A fuel cell ship energy management method based on improved global optimization algorithm

Abstract

The present invention belongs to the field of energy management and optimization control technology, and discloses a fuel cell ship energy management method based on an improved global optimization algorithm, comprising the following steps: constructing a topological structure of a fuel cell ship system; establishing an objective function based on the hydrogen consumption of the fuel cell ship system and the state of charge of the lithium battery; constructing an equivalent minimum hydrogen consumption control strategy; and optimizing the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy using an improved ant lion optimization algorithm. The present invention adopts the above-mentioned fuel cell ship energy management method based on an improved global optimization algorithm, aiming to combine the advantages of instantaneous optimization and global optimization, meet the dual needs of high fluctuating loads and fuel economy, and extend the service life of batteries while reducing fuel consumption. By using fuel cells as the main power source, the advantage of zero emissions is fully utilized to achieve carbon emission reduction and energy transformation.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management and optimization control, and in particular to a fuel cell ship energy management method based on an improved global optimization algorithm. Background Art

[0002] In recent years, the application of clean energy in the marine sector has become a research hotspot. Clean energy sources such as hydrogen, wind, and solar energy are widely used in modern marine propulsion systems. In the energy management of multi-energy coupled marine propulsion systems, power allocation under different operating modes is a key factor affecting performance, and therefore the development of relevant energy management strategies has become a research priority. Currently, new energy ships generally utilize hybrid energy storage systems, consisting of batteries and supercapacitors of specified capacities, which are flexibly configured to improve overall efficiency. For high-power vessels (such as tugboats and passenger ships) and for diverse operational scenarios (such as cruising, berthing, and offshore), fuel cell ships require multi-stack hydrogen fuel cell systems (MFCSs) and multiple energy storage systems. Given the unique operational scenarios of ships, hybrid energy storage systems without effective energy management can compromise overall efficiency, service life, and safety. Therefore, developing an optimized multi-electric energy management system (EMS) for ships to achieve optimal hydrogen consumption, ship life, and safety and stability is key to solving the power coordination problem of multiple power sources.

[0003] Currently, energy management strategies are primarily categorized into three types: rule-based, optimization-based, and intelligent optimization algorithm-based. Rule-based energy management strategies select the hybrid system operating mode based on operating conditions, battery status, and parameters to ensure optimal operation. They formulate strategies based on battery status and operating conditions, utilizing batteries to supplement fuel cell output and improve overall efficiency. While rule-based strategies have simple control logic and are easy to implement, they rely on experience and react slowly to system state changes, making optimal control difficult to achieve. Energy management strategies based on intelligent optimization algorithms, such as machine reinforcement learning, deep learning, and optimal control, enable more flexible energy management strategies and utilize battery state of charge as a metric to improve battery life. While intelligent optimization algorithm-based strategies can achieve good control results, they are challenging to design, require high-quality ship models, and have limited training datasets. Therefore, there is still considerable room for exploration. Optimization-based strategies can be categorized into global optimization control strategies and instantaneous optimization control strategies. Global optimization strategies focus on long-term system performance, while instantaneous optimization strategies prioritize optimal performance at a specific moment. The three mainstream management strategies each have their own advantages and disadvantages: rule-based strategies are simple, easy to implement and highly stable, but rely on experience; strategies based on intelligent optimization algorithms are difficult to design and have high technical requirements; and although optimization-based strategies have higher algorithm complexity, they have better optimization effects and greater potential.

[0004] Therefore, the current problem in the development of ship technology is that traditional ships are driven by diesel engines, which cause serious pollution and loud noise, while ships powered by clean energy lack effective energy management and hybrid energy storage systems, which may affect overall efficiency, service life and safety. Summary of the Invention

[0005] The purpose of the present invention is to provide a fuel cell ship energy management method based on an improved global optimization algorithm, which aims to combine the advantages of instantaneous optimization and global optimization to meet the dual needs of high fluctuating loads and fuel economy, while reducing fuel consumption and extending the service life of batteries. By using fuel cells as the main power source, the zero-emission advantage is fully utilized to achieve carbon emission reduction and energy transformation.

[0006] To achieve the above objectives, the present invention provides a fuel cell ship energy management method based on an improved global optimization algorithm, comprising the following steps:

[0007] Step S1, constructing a topological structure of a fuel cell ship system;

[0008] Step S2: establishing an objective function based on the hydrogen consumption and lithium battery state of charge of the fuel cell ship system;

[0009] Step S3: constructing an equivalent minimum hydrogen consumption control strategy ECMS; wherein the input variables are the load power of the fuel cell ship system and the state of charge of the lithium battery; and the output variables are the output power of the fuel cell ship system and the output power of the lithium battery;

[0010] Step S4: Optimize the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy using the improved ant lion optimization algorithm IWALO.

