A Hybrid Power Ship Energy Management Method Based on Genetic Algorithm-Optimized Fuzzy Controller Membership Function

Abstract

This invention discloses a hybrid-powered ship energy management method based on a genetic algorithm to optimize the membership function of a fuzzy controller. The method constructs a multi-objective function and constraints, encodes the membership function parameters of the fuzzy controller, establishes a mathematical model and its constraints, establishes a fitness function J based on the mathematical model, and calculates the fitness value of each sample using a genetic algorithm. This invention improves and optimizes the membership function of the fuzzy controller through a genetic algorithm, further optimizing the power allocation factor, thereby reducing the peak power of the lithium battery module, maximizing the lifespan of the lithium battery module, fully utilizing the high power density characteristics of supercapacitors, and achieving a more rational power allocation in the composite energy storage system.

Description

Technical Field

[0001] This invention relates to the field of energy management technology for hybrid-powered ships, and specifically to an energy management method for hybrid-powered ships based on optimizing the membership function of a fuzzy controller using a genetic algorithm. Background Technology

[0002] Since the beginning of the 21st century, the pollution problems caused by the rapid development of global trade have become increasingly prominent. The international community has also become more aware of the importance of environmental protection, and has consequently formulated increasingly stringent energy conservation and environmental protection standards for various industries. Ships, as the most important mode of transportation in global trade, play an irreplaceable role in promoting economic globalization and social development. However, at the same time, ships consume large amounts of fossil fuels and emit various harmful pollutants during their voyages, making them a significant cause of energy crises and environmental pollution. With increasing public awareness of environmental protection, energy conservation and emission reduction have become important trends for the future development of the shipping industry.

[0003] Against the backdrop of advocating the development of green and intelligent ships, fuel cell-based hybrid ships have received widespread attention and application. Fuel cell-based hybrid ships can efficiently and reliably adapt to complex operating conditions during navigation, and can also significantly reduce the impact of load fluctuations on the ship's power grid and propulsion system. This method allows the fuel cell to operate within a high-efficiency range, maintaining high economic efficiency and good emission performance; that is, storing the fuel cell's surplus power under changing operating conditions and releasing electrical energy to the grid to supplement the ship's power when load demand is high. The composite energy storage system not only effectively smooths out power fluctuations, but can also serve as the ship's main power source in areas with special requirements such as ship air pollution emissions or noise control zones. This allows the main engine to be shut down to meet noise and emission requirements in the route area, aligning with the development concept of green and clean ships.

[0004] In order to fully leverage the advantages of the aforementioned power supply units, there is an urgent need for a control system and control method for fuel cell-based hybrid ships to coordinate and manage the power output of each energy module under different operating conditions of the ship, so as to achieve the optimal state of hybrid ship operating efficiency, emissions and power performance.

[0005] With the development of the shipbuilding industry, hybrid power systems have gradually emerged, combining fuel cells with supercapacitors and lithium batteries for energy storage and power supply. This can significantly save fuel, reduce pollution, and lower operating costs when applied to the energy supply system of ship propulsion systems. However, a common problem with current hybrid power systems is that the peak power of lithium batteries is too high, while the power output of supercapacitors is relatively low. This results in an unreasonable power distribution within the hybrid system, leading to problems such as reduced lithium battery lifespan. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of the above-mentioned technologies by providing a hybrid power ship energy management method based on the genetic algorithm to optimize the membership function of a fuzzy controller, which offers a more reasonable power allocation and maximizes the lifespan of lithium battery modules.

[0007] To achieve the above objectives, this invention provides a hybrid power ship energy management method based on genetic algorithm optimization of fuzzy controller membership function as follows:

[0008] 1) Construct a multi-objective function and its constraints;

[0009] 2) Encoding the membership function parameters of the fuzzy controller;

[0010] 3) Establish a mathematical model and its constraints based on steps 1) and 2);

[0011] 4) Establish the fitness function J based on the mathematical model;

[0012] 5) Calculate the fitness value of each sample based on the genetic algorithm.

