of the fuel cell of the vehicle in the preset control interval, and to regulate the power distribution of the hybrid energy storage device in the control interval; wherein, the preset reference value is the preset state of charge of the electric energy storage device and the preset current value of the fuel cell.

[0064] The model predictive control algorithm is an optimization-based method, which can be applied to a multivariable system, such as a hybrid energy storage device. However, the calculation complexity of the nonlinear MPC is very high, and the accuracy of the linear MPC will be affected. In order to improve the accuracy of the linear MPC, the residual compensation value and the load compensation value are added to the model in the model predictive control algorithm for compensation. The model in the model predictive control algorithm can comprehensively consider the constraints of various spatial state variables, so that the prediction result of the next time is more accurate. The preset reference value is the preset state of charge SOC ref of the super capacitor and the preset current value i fcref of the fuel cell, for example, i fcref can be set to 0, and SOC ref can be set to 50%, it can be understood that the specific value of the preset reference value is not limited here, and those skilled in the art can adaptively set it according to the performance of the fuel cell and the super capacitor.

[0065] Please refer to Figure 1 , Figure 2 and Figure 6Specifically, in an embodiment of the present application, the input of the preset reference value and the load compensation value and the residual compensation value into the model in the model predictive control algorithm, the calculation of the current variation of the fuel cell of the vehicle in the preset control interval, and the regulation of the power distribution of the hybrid energy storage device in the control interval include the following processes:

[0066] S31, input the preset reference value and the load compensation value and the residual compensation value into the model in the model predictive control algorithm, and calculate the current variation of the fuel cell in the control interval according to the minimization of the cost function;

[0067] S32, input the current variation of the fuel cell in the control interval into a preset proportional integral controller, predict the duty cycle of the boost converter in the control interval, and regulate the state variable of the hybrid energy storage device in the control interval based on the duty cycle in the control interval to control the power distribution of the hybrid energy storage device; wherein the boost converter is connected in parallel with the electric energy storage device after being connected in series with the fuel cell.

[0068] Further, in an embodiment of the present application, before obtaining the speed and acceleration of the vehicle at the current time, the method further comprises: constructing a cost function of the hybrid energy storage device according to the preset constraint conditions of the fuel cell and the electric energy storage device and the operation requirement of the hybrid energy storage device to obtain the model in the model predictive control algorithm.

[0069] Specifically, based on the operation constraints of the fuel cell and the supercapacitor, the cost function J of the model predictive control algorithm and the corresponding constraint conditions in the present application are shown in formula (4):

[0070]

[0071] s.t.x k+1 =f(x k ,u k )+g(v k ,a k )+h(x k ,u k )

[0072]

[0073]

[0074] wherein X k is a set of N state variables predicted at k time, U k is a set of N-1 input data predicted at k time, Q is a symmetric positive definite weight coefficient matrix of the state variable, R is a symmetric positive definite weight coefficient matrix of the input data, X refA set of preset reference values for the state variable, χ is the state constraint of the system, U is the input constraint of the system, and N is the prediction step. Specifically, x k|k is the state variable at time k predicted at time k, x k+N|k is the state variable at time k+N predicted at time k predicted at time k, u k|k is the input data at time k predicted at time k, u k+N-1|k is the input data at time k+N-1 predicted at time k predicted at time k, u ref includes a reference to the state variable: X ref =[SOC ref ,0] T , SOC ref is the preset state of charge of the super capacitor. R=r. Wherein, q1, q2, r are preset weight values respectively.