[0011] Preferably, in step S1, the topology structure of the fuel cell ship system is constructed, and the specific process is as follows:

[0012] Step S11: Determine the fuel cell model as follows:

[0013] ;

[0014] in, is the output voltage; is the open circuit voltage; N is the number of cells in the fuel cell; A is the Tafel slope; is the output current; is the exchange current; is the internal resistance;

[0015] Open circuit voltage , as shown below:

[0016] ;

[0017] in, is the fuel cell potential; is the voltage constant;

[0018] The fuel cell potential is given by:

[0019] ;

[0020] Where R is the gas constant; T is the operating temperature; F is the Faraday constant; for partial pressure; for partial pressure;

[0021] Step S12: Determine the relationship between the hydrogen consumption rate and the output power of the fuel cell based on the output voltage of the fuel cell and the efficiency of the fuel cell ship system, as shown below:

[0022] ;

[0023] in, is the fuel cell hydrogen consumption rate, in g / s; Output power for the fuel cell; is the fuel cell single chip voltage; is the Faraday constant; The electrical efficiency of the fuel cell ship system;

[0024] Step S13: lithium battery model, as shown below:

[0025] ;

[0026] in, is the lithium battery voltage; is the cumulative discharge capacity; is the current; is the constant voltage of lithium battery; K is the polarization constant; is the battery capacity; is the filtered lithium battery current; is the amplitude of the exponential region; B is the inverse of the time constant of the exponential region; is the internal resistance of the lithium battery;

[0027] Step S14: The calculation formula for lithium battery hydrogen consumption is as follows:

[0028] ;

[0029] in, is the equivalent hydrogen consumption of lithium batteries; Output power for lithium battery; and are the charge and discharge efficiency of lithium batteries; and are the average charge and discharge efficiency of lithium batteries; is the average hydrogen consumption rate of the fuel cell; is the average output power of the fuel cell.

[0030] Preferably, in step S2, an objective function based on the hydrogen consumption of the fuel cell ship system and the state of charge of the lithium battery is established; with the optimization of the hydrogen consumption of the fuel cell ship system and the maintenance of the battery power output as the optimization objectives, an optimization function of the fuel cell ship system is established, that is, minimizing the equivalent hydrogen consumption objective function. , as shown below:

[0031] ;

[0032] in, is the time interval; is the fixed equivalent coefficient of the equivalent minimum hydrogen consumption control strategy; is the dynamic equivalent coefficient of the equivalent minimum hydrogen consumption control strategy.

[0033] Preferably, in step S3, an equivalent minimum hydrogen consumption control strategy ECMS is constructed; wherein the input variables are the load power of the fuel cell ship system and the state of charge of the lithium battery; and the output variables are the output power of the fuel cell ship system and the output power of the lithium battery;

[0034] From the objective function of minimizing equivalent hydrogen consumption, it can be seen that different dynamic equivalent coefficients will affect the power distribution during operation. The power adjustment function of the dynamic equivalent coefficient is as follows:

[0035] ;

[0036] in, It is the regulation parameter of the lithium battery charge state; and They are the upper and lower limits of the lithium battery state of charge; is the load fluctuation regulation parameter; is the current load fluctuation amplitude; is the total load power demand.

[0037] Preferably, It is the charge state adjustment item of the lithium battery; Used to balance the state of charge of lithium batteries during operation;

[0038] It is used to measure the degree to which the current lithium battery charge state deviates from the center; when the lithium battery charge state is in the middle range, It tends to reduce the power output of lithium batteries; when the state of charge of lithium batteries is close to the upper and lower limits, Tends to reduce lithium battery load.

[0039] Preferably, It is the load fluctuation adjustment item; Used to measure the impact of load fluctuations on the dynamic equivalent coefficient. When the load fluctuation is large, the load fluctuation adjustment item will increase to tend to use more fuel cell power and reduce the pressure on the battery to respond quickly. When the load is stable, the impact of the load fluctuation adjustment item on the dynamic equivalent coefficient will decrease.

[0040] The lithium battery charge state adjustment item controls the output priority of the lithium battery and the fuel cell under different lithium battery charge state ranges; the load fluctuation adjustment item further dynamically adjusts the power distribution ratio of the lithium battery and the fuel cell according to load changes.