[0013] Furthermore, the multi-objective function in step 1) includes:

[0014] The objective function that minimizes the number of drastic power changes in lithium batteries is:

[0015] minf1=rN(ΔP bat )

[0016] In the formula, r is the lifetime decay rate, and ΔP bat It is the power change rate of the lithium battery, N(ΔP) bat This indicates the number of times the lithium battery's power changes during the entire operating condition;

[0017] The objective function for minimizing system energy consumption is:

[0018] minf ２ =E HESS =∫(P bat +P sc )dt

[0019] In the formula, P bat It is the real-time power of the lithium battery module, P sc This is the real-time power of the supercapacitor module;

[0020] Therefore, multi-objective optimization can be further expressed as:

[0021] minf (x) =[f1, f2] =[rN(ΔP) bat ), ∫(Pbat +P sc )dt]

[0022] In the formula, f1 and f2 are the optimization objectives. The smaller the objective function value, the closer the obtained solution is to the optimal solution.

[0023] The constraints that the optimization variables must satisfy include:

[0024] SOC constraints for lithium batteries and supercapacitors

[0025]

[0026] In the formula, SOC bat，min For lithium battery SOC bat The minimum value of SOC bat，max For lithium battery SOC bat The maximum value, SOC sc，min Supercapacitor SOC sc The minimum value of SOC sc，max Supercapacitor SOC sc The maximum value;

[0027] Instantaneous power balance constraint

[0028] P bat +P sc +P fc =P load

[0029] In the formula, P bat It is the instantaneous output power of the lithium battery, P sc It is the instantaneous output power of the supercapacitor, P fc It is the instantaneous output power of the fuel cell, P. load It is the power demand at any given moment;

[0030] Energy constraints:

[0031] E bat +E sc ≥E gap

[0032] In the formula, E bat It is the energy output of the lithium battery, E sc It is the output power of the supercapacitor, E gap This is the total energy required for composite energy storage;

[0033] Supercapacitor voltage constraint

[0034]

[0035] In the formula, U sc，max This represents the maximum operating voltage of the supercapacitor.

[0036] Furthermore, the specific process of encoding the membership function parameters of the fuzzy controller in step 2) is as follows:

[0037] The fuzzy controller includes three input parameters and one output parameter. The three input parameters are the power P of the lithium battery. bat (t) SOC of lithium battery bat (t) and the SOC of the supercapacitor sc (t), the output parameter is the power regulation coefficient k of the lithium battery module, and all four parameters are represented by Gaussmf;

[0038] The expression for the Gaussian membership function is:

[0039]

[0040] x is the independent variable, c represents the position of the Gaussian function in the fuzzy universe of discourse, and σ represents the variation range and coverage of the Gaussian function.

[0041] The output parameter k of the fuzzy controller is divided by 5 Gaussian functions, and the input parameter is the power P of the lithium battery. bat The SOC value of a lithium battery is determined by three Gaussian functions. bat The SOC value of a supercapacitor is determined by five Gaussian functions. sc Divided by 5 Gaussian functions, the optimization parameters c and σ are expressed as follows:

[0042] c = X = [X] 11 X 12 X 13 X 21 X 22 X 23 X 24 X 25 X 31 X 32 X 33 X 34 X 35 X 41 X 42 X 43 X 44 X 45 ]

[0043] σ=Y=[Y 11 Y 12 Y 13 Y 21 Y 22 Y 23 Y 24 Y 25 Y 31 Y 32 Y 33 Y34 Y 35 Y 41 Y 42 Y 43 Y 44 Y 45 ]

[0044] To ensure the relationships between the various fuzzy subsets, enable the fuzzy logic controller to function properly, and reduce computation time, the positional constraints of the Gaussian function are given:

[0045]

[0046] The membership function is encoded using floating-point encoding. Each Gaussian function is determined by two parameters, c and σ. There are a total of 18 Gaussian functions for the one output parameter and three input parameters to be optimized. Therefore, there are a total of 36 parameters to be optimized. Each parameter to be optimized is represented by a 7-bit binary number. Thus, each sample in the population has 7 × 36 = 252 bits.

