A method and system for optimizing the efficiency of a hydrogen battery of an unmanned aerial vehicle

Abstract

The present invention relates to the field of optimization control technology, and specifically to a method and system for optimizing the efficiency of hydrogen batteries of unmanned aerial vehicles, including the following steps: collecting flight state data of the unmanned aerial vehicle currently including speed and altitude, as well as power data of the hydrogen battery, performing standardized processing, and analyzing and determining the energy requirements of each battery unit to construct a preliminary energy distribution model. In the present invention, by analyzing the flight state data of the unmanned aerial vehicle in detail and establishing an initial energy distribution model, the specific energy requirements of each battery unit can be more accurately met, thereby reducing the ineffective consumption of energy. The real-time updated energy transmission path, combined with the dynamic adjustment of flight requirements, improves the energy utilization efficiency. The minimum cut algorithm is introduced to optimize the energy supply of key nodes, ensuring the operation stability and reliability of the system at critical moments. The control parameters optimized by the differential evolution algorithm allow for fine adjustment of the battery discharge rate and power output to adapt to environmental changes and mission requirements.

Description

Technical Field

[0001] The present invention relates to the field of optimization control technology, and particularly to a method and system for optimizing the efficiency of a drone hydrogen battery. Background Art

[0002] Optimization control is a branch of automatic control engineering that focuses on developing and applying algorithms to improve the performance and efficiency of systems. This technology mainly utilizes various mathematical optimization methods, including linear and non - linear programming, dynamic programming, and model - based predictive control, etc. The goal of optimization control is to find the optimal control strategy, so that the system operates in the best performance state, reduces energy consumption, and improves the reliability and stability of the system. It is widely applied in multiple industries, including but not limited to energy systems, aerospace, automotive industry, and manufacturing.

[0003] Among them, the method for optimizing the efficiency of a drone hydrogen battery explores how to improve the efficiency of the drone hydrogen battery through optimization control technology. This optimization method involves adjusting and improving the energy management system of the drone to ensure that the hydrogen battery minimizes energy loss while providing power. This can not only extend the flight time of the drone, but also improve the overall energy utilization efficiency, so that when performing long - term surveillance, search and rescue, or other tasks, the drone can operate more economically and efficiently. In addition, improving the efficiency of the hydrogen battery helps to reduce operating costs and environmental impacts, making an important step for drone technology in sustainable development.

[0004] Although existing optimization control technologies are widely applied in multiple industries, their application in optimizing the efficiency of drone hydrogen batteries shows certain limitations. Existing technologies mainly rely on traditional optimization methods, which are usually inefficient in dealing with dynamic responses under real - time changes and complex environmental conditions. When the drone performs emergency tasks or faces environmental changes, this inflexible optimization strategy may not be able to effectively adjust the energy output, resulting in energy waste or insufficient power supply. In addition, existing systems are often not powerful enough in fault prevention and rapid recovery, unable to effectively identify key nodes or ensure energy supply at critical moments, increasing the operating risk of the system. These limitations not only increase the operating costs, but also exacerbate the environmental burden, affecting the application potential of drones in long - term and complex tasks. The innovative solution effectively overcomes these deficiencies by introducing advanced dynamic adjustment technologies and precise control algorithms, providing a more efficient and reliable solution for drone hydrogen battery management. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art, and to propose a method and system for optimizing the efficiency of a drone hydrogen battery.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A method for optimizing the efficiency of a hydrogen battery of a drone, comprising the following steps:

[0007] S1: Collect the flight state data of the drone including speed and altitude, as well as the power data of the hydrogen battery, perform standardization processing, analyze and determine the energy requirements of each battery unit, and construct a preliminary energy allocation model;

[0008] S2: Dynamically optimize the preliminary energy allocation model, reflect the change of the energy transmission path by real-time updating the graph model structure, and adjust the energy path according to the sudden demand of the flight phase or additional tasks, so as to match the energy allocation with the actual demand of the drone, and obtain an adjusted energy allocation model;

[0009] S3: On the basis of the adjusted energy allocation model, introduce the minimum cut algorithm to identify the key transmission nodes and paths, supply priority energy to the key transmission nodes and paths, and isolate some battery units based on the operating efficiency ranking, and dynamically optimize the overall battery management to obtain an optimized network energy allocation model;

[0010] S4: Based on the optimized network energy allocation model, apply the differential evolution algorithm to optimize the parameters of the fuzzy controller, adjust the shape and width of the membership function, generate adjusted control parameters, and adjust the discharge rate and power output of the battery according to the real-time environment and task requirements.