[0041] Preferably, the constraints of the equivalent minimum hydrogen consumption control strategy ECMS are as follows:

[0042] ;

[0043] in, and is the maximum output power and minimum output power of the fuel cell; and is the maximum output power and minimum output power of the lithium battery; is the current total load power demand; is the output power of the lithium battery at time t; is the output power of the fuel cell at time t.

[0044] Preferably, in step S4, the improved ant lion optimization algorithm IWALO is used to optimize the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy ECMS, that is, to optimize the dynamic equivalent coefficient in the equivalent minimum hydrogen consumption control strategy ECMS. ,The improved ant lion optimization algorithm IWALO is used to seek the optimal solution for the dynamic equivalent coefficient. The specific process is as follows:

[0045] Step S41: First, construct a fitness function as follows:

[0046] ;

[0047] in, is the fitness value; is the output power of the lithium battery at time t; is the output power of the fuel cell at time t; is the dynamic equivalent coefficient, which is used to dynamically adjust the power distribution weight between the fuel cell and the battery; is the time step; T is the total time step;

[0048] Step S42: In the Ant Lion Optimization Algorithm, the initial parameters The positions of ants and ant lions are randomly distributed in the parameter space after initialization; the random walk of ants in nature is shown below:

[0049] ;

[0050] Where D(t) is the set of ant's walking steps; t is the current iteration number; is the maximum number of iterations; rand is a random number in the range [0,1];

[0051] Normalize the number of random walk steps D(t) as follows:

[0052] ;

[0053] in, is the normalized step size to ensure that the value is within a reasonable range; and are the upper and lower bounds of the i-dimensional search space respectively;

[0054] Step S43: Based on the ant lion optimization algorithm, an adaptive factor is introduced to balance early exploration and late exploitation, and to dynamically adjust the ants' reliance on random walks and elite positions; the early stage is mainly exploration, while the late stage is mainly exploitation;

[0055] Adaptive factors are as follows:

[0056] ;

[0057] When t approaches 0, it is an early iteration: the adaptive factor is close to 1, emphasizing exploration, allowing the algorithm to conduct extensive searches in the entire search space;

[0058] When t approaches Tmax, it is the late iteration: the adaptive factor is gradually reduced to 0.2, emphasizing development, so that the algorithm focuses on fine search near the optimal solution;

[0059] Step S44: Select the position of the ant lion according to the fitness value of the ant, so that it moves to a better solution;

[0060] Using the elite strategy, the ant lion with the best fitness value is selected in each iteration, as shown below:

[0061] ;

[0062] in, is the fitness value of the qth ant at the tth time; It is the optimal solution for fitness in the current iteration; the optimal ant lion position always participates in the population update of the next iteration;

[0063] Step S45: Under the combined effects of the roulette wheel method and the elite strategy, combined with the adaptive factor, the positions of the ants are updated as follows:

[0064] ;

[0065] in, is the new position of the qth ant in the t+1 iteration; is the position of the ant at the tth iteration, when it moves around the antlion according to the roulette wheel method; is the position of the ant at the tth iteration when it moves around the ant lion according to the elite strategy;

[0066] Step S46: Determine whether the algorithm has converged to the optimal solution: When the maximum number of iterations is reached or the change in fitness value meets the convergence condition, the iteration is stopped; when the fitness value of the ant lion exceeds that of the ant, the ant lion will capture the ant and output the optimal parameters;

[0067] Step S47: Based on the above optimization process, the fuel cell output power is matched with the lithium battery output power through the optimized dynamic equivalent coefficient to achieve the global real-time pursuit of the minimum consumption objective function. , as shown below:

[0068] ;

[0069] in, is the total hydrogen consumption of the fuel cell ship system; is the total hydrogen consumption of the fuel cell, which is calculated by accumulating the hydrogen consumption rate of the fuel cell get; is the lower calorific value of hydrogen; is the optimal equivalent coefficient; and The optimal output power for fuel cells and lithium batteries respectively.

[0070] Therefore, the present invention adopts the above-mentioned fuel cell ship energy management method based on the improved global optimization algorithm, and the beneficial effects are as follows:

[0071] (1) The present invention ensures the accuracy of system optimization by constructing the topological structure and mathematical model of the fuel cell ship system;

[0072] (2) The present invention considers factors such as fuel consumption and lithium battery status to establish an objective function, thereby improving the economy and durability of the system and ensuring that energy management is more efficient and reliable.

