Multi-objective optimization dual fuzzy control method for energy of fuel cell hybrid electric vehicle

Optimizing the energy management of fuel cell hybrid vehicles through dual fuzzy control and non-dominant sorting genetic algorithms, solving the complexity and life problems of energy management in the existing technology, and achieving dual optimization of fuel economy and life.

CN118386950BActive Publication Date: 2024-11-19GUANGXI UNIVERSITY OF TECHNOLOGY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410404758.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-03
Publication Date
2024-11-19
Estimated Expiration
2044-04-03

AI Technical Summary

Technical Problem

The existing fuel cell hybrid vehicle energy management strategies are difficult to optimize fuel economy and energy source life simultaneously, and the single fuzzy control strategy is designed in a complex manner and depends on experience, so it cannot adapt to complex driving conditions.

Method used

Using the dual fuzzy control method, the dual fuzzy controller is optimized to achieve multi-objective optimization by establishing equivalent energy consumption models and performance decay models of fuel cells and power cells, combined with improved non-dominant sorting genetic algorithms.

Benefits of technology

Real-time optimization and control of fuel cell hybrid vehicles is realized, the fuel economy and energy source life of the entire vehicle are improved, the fuel economy and energy source life are adapted to different vehicle operating scenarios, and the fuzzy rule design is simplified.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118386950B_ABST
    Figure CN118386950B_ABST
Patent Text Reader

Abstract

The present invention discloses a multi-objective optimized dual fuzzy control method for energy of a fuel cell hybrid electric vehicle, and relates to the technical field of energy management of new energy vehicles. The present invention comprises the following steps: S1, establishing an energy management model of a fuel cell hybrid electric vehicle based on dual fuzzy control, a main fuzzy controller with the output power of the fuel cell as the output, and a sub-fuzzy controller with the output power of the power battery as the output; S2, establishing a multi-objective function for optimization calculation, including an equivalent energy consumption model based on three energy sources and an equivalent performance degradation model based on two energy sources; S3, using an improved non-dominated sorting genetic algorithm, and optimizing the dual fuzzy controller of step S1 based on the Pareto multi-objective optimization method. The present invention can simplify the complexity of fuzzy control strategy design by formulating a dual fuzzy controller for energy management of a three-energy source fuel cell hybrid electric vehicle with multi-objective optimization.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of energy management of new energy vehicles, and in particular, relates to a multi-objective optimized dual fuzzy control method for energy of a fuel cell hybrid electric vehicle. Background Art

[0002] As a clean and efficient energy source, hydrogen fuel cells have great development potential. As a unidirectional energy flow system, they are usually paired with power batteries (lithium batteries) and / or supercapacitors to form a fuel cell vehicle hybrid power system (FCHPS). Formulating an energy management strategy (EMS) for hybrid vehicles is one of the research focuses. Considering the three aspects of vehicle performance, fuel economy and power system cost, the three-energy source FCHPS has higher efficiency. However, due to the multi-energy source structure of FCHPS, the flow of electric energy between the energy sources becomes complicated, and it is also related to energy consumption and energy source performance degradation (economy and energy source durability). It is a multi-objective system, and the optimized objectives are usually mutually exclusive, that is, improving the optimization effect of one objective may reduce the optimization effect of other objectives. Therefore, its energy management is more complicated.

[0003] There are two main types of EMS for fuel cell vehicles: rule-based and optimization-based. Rule-based EMS is further divided into logic-based and fuzzy-based. Conventional logic-based EMS controls the output power of energy sources according to their state, with strong real-time control and simple implementation, but the control effect is not optimized and cannot achieve better results. In addition to the advantages of conventional logic-based EMS, fuzzy-based EMS has a good control effect on complex and multi-variable systems, so it is widely used in automotive energy management. However, its control effect usually depends on the experience of designers. In existing research, the main results are as follows: a single fuzzy control strategy with global optimization is used as the energy management strategy, with equivalent fuel economy as the optimization target, and particle swarm optimization, genetic algorithm and other methods are used to optimize its control parameters and membership functions to improve the fuel economy of the whole vehicle. However, the existing strategy still has many shortcomings that need to be solved:

[0004] 1. For complex FCHPS, designing a single fuzzy control strategy requires designing too many variables and fuzzy rules, and the formulation of fuzzy control strategies is often affected by the designer's own subjective factors. Especially when the fuzzy rule design is not appropriate or perfect, it is easy to seriously affect its performance;

[0005] 2. Currently, more and more people use fuel economy as a single optimization target, ignoring the performance degradation of the energy source, which may lead to increased damage to the energy source and shorten its lifespan;

[0006] 3. Usually the optimization is carried out offline. Even if multi-objective optimization is considered, it is mainly a simple weighted method used in each objective function. This method is not easy to determine the appropriate weight coefficient. Secondly, it is equivalent to only considering specific situations and is difficult to adapt to complex driving conditions of the car. Summary of the invention

[0007] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a multi-objective optimized dual fuzzy control method for energy of a fuel cell hybrid electric vehicle.

[0008] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0009] A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method comprises the following steps:

[0010] S1. Establish a fuel cell hybrid electric vehicle energy management model based on dual fuzzy control, with the main fuzzy controller taking the fuel cell output power as the output and the sub-fuzzy controller taking the power battery output power as the output;

[0011] S2. Establish a multi-objective function for optimization calculation, and establish an equivalent energy consumption model based on three energy sources according to the consumption of hydrogen by fuel cells and the consumption of electricity by power batteries and supercapacitors; consider the performance degradation of fuel cells and lithium batteries, and establish an equivalent performance degradation model based on two energy sources;

[0012] S3. Using an improved non-dominated sorting genetic algorithm, the dual fuzzy controller in step S1 is optimized to establish a fuel cell hybrid electric vehicle energy management model based on dual fuzzy control of Pareto multi-objective optimization.

