Heat dissipation optimization method for electric aircraft based on deep reinforcement learning

Abstract

The present invention discloses an electric aircraft heat dissipation optimization method based on deep reinforcement learning, which specifically relates to the technical field of deep learning analysis, including: building a heat production efficiency prediction model and a heat dissipation efficiency prediction model based on a deep learning model, and obtaining the heat production efficiency fluctuation curve over time and the heat dissipation efficiency and heat dissipation execution parameter mapping table of the electric aircraft based on the deep learning model; based on the heat production efficiency fluctuation curve over time and the average heat dissipation demand efficiency in each time period, the heat dissipation efficiency fluctuation curve is designed with the temperature of the motor in a normal range as a constraint condition, and a heat dissipation efficiency preset value at each time point is obtained; based on the heat dissipation efficiency and the mapping table, a heat dissipation execution parameter preset value at each time point is obtained; the heat dissipation execution parameter preset value is used as the initial setting value of the heat dissipation system; based on the actual response time of the heat dissipation system, the execution time of the heat dissipation execution parameter preset value is obtained, thereby effectively solving the problem that the existing electric aircraft heat dissipation system is not intelligent enough.

Description

Technical Field

[0001] The present invention relates to the technical field of deep learning analysis, and more specifically, to an electric aircraft heat dissipation optimization method based on deep reinforcement learning. Background Art

[0002] The cooling system of electric aircraft is a core part of its design. It is not only directly related to the flight safety and performance stability of the aircraft, but also has a profound impact on the aircraft's endurance and passenger comfort. With the rapid development of electric aviation technology, the design and optimization of the cooling system has become a research hotspot in the industry. Existing electric aircraft cooling systems are generally equipped with advanced sensor networks that can accurately collect key parameters such as motor temperature, ambient temperature, humidity, and wind speed in real time. Based on the collected data, the cooling system can dynamically adjust its operating parameters, such as fan speed, coolant flow, air duct opening, etc., to achieve the best cooling effect.

[0003] However, in actual use, it still has many shortcomings. For example, the existing electric aircraft cooling system is not smart enough. It cannot match the heat generation efficiency prediction to obtain the preset value of the cooling execution parameter at each time point. The cooling system is started based on the temperature threshold, which may easily cause the motor temperature to exceed the standard. Or in order to avoid the motor temperature exceeding the standard, the cooling efficiency is always maintained at a high level, resulting in unsatisfactory cooling effect or increased cooling energy consumption. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an electric aircraft heat dissipation optimization method based on deep reinforcement learning to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an electric aircraft heat dissipation optimization method based on deep reinforcement learning, comprising the following steps:

[0006] Step 1: Build a heat generation efficiency prediction model based on the deep learning model; obtain the motion trajectory, load and operation status of the electric aircraft at each time point, predict the heat generation efficiency of the electric aircraft, and output the heat generation efficiency fluctuation curve of the electric aircraft over time;

[0007] Step 2: Build a heat dissipation efficiency prediction model based on the deep learning model; obtain the basic parameters and heat dissipation execution parameters of the air-cooled motor of the electric aircraft at each time point, predict the heat dissipation efficiency of the air-cooled motor, and establish a mapping table between heat dissipation efficiency and heat dissipation execution parameters;

[0008] Step 3: Divide the heat generation efficiency fluctuation curve over time into several time periods, obtain the predicted heat generation and motor target temperature of each time period, calculate the target heat dissipation in each time period based on the predicted heat generation, motor target temperature and the motor temperature at the starting point of each time period, and use the ratio of the target heat dissipation in each time period to the time length of each time period to represent the average heat dissipation demand efficiency of each time period;

[0009] Step 4: when the average heat dissipation demand efficiency in each time period exceeds the maximum heat dissipation efficiency of the heat dissipation system, a warning is given to the user, and the time period is reallocated so that the average heat dissipation demand efficiency in each time period is less than the maximum heat dissipation efficiency of the heat dissipation system;

[0010] Step 5: Based on the heat generation efficiency fluctuation curve over time and the average heat dissipation demand efficiency in each time period, a heat dissipation efficiency fluctuation curve is designed with the motor temperature in the normal range as a constraint condition to obtain a preset value of the heat dissipation efficiency at each time point;

[0011] Step 6: obtaining a preset value of the heat dissipation execution parameter at each time point based on the heat dissipation efficiency and the mapping table; using the preset value of the heat dissipation execution parameter as an initial setting value of the heat dissipation system;

[0012] Step 7: Obtain the actual response time of the heat dissipation system to reach the preset value of the heat dissipation efficiency after executing the heat dissipation execution parameter, and obtain the execution time of the preset value of the heat dissipation execution parameter.