[0073] (3) This invention constructs a multi-objective optimization model, taking into account the fuel consumption and optimal distribution of battery power of fuel cell ships, aiming to extend the life of fuel cells, maintain the health of lithium batteries, reduce hydrogen costs, and optimize the energy management strategy of fuel cell ships.

[0074] (4) The present invention uses an improved ant lion optimization algorithm to optimize the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy, which speeds up the system's response to rapid load changes, meets real-time power requirements, and effectively improves the accuracy and efficiency of energy management of the fuel cell ship system, ensuring the economy and durability of the system, thereby improving the control quality and economy of the system.

[0075] (5) The present invention ensures that the fuel cell operates in an optimal state, prolongs its service life and reduces maintenance costs, enhances the adaptability and stability of the fuel cell power system in different flight phases and environments, and helps to perform better in various missions.

[0076] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 This is a flow chart of a fuel cell ship energy management method based on an improved global optimization algorithm of the present invention;

[0078] Figure 2 3 is a comparison diagram of convergence curves of different algorithms in the embodiments of the present invention. DETAILED DESCRIPTION

[0079] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0080] like Figure 1 As shown, the present invention provides a fuel cell ship energy management method based on an improved global optimization algorithm, comprising the following steps:

[0081] Step S1, constructing a topological structure of a fuel cell ship system;

[0082] Step S2: establishing an objective function based on hydrogen consumption and lithium battery state of charge (SOC) of the fuel cell ship system;

[0083] Step S3: constructing an equivalent minimum hydrogen consumption control strategy (ECMS); wherein the input variables are the load power of the fuel cell ship system and the state of charge of the lithium battery; and the output variables are the output power of the fuel cell ship system and the output power of the lithium battery;

[0084] Step S4: Optimize the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy (ECMS) using the improved ant lion optimization algorithm (IWALO) to achieve real-time optimal power distribution to meet the dual requirements of high fluctuating load and fuel economy.

[0085] Example

[0086] The present invention provides a fuel cell ship energy management method based on an improved global optimization algorithm, comprising the following steps:

[0087] Step S1: constructing a topological structure of a fuel cell ship system.

[0088] Step S11: Determine the fuel cell model.

[0089] The ideal open-circuit voltage of a fuel cell is calculated using the Nynes equation, which is affected by the partial pressures of the reactants and products, temperature, and reactant concentrations. In practice, the output voltage of a fuel cell is described as a function of the Nynes voltage, activation voltage, ohmic voltage drop, and concentration voltage.

[0090] Therefore, the fuel cell model is as follows:

[0091] ;

[0092] in, is the output voltage (V); is the open circuit voltage (V); N is the number of cells in the fuel cell; A is the Tafel slope; is the output current (A); is the exchange current (A); is the internal resistance (Ω).

[0093] Open circuit voltage , as shown below:

[0094] ;

[0095] in, is the fuel cell potential; is a voltage constant determined by the internal current and the Tafel slope under rated operating conditions, and its value is less than or equal to 1.

[0096] The fuel cell potential is given by:

[0097] ;

[0098] Where R is the gas constant (8.3145 J / mol K); T is the operating temperature (Kelvin); F is the Faraday constant (96485 A s / mol); for partial pressure; for partial pressure.

[0099] Step S12: hydrogen consumption rate of the fuel cell.

[0100] Fuel cells have time-varying and nonlinear characteristics. Based on the fuel cell output voltage and the fuel cell ship system efficiency, the relationship between the fuel cell hydrogen consumption rate and output power is determined as follows:

[0101] ;

[0102] in, is the fuel cell hydrogen consumption rate, in g / s; Output power for the fuel cell; is the fuel cell single chip voltage; is the Faraday constant; is the electrical efficiency of the fuel cell ship system.

[0103] Step S13: lithium battery model, as shown below:

[0104] ;

[0105] in, is the lithium battery voltage; is the cumulative discharge capacity; is the current; is the constant voltage of lithium battery; K is the polarization constant; Q is the battery capacity; is the filtered lithium battery current; is the amplitude of the exponential region; B is the inverse of the time constant of the exponential region; is the internal resistance of the lithium battery.

[0106] Step S14: The calculation formula for lithium battery hydrogen consumption is as follows:

[0107] ;

[0108] in, is the equivalent hydrogen consumption of lithium batteries; Output power for lithium battery; and are the charge and discharge efficiency of lithium batteries; and are the average charge and discharge efficiency of lithium batteries; is the average hydrogen consumption rate of the fuel cell; is the average output power of the fuel cell.

[0109] Step S2: establishing an objective function based on the hydrogen consumption of the fuel cell ship system and the state of charge (SOC) of the lithium battery.