[0075] After the preset reference value and the residual compensation value and the load compensation value at the current time k are input to the model in the model predictive control algorithm, the input data that minimizes the cost function J(X k , U k ) in formula (4) is obtained. According to the following spatial state equation, the predicted value of the state variable in the control interval is calculated. It can be understood that if the prediction step is set to M, the preset reference value and the residual compensation value and the load compensation value at the current time k are input to the model in the model predictive control algorithm, and the model in the model predictive control algorithm will output M fuel cell current changes Δi fc (k+1), Δi fc (k+2),..., Δi fc (k+M), and the current changes of adjacent fuel cells are separated by a fixed sampling period T s . Through the proportional integral (PI) controller, M corresponding duty cycle data D fc (k+1), D fc (k+2),..., D fc (k+M) are obtained to form a duty cycle sequence. The first duty cycle data D fc (k+1) in the sequence is used as an adjustment signal to adjust the duty cycle of the boost converter, so that the current flowing through the fuel cell and the current flowing through the super capacitor change, thereby realizing power regulation of the hybrid energy storage device at future time.

[0076] Further, in the present application, the predicted value of the state variable is obtained through the state space equation, and the model in the model predictive control algorithm can be represented as the space state equation shown in formula (5) in the discrete state space by using Euler discretization:

[0077] x k+1 =f(x k ,u k )+g(v k ,a k )+h(x k ,u k ) (5)

[0078] Wherein, x k+1 is the state variable at k+1 time, f is the nominal part, g is the load compensation, h is the residual compensation, x k is the state variable at k time, u k is the input data at k time, v k is the speed at k time, a k is the acceleration at k time. Further, f, g and h are respectively shown in the following formula (6) to (8):

[0079] f(x k ,u k )=Ax k +B u u k (6)

[0080] g(v k ,a k )=B d d k (7)

[0081] h(x k ,u k )=C1x ep (k)+C2x ec (k) (8)

[0082] Wherein, A is the state coefficient matrix, B u is the input coefficient matrix, B d is the disturbance coefficient matrix, d k is the load disturbance at k time, x ep (k) is the offline residual at k time, x ec (k) is the online residual at k time, C1 is the compensation factor of offline residual, C2 is the compensation factor of online residual. Substituting formula (6) to (8) into formula (5), the state space equation is obtained: x k+1 =Ax k +B u u k +Bd d k +C1x ep +C2x ec It can be understood that C1 and C2 can be adaptively set by those skilled in the art according to actual needs, and no requirement is made herein. For the specific MPC scheme considered in the present application, the state variables, input data and load disturbance at the k moment are respectively represented as shown in formulas (9) to (11):

[0083] x k =[SOC(k), i fc (k)] T (9)

[0084] u k =Δi fc (k) (10)

[0085] d k =i load (k) (11)

[0086] wherein SOC(k) is the charge state of the super capacitor at the k moment, i fc (k) is the current of the fuel cell at the k moment, Δi fc (k) is the current change amount of the fuel cell at the k moment, i load (k) is the load current at the k moment. Further, each of the above coefficient matrices is respectively represented as shown in formulas (12) to (14):

[0087]

[0088]

[0089]

[0090] wherein η boost is the efficiency of the boost converter, v fc (k) is the voltage value of the fuel cell at the k moment, T s is the preset sampling period, C sc is the capacitance value of the super capacitor, v sc (k) is the output voltage of the super capacitor at the k moment, v max is the maximum working voltage of the fuel cell. By substituting formulas (6) to (14) into formula (5), the predicted value of the state variable at the future moment can be calculated. It should be noted that the state variable x k+1 in the present application includes the predicted current value of the fuel cell at the k+1 moment and the predicted charge state of the super capacitor, and the corresponding x k is the current value of the fuel cell at the k moment and the charge state of the super capacitor.

[0091] To avoid high computational costs, in one embodiment of the present invention, the load compensation model and the residual compensation model are Gaussian process regression models trained offline. The Gaussian process regression model is constructed as follows: Assume there is a set of output data y that conforms to a multivariate normal distribution: y 1:n+1 ~N(μ, K), where μ∈R n+1 and K∈R n+1，n+1 Element K i，j By kernel function K(t) i , t j (Given). By marginalizing y 1:n Obtain the unknown output y from a multivariate normal distribution n+1 The prediction. Since the normal distribution is true for any input t n+1 Predict y n+1 Therefore, this process produces the distribution function y. n+1 =gt n+1 It is a Gaussian process, and the model constructed from it is the Gaussian process regression model. Furthermore, for a variance of... For data with a size matrix of L, the kernel function used in this invention is shown in formula (15):

[0092]

[0093] Furthermore, according to the law of power conservation, the load power is as shown in formula (16):

[0094] P load =η boost P fc +P sc (16)

[0095] Among them, P fc For the power of the fuel cell, P sc P represents the power of the supercapacitor. load This represents the load power. Therefore, by changing the current in the fuel cell and the supercapacitor, the overall power distribution of the hybrid energy storage device can be altered.