[0011] As a further solution of the present invention, the obtaining steps of the preliminary energy allocation model are specifically as follows:

[0012] S111: Collect the speed, altitude of the drone and the power data of the hydrogen battery, and use the formula:

[0013] ;

[0014] Generate standardized flight state and power data;

[0015] Wherein, is the standardized flight state and power data, represents the original data point, represents the average value of the data, represents the standard deviation of the data, is a small quantity, is an adjustment factor;

[0016] S112: According to the standardized flight state and power data, combined with the task characteristics of the drone, use a multiple linear regression model to predict the energy consumption of each battery unit, and use the formula:

[0017] ;

[0018] Generate the energy demand of each battery cell;

[0019] Among them, represents the energy demand, is the regression coefficient, 、 represent the standardized speed and altitude data, Increase the complexity of the model to match the non-linear change;

[0020] S113: Utilize the energy demand of each battery cell, apply the allocation formula to optimize the energy allocation, using the formula:

[0021] ;

[0022] Generate a preliminary energy allocation model;

[0023] Among them, represents the energy allocation percentage of the th battery cell, is the predicted energy demand of the th battery cell, is the total number of battery cells, is a factor adjusted according to the battery cell performance.

[0024] As a further solution of the present invention, the steps for obtaining the adjusted energy allocation model are specifically as follows:

[0025] S211: Dynamically optimize the preliminary energy allocation model, update the graph model structure according to real-time flight data, using the formula:

[0026] ;

[0027] Generate an updated graph model;

[0028] Among them, represents the original graph model, represents the time change since the last update, represents the distance change since the last update, represents the speed change amount, is a coefficient for adjusting the graph model update, is the updated graph model;

[0029] S212: Use the updated graph model to recalculate the energy path, using the formula:

[0030] ;

[0031] Generate an optimized energy path;

[0032] Among them, is the energy allocation model optimized based on the current flight state, represents the energy transfer influence factor corresponding to the updated graph model;

[0033] S213: Match the optimized energy path with the actual requirements of the UAV, adjust the energy path, and use the formula:

[0034] ;

[0035]

[0036] Among them, is the adjusted energy allocation model.

[0037] As a further solution of the present invention, the specific steps for obtaining the priority energy supply are:

[0038] S311: Based on the adjusted energy allocation model, use the minimum cut algorithm to identify key transmission nodes and paths, and use the formula:

[0039] ;

[0040] Generate the minimum cut set ;

[0041] Among them, represents the total capacity of the minimum cut, represents the capacity of the edge in the graph, represents the existence of the edge, , are characteristic adjustment factors;

[0042] S312: According to the nodes and paths in the minimum cut set, extract the key transmission nodes and paths, and use the formula:

[0043] ;

[0044] Generate the key transmission node , and connect all the key transmission nodes , and establish the path set ;

[0045] Among them, is the set of nodes in the graph, represents the weight of the edge, is the threshold;

[0046] S313: For the key transmission node and the path set Perform priority energy supply using the energy supply formula:

[0047] ;

[0048] Generate a priority energy supply plan ;

[0049] Among them, represents the energy supplied preferentially, represents the total energy, , represent the priority weight coefficients of the key nodes, is the sensitivity coefficient for adjusting energy distribution.

[0050] As a further solution of the present invention, the steps for obtaining the optimized network energy distribution model are specifically as follows:

[0051] S321: Calculate the operating efficiency of all battery units using the formula:

[0052]

[0053] Based on the operating efficiency of all battery units, establish an efficiency sorting result ;

[0054] Among them, is the operating efficiency of the battery unit, is the power output, is the energy consumption, is the weight parameter, is the adjustment parameter, is the fine-tuning parameter, is the list of battery units sorted according to the operating efficiency;

[0055] S322: According to the efficiency sorting result, isolate the battery units with efficiency lower than the threshold, formula:

[0056] ;

[0057] Generate the set of isolated battery units;

[0058] Among them, represents the set of isolated battery units, is the basic threshold, is the adjustment threshold, is the parameter for fine-tuning the threshold;

[0059] S323: Combine the key transmission nodes and the path set , and the non-isolated battery units, and reconfigure the energy distribution, using the formula:

[0060] ;

[0061] Generate an optimized network energy allocation model;

[0062] Wherein, is the optimized network energy allocation model, is the allocation coefficient, is the sensitivity coefficient for adjusting the energy allocation of multiple battery units.

[0063] As a further solution of the present invention, the obtaining step of the adjusted control parameters is specifically as follows:

[0064] S411: Based on the optimized network energy allocation model, apply the differential evolution algorithm to optimize the parameters of the fuzzy controller, using the formula:

[0065] ;

[0066] Generate the current control parameter set;

[0067] Wherein, is the current control parameter set, is the current optimal solution, , is a randomly selected solution, is the differential weight, is the change amount of the network energy allocation model, is the calculation stability parameter;

[0068] S412: According to the current control parameters, adjust the shape and width of the membership function of the fuzzy controller, using the formula:

[0069] ;

[0070] Generate the adjusted membership function;

[0071] Wherein, is the adjusted membership degree, is the original membership degree, is the adjustment coefficient, is the non-linear function of parameter adjustment;

[0072] S413: Based on the comprehensively adjusted membership function, determine the fuzzy control parameters, using the formula:

[0073] ;

[0074] Generate the adjusted control parameters;

[0075] Wherein, is the adjusted control parameter, is the adjusted membership degree, is the weight coefficient.