[0110] With the optimization of hydrogen consumption of fuel cell ship system and maintaining battery power output as the optimization goal, the optimization function of fuel cell ship system is established, that is, the objective function of minimizing equivalent hydrogen consumption. , as shown below:

[0111] ;

[0112] in, is the time interval; is the fixed equivalent coefficient of the equivalent minimum hydrogen consumption control strategy (ECMS), used for preliminary power allocation and also responsible for benchmark setting; It is the dynamic equivalent coefficient of the equivalent minimum hydrogen consumption control strategy (ECMS), which is adjusted in real time according to load fluctuations and the charge state of the lithium battery, and plays a role in balancing and optimizing the power distribution between the fuel cell and the lithium battery.

[0113] Step S3: constructing an equivalent minimum hydrogen consumption control strategy (ECMS); wherein the input variables are the load power of the fuel cell ship system and the state of charge of the lithium battery; and the output variables are the output power of the fuel cell ship system and the output power of the lithium battery;

[0114] From the objective function of minimizing equivalent hydrogen consumption, it can be seen that different dynamic equivalent coefficients will affect the power distribution during operation. Its power adjustment function is as follows:

[0115] ;

[0116] in, It is the regulation parameter of the lithium battery charge state; and They are the upper and lower limits of the lithium battery state of charge; is the load fluctuation regulation parameter; is the current load fluctuation amplitude; is the total load power demand.

[0117] It is the charge state adjustment item of the lithium battery; Used to balance the state of charge of lithium batteries during operation; It is used to measure the degree to which the current lithium battery charge state deviates from the center; when the lithium battery charge state is in the middle range, It tends to reduce the power output of lithium batteries; when the state of charge of lithium batteries is close to the upper and lower limits, Tends to reduce lithium battery load.

[0118] It is the load fluctuation adjustment item; Used to measure the impact of load fluctuations on the dynamic equivalent coefficient. When the load fluctuation is large, the load fluctuation adjustment item will increase to tend to use more fuel cell power and reduce the pressure of rapid battery response. When the load is stable, the impact of the load fluctuation adjustment item on the dynamic equivalent coefficient will decrease.

[0119] The lithium battery charge state adjustment item controls the output priority of the lithium battery and the fuel cell under different lithium battery charge state ranges. The load fluctuation adjustment item further dynamically adjusts the power distribution ratio of the lithium battery and the fuel cell according to load changes.

[0120] Among them, the constraints of the equivalent minimum hydrogen consumption control strategy (ECMS) are as follows:

[0121] ;

[0122] in, and is the maximum output power and minimum output power of the fuel cell; and The maximum output power and minimum output power of the lithium battery; the maximum charge state of the lithium battery is 80%, and the minimum is 30%; is the current total load power demand; is the output power of the lithium battery at time t; is the output power of the fuel cell at time t.

[0123] According to the ECMS strategy, the optimal fuel cell power can be calculated as a function of the load power and the lithium battery state of charge. This optimal fuel cell power is limited between the minimum and maximum fuel cell powers to avoid operation in the low efficiency area.

[0124] Subtracting the calculated fuel cell power from the required load power determines the battery charge. The required current for each system is then calculated by dividing the fuel cell power and battery power by the voltage.

[0125] Step S4: Optimize the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy (ECMS) using the improved ant lion optimization algorithm (IWALO) to achieve real-time optimal power distribution to meet the dual requirements of high fluctuating load and fuel economy.

[0126] To optimize the dynamic equivalent coefficient in the equivalent minimum hydrogen consumption control strategy (ECMS) ,The improved ant lion optimization algorithm (IWALO) is used to seek the optimal solution for the dynamic equivalent coefficient.

[0127] Step S41: First, construct a fitness function as follows:

[0128] ;

[0129] in, is the fitness value; is the output power of the lithium battery at time t; is the output power of the fuel cell at time t; is the dynamic equivalent coefficient, which is used to dynamically adjust the power distribution weight between the fuel cell and the battery; is the time step; T is the total time step.

[0130] Step S42: In the Ant Lion Optimization Algorithm, the initial parameters The positions of ants and ant lions need to be randomly distributed in the parameter space after initialization. The random walk of ants in nature is as follows:

[0131] ;

[0132] Where D(t) is the set of ant's walking steps; t is the current iteration number; is the maximum number of iterations; rand is a random number in the range [0,1].