[0047] Furthermore, step 3) establishes the mathematical model and its constraints based on steps 1) and 2) as follows:

[0048]

[0049] In the formula, k1 and k2 are weighting coefficients.

[0050] Furthermore, in step 4), the fitness function J is established as follows:

[0051] J = k1(f1) + k2(f2).

[0052] Furthermore, step 5) specifically involves: calculating the optimization index value J for each sample based on a genetic algorithm, a mathematical model, and constraints. i Let i = 1, 2, and then sort the results according to their size. The smaller the J value, the better the sample.

[0053] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention improves and optimizes the membership degree of the fuzzy controller through a genetic algorithm, further optimizes the power allocation factor, thereby reducing the peak power of the lithium battery module, maximizing the service life of the lithium battery module, giving full play to the high power density characteristics of the supercapacitor, and realizing a more reasonable power allocation of the composite energy storage system. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the topology of the marine hybrid power system of the present invention;

[0055] Figure 2 This is a schematic diagram of the membership function of lithium battery power.

[0056] Figure 3 A schematic diagram of the membership function for the remaining state of charge (SOC) of a lithium battery;

[0057] Figure 4 A schematic diagram of the membership function for the remaining state of charge (SOC) of a super battery;

[0058] Figure 5 A schematic diagram of the membership function for the adjustment coefficient k;

[0059] Figure 6 A fuzzy rule-based 3D surface plot;

[0060] Figure 7 This is a flowchart of the optimization algorithm of the present invention;

[0061] Figure 8 A comparison chart of real-time power output of lithium batteries;

[0062] Figure 9 A comparison chart of lithium battery output current;

[0063] Figure 10 A comparison chart of the State of Charge (SOC) of lithium batteries;

[0064] Figure 11 This is a comparison chart of the State of Charge (SOC) of supercapacitors. Detailed Implementation

[0065] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments to facilitate a clearer understanding of the invention, but these descriptions do not constitute a limitation on the invention.

[0066] Marine hybrid power system topology as follows Figure 1 As shown, the system includes a fuel cell, propeller, propulsion motor, DC / DC converter, supercapacitor, and lithium battery. The battery and supercapacitor together form an auxiliary energy source, combining the high energy density and high power density advantages of both. The fuel cell stack provides most of the power required by the load; the battery serves as the primary auxiliary power source, assisting the fuel cell in powering during startup and acceleration; the supercapacitor serves as the secondary auxiliary power source, providing peak power for short periods. Both the battery and supercapacitor are rechargeable and dischargeable energy sources.

[0067] The designed fuzzy controller contains three inputs: the power P of the lithium battery. bat (t) SOC of lithium battery bat (t) and the SOC of the supercapacitor sc(t), which includes an output quantity, the power regulation coefficient k of the lithium battery module. The power of the lithium battery is fuzzified using quantization factor K1, the SOC of the lithium battery is fuzzified using quantization factor K2, and the SOC of the supercapacitor is fuzzified using quantization factor K3. Based on the fuzzified values, the fuzzified lithium battery power is determined to belong to the fuzzy subset {L, M, H}, and the fuzzified lithium battery SOC is determined to belong to the fuzzy subset {VL, L, M, H, VH}, and the supercapacitor SOC is determined to belong to the fuzzy subset {VL, L, M, H, VH}. The position of the fuzzified input is found in the membership function curves of the three.