[0076] An unmanned aerial vehicle hydrogen battery efficiency optimization system, which is used to execute the above-mentioned unmanned aerial vehicle hydrogen battery efficiency optimization method. The system includes:

[0077] The data collection module is based on the current flight state data of the unmanned aerial vehicle, including speed and altitude, and the power data of the hydrogen battery, performs normalization processing and energy demand analysis, determines the energy demand of each battery unit, and constructs a preliminary energy distribution model;

[0078] The dynamic optimization module is based on the preliminary energy distribution model, reflects the change of the energy transmission path by real-time updating the graph model structure, generates a real-time optimization path, adjusts the energy path according to the flight stage or the sudden demand of additional tasks, matches the energy distribution with the actual demand of the unmanned aerial vehicle, and obtains an adjusted energy distribution model;

[0079] The key node identification module is based on the adjusted energy distribution model, introduces the minimum cut algorithm, identifies the key transmission nodes and paths, supplies priority energy to the key transmission nodes and paths, and isolates some battery units based on the operation efficiency ranking, dynamically optimizes the overall battery management, and obtains an optimized network energy distribution model;

[0080] The control parameter optimization module is based on the optimized network energy distribution model, applies the differential evolution algorithm to optimize the parameters of the fuzzy controller, generates optimized control parameters, adjusts the shape and width of the membership function, and obtains the adjusted control parameter.

[0081] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0082] In the present invention, by analyzing the flight state data of the unmanned aerial vehicle in detail and establishing an initial energy distribution model, the specific energy requirements of each battery unit can be more accurately met, thereby reducing the ineffective consumption of energy. The real-time updated energy transmission path, combined with the dynamic adjustment of flight requirements, ensures that the energy distribution is always synchronized with the actual demand, improving the energy utilization efficiency. The minimum cut algorithm is introduced to optimize the energy supply of key nodes, ensuring the operation stability and reliability of the system at critical moments. The control parameters optimized by the differential evolution algorithm allow for fine adjustment of the battery discharge rate and power output, adapting to environmental changes and mission requirements, extending the battery life and optimizing the energy management, reducing the environmental impact and enhancing the economic benefits of the unmanned aerial vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 is a schematic diagram of the working process of the present invention;

[0084] Figure 2 It is a flowchart of the acquisition steps of the preliminary energy distribution model of the present invention;

[0085] Figure 3 It is a flowchart of the acquisition steps of the adjusted energy distribution model of the present invention;

[0086] Figure 4 It is a flowchart of the acquisition steps of the priority energy supply of the present invention;

[0087] Figure 5 It is a flowchart of the acquisition steps of the optimized network energy distribution model of the present invention;

[0088] Figure 6 It is a flowchart of the acquisition steps of the adjusted control parameters of the present invention. Specific Embodiments

[0089] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0090] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise specifically defined.

[0091] Embodiment 1

[0092] Please refer to Figure 1 , the present invention provides a technical solution: a method for optimizing the efficiency of a drone hydrogen battery, including the following steps:

[0093] S1: Collect the flight state data of the drone currently including speed and altitude, as well as the power data of the hydrogen battery, perform standardized processing, analyze and determine the energy requirements of each battery unit, and construct a preliminary energy distribution model;

[0094] S2: Dynamically optimize the preliminary energy distribution model, reflect the changes in the energy transmission path by real-time updating the graph model structure, and adjust the energy path according to the sudden demands of the flight stage or additional tasks to match the energy distribution with the actual needs of the drone, and obtain the adjusted energy distribution model;

[0095] S3: Based on the adjusted energy allocation model, introduce the minimum cut algorithm to identify key transmission nodes and paths, supply energy preferentially to the key transmission nodes and paths, isolate some battery units based on the operating efficiency ranking, and dynamically optimize the overall battery management to obtain the optimized network energy allocation model;

[0096] S4: Based on the optimized network energy allocation model, apply the differential evolution algorithm to optimize the parameters of the fuzzy controller, adjust the shape and width of the membership function, generate the adjusted control parameters, and adjust the discharge rate and power output of the battery according to the real-time environment and task requirements.

[0097] The preliminary energy allocation model specifically includes energy demand analysis, battery unit allocation ratio, and real-time data integration. The adjusted energy allocation model includes path dynamic adjustment, demand response matching, and update frequency. The optimized network energy allocation model includes a list of key nodes, optimized energy supply decisions, and isolation schemes. The adjusted control parameters include parameter optimization range, adjusted function characteristics, and control strategy accuracy.