[0133] In the Ant Lion Optimization Algorithm, D(t) is unbounded in its original form. Therefore, if used directly, it may cause the ants' positions to exceed the parameter boundaries defined by the problem. Therefore, normalization is required to prevent the ants from wandering beyond the boundaries. To solve this problem, the number of random walk steps D(t) is normalized to the current boundary range, as shown below:

[0134] ;

[0135] in, is the normalized step size to ensure that the value is within a reasonable range; and are the upper and lower bounds of the i-dimensional search space respectively.

[0136] Step S43: To meet the requirements of ship energy management, an adaptive factor is introduced based on the ant lion optimization algorithm to balance early exploration and late exploitation, and to dynamically adjust the ants' dependence on random walks and elite positions; the early stage is mainly exploration, while the late stage is mainly exploitation.

[0137] Adaptive factors are as follows:

[0138] ;

[0139] When t≈0, it is an early iteration: the adaptation factor is close to 1, emphasizing exploration and enabling the algorithm to search extensively throughout the search space.

[0140] When t≈Tmax, it is the late iteration: the adaptive factor is gradually reduced to 0.2, emphasizing development, so that the algorithm focuses on fine search near the optimal solution.

[0141] Through testing of a large number of benchmark functions (such as Sphere, Ackley, and Rastrigin functions), the present invention shows that the performance of an adaptive factor of 0.8 is better than other coefficients in terms of the balance between exploration and exploitation.

[0142] Step S44: Select the position of the ant lion according to the fitness value of the ant, so that it moves to a better solution.

[0143] Using the elite strategy, the ant lion with the best fitness value is selected in each iteration, as shown below:

[0144] ;

[0145] in, is the fitness value of the qth ant at the tth time; Is the optimal solution for fitness in the current iteration. The optimal ant lion position always participates in the population update of the next iteration.

[0146] Step S45: Under the combined effects of the roulette wheel method and the elite strategy, combined with the adaptive factor, the positions of the ants are updated as follows:

[0147] ;

[0148] in, is the new position of the qth ant in the t+1 iteration; is the position of the ant at the tth iteration, when it moves around the antlion according to the roulette wheel method; It is the position of the ant at the tth iteration when it moves around the ant lion according to the elite strategy.

[0149] Step S46: Determine whether the algorithm has converged to the optimal solution. If the maximum number of iterations is reached or the change in fitness value meets the convergence condition, the iteration stops. When the fitness value of the ant lion exceeds that of the ant, the ant lion will capture the ant and output the optimal parameters.

[0150] Step S47: Based on the above optimization process, the fuel cell output power is matched with the lithium battery output power through the optimized dynamic equivalent coefficient to achieve the global real-time pursuit of the minimum consumption objective function. , as shown below:

[0151] ;

[0152] in, is the total hydrogen consumption of the fuel cell ship system; is the total hydrogen consumption of the fuel cell, which is calculated by accumulating the hydrogen consumption rate of the fuel cell get; is the lower calorific value of hydrogen; is the optimal equivalent coefficient; and The optimal output power for fuel cells and lithium batteries respectively.

[0153] Under traditional ECMS control strategies, the fixed equivalent coefficient is typically a constant value. Different dynamic equivalent coefficients can significantly impact the control effectiveness of the ECMS strategy. Therefore, selecting the appropriate dynamic equivalent coefficient for different operating conditions and modes is a key factor in achieving optimal control with the ECMS strategy. Effective dynamic equivalent coefficients can yield optimal allocation solutions under different constraints.

[0154] like Figure 2 Figure 2 shows a comparison of the convergence curves of the improved IWALO algorithm proposed in this paper with those of traditional ALO, PSO, GA, and SSA optimization algorithms under the same test conditions. IWALO, represented by the solid red line, rapidly converges to 8.08e-13 after 250 optimization iterations, a six-order-of-magnitude improvement in convergence accuracy compared to ALO's 3.12e-7. Furthermore, compared to traditional intelligent optimization algorithms such as PSO, GA, and SSA, IWALO maintains a superior convergence trend, avoiding the oscillation observed in SSA and GA in the late stages of convergence.

[0155] Therefore, the present invention adopts the improved IWALO algorithm as the optimization solution strategy, which improves the optimization effect of the energy management strategy and reduces the hydrogen consumption of the fuel cell.