[0096] Furthermore, during simulation, it is necessary to construct fuel cell models, supercapacitor models, and boost converter models. For the fuel cell model, since the voltage of a single fuel cell is relatively low and varies with the load current, in practical applications, multiple fuel cells are usually connected in series to form a stack to obtain a certain output voltage. The output voltage of the fuel cell at time k is shown in formula (17):

[0097] v fc (k)=V ofc -R ofc i fc(k) (17)

[0098] where V ofc is the open circuit voltage of the fuel cell, R ofc is the equivalent resistance of the fuel cell, v fc (k) is the output voltage of the fuel cell at continuous time domain k, i fc (k) is the current of the fuel cell at continuous time domain k. Obviously, if the current i fc (k) of the fuel cell at k moment increases, the output voltage of the corresponding fuel cell will decrease accordingly, thus affecting the power of the fuel cell at k moment.

[0099] For the super capacitor model, since the super capacitor is characterized by the rate of change of the state of charge (SOC), which is related to the output current of the super capacitor, as shown in equation (18):

[0100]

[0101] where C sc is the capacitance of the super capacitor, v max is the maximum operating voltage of the fuel cell, i sc (k) is the output current of the super capacitor at k moment.

[0102] Please refer to Figure 1 For the boost converter model, since the boost converter is used to bridge the voltage difference between the fuel cell and the super capacitor, the output current of the boost converter is as shown in equation (19):

[0103]

[0104] where v sc (k) is the output voltage of the super capacitor at k moment, i dc (k) is the output current of the boost converter at k moment, η boost is the efficiency of the boost converter. Obviously, as shown in equation (20), the load current i load (k) at k moment is composed of the output current i dc (k) of the boost converter at k moment, and the output current i sc (k) of the super capacitor at k moment:

[0105] i load (k) = i sc (k) + i dc (k) (20)

[0106] Therefore, substituting equations (19) and (20) into equation (18), equation (18) can be rewritten as shown in equation (21):

[0107]

[0108] From the following table, compared with the speed MSE of 0.1142 in the original MPC scheme, the speed MSE of the MPC with load prediction and the GPRC-MPC of the application is significantly reduced to 0.0287, which can more accurately realize speed prediction. In terms of hydrogen consumption, compared with the previous two methods, the hydrogen consumption of the application is significantly reduced. During vehicle driving, the change amount of the state of charge of the super capacitor is significantly reduced, which is reduced by 81.15% compared with the original MPC model. In addition, compared with the MPC with load prediction (the maximum current change rate of the fuel cell is 20.6662 A / s), the maximum current change rate of the fuel cell of the application is as low as 18.1322 A / s. Therefore, the GPRC-MPC scheme proposed in the application is superior to the previous two schemes in reducing hydrogen consumption, relieving fuel cell degradation, and maintaining the change amount of the state of charge of the super capacitor.