[0098] Please refer to Figure 2 , and the specific steps for obtaining the preliminary energy allocation model are as follows:

[0099] S111: Collect the speed, altitude of the unmanned aerial vehicle (UAV) and the power data of the hydrogen battery, and use the formula:

[0100] ;

[0101] Generate the standardized flight status and power data;

[0102] Among them, is the standardized flight status and power data, represents the original data point, represents the average value of the data, represents the standard deviation of the data, is a small quantity, is the adjustment factor;

[0103] S112: According to the standardized flight status and power data, combined with the task characteristics of the UAV, use the multiple linear regression model to predict the energy consumption of each battery unit, and use the formula:

[0104] ;

[0105] Generate the energy demand of each battery unit;

[0106] Among them, represents the energy demand, is the regression coefficient, , Represent the speed and altitude data after standardization, Increase the complexity of the model to match the non-linear changes;

[0107] S113: Utilize the energy requirements of each battery cell and apply the allocation formula to optimize the energy allocation, using the formula:

[0108] ;

[0109] Generate a preliminary energy allocation model;

[0110] Wherein, Represents the energy allocation percentage of the th battery cell, Is the predicted energy requirement of the th battery cell, Is the total number of battery cells, Is a factor adjusted according to the battery cell performance.

[0111] Step 1: Data collection and standardization

[0112] Formula and parameters

[0113] ;

[0114] Parameter explanation:

[0115] : The original data point, assuming the speed data of the drone is taken.

[0116] : The average value of the data, calculated based on the collected data.

[0117] : The standard deviation of the data, calculated based on the collected data.

[0118] : A small quantity, set to , to avoid a zero denominator.

[0119] : The adjustment factor, used to enhance the sensitivity of standardization, generally set to 1.1.

[0120] Calculation process and example

[0121] Suppose there are the following drone speed data points (unit: ): .

[0122] Calculate the average value

[0123] Calculate the standard deviation :

[0124]

[0125] Calculate a data point using a formula The standardized result of:

[0126]

[0127] This standardized value represents the deviation of the data point from the average speed, taking into account the adjustment factor, which makes the result more sensitive to outliers.

[0128] Step 2: Energy demand prediction

[0129] Formula and parameters

[0130]

[0131] Parameter explanation:

[0132] : Predicted energy demand.

[0133] : Standardized values of the UAV speed and altitude respectively.

[0134] : Regression coefficient obtained through statistical analysis of historical data.

[0135] Calculation process and example

[0136] Assume:

[0137] The regression coefficient has been obtained through data analysis: .

[0138] Standardized speed (obtained from Step 1), altitude (also obtained through standardization).

[0139] Calculate the energy demand:

[0140]

[0141]

[0142] This value represents the estimated energy demand of the UAV in the given flight state, in energy units (such as kilowatt-hours).

[0143] Step 3: Construction of energy allocation model

[0144] Formula and parameters

[0145]

[0146] Parameter Explanation:

[0147] : The energy distribution percentage of the th battery cell.

[0148] : The predicted energy demand of the th battery cell.

[0149] : The total number of battery cells.

[0150] : The adjustment factor, set according to battery performance, assuming .

[0151] Calculation Process and Example

[0152] Suppose there are three battery cells, and the energy demand of each cell is .

[0153] Calculate the denominator :

[0154]

[0155] For the first battery , the distribution percentage:

[0156]

[0157] This percentage means that the first battery cell should be allocated approximately of the energy, adjusted according to its performance.

[0158] Please refer to Figure 3 for the specific steps to obtain the adjusted energy distribution model:

[0159] S211: Dynamically optimize the preliminary energy distribution model, update the graph model structure according to real-time flight data, and use the formula:

[0160] ;

[0161] Generate the updated graph model;

[0162] Among them, represents the original graph model, represents the time change since the last update, represents the distance change since the last update, represents the change in speed, is the coefficient for adjusting the update of the graph model, is the updated graph model;

[0163] S212: Use the updated graph model to recalculate the energy path using the formula:

[0164] ;

[0165] Generate the optimized energy path;

[0166] where, is the energy distribution model optimized based on the current flight state, represents the energy transfer impact factor corresponding to the updated graph model;

[0167] S213: Match the optimized energy path with the actual requirements of the UAV and adjust the energy path using the formula:

[0168]

[0169]

[0170] where, .

[0171] Step 1: Update the graph model structure

[0172] Formula:

[0173]

[0174] Detailed parameter explanation:

[0175] : The original graph model value. Assume the original value is 100.

[0176] : The time variation. For example, 5 minutes have passed since the last update. Assume it is 5.