[0156] Therefore, the present invention adopts the above-mentioned fuel cell ship energy management method based on the improved global optimization algorithm, and ensures the accuracy of system optimization by constructing the topological structure and mathematical model of the fuel cell ship system; considers factors such as fuel consumption and lithium battery status, establishes an objective function, thereby improving the economy and durability of the system, and ensuring that energy management is more efficient and reliable; by constructing a multi-objective optimization model, considers the fuel cell ship's fuel consumption and the optimal distribution of battery power, aiming to extend the life of the fuel cell, maintain the health of the lithium battery, and reduce the hydrogen cost, and optimize the energy management strategy of the fuel cell ship; adopts the improved ant lion optimization algorithm to optimize the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy, accelerates the system's response speed to rapid load changes, meets real-time power requirements, effectively improves the accuracy and efficiency of the fuel cell ship system's energy management, ensures the economy and durability of the system, and thus improves the system's control quality and economy; the present invention ensures that the fuel cell operates in the best state, extends its service life and reduces maintenance costs, enhances the adaptability and stability of the fuel cell power system in different flight phases and environments, and helps to perform better in various tasks.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A fuel cell ship energy management method based on an improved global optimization algorithm, characterized in that: The following steps are involved: Step S1, constructing a topological structure of a fuel cell ship system; Step S2: establishing an objective function based on the hydrogen consumption and lithium battery state of charge of the fuel cell ship system; Step S3: constructing an equivalent minimum hydrogen consumption control strategy ECMS; wherein the input variables are the load power of the fuel cell ship system and the state of charge of the lithium battery; and the output variables are the output power of the fuel cell ship system and the output power of the lithium battery; From the objective function of minimizing equivalent hydrogen consumption, it can be seen that different dynamic equivalent coefficients will affect the power distribution during operation. The power adjustment function of the dynamic equivalent coefficient is as follows: ； in, It is the regulation parameter of the lithium battery charge state; and They are the upper and lower limits of the lithium battery state of charge; is the load fluctuation regulation parameter; is the current load fluctuation amplitude; is the total load power demand; Step S4: Optimize the equivalent coefficient of the equivalent minimum hydrogen consumption control strategy using the improved ant lion optimization algorithm IWALO, that is, optimize the dynamic equivalent coefficient in the equivalent minimum hydrogen consumption control strategy ECMS. ,The improved ant lion optimization algorithm IWALO is used to seek the optimal solution for the dynamic equivalent coefficient. The specific process is as follows: Step S41: First, construct a fitness function as follows: ； in, is the fitness value; is the output power of the lithium battery at time t; is the output power of the fuel cell at time t; is the dynamic equivalent coefficient, which is used to dynamically adjust the power distribution weight between the fuel cell and the battery; is the time step; T is the total time step; Step S42: In the Ant Lion Optimization Algorithm, the initial parameters The positions of ants and ant lions are randomly distributed in the parameter space after initialization; the random walk of ants in nature is shown below: ； Where D(t) is the set of ant's walking steps; t is the current iteration number; is the maximum number of iterations; rand is a random number in the range [0,1]; Normalize the number of random walk steps D(t) as follows: ； in, is the normalized step size to ensure that the value is within a reasonable range; and are the upper and lower bounds of the i-dimensional search space respectively; Step S43: Based on the ant lion optimization algorithm, an adaptive factor is introduced to balance early exploration and late exploitation, and to dynamically adjust the ants' reliance on random walks and elite positions; the early stage is mainly exploration, while the late stage is mainly exploitation; Adaptive factors are as follows: ； When t approaches 0, it is an early iteration: the adaptive factor is close to 1, emphasizing exploration, allowing the algorithm to conduct extensive searches in the entire search space; When t approaches Tmax, it is the late iteration: the adaptive factor is gradually reduced to 0.2, emphasizing development, so that the algorithm focuses on fine search near the optimal solution; Step S44: Select the position of the ant lion according to the fitness value of the ant, so that it moves to a better solution; Using the elite strategy, the ant lion with the best fitness value is selected in each iteration, as shown below: ； in, is the fitness value of the qth ant at the tth time; It is the optimal solution for fitness in the current iteration; the optimal ant lion position always participates in the population update of the next iteration; Step S45: Under the combined effects of the roulette wheel method and the elite strategy, combined with the adaptive factor, the positions of the ants are updated as follows: ； in, is the new position of the qth ant in the t+1 iteration; is the position of the ant at the tth iteration, when it moves around the antlion according to the roulette wheel method. is the position of the ant at the tth iteration when it moves around the ant lion according to the elite strategy; Step S46: Determine whether the algorithm has converged to the optimal solution: When the maximum number of iterations is reached or the change in fitness value meets the convergence condition, the iteration is stopped; when the fitness value of the ant lion exceeds that of the ant, the ant lion will capture the ant and output the optimal parameters; Step S47: Based on the above optimization process, the fuel cell output power is matched with the lithium battery output power through the optimized dynamic equivalent coefficient to achieve the global real-time pursuit of the minimum consumption objective function. , as shown below: ； in, is the total hydrogen consumption of the fuel cell ship system; is the total hydrogen consumption of the fuel cell, which is calculated by accumulating the hydrogen consumption rate of the fuel cell get; is the lower calorific value of hydrogen; is the optimal equivalent coefficient; and The optimal output power for fuel cells and lithium batteries respectively.