[0013] Optionally, the step S1 specifically includes the following sub-steps:

[0014] S11. Establish the main fuzzy controller for energy management of fuel cell hybrid electric vehicles;

[0015] S12. Establish a sub-fuzzy controller for energy management of fuel cell hybrid electric vehicles.

[0016] Optionally, the step S11 specifically includes the following sub-steps:

[0017] S111, setting the input variables and output variables of the main fuzzy controller, setting the input variables of the main fuzzy controller to the required power of the whole vehicle and the state of charge of the lithium battery, and the output variable to the output power of the fuel cell;

[0018] S112. The input variables and output variables of the main fuzzy controller are fuzzified. The fuzzy subset of the vehicle's required power is {zero, extremely small, very small, small, small, medium, large, large, very large}, expressed as {Z, PES, PVS, PCS, PS, PM, PL, PCL, PVL}; the fuzzy subset of the lithium battery's state of charge is {very low, low, medium, high, very high}, expressed as {PVS, PS, M, PL, PVL}; the fuzzy subset of the fuel cell's output power is {zero, very small, very small, small, small, medium, large, large, very large}, expressed as {Z, PES, PVS, PCS, PS, PM, PL, PCL, PVL, PEL};

[0019] S113, setting fuzzy rules of the main fuzzy controller and performing fuzzy reasoning;

[0020] S114, defuzzifying the output variable of the main fuzzy controller to obtain the power that should be output by the fuel cell.

[0021] Optionally, the step S12 specifically includes the following sub-steps:

[0022] S121, setting the input variables and output variables of the sub-fuzzy controller, setting the input variables of the sub-fuzzy controller to the state of charge of the lithium battery, the state of charge of the supercapacitor, and the remaining power, wherein the remaining power is the difference between the required power of the whole vehicle and the output power of the fuel cell, and the output variable is the output power of the lithium battery;

[0023] S122, the input variables and output variables of the auxiliary fuzzy controller are fuzzified. The fuzzy subset of the state of charge of the lithium battery is the same as that of the main fuzzy controller. The fuzzy subset of the state of charge of the supercapacitor is {very low, low, medium, high}, expressed as {PVS, PS, PM, PL}; the fuzzy subset of the residual power is {negative large, negative small, zero, small, medium, large, very large}, expressed as {NL, NS, Z, PS, PM, PL, PVL}; the fuzzy subset of the output power of the lithium battery is {negative large, negative small, very small negative, zero, small, medium, large, very large}, expressed as {NL, NS, NVS, Z, PS, PM, PL, PVL};

[0024] S123, setting fuzzy rules of the sub-fuzzy controller and performing fuzzy reasoning;

[0025] S124, defuzzifying the output variable of the sub-fuzzy controller to obtain the power that should be output by the power lithium battery, and subtracting the power that should be output by the power lithium battery from the residual power to obtain the power that should be output by the supercapacitor.

[0026] Optionally, in step S112 and step S122, the membership functions of the designed input variables and output variables are composed of half-trapezoidal membership functions and triangular membership functions.

[0027] Optionally, the equivalent energy consumption model based on three energy sources in step S2 is specifically:

[0028]

[0029] Among them, J H2 is the lower calorific value of hydrogen; P fc , P bat , P cap are the output power of fuel cell, power battery and supercapacitor respectively. They are the working efficiency of fuel cells, lithium batteries, and supercapacitors, These are the working efficiencies of the DC / DC converters connected to lithium batteries and supercapacitors, respectively.

[0030] Optionally, the equivalent performance degradation model based on two energy sources in step S2 is specifically:

[0031] minf2(x)=f fc (x)+f bat (x)

[0032]

[0033]

[0034] Among them, f fc (x) is the equivalent performance degradation of the fuel cell, f bat (x) is the equivalent performance degradation of the power battery, k is the acceleration coefficient of the fuel cell life degradation, ΔV is the allowable voltage drop within the fuel cell life range, n1 is the number of fuel cell start-stop times, n2 is the number of fuel cell power changes, t1 is the idle time, t2 is the high-power operation time, V1, V2, U1, and U2 represent the performance degradation values ​​of the fuel cell start-stop, power change, idle, and high-power operation, respectively, n bat Indicates the number of charge and discharge cycles of lithium batteries at different discharge depths, LC bat Indicates the maximum allowable number of charge and discharge cycles of lithium batteries at different discharge depths.

[0035] Optionally, the step S3 specifically includes the following sub-steps:

[0036] S31. Establish a simulation experimental platform, build and run a fuel cell hybrid vehicle dynamics model, energy management strategy based on dual fuzzy control, optimization calculation, and vehicle standard test conditions;

[0037] S32, setting the parameters required to be optimized for the membership functions of the input variables and output variables of the main fuzzy controller and the sub-fuzzy controller to form an initial population;

[0038] S33, setting initial operation parameters of the improved non-dominated sorting genetic algorithm, wherein the initial operation parameters include the number of initial populations, the maximum number of iterations, the mutation probability and the crossover probability.

[0039] Optionally, according to the steps S31-33, the following operation steps are performed:

[0040] S34, generating the first generation subgroup as the first generation new population;

[0041] S35, determine whether to establish a new parent population, if established, proceed to the next step, otherwise, use the equivalent energy consumption model based on three energy sources and the equivalent performance degradation model based on two energy sources in step S2 as the fitness function of the non-dominated sorting genetic algorithm, calculate the fitness value, determine the dominance and non-domination relationship of individuals in the population, calculate the crowding degree of individuals, and repeat the operation until a new parent population is generated;

[0042] S36, performing selection, crossover, and mutation operations to generate a new offspring population;

[0043] S37, judging whether the set maximum number of iterations has been reached, if so, all non-dominated individuals are stored in the non-dominated set, and a Pareto non-inferior solution set is obtained, otherwise, the parent population and the child population are merged into a new population, and step S35 is re-executed.