[0013] Preferably, the electric aircraft heat dissipation optimization method includes data collection and preprocessing steps, including:

[0014] Data acquisition: Install high-precision temperature sensors, current sensors, and voltage sensors in the electric aircraft's motor and its cooling system to collect real-time data on the motor's temperature, current, voltage, and cooling system status; Data preprocessing: Clean, denoise, and calibrate the collected raw data to ensure data accuracy and reliability; Perform time synchronization on the collected data for subsequent analysis.

[0015] Preferably, the heat production efficiency prediction model building process includes the following steps:

[0016] Step S11, data collection and preprocessing: collecting historical motion trajectory data, load data and operating environment data of the electric aircraft; collecting actual measurement values ​​of heat generation efficiency as a supervision signal of the model;

[0017] Step S12, model selection and initialization: select a deep learning algorithm model to set the initial parameters of the model and configure the model architecture, including the number of layers, number of neurons, and activation function;

[0018] Step S13, model training: divide the preprocessed data into a training set and a validation set, use the training set data to train the model, and output a heat production efficiency prediction model; use the back propagation algorithm and optimizer to adjust the heat production efficiency prediction model parameters, and after each iteration cycle, use the validation set to evaluate the heat production efficiency prediction model performance to avoid overfitting; adjust the learning rate and add regularization terms as needed to optimize the training process;

[0019] Step S14, model evaluation and optimization: using the test set to evaluate the prediction accuracy of the heat production efficiency prediction model;

[0020] Step S15, model deployment and application: deploy the trained heat generation efficiency prediction model to the actual application environment, obtain the motion trajectory, load data and operating environment information of the electric aircraft in real time; input the motion trajectory, load data and operating environment information into the model, and output the heat generation efficiency of the electric aircraft.

[0021] Preferably, the process of building the heat dissipation efficiency prediction model includes the following steps:

[0022] Step S21, collecting basic parameters, operating parameters and actual values ​​of heat dissipation efficiency of the air-cooled motor to obtain historical data of the air-cooled motor;

[0023] Step S22, dividing the historical data into a training set and a test set;

[0024] Step S23, initialize the deep learning model, input the training set into the deep learning model, optimize the deep learning model by adjusting model parameters (such as learning rate, number of hidden layers, number of neurons) and training strategies (such as regularization processing), and evaluate the model performance through the test set;

[0025] Step S24, training until the loss function or maximum number of iterations of the deep learning model meets the requirements, and outputting the trained heat dissipation efficiency prediction model;

[0026] Step S25, deploying and applying the trained heat dissipation efficiency prediction model, inputting the acquired basic parameters and operating parameters of the air-cooled motor into the heat dissipation efficiency prediction model, and outputting a heat dissipation efficiency prediction value.

[0027] Preferably, the mapping method of establishing the heat dissipation efficiency and the heat dissipation execution parameters is: obtaining the relationship between the heat dissipation execution parameters and the heat dissipation efficiency through experimental simulation; establishing a mapping relationship lookup table to convert the predicted heat dissipation efficiency into the corresponding heat dissipation execution parameter preset value.

[0028] Preferably, the preset value of the heat dissipation efficiency at each time point is replaced by a modified preset value of the heat dissipation efficiency, and the modified preset value of the heat dissipation efficiency refers to the product of the preset value of the heat dissipation efficiency and the redundancy coefficient. The value of the redundancy coefficient is greater than 1. The redundancy coefficient is obtained by setting it according to the growth rate of the heat production efficiency at each time point, that is, the greater the growth rate of the heat production efficiency, the greater the redundancy coefficient.

[0029] Preferably, the electric aircraft heat dissipation optimization method further includes:

[0030] Monitor the operating status of the electric aircraft in real time, obtain the real-time temperature, actual heat dissipation efficiency and actual heat generation efficiency of the motor at each time point, and calculate the real-time temperature anomaly risk index; allocate the cooling system monitoring resources of the electric aircraft based on the real-time temperature anomaly risk index.