[0177] : The distance variation. For example, the UAV has flown 2 kilometers. Assume it is 2.

[0178] : The speed variation. For example, the speed changes from to , and assume the increase is .

[0179] : The adjustment coefficient. Assume it is .

[0180] Calculation process:

[0181] Calculate the combined impact of time and distance:

[0182]

[0183] Update the graph model structure:

[0184]

[0185] Step 2: Recalculate the energy transmission path

[0186] Formula:

[0187]

[0188] Detailed parameter explanation:

[0189] : Original energy distribution, assumed to be 30.

[0190] : Value obtained from Step 1, which is 154.5.

[0191] : Total number of units, assumed to be 3, and other units are assumed to be 20.

[0192] Calculation process:

[0193] Calculate the denominator part of the new energy path (assuming the updated graph models of all units are the same):

[0194]

[0195] Calculate the numerator part of the new energy path:

[0196]

[0197] Update the energy path:

[0198]

[0199] Step 3: Match the energy distribution with the actual demand

[0200] Formula:

[0201]

[0202] Detailed parameter explanation:

[0203] : Value calculated from Step 2, which is 0.5.

[0204] : Actual energy demand, assuming the drone demand is 0.4.

[0205] Calculation process:

[0206] Compare and :

[0207]

[0208] Result description

[0209] It indicates that the adjusted energy distribution model will be adjusted according to the actual demand to ensure that the UAV energy supply does not exceed the actual demand and optimize the energy use efficiency.

[0210] Please refer to Figure 4 , and the specific steps for obtaining the priority energy supply are as follows:

[0211] S311: Based on the adjusted energy distribution model, use the minimum cut algorithm to identify the key transmission nodes and paths, and use the formula:

[0212] ;

[0213] Generate the minimum cut set ;

[0214] Among them, represents the total capacity of the minimum cut, represents the capacity of the edge in the graph, represents the existence of the edge, , are characteristic adjustment factors;

[0215] S312: According to the nodes and paths in the minimum cut set, extract the key transmission nodes and paths, and use the formula:

[0216] ;

[0217] Generate the key transmission nodes , and connect all the key transmission nodes to establish a path set ;

[0218] Among them, is the set of nodes in the graph, represents the weight of the edge, is the threshold;

[0219] S313: Perform priority energy supply to the key transmission nodes and the path set , and use the energy supply formula:

[0220] ;

[0221] Generate a priority energy supply plan ;

[0222] Among them, represents the energy supplied preferentially, represents the total energy, , represents the priority weight coefficient of the key node, is the sensitivity coefficient for adjusting energy distribution.

[0223] Step 1: Introduce the minimum cut algorithm

[0224] Formula:

[0225]

[0226] Parameter explanation and derivation:

[0227] : Capacity of edge , representing the maximum transmission capacity from node to node . Assume the value is 10.

[0228] : Existence indicator function of edge , which is 1 if the edge exists, otherwise 0.

[0229] : Characteristic coefficients of nodes and , which can represent the load or priority of the nodes. Assume and .

[0230] Calculation process:

[0231] Assume there is an edge in the network and this edge currently exists, i.e., .

[0232] Calculate the adjusted capacity of this edge:

[0233]

[0234] The value 5.56 represents the effective transmission capacity of edge after considering the node load and priority.

[0235] Step 2: Identify key transmission nodes and paths

[0236] Formula:

[0237]

[0238] Parameter Explanation and Deduction:

[0239] : Edge The weight of, which can be assigned based on traffic or importance, is assumed to be 8.

[0240] : The threshold of the weight, used to filter important edges, is assumed to be 5.

[0241] Calculation Process:

[0242] For each edge , check whether its weight is greater than the threshold .

[0243] For the in the assumption, since , the edge is considered a critical transmission path.

[0244] Therefore, the nodes and are marked as critical transmission nodes.

[0245] Step 3: Priority Energy Supply

[0246] Formula:

[0247]

[0248] Parameter Explanation and Deduction:

[0249] : The total energy of the system, assumed to be 100 units.

[0250] : The priority weight coefficient of critical nodes, assumed , and the sum of all critical nodes is assumed to be 2.

[0251] : The coefficient to adjust the sensitivity of energy distribution, assumed to be 2.

[0252] Calculation Process:

[0253] Calculate the total denominator of the priority weight:

[0254]

[0255] Calculate the proportion of priority energy supply:

[0256]

[0257] Apply the adjustment coefficient to adjust the supplied energy:

[0258]

[0259] The value 6.32 represents the energy unit allocated to the key nodes after adjustment under the given parameters.