[0044] Optionally, three groups of solutions are extracted from the Pareto non-inferior solution set, which can respectively adapt to three vehicle operation scenarios, namely, fuel cell storage of less energy, fuel cell storage of moderate energy, and fuel cell storage of more energy, to form a dual fuzzy control table for energy management of each vehicle, and perform energy distribution in real time by real-time table lookup while the vehicle is driving.

[0045] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art. Of course, any product implementing the present invention does not necessarily need to achieve all the advantages described below at the same time:

[0046] By developing a dual fuzzy controller for energy management for three-energy-source fuel cell hybrid vehicles, the problems of single fuzzy control strategies requiring the design of too many variables and fuzzy rules, and the difficulty in designing appropriate fuzzy rules, are avoided; an improved non-dominated sorting genetic algorithm is used to optimize the dual fuzzy controller with the equivalent energy consumption and equivalent energy performance degradation of three-energy-source fuel cell hybrid vehicles as optimization targets, thus achieving real-time optimization control and comprehensively improving the fuel economy of the whole vehicle and the life of the energy source; a Pareto non-inferior solution set is obtained through optimization calculation, and the corresponding optimized dual fuzzy controller for vehicle energy management can be obtained, which can adapt to three vehicle operation scenarios, namely, fuel cell with less energy storage, fuel cell with moderate energy storage, and fuel cell with more energy storage, thus expanding the application scenarios of energy management strategy optimization.

[0047] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings described below are only some embodiments. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0049] Figure 1 A topological structure diagram of a fuel cell hybrid vehicle according to an embodiment of the present invention;

[0050] Figure 2 It is an overall framework diagram of an energy management strategy based on Pareto multi-objective optimization and dual fuzzy control according to an embodiment of the present invention;

[0051] Figure 3 A flow chart of a multi-objective optimized fuel cell hybrid electric vehicle energy dual fuzzy control method according to an embodiment of the present invention;

[0052] Figure 4 This is a framework diagram of energy dual fuzzy control of a fuel cell hybrid electric vehicle according to an embodiment of the present invention;

[0053] Figure 5 The input and output variable membership function diagram of the main fuzzy controller of an embodiment of the present invention, wherein (a) is the required power P req The membership function of the lithium battery SOC b The membership function of the fuel cell is shown in Figure (c). fc The membership function of

[0054] Figure 6 FIG. 1 is a graph of the membership function of the input and output variables of the sub-fuzzy controller according to an embodiment of the present invention, wherein FIG. (a) is the residual power P w The membership function of the lithium battery SOC b The membership function of the supercapacitor SOC is shown in Figure (c). c The membership function of the lithium battery is shown in Figure (d). bat The membership function of

[0055] Figure 7 A Pareto multi-objective optimization flow chart of a fuzzy control membership function based on an improved non-dominated sorting genetic algorithm according to an embodiment of the present invention;

[0056] Figure 8The figure is a result diagram of optimizing the objective function based on the improved non-dominated sorting genetic algorithm according to one embodiment of the present invention, wherein figure (a) is a Pareto non-inferior solution set obtained after 200 iterations, figure (b) is a partition diagram, and figure (c) is nine representative non-inferior solution data selected from the Pareto non-inferior solution set;

[0057] Fig. 9 The optimized main fuzzy controller input and output variable membership function diagram of one embodiment of the present invention, wherein (a) is the required power P req The membership function of the lithium battery SOC b The membership function of the fuel cell is shown in Figure (c). fc The membership function of

[0058] Fig.10 A comparison diagram of hydrogen consumption of a fuel cell under a control strategy of an embodiment of the present invention and a related control strategy;

[0059] Fig.11 This is a diagram showing the power output results of each energy source under the optimized fuel cell hybrid vehicle energy management strategy of an embodiment of the present invention, wherein (a) shows the fuel cell output before and after the dual fuzzy control strategy is optimized, and (b) shows the power output of each energy source after the control strategy is optimized.

[0060] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0061] The present invention will now be described in further detail with reference to the accompanying drawings.

[0062] See also Figure 1-11 As shown, in this embodiment, a multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method is provided, comprising the following steps:

[0063] S1. Establish a fuel cell hybrid electric vehicle energy management model based on dual fuzzy control. The main fuzzy controller uses the output power of the fuel cell as the output, and the sub-fuzzy controller uses the output power of the power battery (lithium battery) as the output. Two simple fuzzy controllers are used to replace the conventional complex single fuzzy controller. The control logic becomes simple and clear, which is conducive to the formulation of fuzzy rules, such as Figure 4 As shown;

[0064] S2. Establish a multi-objective function for optimization calculation. According to the consumption of hydrogen by fuel cells and the consumption of electricity by power batteries and supercapacitors, establish an equivalent energy consumption model based on three energy sources; considering the performance degradation of fuel cells and lithium batteries, establish an equivalent performance degradation model based on two energy sources. In order to make the energy management strategy of fuel cell hybrid vehicles achieve "optimization", "economy" and "durability", it is necessary to establish an objective function for optimization calculation;

[0065] S3. An improved non-dominated sorting genetic algorithm (NSGA-Ⅱ) is used to optimize the dual fuzzy controller in step S1, and a fuel cell hybrid electric vehicle energy management model of dual fuzzy control based on Pareto multi-objective optimization is established. For example, the dual fuzzy controller in step S1, whether it is fuzzy rules or parameters of membership functions, is designed based on human experience, and its effect is not optimized. Therefore, with respect to the parameters of its membership function, the objective function of step S2 is optimized and calculated to obtain an optimized dual fuzzy controller, so that the energy management strategy of the fuel cell hybrid electric vehicle meets the requirements.