[0109] Table 1 Comparison of three power energy management strategies

[0110]

[0111] In summary, the power regulation method of the vehicle hybrid energy storage device in the application is as follows: the speed v(k) and acceleration a(k) of the vehicle at the current time k are obtained, as well as the state variables x k of the vehicle at the current time k, i.e. the state of charge SOC(k) of the super capacitor and the current value I fc (k) of the fuel cell. The speed v(k) and acceleration a(k) at time k are input into the pre-trained load compensation model to obtain the load compensation value at the future time. The state variables at the current time k and the current change amount of the fuel cell given at the previous time are input into the residual compensation model to obtain the offline residual x ep (k) at the future time. When the entire system is running online, the duty cycle of the boost converter in the control interval is predicted, and the entire hybrid energy storage device is regulated to obtain the actual value of the state variable at the current time. The actual value of the state variable at the current time and the state variable predicted value predicted by the above model predictive control algorithm at the previous time are subtracted to obtain the online residual x ec (k) at the future time. The online residual and the offline residual are weighted and summed to obtain the residual compensation value. The preset state of charge SOC ref of the super capacitor and the preset current value i fcref of the fuel cell, the load compensation value d kThe residual compensation value is input into the MPC model together with the load compensation value to predict the current variation of the fuel cell in the control interval and the predicted value of the state variable in the control interval. The obtained current variation of the fuel cell is input into the PI controller to obtain the duty ratio of the boost converter. The duty ratio is used to control the boost converter to change the current value of the fuel cell and the state of charge of the super capacitor at the future time, thereby realizing the power regulation of the hybrid energy storage device.

[0112] The step division of the above method is only for the purpose of clear description. In actual implementation, the steps can be combined into one step or split into multiple steps. As long as the same logical relationship is included, it is within the protection scope of the present application. Irrelevant modifications or irrelevant designs can be added to the algorithm or the flow, but the core design of the algorithm and the flow is within the protection scope of the present application.

[0113] Please refer to Figure 7 . The power regulation system 100 of the vehicle hybrid energy storage device includes a load compensation obtaining module 110, a residual compensation obtaining module 120, and a power control module 130. The load compensation obtaining module 110 is configured to obtain the speed and acceleration of the vehicle at the current time, and predict the load compensation value of the hybrid energy storage device at the future time. The residual compensation obtaining module 120 is configured to obtain the state variable of the hybrid energy storage device at the current time and the current variation of the fuel cell at the previous time, and obtain the residual compensation value of the hybrid energy storage device at the future time. The state variable is the current value of the fuel cell and the state of charge of the electric energy storage device. The power control module 130 is configured to input the preset reference value, the load compensation value, and the residual compensation value into the model in the model predictive control algorithm to obtain the current variation of the fuel cell of the vehicle in the preset control interval, and regulate the power distribution of the hybrid energy storage device in the control interval. The preset reference value is the preset state of charge of the electric energy storage device and the preset current value of the fuel cell.

[0114] It should be noted that, in order to highlight the innovative part of the present application, modules not closely related to solving the technical problems proposed in the present application are not introduced in the present embodiment, but this does not mean that there are no other modules in the present embodiment.

[0115] In addition, those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the system described above can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here. In the embodiments provided by the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, and the division of the modules is only a logical function division, and there can be another division in actual implementation, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the modules shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0116] The modules described as separate components can or can not be physically separate, and the components shown as modules can or can not be physical modules, i.e. can be located in one place, or can be distributed to multiple network modules. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0117] In addition, each functional module in each embodiment of the present application can be integrated into a processing module, or each module can exist physically independently, or two or more modules can be integrated into one module. The integrated module can be realized in the form of hardware or in the form of a software functional unit.

[0118] The embodiment further provides a power control device of a vehicle hybrid energy storage device, which comprises a processor and a memory coupled with the processor, and the memory stores program instructions which, when executed by the processor, implement the above task management method. The processor can be a general processor, including a central processing unit (CPU), a network processor (NP), etc.; or a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component; and the memory can include a random access memory (RAM) and can also include a non-volatile memory (Non-Volatile Memory), such as at least one disk memory. The memory can be an internal memory of a random access memory (RAM) type, and the processor and the memory can be integrated into one or more independent circuits or hardware, such as an application specific integrated circuit (ASIC). It should be noted that the computer program in the above memory can be implemented in the form of a software functional unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application.