[0031] Preferably, the temperature anomaly risk index is obtained in the following manner:

[0032] According to the real-time temperature, heat generation efficiency and heat dissipation efficiency of the motor, the heating rate of the motor is calculated using the heat balance equation;

[0033] Based on the real-time temperature and heating rate of the motor, the time for the motor to reach the motor target temperature is calculated; based on the time for the motor to reach the motor target temperature and the safety time, the safety time refers to the response time of the cooling system from being turned on to achieving maximum cooling efficiency;

[0034] The time it takes for the motor to reach the target temperature is recorded as YT and the safety time is AT.

[0035] By formula The temperature anomaly risk index Ft is calculated, where sw represents the real-time temperature of the electric aircraft, yw represents the motor target temperature of the electric aircraft, f1 represents the time risk coefficient, f2 represents the temperature risk coefficient, and f1+f2=1.0, α represents the time impact index, β represents the temperature impact index, and Form(·) represents the linear normalization function, which is used to convert f1*(AT-YT) α +f2*(yw-sw) β The value is controlled in the range of 0 to 1.

[0036] Preferably, when the temperature anomaly risk index exceeds a preset value, the temperature anomaly risk index is brought within a threshold by adjusting the actual heat dissipation efficiency and / or the actual heat generation efficiency.

[0037] Technical effects and advantages of the present invention:

[0038] (1) The electric aircraft heat dissipation optimization method of the present invention is based on deep reinforcement learning. Based on deep reinforcement learning, a heat dissipation efficiency and heat generation efficiency prediction model of the electric aircraft is built to obtain a preset value of the heat dissipation efficiency at each time point, which can more effectively control the risk of temperature anomalies and ensure the safe operation of the electric aircraft. Through the four steps of prediction, monitoring, adjustment and optimization, a closed-loop heat dissipation optimization system is formed; it effectively solves the problem that the electric aircraft heat dissipation system in the prior art is not intelligent enough.

[0039] (2) The electric aircraft heat dissipation optimization method based on deep reinforcement learning of the present invention predicts the heat dissipation efficiency of the air-cooled motor in the manned electric aircraft through advanced machine learning technology, processes and analyzes a large amount of real-time data, such as motor temperature and environmental conditions, and adjusts the operating parameters of the heat dissipation system (such as fan speed and cooling medium flow rate) accordingly, which can ensure that the motor can maintain the optimal operating temperature under various flight conditions, improve the performance and safety of the aircraft, and reduce energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of the electric aircraft heat dissipation optimization method based on deep reinforcement learning of the present invention.

[0041] Figure 2 This is a flow chart for building a heat dissipation efficiency prediction model of the present invention. DETAILED DESCRIPTION

[0042] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0043] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0044] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present application, its application, or uses.

[0045] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0046] Example 1

[0047] See also Figure 1 A flow chart of a method for optimizing heat dissipation of an electric aircraft based on deep reinforcement learning. The present invention provides the following Figure 1 A method for optimizing heat dissipation of an electric aircraft based on deep reinforcement learning is shown, comprising the following steps:

[0048] Step 1: Build a heat generation efficiency prediction model based on the deep learning model, obtain the motion trajectory, load and operation status of the electric aircraft at each time point, predict the heat generation efficiency of the electric aircraft, and output the heat generation efficiency fluctuation curve of the electric aircraft over time;

[0049] Explanation: The heat generation efficiency fluctuation curve over time reflects the heat generation of the electric aircraft at different time points;

[0050] Step 2: Build a heat dissipation efficiency prediction model based on the deep learning model, obtain the basic parameters and heat dissipation execution parameters of the air-cooled motor of the electric aircraft at each time point, predict the heat dissipation efficiency of the air-cooled motor, and establish a mapping table between heat dissipation efficiency and heat dissipation execution parameters;

[0051] Step 3: Divide the heat generation efficiency fluctuation curve over time into several time periods, obtain the predicted heat generation and motor target temperature of each time period, calculate the target heat dissipation in each time period based on the predicted heat generation, motor target temperature and the motor temperature at the starting point of each time period, and use the ratio of the target heat dissipation in each time period to the time length of each time period to represent the average heat dissipation demand efficiency of each time period;

[0052] Explanation: The motor target temperature refers to setting a reasonable motor target temperature based on the motor's design specifications, material properties and safety standards to ensure that the motor will not be damaged or its performance will not be affected due to overheating during long-term operation.