[0066] Step S1 of this embodiment specifically includes the following sub-steps:

[0067] S11. Establish the main fuzzy controller for energy management of fuel cell hybrid electric vehicles;

[0068] S12. Establish a sub-fuzzy controller for energy management of fuel cell hybrid electric vehicles.

[0069] Step S11 of this embodiment specifically includes the following sub-steps:

[0070] S111, set the input variables and output variables of the main fuzzy controller, such as Figure 4 As shown in the figure, the input variable of the main fuzzy controller is set to the vehicle required power (P req ) and lithium battery state of charge (SOC b ), where the output variable is the fuel cell output power (P fc ), according to the hybrid electric vehicle model, the range of input and output variables is:

[0071]

[0072] S112. The input variables and output variables of the main fuzzy controller are fuzzified. The fuzzy subset of the vehicle's required power is {zero, very small, very small, small, small, medium, large, large, very large}, expressed as {Z, PES, PVS, PCS, PS, PM, PL, PCL, PVL}, that is, there are 9 fuzzy variables; the state of charge of the lithium battery (SOC b) is {very low, low, medium, high, very high}, expressed as {PVS, PS, M, PL, PVL}; the fuzzy subset of fuel cell output power is {zero, very small, very small, small, small, medium, large, large, very large}, expressed as {Z, PES, PVS, PCS, PS, PM, PL, PCL, PVL, PEL};

[0073] S113, setting fuzzy rules of the main fuzzy controller and performing fuzzy reasoning. The fuzzy rules of the main fuzzy controller are set based on the operation history data of the fuel cell hybrid vehicle. There are 45 fuzzy rules in total, as shown in the following table:

[0074] Table 1 Main fuzzy control rules table

[0075]

[0076] The fuzzy reasoning is performed using the "IF-THEN" language expression, and the fuzzy rules are specifically expressed as:

[0077] R1:if P req =Z and SOC b =PVS, then P fc =PS

[0078] R2:if P req =PES and SOC b =PVS, then P fc =PM ......

[0080] R45:if P req =PVL and SOC b =PVL, then P fc =PCS

[0081] S114, defuzzify the output variable of the main fuzzy controller to obtain the power to be output by the fuel cell. The actual input value of the fuzzy controller is converted into the corresponding fuzzy variable through fuzzification. By querying the fuzzy rule base in Table 2 and performing fuzzy reasoning, the output fuzzy value of the fuzzy controller is obtained, that is, the fuzzy subvalue on the domain of fuel cell output power. Only by converting it into an accurate quantity can it be applied to the object. The designed fuzzy controller is defuzzified by the centroid method, that is, the membership function of the output fuzzy set is set to u(x), x is an element of a set in the domain, and the output clarity value can be expressed as:

[0082]

[0083] Step S12 of this embodiment specifically includes the following sub-steps:

[0084] S121, set the input variables and output variables of the sub-fuzzy controller, such as Figure 4 As shown in the figure, the input variable of the sub-fuzzy controller is set to the state of charge (SOC b ), supercapacitor state of charge (SOC c ), remaining power (P w ), where the residual power is the difference between the vehicle demand power and the fuel cell output power, and the output variable is the lithium battery output power (P bat ), according to the hybrid electric vehicle model, the range of input and output variables is:

[0085]

[0086] S122, fuzzification of input variables and output variables of the sub-fuzzy controller, state of charge of the lithium battery (SOC b ) is the same as the main fuzzy controller, the supercapacitor state of charge (SOC c ) is {very low, low, medium, high}, expressed as {PVS, PS, PM, PL}; the remaining power (P w ) is {negative large, negative small, zero, small, medium, large, very large}, expressed as {NL, NS, Z, PS, PM, PL, PVL}; lithium battery output power (P bat ) The fuzzy subsets are {negative large, negative small, negative very small, zero, small, medium, large, very large}, denoted as {NL, NS, NVS, Z, PS, PM, PL, PVL};

[0087] S123, setting fuzzy rules of the sub-fuzzy controller and performing fuzzy reasoning. The fuzzy rules of the sub-fuzzy controller are set based on the operation history data of the fuel cell hybrid vehicle. There are 140 fuzzy rules in total, as shown in the following table:

[0088] Table 2 Sub-fuzzy control rules table

[0089]

[0090] Similarly, the “IF-THEN” language expression is used for fuzzy reasoning, and the fuzzy rules are specifically expressed as:

[0091] R1:if P W =NL,SOC b =PVS and SOC c =PVS, then P bat =NVS

[0092] R2:if P W =NS, SOC b=PVS and SOC c =PVS, then P bat =NVS ......

[0094] R140:if P W =PVL, SOC b =PVL and SOC c =PL, then P bat =PS

[0095] S124. Defuzzify the output variable of the sub-fuzzy controller to obtain the power that should be output by the power lithium battery, and subtract the power that should be output by the power lithium battery from the remaining power to obtain the power that should be output by the supercapacitor. Use the same method as the defuzzification of the output variable of the main fuzzy controller to obtain the precise value of the output variable of the sub-fuzzy controller, that is, to obtain the power that should be output by the lithium battery, and then obtain the power that should be output by the supercapacitor.

[0096] In step S112 and step S122 of this embodiment, the membership functions of the designed input variables and output variables are composed of a half-trapezoidal membership function and a triangular membership function, such as Figure 5 , 6 As shown, the horizontal axis represents the input variables and output variables, and the vertical axis represents the degree of membership. For fuzzy subsets with more fuzzy variables, in order to reduce the design parameters without affecting the performance, the fuzzy variables of each membership function can be reduced in crossover, the number of membership function parameters can be reduced, and the complexity of the optimization calculation described below can be reduced.