[0260] Please refer to Figure 5 , and the steps for obtaining the optimized network energy allocation model are specifically as follows:

[0261] S321: Calculate the operating efficiency of all battery cells using the formula:

[0262]

[0263] Establish an efficiency ranking result based on the operating efficiency of all battery cells ;

[0264] Among them, is the operating efficiency of the battery cell, is the power output, is the energy consumption, is the weight parameter, is the adjustment parameter, is the fine-tuning parameter;

[0265] S322: According to the efficiency ranking result, isolate the battery cells with efficiency lower than the threshold, formula:

[0266] ;

[0267] Generate a set of isolated battery cells;

[0268] Among them, represents the set of isolated battery cells, is the basic threshold, is the adjusted threshold, is the parameter for fine-tuning the threshold, is the list of battery cells sorted according to the operating efficiency.

[0269] S323: Combine the key transmission nodes and the path set , and the non-isolated battery cells, and reconfigure the energy allocation using the formula:

[0270] ;

[0271] Generate the optimized network energy allocation model;

[0272] Among them, is the optimized network energy allocation model, is the allocation coefficient, is the sensitivity coefficient for adjusting the energy distribution of multiple battery cells.

[0273] Step 1: Calculate the operating efficiency of the overall battery cell

[0274] Formula:

[0275]

[0276] : Power output of the battery cell (Assumption: 100W)

[0277] : Performance weighting factor (Assumption: 1.2)

[0278] : Energy consumption of the battery cell (Assumption: 50W)

[0279] : Energy consumption adjustment factor (Assumption: 5W)

[0280] : Fine-tuning parameter for stable calculation (Assumption: 0.5)

[0281] Calculation process

[0282] First calculate the numerator:

[0283] Calculate the denominator:

[0284] Finally calculate the efficiency:

[0285] Step 2: Isolate the battery cells with low efficiency

[0286] Formula:

[0287]

[0288] : Efficiency base threshold (Assumption: 3)

[0289] : Dynamic factor for adjusting the threshold (Assumption: 1.1)

[0290] : Additional downward adjustment coefficient (Assumption: 0.2)

[0291] Calculation process

[0292] First calculate the threshold:

[0293] Compare the efficiency of each battery cell with the threshold: If , then the battery cell is added to In the example, , so it is not isolated.

[0294] Step 3: Dynamically optimize the overall battery management

[0295] Formula:

[0296]

[0297] : Distribution coefficient (assumption: 0.9)

[0298] : Energy distribution sensitivity adjustment coefficient (assumption: 0.1)

[0299] : Power output of the battery cell (as before, 100W)

[0300] Calculation process

[0301] Calculate the allocated energy of each battery cell:

[0302] Accumulate the energy output of all non-isolated batteries to obtain .

[0303] Please refer to Figure 6 , and the specific steps for obtaining the adjusted control parameters are as follows:

[0304] S411: Based on the optimized network energy distribution model, apply the differential evolution algorithm to optimize the parameters of the fuzzy controller, using the formula:

[0305] ;

[0306] Generate the current control parameter set;

[0307] Among them, is the current control parameter set, is the current optimal solution, , are randomly selected solutions, is the differential weight, is the change amount of the network energy distribution model, is the calculation stability parameter;

[0308] S412: According to the current control parameters, adjust the shape and width of the membership function of the fuzzy controller, using the formula:

[0309] ;

[0310] Generate the adjusted membership function;

[0311] Among them, is the adjusted membership degree, is the original membership degree, is the adjustment coefficient, is the non-linear function for parameter adjustment;

[0312] S413: Based on the comprehensively adjusted membership function, determine the fuzzy control parameters using the formula:

[0313] ;

[0314] Generate the adjusted control parameters;

[0315] Among them, is the adjusted control parameter, is the adjusted membership degree, is the weight coefficient.

[0316] Step 1: Apply the differential evolution algorithm

[0317] Formula:

[0318]

[0319] Parameter explanation and example derivation:

[0320] : The current optimal solution, set as

[0321] : Two randomly selected sets of control parameters, assume

[0322] : The differential weight, which controls the degree of variation of the control parameters, assume

[0323] : The total value of the network energy distribution model, assume it is 1000 units

[0324] The change amount of, assume it is 50 units

[0325] : Calculate the stability parameter, used to prevent the denominator from being zero, set as 1

[0326] Derivation process:

[0327] Calculate

[0328] Apply the differential weight:

[0329] Calculate

[0330] Final

[0331] Explanation of calculation result:

[0332] It is approximately 3.65, representing the parameter value of the newly generated candidate solution in this iteration.

[0333] Step 2: Adjust the shape and width of the membership function

[0334] Formula:

[0335]

[0336] Parameter explanation and example derivation:

[0337] : Original membership degree, assumed to be 0.5

[0338] : 3.65 obtained from the previous calculation

[0339] : Adjustment coefficient, assumed to be 0.3

[0340] Derivation process:

[0341] Calculate

[0342] Calculate

[0343] Final

[0344] Explanation of calculation result:

[0345] It is 0.725, indicating that the membership degree after adjustment increases, and the membership function becomes wider or smoother.