[0068] The quantization factors K1, K2, and K3 represent the conversion of the physical domains of lithium battery power, remaining lithium battery capacity, and remaining supercapacitor capacity into fuzzy domains, respectively, and the establishment of fuzzy subsets based on these fuzzy domains; SOC represents the remaining battery capacity value; VL, L, M, H, and VH represent that the fuzzy subsets are divided into four levels, corresponding to four states: extremely low, low, medium, high, and extremely high; the membership function is a non-uniform membership function selected to improve the sensitivity of fuzzy logic control, which is composed of a triangular function and a trapezoidal function.

[0069] Traditionally, the relationship between elements and sets is simply "belongs to" or "does not belong to". However, in fuzzy control, any element in a variable can be related to a certain fuzzy subset to varying degrees. This degree of correlation between an element and the fuzzy subset is called the membership degree. The domain of the membership degree is [0, 1]. The closer the membership degree is to 1, the higher the degree to which the element belongs to the fuzzy subset; conversely, the closer it is to 1, the lower the degree to which the element belongs to the fuzzy subset.

[0070] The membership function curves of the input-output fuzzy sets, constructed using Matlab, are shown below. The power P of the lithium battery... bat The membership function curve of (t) is as follows: Figure 2 lithium battery state of charge (SOC) bat (t) Membership function curve as shown Figure 3 SOC of supercapacitors sc (t) Membership function curve as shown Figure 4 The membership function curve of the adjustment coefficient k is as follows: Figure 5 ;

[0071] Fuzzy control rule formulation:

[0072] The hybrid power propulsion control system for ships generates control rules based on a fuzzy model of the process, and pays attention to the rationality, consistency, and completeness of the number of rules during the design process. The design of fuzzy control rules for hybrid power ships is mainly based on the following criteria:

[0073] (1) Lithium batteries handle power with relatively low frequency and high energy, while supercapacitors handle power fluctuations with higher frequency and lower energy, thus giving full play to the working characteristics of both.

[0074] (2) Lithium batteries have a short cycle life and should not be overcharged, over-discharged, or have their power exceeded. Therefore, the SOC value of lithium batteries needs to be maintained in a high range.

[0075] (3) When the working condition allows, the supercapacitor should share the peak power of the lithium battery as much as possible to reduce the power fluctuation of the lithium battery.

[0076] Based on meeting the ship's power and energy requirements, 25 x 3 = 75 fuzzy rules were created for the designed three-input single-output fuzzy controller, thus formulating the corresponding fuzzy rule table.

[0077] Table 1 P bat Fuzzy rule table for L

[0078]

[0079] Table 2 P bat Fuzzy rule table for M

[0080]

[0081] Table 3 P bat Fuzzy rule table for H

[0082]

[0083] By utilizing fuzzy control reasoning relationships, the value of k is determined using the "if A and B then C" condition.

[0084] As shown in Table 3, there is a relationship ifSOC. sc =VL and SOC bat =VL then k=S.

[0085] At this point, based on the established fuzzy rule control table, the fuzzy subset in which the current power adjustment coefficient k lies is determined, which is one of {VS, S, M, B, VB}. Here, VS represents minimum, S represents small, M represents medium, B represents large, and VB represents maximum.

[0086] Using fuzzy rule tables and fuzzy control inference relationships, MATLAB can be used to generate fuzzy logic rule observation tables, and then fuzzy rule three-dimensional surface plots can be obtained, such as... Figure 6 As shown in a and 6b.

[0087] Based on the fuzzy subset of the output lithium battery power regulation coefficient k, a weighted average method is used to determine the lithium battery power, and then the obtained lithium battery power regulation coefficient k, P′ is used. bat (t)=P bat (t)×(1-k) can be used to calculate the power required by the adjusted lithium battery, which can then be used to calculate P′. sc (t)=P sc (t)+P bat (t)×k, and then the power required by the supercapacitor can be obtained.

[0088] Where P′ bat (t), P bat (t), P′ sc (t), P sc (t) and t represent the lithium battery power after fuzzy rule judgment, the lithium battery power before fuzzy rule judgment, the supercapacitor power after fuzzy rule judgment, the supercapacitor power before fuzzy rule judgment, and the working time, respectively.