[0097] In the fuel cell hybrid vehicle of this embodiment, the fuel cell consumes hydrogen, and the lithium battery and supercapacitor consume electrical energy. The hydrogen consumption and electrical energy consumption must be converted into a consistent evaluation index. Therefore, when performing the total amount of hydrogen analysis, it is necessary to convert the electrical energy consumed by the lithium battery and supercapacitor into equivalent hydrogen consumption under a standard test condition, and calculate the equivalent fuel consumption of the whole vehicle. The equivalent energy consumption model based on the three energy sources in step S2 is specifically:

[0098]

[0099] Among them, J H2 is the lower calorific value of hydrogen; P fc , P bat , P cap They are the output power of fuel cells, power batteries (lithium batteries), and supercapacitors. They are the working efficiency of fuel cells, lithium batteries, and supercapacitors, These are the working efficiencies of the DC / DC converters connected to lithium batteries and supercapacitors, respectively.

[0100] During the driving process of the fuel cell hybrid vehicle of this embodiment, the three energy sources will cause different degrees of performance degradation according to different vehicle driving conditions. When establishing the energy source equivalent performance degradation model, since the performance degradation of the supercapacitor is very small and can be ignored, usually only the performance degradation of the fuel cell and the lithium battery is considered. The equivalent performance degradation objective function is jointly evaluated by the performance degradation rate of the fuel cell and the lithium battery. The equivalent performance degradation model based on the two energy sources in step S2 is specifically:

[0101] minf2(x)=f fc (x)+f bat (x)

[0102]

[0103]

[0104] Among them, f fc (x) is the equivalent performance degradation of the fuel cell, f bat (x) is the equivalent performance degradation of the power battery, k is the acceleration coefficient of the fuel cell life degradation, ΔV is the allowable voltage drop within the fuel cell life range, n1 is the number of fuel cell start and stop times, n2 is the number of fuel cell power changes, t1 is the idle time, t2 is the high power operation time, V1, V2, U1, U2 represent the performance degradation values ​​of the fuel cell start and stop, power change, idle, and high power operation respectively, and the values ​​are shown in the following table:

[0105] Table 3 Fuel cell voltage attenuation values

[0106]

[0107] n bat Indicates the number of charge and discharge cycles of lithium batteries at different discharge depths, LC bat Indicates the maximum allowable number of charge and discharge cycles of lithium batteries at different discharge depths. The values ​​are shown in the following table:

[0108] Table 4 Number of cycles allowed at different discharge depths

[0109]

[0110]

[0111] Step S3 of this embodiment specifically includes the following sub-steps:

[0112] S31. Establish a simulation experiment platform, build and run the fuel cell hybrid vehicle dynamics model, energy management strategy based on dual fuzzy control, optimization calculation, and automobile standard test conditions. Use SIMULINK in MATLAB and advanced vehicle simulation software ADVISOR to jointly build a simulation platform. SIMULINK is used to build an energy management strategy based on dual fuzzy control, and the other parts are constructed by ADVISOR, including vehicle dynamics module, vehicle control module, wheel and half-axle module, main reducer module, transmission module, motor and control module, fuel cell module, power lithium battery module, supercapacitor module, DC-DC converter module, power bus module, automobile standard test conditions, etc. The global unified light vehicle test cycle (WLTC) is used as the test condition, such as Figure 2 As shown, under this test condition, the simulation experimental platform runs the fuel cell hybrid vehicle dynamics model, the energy management strategy based on dual fuzzy control, the optimization calculation, etc., and finally obtains the optimized dual fuzzy control (control table);

[0113] S32, setting the parameters required to be optimized for the membership functions of the input variables and output variables of the main fuzzy controller and the sub-fuzzy controller to form an initial population, such as Figure 5 , 6 As shown in the figure, the membership functions of the input variables and output variables of the main fuzzy controller and the sub-fuzzy controller are composed of triangular functions and semi-trapezoidal functions. The triangular function has three parameters corresponding to the horizontal axis, namely the starting point, the peak point and the end point. The semi-trapezoidal function has two parameters, namely the starting point or the end point and the peak point. The parameters that need to be set for the membership function are shown in the following table:

[0114] Table 5 Fuzzy control optimization parameter quantity table

[0115]

[0116] A total of 70,the parameters initially set are used as the initial population for,optimization calculations;

[0117] S33, setting the initial operation parameters of the improved non-dominated sorting genetic algorithm (NSGA-Ⅱ), wherein the initial operation parameters include the initial population number, the maximum number of iterations, the mutation probability and the crossover probability. The process of the non-dominated sorting genetic algorithm (NSGA-Ⅱ) is as follows: Figure 7 As shown, the initial parameters of the algorithm are set as follows: the initial population size is 70, the maximum number of iterations is 200, the mutation probability and the crossover probability are 0.3 and 0.9 respectively.

[0118] According to steps S31-33 of this embodiment, the following operation steps are performed:

[0119] S34, generating a first-generation subgroup as a first-generation new population, and generating a first-generation subgroup as a first-generation new population through selection, crossover, and mutation operations;

[0120] S35, determine whether to establish a new parent population. If established, proceed to the next step. Otherwise, use the equivalent energy consumption model based on three energy sources and the equivalent performance degradation model based on two energy sources in step S2 as the fitness function of NSGA-Ⅱ, calculate the fitness value, determine the dominance and non-domination relationship of individuals in the population, calculate the crowding degree of individuals, and loop the operation until a new parent population is generated. The determination of the dominance and non-domination relationship of individuals in the population uses a fast non-dominated sorting algorithm. This embodiment is specifically as follows: If the two objective functions meet:

[0121]

[0122] In the above formula, individual x1 is said to dominate individual x2. By continuously comparing all individuals in the population, all non-dominated individuals (not dominated by other individuals) are finally stored in the non-dominated set, which is called the Pareto non-inferior solution set. Obviously, the algorithm reduces the computational complexity. The crowding degree is the density of other individuals around the individual. The higher the crowding degree, the lower the similarity between the individual and other individuals, and the better the individual. Retention can ensure the diversity of the population. The calculation method is: sort the fitness function value of each individual in ascending order. A certain individual forms a rectangle around the two closest individuals on both sides of the individual. The perimeter of the rectangle is calculated, which is the crowding degree of the individual.