[0053] Step 4: when the average heat dissipation demand efficiency in each time period exceeds the maximum heat dissipation efficiency of the heat dissipation system, a warning is given to the user, and the time period is reallocated so that the average heat dissipation demand efficiency in each time period is less than the maximum heat dissipation efficiency of the heat dissipation system;

[0054] Explanation: When the average heat dissipation demand efficiency in each time period is less than the maximum heat dissipation efficiency, it indicates that the regional division is effective; when the average heat dissipation demand efficiency in each time period is not less than the maximum heat dissipation efficiency, it indicates that the regional division is invalid; a warning is given to the user, and the time periods are re-divided so that the average heat dissipation demand efficiency in each time period is less than the maximum heat dissipation efficiency of the cooling system.

[0055] Step 5: Based on the heat generation efficiency fluctuation curve over time and the average heat dissipation demand efficiency in each time period, a heat dissipation efficiency fluctuation curve is designed with the motor temperature in the normal range as a constraint condition to obtain a preset value of the heat dissipation efficiency at each time point;

[0056] Step 6: obtaining a preset value of the heat dissipation execution parameter at each time point based on the heat dissipation efficiency and the mapping table; using the preset value of the heat dissipation execution parameter as an initial setting value of the heat dissipation system;

[0057] Step 7: Obtain the actual response time of the heat dissipation system to reach the preset value of the heat dissipation efficiency after executing the heat dissipation execution parameter, and obtain the execution time of the preset value of the heat dissipation execution parameter.

[0058] It needs to be further explained in the embodiment of the present invention that the electric aircraft heat dissipation optimization method includes data collection and preprocessing steps, including:

[0059] Data collection: Install high-precision temperature sensors, current sensors, and voltage sensors in the motors and cooling systems of electric aircraft to collect real-time data on the motor's temperature, current, voltage, and the status of the cooling system (such as fan speed, cooling medium flow rate, etc.);

[0060] Data preprocessing: Clean, denoise and calibrate the collected raw data to ensure the accuracy and reliability of the data; perform time synchronization on the collected data for subsequent analysis.

[0061] What needs to be further explained in the embodiments of the present invention is that the method of predicting the heat generation efficiency of the electric aircraft is: using deep learning algorithms (such as LSTM, GRU and other time series models), combined with the historical motion trajectory, load data and operating environment (such as temperature, humidity, air pressure, etc.) of the electric aircraft, to establish a heat generation efficiency prediction model.

[0062] In a further design, the heat production efficiency prediction model building process includes the following steps:

[0063] Step S11, data collection and preprocessing: collect historical motion trajectory data of the electric aircraft, including speed, altitude, and attitude; collect load data, such as battery power consumption, motor current, and voltage; collect operating environment data, such as temperature, humidity, air pressure, and wind speed; collect actual measured values ​​of heat generation efficiency as a supervision signal for the model; clean the data to remove outliers and missing values; standardize or normalize the data to eliminate the influence of different dimensions;

[0064] Step S12, model selection and initialization: select a deep learning algorithm model and set the initial parameters of the model, such as learning rate, batch size, and number of iterations; configure the model architecture, including the number of layers, number of neurons, and activation function;

[0065] Step S13, model training: divide the preprocessed data into a training set and a validation set, use the training set data to train the model, and output a heat production efficiency prediction model; use the back propagation algorithm and optimizer to adjust the heat production efficiency prediction model parameters, and after each iteration cycle, use the validation set to evaluate the heat production efficiency prediction model performance to avoid overfitting; adjust the learning rate and add regularization terms as needed to optimize the training process;

[0066] Step S14, model evaluation and optimization: using the test set to evaluate the prediction accuracy of the heat production efficiency prediction model; commonly used indicators include mean square error, root mean square error, and mean absolute error; adjusting the model parameters or structure according to the evaluation results;

[0067] Step S15, model deployment and application: deploy the trained heat generation efficiency prediction model to the actual application environment, such as the control system of the electric aircraft; obtain the motion trajectory, load data and operating environment information of the electric aircraft in real time; input the motion trajectory, load data and operating environment information into the model, and output the heat generation efficiency of the electric aircraft.