[0119] The embodiment also provides a computer readable storage medium storing computer instructions for causing a computer to execute the power regulation method of the vehicle hybrid energy storage device. The storage medium can be an electronic medium, a magnetic medium, an optical medium, an electromagnetic medium, an infrared medium, or a semiconductor system or a propagation medium. The storage medium can also include a semiconductor or solid-state memory, a magnetic tape, a removable computer disk, a random access memory (RAM), a read-only memory (ROM), a hard disk, and an optical disk. The optical disk can include a compact disk-read only memory (CD-ROM), a compact disk-read / write (CD-RW), and a DVD.

[0120] In summary, in the present application, the target of the power management strategy is to reduce hydrogen consumption, maintain the state of charge of the electrical energy storage device stable, and alleviate the degradation of the fuel cell. These targets are translated into the minimization of three variables in a multi-variable optimization framework: the fuel cell current, the state of charge fluctuation of the electrical energy storage device, and the fuel cell current rate of change. The present application starts from this target, and by constructing the GPRC-MPC scheme, the linear accuracy of the model in the final model predictive control algorithm is effectively improved, and effective power distribution can be achieved, thereby greatly improving the service life of the fuel cell and keeping the state of charge of the electrical energy storage device within a reasonable range. Based on the semi-active topology, a Gaussian process regression compensation is introduced on the basis of the model predictive control algorithm. The present application has better performance than the traditional model predictive control algorithm in terms of hydrogen consumption, state of charge stability of the electrical energy storage device, and alleviating the degradation of the fuel cell. Therefore, the present application effectively overcomes some practical problems in the prior art and has high utilization value and use significance.

[0121] The above embodiments only exemplarily illustrate the principles and effects of the present application, and are not used to limit the present application. Any person skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by those skilled in the art without departing from the spirit and technical thought disclosed by the present application should be covered by the claims of the present application.

Claims

1. A power regulation method for a vehicle hybrid energy storage device, characterized in that, The hybrid energy storage device of a hybrid vehicle includes a fuel cell and an electrical energy storage device, wherein the electrical energy storage device includes one or more of a lithium-ion battery, a supercapacitor, and a lithium-ion capacitor, and the method includes the following processes: The vehicle's speed and acceleration at the current moment are obtained, and the load compensation value of the hybrid energy storage device at future moments is predicted. The state variables of the hybrid energy storage device at the current moment and the current change of the fuel cell given at the previous moment are obtained to obtain the residual compensation value of the hybrid energy storage device at future moments; wherein, the state variables are the current value of the fuel cell and the state of charge of the energy storage device. The preset reference value, load compensation value, and residual compensation value are input into the model in the model predictive control algorithm to obtain the change in fuel cell current of the vehicle within the preset control range, and to regulate the power distribution of the hybrid energy storage device within the control range; wherein, the preset reference value is the preset charge state of the energy storage device and the preset current value of the fuel cell. The step of obtaining the vehicle's current speed and acceleration, and predicting the load compensation value of the hybrid energy storage device at future times, includes: The vehicle's current speed and acceleration are input into a pre-trained load compensation model to predict the vehicle's speed and acceleration at future times; Based on the velocity and acceleration at the future moment, the load compensation value is obtained; The step of obtaining the state variables of the hybrid energy storage device at the current moment and the current change of the fuel cell given at the previous moment, and obtaining the residual compensation value of the hybrid energy storage device at future moments, includes: The current state variable and the current change of the fuel cell given in the previous time step are input into the pre-trained residual compensation model to obtain the offline residual. Based on the actual and predicted values ​​of the state variables at the current moment, the online residual is obtained, and the online residual is weighted and summed with the offline residual to obtain the residual compensation value of the hybrid energy storage device; wherein, the predicted value is obtained by solving the model in the model predictive control algorithm at the previous moment.

2. The power regulation method for a vehicle hybrid energy storage device according to claim 1, characterized in that, The predicted values ​​of the state variables are obtained through a state-space equation, which is: x k+1 =Ax k +B u u k +B d d k +C1x ep +C2x ec Where A is the state coefficient matrix, B u For the input coefficient matrix, B d Let C1 be the perturbation coefficient matrix, C2 be the offline residual compensation factor, and C3 be the online residual compensation factor. k Let u be the state variable at time k. k Input data at time k, d k Let x be the load disturbance at time k. ep Let x be the offline residual at time k. ec Let x be the online residual at time k. k+1 Let k+1 be the state variable.