[0123] S36, performing selection, crossover, and mutation operations to generate a new offspring population, using an iterative selection method, that is, in each iteration, calculating the fitness function values ​​of the parent and offspring, selecting some individuals with good fitness as the parent, and obtaining new individuals through crossover and mutation, and calculating the fitness function of the individual. If the fitness function value is better than that of the parent population, it becomes the new parent population, and the other individuals become the new offspring population through crossover combination or mutation;

[0124] S37, judging whether the set maximum number of iterations has been reached, if so, all non-dominated individuals are stored in the non-dominated set, and a Pareto non-inferior solution set is obtained, otherwise, the parent population and the child population are merged into a new population, and step S35 is executed again;

[0125] After 200 iterations, the Pareto non-inferior solution set is obtained, such as Figure 8 (a); for further explanation, it is divided into four areas: A, B, C, and D. Figure 8As shown in (b); Region D represents the unoptimized situation, where point I indicates that the initial population can achieve a lower equivalent energy consumption value, and point II indicates that the initial population can achieve a lower energy source equivalent performance degradation value. The equivalent energy consumption corresponding to all Pareto non-inferior solutions in region C is better than point I, but the equivalent energy source performance degradation value is larger, higher than point II. The energy source equivalent performance degradation of all Pareto non-inferior solutions in region B is better than point II, but its equivalent energy consumption is higher than point I. Therefore, the values ​​of the two objective functions corresponding to the Pareto non-inferior solution sets in regions B and C are contradictory, that is, if the optimization of energy consumption is mainly considered, the energy source performance degradation will be aggravated; on the contrary, if the optimization of energy source performance degradation is mainly considered, the energy consumption will be larger. The Pareto non-inferior solution in region A can compromise the optimization of the two performance indicators, that is, energy consumption and energy source performance degradation are optimized to different degrees at the same time, but the unilateral optimization effect may not be as good as the Pareto non-inferior solution in region C or region B; Figure 8 As shown in (c), nine representative non-inferior solution data are selected from the Pareto frontier distribution, corresponding to nine optimization schemes such as ①-⑨.

[0126] In this embodiment, three groups of solutions are extracted from the Pareto non-inferior solution set, which can respectively adapt to three vehicle operation scenarios, namely, a fuel cell with less energy storage, a fuel cell with moderate energy storage, and a fuel cell with more energy storage, to form a dual fuzzy control table for energy management of each vehicle. During the driving of the vehicle, energy is obtained by real-time table lookup and real-time energy distribution is performed. The selection is based on the amount of hydrogen stored in the hydrogen storage tank of the fuel cell vehicle, such as Figure 8 For the control schemes ①-⑨ in the Pareto frontier in (c), when the stored hydrogen is small, the Pareto non-inferior solutions numbered ①-③ can be applied; when the stored hydrogen is moderate, the Pareto non-inferior solutions numbered ④-⑥ can be applied; when the stored hydrogen is large, the Pareto non-inferior solutions numbered ⑦-⑨ can be applied.

[0127] In the embodiment of the present invention, the energy management strategy is optimized for the power system of a fuel cell hybrid vehicle equipped with three energy sources. The actual reference object is a four-wheel vehicle, and its power system topology is as follows: Figure 1 The parameters of the three-energy-source fuel cell hybrid vehicle and its main components are shown in the following table:

[0128] Table 6 Fuel cell vehicles and their main basic parameters

[0129]

[0130]

[0131] exist Figure 1In the invention, the power system assembly of a fuel cell hybrid vehicle can be divided into two parts: an energy source subsystem and a traction subsystem, wherein the control strategy of the energy source subsystem includes an energy management strategy and a control strategy of a DC / DC converter, and the control method of the traction subsystem implements control on an inverter (DC / AC) to achieve the goal of controlling the traction subsystem, including regenerative braking control of a fuel cell vehicle. The energy management strategy serves as an upper-level control to allocate output power and / or absorb braking feedback power to three energy sources, and accordingly implements corresponding control on the DC / DC converter. The present invention mainly focuses on the energy management strategy and optimization of the energy source subsystem.

[0132] The present invention sets the requirements to be met by the energy management strategy of fuel cell hybrid electric vehicles into four aspects:

[0133] (1) Real-time performance: the energy management strategy should meet the real-time energy allocation requirements of the fuel cell hybrid vehicle under the current actual working conditions;

[0134] (2) Optimization: The energy management strategy should optimize the energy distribution of the three energy sources as much as possible for the current actual operating conditions of fuel cell hybrid vehicles;

[0135] (3) Economy: The energy management strategy should try to make the fuel cell, power battery and supercapacitor operate in the high efficiency range to maximize the equivalent fuel economy of the vehicle;

[0136] (4) Durability. The energy management strategy should make the fuel cell and power battery operate in the ideal area as much as possible to avoid frequent sudden changes in the current of the energy source, which would cause its performance degradation; at the same time, minimize the number of start / stop cycles of the fuel cell, thereby extending the service life of the fuel cell.