[0089] By employing genetic algorithms to improve the fuzzy controller, and using energy efficiency and battery durability as optimization indicators, the high power density characteristics of supercapacitors can be fully utilized to achieve a more rational power allocation in the composite energy storage system. Figure 7 The optimization steps shown are as follows:

[0090] 1) Constructing multi-objective functions and constraints

[0091] The objective function that minimizes the number of drastic power changes in lithium batteries is:

[0092] minf1=rN(ΔP bat )

[0093] In the formula, r is the lifetime decay rate, and ΔP bat It is the power change rate of the lithium battery, N(ΔP) bat This indicates the number of times the lithium battery's power changes during the entire operating condition;

[0094] The objective function for minimizing system energy consumption is:

[0095] minf2=E HESS =∫(P bat +P sc )dt

[0096] In the formula, P bat It is the real-time power of the lithium battery module, P sc This is the real-time power of the supercapacitor module;

[0097] Therefore, multi-objective optimization can be further expressed as:

[0098] minf(x) =[f1, f2] =[rN(ΔP) bat ), ∫(P bat +P sc )dt]

[0099] In the formula, f1 and f2 are the optimization objectives. The smaller the objective function value, the closer the obtained solution is to the optimal solution.

[0100] The constraints that the optimization variables must satisfy include:

[0101] SOC constraints for lithium batteries and supercapacitors

[0102]

[0103] In the formula, SOC bat，min For lithium battery SOC bat The minimum value of SOC bat，max For lithium battery SOC bat The maximum value, SOC sc，min Supercapacitor SOC sc The minimum value of SOC sc，max Supercapacitor SOC sc Maximum value; Lithium battery SOC bat The range is 0.2 to 0.8, and the SOC of the supercapacitor is... sc The range is 0.3 to 0.9;

[0104] Instantaneous power balance constraint

[0105] P bat +P sc +P fc =P load

[0106] In the formula, P bat It is the instantaneous output power of the lithium battery, P sc It is the instantaneous output power of the supercapacitor, P fc It is the instantaneous output power of the fuel cell, P. load It is the power demand at any given moment;

[0107] Energy constraints:

[0108] E bat +E sc ≥E gap

[0109] In the formula, E bat It is the energy output of the lithium battery, E sc It is the output power of the supercapacitor, E gap This is the total energy required for composite energy storage;

[0110] Supercapacitor voltage constraint

[0111]

[0112] In the formula, U sc，max This represents the maximum operating voltage of the supercapacitor.

[0113] 2) Encoding the membership function parameters of the fuzzy controller

[0114] The fuzzy controller includes three input parameters and one output parameter. The three input parameters are the power P of the lithium battery. bat (t) SOC of lithium battery bat (t) and the SOC of the supercapacitor sc (t), the output parameter is the power regulation coefficient k of the lithium battery module, and all four parameters are represented by Gaussmf;

[0115] The expression for the Gaussian membership function is:

[0116]

[0117] x is the independent variable, c represents the position of the Gaussian function in the fuzzy universe of discourse, and σ represents the variation range and coverage of the Gaussian function.

[0118] The output parameter k of the fuzzy controller is divided by 5 Gaussian functions, and the input parameter is the power P of the lithium battery. bat The SOC value of a lithium battery is determined by three Gaussian functions. bat The SOC value of a supercapacitor is determined by five Gaussian functions. sc Divided by 5 Gaussian functions, the optimization parameters c and σ are expressed as follows:

[0119] c = X = [X] 11 X 1２ X 1３ X ２1 X ２２ X 23 X 24 X 25 X 31 X 32 X 33 X 34 X 35 X 41 X 42 X 43 X 44 X 45 ]

[0120] σ=Y=[Y 11 Y 12 Y 13 Y 21 Y22 Y 23 Y 24 Y 25 Y 31 Y 32 Y 33 Y 34 Y 35 Y 41 Y 42 Y 43 Y 44 Y 45 ]