3. The power regulation method for a vehicle hybrid energy storage device according to claim 1, characterized in that, The step of inputting preset reference values, load compensation values, and residual compensation values ​​into the model in the model predictive control algorithm to calculate the current change of the fuel cell in the vehicle within a preset control range, and regulating the power distribution of the hybrid energy storage device within the control range, includes: The preset reference value, load compensation value, and residual compensation value are input into the model in the model predictive control algorithm. Based on the cost function minimization, the change in current of the fuel cell within the control interval is obtained. The current change of the fuel cell within the control range is input to a preset proportional-integral controller to predict the duty cycle of the boost converter within the control range. Based on the duty cycle within the control range, the state variables of the hybrid energy storage device within the control range are adjusted to control the power distribution of the hybrid energy storage device. The boost converter is connected in series with the fuel cell and then connected in parallel with the energy storage device.

4. The power regulation method for a vehicle hybrid energy storage device according to claim 1, characterized in that, Before obtaining the vehicle's speed and acceleration at the current moment, the method further includes: constructing a cost function for the hybrid energy storage device based on the preset constraints of the fuel cell and the energy storage device and the operating requirements of the hybrid energy storage device, thereby obtaining the model in the model predictive control algorithm.

5. The power regulation method for a vehicle hybrid energy storage device according to claim 1, characterized in that, The load compensation model and the residual compensation model are Gaussian process regression models trained offline.

6. A power regulation system for a vehicle hybrid energy storage device, characterized in that, The hybrid energy storage device of a hybrid vehicle includes a fuel cell and an electrical energy storage device, wherein the electrical energy storage device includes one or more of a lithium-ion battery, a supercapacitor, and a lithium-ion capacitor, and the system includes: The load compensation acquisition module is used to acquire the speed and acceleration of the vehicle at the current moment and predict the load compensation value of the hybrid energy storage device at future moments. The residual compensation acquisition module is used to acquire the state variables of the hybrid energy storage device at the current moment and the current change of the fuel cell given at the previous moment, and obtain the residual compensation value of the hybrid energy storage device at the current moment; wherein, the state variables are the current value of the fuel cell and the state of charge of the energy storage device. The power control module is used to input preset reference values, load compensation values, and residual compensation values ​​into the model in the model predictive control algorithm to obtain the current change of the fuel cell in the vehicle within a preset control range, and to regulate the power distribution of the hybrid energy storage device within the control range; wherein, the preset reference values ​​are the preset charge state of the energy storage device and the preset current value of the fuel cell. The step of obtaining the vehicle's current speed and acceleration, and predicting the load compensation value of the hybrid energy storage device at future times, includes: The vehicle's current speed and acceleration are input into a pre-trained load compensation model to predict the vehicle's speed and acceleration at future times; Based on the velocity and acceleration at the future moment, the load compensation value is obtained; The step of obtaining the state variables of the hybrid energy storage device at the current moment and the current change of the fuel cell given at the previous moment, and obtaining the residual compensation value of the hybrid energy storage device at future moments, includes: The current state variable and the current change of the fuel cell given in the previous time step are input into the pre-trained residual compensation model to obtain the offline residual. Based on the actual and predicted values ​​of the state variables at the current moment, the online residual is obtained, and the online residual is weighted and summed with the offline residual to obtain the residual compensation value of the hybrid energy storage device; wherein, the predicted value is obtained by solving the model in the model predictive control algorithm at the previous moment.

7. A power control device for a vehicle hybrid energy storage device, characterized in that: The method includes a processor coupled to a memory storing program instructions, which, when executed by the processor, implement the method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: Includes a program that, when run on a computer, performs the method as described in any one of claims 1 to 5.

Images