[0068] See also Figure 2 The process of building a heat dissipation efficiency prediction model includes the following steps:

[0069] Step S21, collecting basic parameters, operating parameters and actual values ​​of heat dissipation efficiency of the air-cooled motor to obtain historical data of the air-cooled motor;

[0070] Furthermore, the basic parameters include at least: the rated voltage, rated power, rated current, insulation grade, rotor type, stator structure, and specific heat capacity of the coolant of the motor; the operating parameters include at least: the real-time voltage, real-time current, speed, load rate, ambient temperature, cooling medium temperature and flow rate when the motor is running; the actual value of the heat dissipation efficiency refers to the ratio of the heat removed by the motor to the heat generated by the motor per unit time;

[0071] Step S22, dividing the historical data into a training set and a test set;

[0072] Step S23, initialize the deep learning model, input the training set into the deep learning model, optimize the deep learning model by adjusting model parameters (such as learning rate, number of hidden layers, number of neurons) and training strategies (such as regularization processing), and evaluate the model performance through the test set;

[0073] Step S24, training until the loss function or maximum number of iterations of the deep learning model meets the requirements, and outputting the trained heat dissipation efficiency prediction model;

[0074] Step S25, deploying and applying the trained heat dissipation efficiency prediction model, inputting the acquired basic parameters and operating parameters of the air-cooled motor into the heat dissipation efficiency prediction model, and outputting a heat dissipation efficiency prediction value.

[0075] What needs to be further explained in the embodiments of the present invention is that the method of establishing the mapping between the heat dissipation efficiency and the heat dissipation execution parameters is: obtaining the relationship between the heat dissipation execution parameters (such as fan speed, cooling medium flow rate) and the heat dissipation efficiency through experimental simulation; establishing a mapping relationship lookup table to convert the predicted heat dissipation efficiency into the corresponding heat dissipation execution parameter preset values.

[0076] What needs to be further explained in the embodiment of the present invention is that the preset value of the heat dissipation efficiency at each time point is replaced by a modified preset value of the heat dissipation efficiency, and the modified preset value of the heat dissipation efficiency refers to the product of the preset value of the heat dissipation efficiency and the redundancy coefficient, and the value of the redundancy coefficient is greater than 1.

[0077] By setting the redundancy coefficient, a preset value of the corrected heat dissipation efficiency is obtained to cope with emergencies or system errors. The redundancy coefficient is obtained in the following manner: it is set according to the growth rate of the heat production efficiency at each time point, that is, the greater the growth rate of the heat production efficiency, the greater the redundancy coefficient; the value of the redundancy coefficient is greater than 1.

[0078] Summary: This embodiment uses a deep learning model to predict heat generation and heat dissipation efficiency, combined with dynamic adjustment of the heat dissipation strategy, to ensure that the motor temperature of the electric aircraft is within a safe range and optimize the performance of the heat dissipation system. This method can respond to heat changes in the operation of the electric aircraft in real time and improve the efficiency and reliability of the heat dissipation system.

[0079] Example 2

[0080] The embodiment of the present invention provides an electric aircraft heat dissipation optimization method based on deep reinforcement learning. The difference from the embodiment 1 is that the electric aircraft heat dissipation optimization method further includes:

[0081] Monitor the operating status of the electric aircraft in real time, obtain the real-time temperature, actual heat dissipation efficiency and actual heat generation efficiency of the motor at each time point, and calculate the real-time temperature anomaly risk index; allocate the cooling system monitoring resources of the electric aircraft based on the real-time temperature anomaly risk index.

[0082] It needs to be further explained in the embodiment of the present invention that the temperature anomaly risk index is obtained in the following manner:

[0083] According to the real-time temperature, heat generation efficiency and heat dissipation efficiency of the motor, the heating rate of the motor is calculated using the heat balance equation;

[0084] Based on the real-time temperature and heating rate of the motor, the time for the motor to reach the motor target temperature is calculated; based on the time for the motor to reach the motor target temperature and the safety time, the safety time refers to the response time of the cooling system from being turned on to achieving maximum cooling efficiency;

[0085] The time it takes for the motor to reach the target temperature is recorded as YT and the safety time is AT.