[0137] Figure 2 The overall implementation process of the energy control method of the present invention is given, including an offline part and an online part. Through the design and operation of the offline part, three sets of dual fuzzy controller control tables (note: key work) are obtained, corresponding to three scenarios such as sufficient, moderate and less hydrogen in the fuel cell. The online part refers to the actual operation of the on-board energy management system (EMS). The power demand of the whole vehicle is obtained according to the vehicle driving condition, and the SOC of the lithium battery and the supercapacitor are detected at the same time. The three parameter values ​​are used as the input of the dual fuzzy controller control table, and the control scheme is obtained by looking up the table, and finally the output power of the three energy sources is obtained (P in the figure). fc , P bat , P cap The output power of fuel cells, power lithium batteries, and super capacitors) or recovery power (only for power lithium batteries and super capacitors)

[0138] The optimization results of the energy management of fuel cell hybrid electric vehicles implemented by the present invention under WLTC conditions are analyzed below:

[0139] (1) Optimizing dual fuzzy controller:

[0140] Only representative Figure 8 Taking the Pareto non-inferior solution ⑤ in (c) as an example, the dual fuzzy controller is optimized. Compared with before optimization, the membership functions of the main and auxiliary fuzzy controllers have changed. Only the membership function of the optimized main fuzzy controller is given as an example. Fig. 9 As shown, Fig. 9 (b) Lithium battery SOC b The starting point, midpoint and end point of PS, PM and PL in the triangular membership function are all changed, the third and fourth points of the trapezoidal membership function PVS, and the first and second points of the trapezoidal membership function PVL are all shifted. The optimized fuzzy controller will enable fuel cell hybrid vehicles to achieve better energy management effects.

[0141] (2) The running results of each Pareto non-inferior solution:

[0142] Select Figure 8 The simulation results of the 9 Pareto non-inferior solutions ①-⑨ in (c) are shown in the following table:

[0143] Table 7 Energy optimization results and applicable scenarios of different Pareto non-inferior solutions for dual fuzzy control

[0144]

[0145] The Pareto non-inferior solutions ①-③ in region C have better optimization effect on equivalent energy consumption and are more suitable for situations where less hydrogen is stored; the Pareto non-inferior solutions ④-⑥ in region A can achieve the purpose of optimizing equivalent energy consumption and also obtain better optimization effect on equivalent energy source performance degradation, which is suitable for situations where moderate hydrogen is stored; the Pareto non-inferior solutions ④-⑥ in region B, although the equivalent energy consumption is large, have better optimization effect on equivalent energy source performance degradation, which is suitable for situations where more hydrogen is stored.

[0146] (3) Results and comparative analysis of Pareto non-inferior solution ⑤ after optimization:

[0147] Furthermore, the energy management effect of the dual fuzzy control after optimization by the Pareto non-inferior solution ⑤ is compared with the effects of the power following control strategy and the unoptimized dual fuzzy control strategy.

[0148] The energy management effect of each control strategy is mainly evaluated by factors such as fuel consumption and fuel cell output power, as well as equivalent energy consumption and energy source equivalent performance degradation obtained by calculation. Fig.10 It can be seen that under the WLTC condition, the hydrogen consumption of the hydrogen fuel cell vehicle is minimized, the unoptimized dual fuzzy control strategy is second, and the power following control strategy is the largest.

[0149] Fig.11 (a) illustrates the output power of the fuel cell under the dual fuzzy control strategy before and after optimization. It is obvious that the fuel cell is turned on and off three times under the control strategy before optimization, and its output power fluctuates greatly. However, the fuel cell is turned on and off once under the control strategy after optimization, that is, there is only one switch at the beginning, and no switch operation is performed under the entire working condition, and its output power fluctuates less. The performance degradation of the fuel cell is mainly related to the number of switches, fluctuations and high-rate operation time. Therefore, the present invention is conducive to extending the service life of the fuel cell. Fig.11 (b) shows the power output of each energy source under the control of the optimized dual fuzzy control strategy. It can be seen from the figure that under the entire working condition, the fuel cell is the main energy source, and it can output power steadily and reasonably according to the specific driving conditions of the vehicle. For example, when the vehicle requires a large amount of power, its output power will increase to meet the power demand of the vehicle; while the power lithium battery and supercapacitor, as auxiliary power sources, output power or recover braking power according to the actual driving conditions and their own charge state, or are charged by the fuel cell. The lithium battery can provide stable power output in time when the vehicle's power demand increases, and the supercapacitor has the effect of "peak shaving and valley filling", effectively reducing the strong current impact on the lithium battery.

[0150] Through the above comparative analysis, the embodiment of the present invention can reasonably and optimally distribute power among the three energy sources, realize the multi-objective optimization of equivalent energy consumption and equivalent performance degradation of fuel cell hybrid vehicles, and verify the superiority of the present invention.

[0151] The present invention is not limited to the above-mentioned embodiments. Anyone should be aware that any structural changes made under the enlightenment of the present invention, and any technical solutions that are the same or similar to the present invention, fall within the protection scope of the present invention. The technology, shape, and structural parts not described in detail in the present invention are all well-known technologies.