2. A fuel cell ship energy management method based on an improved global optimization algorithm according to claim 1, characterized in that: In step S1, the topology of the fuel cell ship system is constructed. The specific process is as follows: Step S11: Determine the fuel cell model as follows: ； in, is the output voltage; is the open circuit voltage; N is the number of cells in the fuel cell; A is the Tafel slope; is the output current; is the exchange current; is the internal resistance; Open circuit voltage , as shown below: ； in, is the fuel cell potential; is the voltage constant; The fuel cell potential is given by: ； Where R is the gas constant; T is the operating temperature; F is the Faraday constant; for partial pressure; for partial pressure; Step S12: Determine the relationship between the hydrogen consumption rate and the output power of the fuel cell based on the output voltage of the fuel cell and the efficiency of the fuel cell ship system, as shown below: ； in, is the fuel cell hydrogen consumption rate, in g / s; Output power for the fuel cell; is the fuel cell single chip voltage; is the Faraday constant; The electrical efficiency of the fuel cell ship system; Step S13: lithium battery model, as shown below: ； in, is the lithium battery voltage; is the cumulative discharge capacity; is the current; is the constant voltage of lithium battery; K is the polarization constant; Q is the battery capacity; is the filtered lithium battery current; is the amplitude of the exponential region; B is the inverse of the time constant of the exponential region; is the internal resistance of the lithium battery; Step S14: The calculation formula for lithium battery hydrogen consumption is as follows: ； in, is the equivalent hydrogen consumption of lithium batteries; Output power for lithium battery; and are the charge and discharge efficiency of lithium batteries; and are the average charge and discharge efficiency of lithium batteries; is the average hydrogen consumption rate of the fuel cell; is the average output power of the fuel cell.

3. The fuel cell ship energy management method based on the improved global optimization algorithm according to claim 1 is characterized in that: In step S2, an objective function based on the hydrogen consumption of the fuel cell ship system and the state of charge of the lithium battery is established; with the optimization of the hydrogen consumption of the fuel cell ship system and the maintenance of the battery power output as the optimization goal, an optimization function of the fuel cell ship system is established, that is, the objective function of minimizing the equivalent hydrogen consumption is established. , as shown below: ； in, is the time interval; is the fixed equivalent coefficient of the equivalent minimum hydrogen consumption control strategy; is the dynamic equivalent coefficient of the equivalent minimum hydrogen consumption control strategy.

4. The fuel cell ship energy management method based on the improved global optimization algorithm according to claim 1 is characterized in that: It is the charge state adjustment item of the lithium battery; Used to balance the state of charge of lithium batteries during operation; Used to measure the degree to which the current lithium battery charge state deviates from the center; When the lithium battery state of charge is in the middle range, Tends to reduce lithium battery power output; When the lithium battery charge state is close to the upper and lower limits, Tends to reduce lithium battery load.

5. The fuel cell ship energy management method based on the improved global optimization algorithm according to claim 1 is characterized in that: It is the load fluctuation adjustment item; Used to measure the impact of load fluctuations on the dynamic equivalent coefficient. When the load fluctuation is large, the load fluctuation adjustment item will increase to tend to use more fuel cell power and reduce the pressure on the battery to respond quickly. When the load is stable, the impact of the load fluctuation adjustment item on the dynamic equivalent coefficient will decrease. The lithium battery charge state adjustment item controls the output priority of the lithium battery and the fuel cell under different lithium battery charge state ranges; the load fluctuation adjustment item further dynamically adjusts the power distribution ratio of the lithium battery and the fuel cell according to load changes.

6. The fuel cell ship energy management method based on the improved global optimization algorithm according to claim 1 is characterized in that: The constraints of the equivalent minimum hydrogen consumption control strategy ECMS are as follows: ； in, and is the maximum output power and minimum output power of the fuel cell; and is the maximum output power and minimum output power of the lithium battery; is the current total load power demand; is the output power of the lithium battery at time t; is the output power of the fuel cell at time t.

Images