[0086] By formula The temperature anomaly risk index Ft is calculated, where sw represents the real-time temperature of the electric aircraft, yw represents the motor target temperature of the electric aircraft, f1 represents the time risk coefficient, f2 represents the temperature risk coefficient, and f1+f2=1.0, α represents the time impact index, β represents the temperature impact index, and Form(·) represents the linear normalization function, which is used to convert f1*(AT-YT) α +f2*(yw-sw) β The value is controlled in the range of 0 to 1.

[0087] What needs to be further explained in the embodiment of the present invention is that when the temperature anomaly risk index exceeds a preset value, the temperature anomaly risk index is made within a threshold value by adjusting the actual heat dissipation efficiency and / or the actual heat generation efficiency.

[0088] What needs to be further explained in the embodiments of the present invention is that when the temperature anomaly risk index exceeds the preset value, the actual heat dissipation efficiency or the actual heat generation efficiency is automatically adjusted according to the preset priority and strategy. For example, the fan speed or the cooling medium flow rate is increased to improve the heat dissipation efficiency; if the heat dissipation capacity has reached its limit, consider reducing the motor load or output power to reduce heat generation.

[0089] Summary: According to the size of the real-time temperature anomaly risk index, the allocation of cooling system monitoring resources can be adjusted automatically or manually. For example, for areas or components with higher risks, the monitoring frequency can be increased or the monitoring accuracy can be improved. Potential overheating, poor cooling and other problems can be discovered and handled in a timely manner, thereby ensuring the safe operation of electric aircraft.

[0090] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An electric aircraft heat dissipation optimization method based on deep reinforcement learning, characterized in that: The following steps are involved: Step 1: Build a heat generation efficiency prediction model based on the deep learning model; obtain the motion trajectory, load and operation status of the electric aircraft at each time point, predict the heat generation efficiency of the electric aircraft, and output the heat generation efficiency fluctuation curve of the electric aircraft over time; Step 2: Build a heat dissipation efficiency prediction model based on the deep learning model; obtain the basic parameters and heat dissipation execution parameters of the air-cooled motor of the electric aircraft at each time point, predict the heat dissipation efficiency of the air-cooled motor, and establish a mapping table between heat dissipation efficiency and heat dissipation execution parameters; Step 3: Divide the heat generation efficiency fluctuation curve over time into several time periods, obtain the predicted heat generation and motor target temperature of each time period, calculate the target heat dissipation in each time period based on the predicted heat generation, motor target temperature and the motor temperature at the starting point of each time period, and use the ratio of the target heat dissipation in each time period to the time length of each time period to represent the average heat dissipation demand efficiency of each time period; Step 4: when the average heat dissipation demand efficiency in each time period exceeds the maximum heat dissipation efficiency of the heat dissipation system, a warning is given to the user, and the time period is reallocated so that the average heat dissipation demand efficiency in each time period is less than the maximum heat dissipation efficiency of the heat dissipation system; Step 5: Based on the heat generation efficiency fluctuation curve over time and the average heat dissipation demand efficiency in each time period, a heat dissipation efficiency fluctuation curve is designed with the motor temperature in the normal range as a constraint condition to obtain a preset value of the heat dissipation efficiency at each time point; Step 6: obtaining a preset value of the heat dissipation execution parameter at each time point based on the heat dissipation efficiency and the mapping table; using the preset value of the heat dissipation execution parameter as an initial setting value of the heat dissipation system; Step 7: Obtain the actual response time of the heat dissipation system to reach the preset value of the heat dissipation efficiency after executing the heat dissipation execution parameter, and obtain the execution time of the preset value of the heat dissipation execution parameter.

2. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 1 is characterized in that: The electric aircraft heat dissipation optimization method includes data collection and preprocessing steps, including: Data acquisition: Install high-precision temperature sensors, current sensors, and voltage sensors in the electric aircraft's motor and its cooling system to collect real-time data on the motor's temperature, current, voltage, and cooling system status; Data preprocessing: Clean, denoise, and calibrate the collected raw data to ensure data accuracy and reliability; Perform time synchronization on the collected data for subsequent analysis.

3. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 1 is characterized in that: The heat production efficiency prediction model building process includes the following steps: Step S11, data collection and preprocessing: collecting historical motion trajectory data, load data and operating environment data of the electric aircraft; collecting actual measurement values ​​of heat generation efficiency as a supervision signal of the model; Step S12, model selection and initialization: select a deep learning algorithm model to set the initial parameters of the model and configure the model architecture, including the number of layers, number of neurons, and activation function; Step S13, model training: divide the preprocessed data into a training set and a validation set, use the training set data to train the model, and output a heat production efficiency prediction model; use the back propagation algorithm and optimizer to adjust the heat production efficiency prediction model parameters, and after each iteration cycle, use the validation set to evaluate the heat production efficiency prediction model performance to avoid overfitting; adjust the learning rate and add regularization terms as needed to optimize the training process; Step S14, model evaluation and optimization: using the test set to evaluate the prediction accuracy of the heat production efficiency prediction model; Step S15, model deployment and application: deploy the trained heat generation efficiency prediction model to the actual application environment, obtain the motion trajectory, load data and operating environment information of the electric aircraft in real time; input the motion trajectory, load data and operating environment information into the model, and output the heat generation efficiency of the electric aircraft.

4. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 1 is characterized in that: The process of building the heat dissipation efficiency prediction model includes the following steps: Step S21, collecting basic parameters, operating parameters and actual values ​​of heat dissipation efficiency of the air-cooled motor to obtain historical data of the air-cooled motor; Step S22, dividing the historical data into a training set and a test set; Step S23, initialize the deep learning model, input the training set into the deep learning model, optimize the deep learning model by adjusting model parameters and training strategy, and evaluate the model performance through the test set; Step S24, training until the loss function or maximum number of iterations of the deep learning model meets the requirements, and outputting the trained heat dissipation efficiency prediction model; Step S25, deploying and applying the trained heat dissipation efficiency prediction model, inputting the acquired basic parameters and operating parameters of the air-cooled motor into the heat dissipation efficiency prediction model, and outputting a heat dissipation efficiency prediction value.

5. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 1, characterized in that: The mapping method of establishing the heat dissipation efficiency and the heat dissipation execution parameters is: obtaining the relationship between the heat dissipation execution parameters and the heat dissipation efficiency through experimental simulation; establishing a mapping relationship lookup table to convert the predicted heat dissipation efficiency into the corresponding heat dissipation execution parameter preset value.

6. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 1, characterized in that: The preset value of the heat dissipation efficiency at each time point is replaced by a modified preset value of the heat dissipation efficiency, and the modified preset value of the heat dissipation efficiency refers to the product of the preset value of the heat dissipation efficiency and the redundancy coefficient. The value of the redundancy coefficient is greater than 1. The redundancy coefficient is obtained in the following manner: it is set according to the growth rate of the heat production efficiency at each time point, that is, the greater the growth rate of the heat production efficiency, the greater the redundancy coefficient.

7. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 1 is characterized in that: The electric aircraft heat dissipation optimization method further includes: Monitor the operating status of the electric aircraft in real time, obtain the real-time temperature, actual heat dissipation efficiency and actual heat generation efficiency of the motor at each time point, and calculate the real-time temperature anomaly risk index; allocate the cooling system monitoring resources of the electric aircraft based on the real-time temperature anomaly risk index.

8. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 7 is characterized in that: The temperature anomaly risk index is obtained as follows: According to the real-time temperature, heat generation efficiency and heat dissipation efficiency of the motor, the heating rate of the motor is calculated using the heat balance equation; Based on the real-time temperature and heating rate of the motor, the time for the motor to reach the target temperature is calculated; Based on the time for the motor to reach the motor target temperature and the safety time, wherein the safety time refers to the response time of the cooling system from being turned on to achieving maximum cooling efficiency; The time it takes for the motor to reach the target temperature is recorded as YT and the safety time is AT. By formula The temperature anomaly risk index Ft is calculated, where sw represents the real-time temperature of the electric aircraft, yw represents the motor target temperature of the electric aircraft, f1 represents the time risk coefficient, f2 represents the temperature risk coefficient, and f1+f2=1.0, α represents the time impact index, β represents the temperature impact index, and Form(·) represents the linear normalization function, which is used to convert f1*(AT-YT) α +f2*(yw-sw) β The value is controlled in the range of 0 to 1.

9. The electric aircraft heat dissipation optimization method based on deep reinforcement learning according to claim 7, characterized in that: When the temperature anomaly risk index exceeds the preset value, the temperature anomaly risk index is kept within the threshold by adjusting the actual heat dissipation efficiency or the actual heat generation efficiency.

Images