Claims

1. A multi-objective optimization dual fuzzy control method for energy of fuel cell hybrid electric vehicles. It is characterized in that The following steps are involved: S1. Establish a fuel cell hybrid electric vehicle energy management model based on dual fuzzy control, with a main fuzzy controller taking the vehicle's required power and the lithium battery's state of charge as input variables and the fuel cell's output power as output variable; a secondary fuzzy controller taking the lithium battery's state of charge, the supercapacitor's state of charge, and the remaining power as input variables and the lithium battery's output power as output variable, wherein the remaining power is the difference between the vehicle's required power and the fuel cell's output power; S2. Establish a multi-objective function for optimization calculation. According to the consumption of hydrogen by fuel cells and the consumption of electricity by power batteries and supercapacitors, establish an equivalent energy consumption model based on three energy sources. Specifically: Among them, J H2 is the lower calorific value of hydrogen; P fc , P bat , P cap are the output power of fuel cell, power battery and supercapacitor respectively. They are the working efficiency of fuel cells, lithium batteries, and supercapacitors, They are the working efficiency of the DC / DC converter connected to the lithium battery and supercapacitor respectively; Considering the performance degradation of fuel cells and lithium batteries, an equivalent performance degradation model based on the two energy sources is established, specifically: minf 2 (x)=f fc (x)+f bat (x) Among them, f fc (x) is the equivalent performance degradation of the fuel cell, f bat (x) is the equivalent performance degradation of the power battery, k is the acceleration coefficient of the fuel cell life degradation, ΔV is the allowable voltage drop within the fuel cell life range, n 1 is the number of fuel cell starts and stops, n 2 is the number of fuel cell power changes, t 1 is the idling time, t 2 For high power operation time, V 1 、V 2 , U 1 , U 2 They represent the performance attenuation values ​​of the fuel cell under start-stop, power change, idling, and high-power operation conditions, respectively. bat Indicates the number of charge and discharge cycles of lithium batteries at different discharge depths, LC bat Indicates the maximum allowable number of charge and discharge cycles of lithium batteries at different discharge depths; S3. Using an improved non-dominated sorting genetic algorithm, the dual fuzzy controller in step S1 is optimized to establish a fuel cell hybrid electric vehicle energy management model based on dual fuzzy control of Pareto multi-objective optimization.

2. A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method according to claim 1, It is characterized in that The step S1 specifically includes the following sub-steps: S11. Establish the main fuzzy controller for energy management of fuel cell hybrid electric vehicles; S12. Establish a sub-fuzzy controller for energy management of fuel cell hybrid electric vehicles.

3. A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method according to claim 2, It is characterized in that The step S11 specifically The following steps are included: S111, setting the input variables and output variables of the main fuzzy controller; S112. The input variables and output variables of the main fuzzy controller are fuzzified. The fuzzy subset of the vehicle's required power is {zero, extremely small, very small, small, small, medium, large, large, very large}, expressed as {Z, PES, PVS, PCS, PS, PM, PL, PCL, PVL}; the fuzzy subset of the lithium battery's state of charge is {very low, low, medium, high, very high}, expressed as {PVS, PS, M, PL, PVL}; the fuzzy subset of the fuel cell's output power is {zero, very small, very small, small, small, medium, large, large, very large}, expressed as {Z, PES, PVS, PCS, PS, PM, PL, PCL, PVL, PEL}; S113, setting fuzzy rules of the main fuzzy controller and performing fuzzy reasoning; S114, defuzzifying the output variable of the main fuzzy controller to obtain the power that should be output by the fuel cell.

4. A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method according to claim 2, It is characterized in that The step S12 specifically The following steps are included: S121, setting the input variables and output variables of the sub-fuzzy controller; S122, the input variables and output variables of the auxiliary fuzzy controller are fuzzified. The fuzzy subset of the state of charge of the lithium battery is the same as that of the main fuzzy controller. The fuzzy subset of the state of charge of the supercapacitor is {very low, low, medium, high}, expressed as {PVS, PS, PM, PL}; the fuzzy subset of the residual power is {negative large, negative small, zero, small, medium, large, very large}, expressed as {NL, NS, Z, PS, PM, PL, PVL}; the fuzzy subset of the output power of the lithium battery is {negative large, negative small, very small negative, zero, small, medium, large, very large}, expressed as {NL, NS, NVS, Z, PS, PM, PL, PVL}; S123, setting fuzzy rules of the sub-fuzzy controller and performing fuzzy reasoning; S124, defuzzifying the output variable of the sub-fuzzy controller to obtain the power that should be output by the power lithium battery, and subtracting the power that should be output by the power lithium battery from the residual power to obtain the power that should be output by the supercapacitor.

5. A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method according to claim 3 or 4, It is characterized in that In step S112 and step S122, the membership functions of the designed input variables and output variables are composed of a half-trapezoidal membership function and a triangular membership function.

6. A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method according to claim 1, It is characterized in that The step S3 specifically includes the following sub-steps: S31. Establish a simulation experimental platform, build and run a fuel cell hybrid vehicle dynamics model, energy management strategy based on dual fuzzy control, optimization calculation, and vehicle standard test conditions; S32, setting the parameters required to be optimized for the membership functions of the input variables and output variables of the main fuzzy controller and the sub-fuzzy controller to form an initial population; S33, setting initial operation parameters of the improved non-dominated sorting genetic algorithm, wherein the initial operation parameters include the number of initial populations, the maximum number of iterations, the mutation probability and the crossover probability.

7. A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method according to claim 6, It is characterized in that According to the steps S31-33, the following operation steps are performed: S34, generating the first generation subgroup as the first generation new population; S35, determine whether to establish a new parent population, if established, proceed to the next step, otherwise, use the equivalent energy consumption model based on three energy sources and the equivalent performance degradation model based on two energy sources in step S2 as the fitness function of the non-dominated sorting genetic algorithm, calculate the fitness value, determine the dominance and non-domination relationship of individuals in the population, calculate the crowding degree of individuals, and repeat the operation until a new parent population is generated; S36, performing selection, crossover, and mutation operations to generate a new offspring population; S37, judging whether the set maximum number of iterations has been reached, if so, all non-dominated individuals are stored in the non-dominated set, and a Pareto non-inferior solution set is obtained, otherwise, the parent population and the child population are merged into a new population, and step S35 is re-executed.

8. A multi-objective optimization fuel cell hybrid electric vehicle energy dual fuzzy control method according to claim 7, It is characterized in that Three groups of solutions are extracted from the Pareto non-inferior solution set, which can respectively adapt to three vehicle operation scenarios: fuel cell storage of less energy, fuel cell storage of moderate energy, and fuel cell storage of more energy. A dual fuzzy control table for energy management of each vehicle is formed, and energy is distributed in real time through real-time table lookup while the vehicle is driving.