[0121] To ensure the relationships between the various fuzzy subsets, enable the fuzzy logic controller to function properly, and reduce computation time, the positional constraints of the Gaussian function are given:

[0122]

[0123] The membership function is encoded using floating-point encoding. Each Gaussian function is determined by two parameters, c and σ. There are a total of 18 Gaussian functions for the one output parameter and three input parameters to be optimized. Therefore, there are a total of 36 parameters to be optimized. Each parameter to be optimized is represented by a 7-bit binary number. Thus, each sample in the population has 7 × 36 = 252 bits.

[0124] Establish the mathematical model and its constraints based on steps 1) and 2).

[0125]

[0126] In the formula, k1 and k2 are weighting coefficients.

[0127] A fitness function J is established based on the mathematical model.

[0128] J = k1(f1) + k2(f2).

[0129] 5) Calculate the optimization index value J for each sample based on the genetic algorithm, mathematical model, and constraints. i Let i = 1, 2, and then sort the results according to their size. The smaller the J value, the better the sample.

[0130] Genetic parameter settings are shown in the table.

[0131]

[0132] To compare the differences between the fuzzy control strategy and the optimized energy management strategy, a simulation model of the hybrid power system was built in Matlab / Simulink software in this application. The superiority of the algorithm of this invention was highlighted by comparing the real-time power of the lithium battery, the output current of the lithium battery, and the state of charge (SOC) values ​​of the lithium battery and the supercapacitor.

[0133] Real-time power comparison chart of lithium batteries as shown below Figure 8 .from Figure 8 As can be seen from the optimized curve, the peak power variation of the lithium battery has been significantly reduced, and the fluctuations have become smoother, which provides better protection for the lithium battery and helps to slow down the aging rate of the lithium battery.

[0134] A comparison chart of lithium battery output current is shown below. Figure 9 The improved lithium battery exhibits a further reduction in maximum current variation, which significantly suppresses current fluctuations and enhances battery durability, thus extending battery lifespan.

[0135] A comparison chart of the state of charge (SOC) values ​​of lithium batteries is shown below. Figure 10 The improved lithium battery exhibits a more gradual decrease in discharge current around 190 seconds, demonstrating that the optimized energy management strategy can better allocate energy to the composite energy storage system.

[0136] The comparison chart of the State of Charge (SOC) values ​​of supercapacitors is shown below. Figure 11 The improved supercapacitor responded quickly to the required energy, demonstrating that the designed GA-Fuzzy energy management strategy can more efficiently control the composite energy storage system and ensure that the composite energy storage system operates stably within a reasonable operating range.

[0137] Traditional ships use power batteries as energy storage devices. When high-frequency fluctuations in load power are borne solely by the power battery pack, it can damage the battery's lifespan and hinder the maintenance of good dynamic response and economy in hybrid-powered ships. This invention incorporates supercapacitors to form a composite energy storage system. The resulting ship propulsion system topology avoids the power battery bearing excessive high-frequency power, thus better responding to dynamic changes in the system.

[0138] This invention's fuzzy control achieves good control of the controlled object without requiring a precise mathematical model, making it suitable for controlled objects with significantly changing dynamic characteristics that make it difficult to build a precise mathematical model. Furthermore, fuzzy control is a complete control strategy summarized by the operator based on years of experience, offering advantages such as ease of use, high optimization potential, good robustness, and strong fault tolerance.

[0139] Traditional fuzzy control does not analyze optimization indicators, and its power allocation has limitations. This invention improves and optimizes the membership degree of the fuzzy controller through a genetic algorithm, further optimizing the power allocation factor, thereby reducing the peak power of the lithium battery module, maximizing the service life of the lithium battery module, giving full play to the high power density characteristics of supercapacitors, and achieving a more reasonable power allocation of the composite energy storage system.