[0346] Step 3: Generate the adjusted control parameters

[0347] Formula:

[0348]

[0349] Parameter explanation and example derivation:

[0350] : Adjusted membership degree, assuming there are two control parameters, both being 0.725

[0351] : Weights of the parameters, assumed to be 0.6 and 0.4

[0352] Derivation process:

[0353] For each

[0354] The first parameter:

[0355] The second parameter:

[0356] Final

[0357] Explanation of the calculation result:

[0358] The value is 1.2, which represents the control parameter value after comprehensively considering the adjusted membership degree and weight, and is used in the fuzzy control system to improve the response effect and accuracy of the system.

[0359] An optimization system for the efficiency of an unmanned aerial vehicle (UAV) hydrogen battery. The UAV hydrogen battery efficiency optimization system is used to execute the above-mentioned UAV hydrogen battery efficiency optimization method. The system includes:

[0360] The data collection module, based on the current flight state data of the UAV, including speed and altitude, and the power data of the hydrogen battery, performs standardization processing and energy demand analysis, determines the energy demand of each battery unit, and constructs a preliminary energy distribution model;

[0361] The dynamic optimization module, based on the preliminary energy distribution model, reflects the changes in the energy transmission path by real-time updating the graph model structure, generates a real-time optimization path, adjusts the energy path according to the sudden demands of the flight phase or additional tasks, and matches the energy distribution with the actual demands of the UAV to obtain an adjusted energy distribution model;

[0362] The key node identification module, based on the adjusted energy distribution model, introduces the minimum cut algorithm to identify the key transmission nodes and paths, supplies priority energy to the key transmission nodes and paths, and isolates some battery units based on the operating efficiency ranking to dynamically optimize the overall battery management and obtain an optimized network energy distribution model;

[0363] The control parameter optimization module, based on the optimized network energy distribution model, applies the differential evolution algorithm to optimize the parameters of the fuzzy controller, generates optimized control parameters, and adjusts the shape and width of the membership function to obtain adjusted control parameters.

[0364] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for optimizing the efficiency of hydrogen batteries for unmanned aerial vehicles, characterized in that: The following steps are involved: Collect the current flight status data of the drone, including speed and altitude, and the power data of the hydrogen battery, perform standardization, analyze and determine the energy requirements of each battery unit, and build a preliminary energy distribution model; Dynamically optimize the preliminary energy allocation model, reflect changes in the energy transmission path by updating the graph model structure in real time, and adjust the energy path according to sudden demands of the flight phase or additional tasks, so as to match the energy allocation with the actual needs of the UAV, and obtain an adjusted energy allocation model; Based on the adjusted energy allocation model, a minimum cut algorithm is introduced to identify key transmission nodes and paths, prioritize energy supply to the key transmission nodes and paths, and isolate some battery cells based on operating efficiency, dynamically optimize overall battery management, and obtain an optimized network energy allocation model; Based on the optimized network energy allocation model, the differential evolution algorithm is applied to optimize the parameters of the fuzzy controller, the shape and width of the membership function are adjusted, and the adjusted control parameters are generated to adjust the discharge rate and power output of the battery according to the real-time environment and task requirements; Using the energy requirements of each battery cell, an allocation formula is applied to optimize energy allocation, using the formula: ； Generate a preliminary energy distribution model; in, Representative The energy distribution percentage of each battery cell, It is The predicted energy demand of each battery cell, is the total number of battery cells, is a factor adjusted according to the battery cell performance; Match the optimized energy path with the actual needs of the drone and adjust the energy path using the formula: ； An adjusted energy distribution model is obtained; in, is the adjusted energy allocation model, It is an energy allocation model optimized based on the current flight status; Combining the key transmission nodes and the set of paths, as well as the battery cells that are not isolated, the energy distribution is reconfigured using the formula: ； Generate an optimized network energy allocation model; in, is the optimized network energy allocation model, is the distribution coefficient, is the sensitivity coefficient for adjusting the energy distribution of multiple battery cells, Represents an isolated set of battery cells, is a list of battery cells sorted by operating efficiency; Comprehensively adjust the membership function to determine the fuzzy control parameters, using the formula: ； generating adjusted control parameters; in, is the adjusted control parameter, It is The membership degree after the control parameters are adjusted is It is The weight coefficient of the control parameter.

2. The method for optimizing the efficiency of hydrogen batteries for unmanned aerial vehicles according to claim 1, characterized in that: The steps for obtaining the preliminary energy allocation model are specifically as follows: Collect the speed, altitude and hydrogen battery power data of the drone, using the formula: ； Generate standardized flight status and power data; in, To standardize flight status and power data, represents the original data point, represents the average value of the data, represents the standard deviation of the data, It is a tiny amount, is the adjustment factor; Based on the standardized flight status and power data, a multivariate linear regression model is used to predict the energy consumption of each battery cell using the formula: ； Generates the energy requirements of each battery cell; in, represents the energy demand, is the regression coefficient, , represents the normalized speed and altitude data, Increase the complexity of the model to match nonlinear changes.