Claims

1. A hybrid power ship energy management method based on genetic algorithm to optimize the membership function of a fuzzy controller, characterized in that: The management methods are as follows: 1) Construct a multi-objective function and its constraints; 2) Encoding the membership function parameters of the fuzzy controller; 3) Establish a mathematical model and its constraints based on steps 1) and 2); 4) Establish the fitness function based on the mathematical model. ; 5) Calculate the fitness value of each sample based on a genetic algorithm; The multi-objective function in step 1) includes: The objective function that minimizes the number of drastic power changes in lithium batteries is: In the formula, r is the lifetime decay rate. It is the power change rate of the lithium battery. This indicates the number of times the lithium battery's high power changes during the entire operating condition; The objective function for minimizing system energy consumption is: minf2=E HESS =∫(P bat +P sc )dt In the formula, It is the real-time power of the lithium battery module. This is the real-time power of the supercapacitor module; Therefore, multi-objective optimization can be further expressed as: minf (x) =[f1,f2]=[rN(ΔP bat ),∫(P bat +P sc )dt] In the formula, and To optimize the objective, the smaller the objective function value, the closer the obtained solution is to the optimal solution; The constraints that the optimization variables must satisfy include: SOC constraints for lithium batteries and supercapacitors In the formula, lithium battery The minimum value, lithium battery The maximum value, For supercapacitors The minimum value, For supercapacitors The maximum value; Instantaneous power balance constraint In the formula, It is the instantaneous output power of the lithium battery. It is the instantaneous output power of the supercapacitor. It is the instantaneous output power of the fuel cell. It is the power demand at any given moment; Energy Constraints: In the formula, It is the energy output of the lithium battery. It is the output power of the supercapacitor. This is the total energy required for composite energy storage; Supercapacitor voltage constraint In the formula, This represents the maximum operating voltage of the supercapacitor; The specific process of encoding the membership function parameters of the fuzzy controller in step 2) is as follows: The fuzzy controller includes three input parameters and one output parameter. The three input parameters are the power of the lithium battery, etc. lithium batteries and supercapacitors The output parameter is the power regulation coefficient k of the lithium battery module, and all four parameters are represented by Gaussmf. The expression for the Gaussian membership function is: x Let c be the independent variable, and let c represent the position of the Gaussian function in the fuzzy universe of discourse. This indicates the magnitude and coverage of the Gaussian function. The output parameter k of the fuzzy controller is divided by 5 Gaussian functions, and the input parameter is the power of the lithium battery. The SOC value of lithium batteries is divided by three Gaussian functions. The SOC value of a supercapacitor is determined by five Gaussian functions. Divided by 5 Gaussian functions, therefore the optimization parameters c, Represented as: To ensure the relationships between the various fuzzy subsets, enable the fuzzy logic controller to function properly, and reduce computation time, the positional constraints of the Gaussian function are given: Floating-point encoding is used to encode each parameter of the membership function. Each Gaussian function is encoded by c and... With two parameters determined, and a total of 18 Gaussian functions for the 1 output parameter and 3 input parameters to be optimized, there are a total of 36 parameters to be optimized. If each parameter to be optimized is represented by a 7-bit binary number, then each sample in the population has 7 × 36 = 252 bits.

2. The hybrid power ship energy management method based on genetic algorithm optimization of fuzzy controller membership function as described in claim 1, characterized in that: Step 3) establishes the mathematical model and its constraints based on steps 1) and 2) as follows: In the formula, , These are the weighting coefficients.

3. The hybrid power ship energy management method based on genetic algorithm optimization of fuzzy controller membership function as described in claim 2, characterized in that: Step 4) Establishing the fitness function for 。 4. The hybrid power ship energy management method based on genetic algorithm optimization of fuzzy controller membership function according to claim 3, characterized in that: Step 5) specifically involves: calculating the optimization index value for each sample based on the genetic algorithm, mathematical model, and constraints. Let i = 1, 2, and then sort the results according to their size. The smaller the value, the better the sample.

Images