3. The method for optimizing the efficiency of hydrogen batteries for unmanned aerial vehicles according to claim 1, characterized in that: The steps for obtaining the adjusted energy allocation model are specifically as follows: The preliminary energy allocation model is dynamically optimized, and the graph model structure is updated according to real-time flight data, using the formula: ； Generate an updated graph model; in, Represents the original image model, Indicates the time change since the last update, represents the change in distance since the last update, Indicates the speed change, is the coefficient that adjusts the update of the graph model, is the updated graph model; Using the updated graph model, the energy path is recalculated using the formula: ； Generate optimized energy paths; in, It is an energy allocation model optimized based on the current flight status. Represents the energy transfer impact factor corresponding to the updated graphical model.

4. The method for optimizing the efficiency of hydrogen batteries for unmanned aerial vehicles according to claim 1, characterized in that: The steps for obtaining the priority energy supply are specifically as follows: Based on the adjusted energy allocation model, the minimum cut algorithm is used to identify key transmission nodes and paths, using the formula: ； Generate Minimum Cut Set ; in, represents the total capacity of the minimum cut, represents the capacity of the edges in the graph, Indicates the existence of an edge, , is a characteristic modifier; According to the nodes and paths in the minimum cut set, the key transmission nodes and paths are extracted using the formula: ； Generate key transmission nodes , and connect all key transmission nodes , create a set of paths ; in, is the set of nodes in the graph, represents the weight of the edge, is the threshold value; For the key transmission node and the path set Prioritize energy supply and use the energy supply formula: ； Generate priority energy supply plan ; in, Indicates the energy that is supplied first, represents the total energy, , represents the priority weight coefficient of the key node, is the sensitivity coefficient for adjusting energy distribution.

5. The method for optimizing the efficiency of hydrogen batteries for unmanned aerial vehicles according to claim 1, characterized in that: The steps for obtaining the optimized network energy allocation model are specifically as follows: Calculate the operating efficiency of all battery cells using the formula: ； Establish efficiency ranking results based on the operating efficiency of all battery cells; in, is the operating efficiency of the battery cell, is the power output, It is energy consumption. is the weight parameter, To adjust the parameters, is a fine-tuning parameter; According to the efficiency ranking results, the battery cells with efficiency lower than the threshold are isolated, and the formula is: ； generating an isolated set of battery cells; in, Represents an isolated set of battery cells, is the basic threshold, To adjust the threshold, is the parameter for fine-tuning the threshold, is a list of battery cells sorted by operating efficiency.

6. The method for optimizing the efficiency of hydrogen batteries for unmanned aerial vehicles according to claim 1, characterized in that: The steps for obtaining the adjusted control parameters are specifically as follows: Based on the optimized network energy allocation model, the differential evolution algorithm is applied to optimize the parameters of the fuzzy controller using the formula: ； Generate a current control parameter set; in, is the current set of control parameters, is the current optimal solution, , is a randomly chosen solution, is the difference weight, is the change in the network energy allocation model, is the calculation stability parameter; According to the current control parameters, the shape and width of the fuzzy controller membership function are adjusted using the formula: ； generating adjusted membership functions; in, is the adjusted membership degree, is the original membership, is the adjustment factor, is a nonlinear function of parameter adjustment.

7. A drone hydrogen battery efficiency optimization system, characterized in that: According to the method for optimizing the efficiency of a hydrogen battery for a drone according to any one of claims 1 to 6, the system comprises: The data collection module performs standardized processing and energy demand analysis based on the current flight status data of the drone, including speed and altitude, as well as the power data of the hydrogen battery, determines the energy demand of each battery unit, and builds a preliminary energy allocation model; The dynamic optimization module is based on the preliminary energy allocation model, reflects the changes in the energy transmission path by updating the graph model structure in real time, generates a real-time optimization path, adjusts the energy path according to the sudden demand of the flight phase or additional tasks, matches the energy allocation with the actual demand of the UAV, and obtains the adjusted energy allocation model; The key node identification module introduces a minimum cut algorithm based on the adjusted energy allocation model to identify key transmission nodes and paths, prioritizes energy supply to the key transmission nodes and paths, and isolates some battery units based on operating efficiency, dynamically optimizes overall battery management, and obtains an optimized network energy allocation model; The control parameter optimization module optimizes the parameters of the fuzzy controller based on the optimized network energy distribution model, generates optimized control parameters, adjusts the shape and width of the membership function, and obtains adjusted control parameters.

Images