Heat dissipation method, device, equipment and storage medium for electric scooter drive system

Abstract

The present invention relates to the technical field of heat dissipation of a drive system, and discloses a heat dissipation method, device, equipment and storage medium for a drive system of an electric scooter, wherein the method comprises: inputting heat source distribution data and space constraint data of a drive system in an electric scooter into a two-stage optimization model to perform heat dissipation structure calculation to obtain heat dissipation channel parameters; performing zone sampling and real-time synchronization of heat source temperature and radiator temperature to obtain heat flow distribution characteristics; performing heat conduction model calculation and multi-region temperature classification to obtain temperature dynamic distribution data and performing heat flow prediction calculation to obtain heat flow prediction data; performing thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters; performing heat load division and heat dissipation power optimization on a driving condition according to the heat dissipation compensation parameters to obtain heat dissipation control parameters. The present invention solves the problem of low space utilization of traditional heat dissipation solutions and improves the heat dissipation effect of the drive system.

Description

Technical Field

[0001] The present invention relates to the technical field of heat dissipation of a drive system, and in particular to a heat dissipation method, device, equipment and storage medium for a drive system of an electric scooter. Background Art

[0002] Traditional electric scooter drive systems mainly use natural heat dissipation or simple air cooling. Under high load and high temperature conditions, insufficient heat dissipation is prone to occur, causing the motor and controller to overheat, affecting system performance and service life. Existing heat dissipation methods generally have problems such as unreasonable radiator layout, inaccurate temperature field monitoring, and single heat dissipation control strategy, which are difficult to meet the heat dissipation needs of electric scooters under complex working conditions.

[0003] The current heat dissipation scheme of the electric scooter drive system lacks the ability to accurately analyze and dynamically predict the characteristics of heat flow distribution, resulting in uneven distribution of heat dissipation resources and low heat dissipation efficiency. At the same time, due to the small size and compact structure of electric scooters, traditional heat dissipation schemes have great limitations in spatial layout and structural design, and it is difficult to achieve the optimal heat dissipation effect in a limited space. Especially under complex working conditions such as frequent start-stop and rapid acceleration, the temperature fluctuates violently, and the existing heat dissipation control methods are difficult to respond in time and accurately compensate. In addition, the existing heat dissipation systems generally lack the ability to effectively identify and suppress thermal interference, and external factors such as ambient temperature changes and load fluctuations can easily cause significant fluctuations in system temperature. Traditional fixed parameter control strategies are difficult to adapt to changing working environments, and cannot be dynamically adjusted according to the heat dissipation requirements under different working conditions. Summary of the invention

[0004] The present invention provides a heat dissipation method, device, equipment and storage medium for an electric scooter driving system, which solves the problem of low space utilization of traditional heat dissipation solutions and improves the heat dissipation effect of the driving system.

[0005] In a first aspect, the present invention provides a heat dissipation method for an electric scooter driving system, the heat dissipation method for an electric scooter driving system comprising:

[0006] The heat source distribution data and space constraint data of the driving system in the electric scooter are input into the two-stage optimization model to calculate the heat dissipation structure and obtain the heat dissipation channel parameters;

[0007] Based on the heat dissipation channel parameters, the heat source temperature and the heat sink temperature are sampled and synchronized in real time to obtain heat flow distribution characteristics;

[0008] Performing heat conduction model calculation and multi-region temperature classification on the heat flow distribution characteristics to obtain temperature dynamic distribution data, and inputting the temperature dynamic distribution data into a temperature field prediction model to perform heat flow prediction calculation to obtain heat flow prediction data;

[0009] Performing thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters;

[0010] The heat load is divided and the heat dissipation power is optimized for the driving condition according to the heat dissipation compensation parameters to obtain the heat dissipation control parameters.

[0011] In a second aspect, the present invention provides a heat dissipation device for a driving system of an electric scooter, the heat dissipation device for a driving system of an electric scooter comprising:

[0012] A calculation module, used to input the heat source distribution data and space constraint data of the driving system in the electric scooter into the two-stage optimization model to calculate the heat dissipation structure and obtain the heat dissipation channel parameters;

[0013] A sampling module, used to perform partition sampling and real-time synchronization of the heat source temperature and the heat sink temperature based on the heat dissipation channel parameters to obtain heat flow distribution characteristics;

[0014] A prediction module is used to perform heat conduction model calculation and multi-region temperature classification on the heat flow distribution characteristics to obtain temperature dynamic distribution data, and input the temperature dynamic distribution data into a temperature field prediction model to perform heat flow prediction calculation to obtain heat flow prediction data;

[0015] A compensation module, used for performing thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters;

[0016] The optimization module is used to divide the heat load and optimize the heat dissipation power of the driving condition according to the heat dissipation compensation parameters to obtain the heat dissipation control parameters.

[0017] A third aspect of the present invention provides an electric scooter drive system heat dissipation device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the electric scooter drive system heat dissipation device executes the above-mentioned electric scooter drive system heat dissipation method.

[0018] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned method for dissipating heat of an electric scooter drive system.

[0019] In the technical solution provided by the present invention, through the combination of a two-stage optimization model and a hierarchical constraint generation algorithm, the precise design of the heat dissipation structure is realized, the optimal heat dissipation channel layout is obtained in a limited space, and the problem of low space utilization of traditional heat dissipation solutions is solved; a multi-layer temperature monitoring and real-time synchronization strategy is adopted to establish a complete heat flux distribution feature extraction system, accurately grasp the dynamic change law of the temperature field of the system, and based on the dual guarantee of heat conduction model solution and temperature field prediction model, significantly improve the system's prediction accuracy of temperature changes, so that the heat dissipation control has the ability to respond in advance; through thermal disturbance analysis and adaptive compensation mechanism, a complete interference suppression system is established, which effectively reduces the impact of external environmental changes on the system temperature; a condition-adaptive heat load division method is adopted to achieve accurate distribution of heat dissipation power, greatly improving the heat dissipation efficiency of the system under different working conditions; a complete thermal management control system is constructed, and through multi-objective optimization and dynamic adjustment, the system has intelligent heat dissipation control capabilities.

[0020] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic diagram of an embodiment of a heat dissipation method for an electric scooter driving system according to an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of an embodiment of a heat dissipation device for a driving system of an electric scooter according to an embodiment of the present invention;

[0024] Figure 3 Schematic diagram of an embodiment of a heat dissipation device for an electric scooter driving system in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0027] To facilitate understanding of this embodiment, a heat dissipation method for an electric scooter driving system disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps:

[0028] 101. Input the heat source distribution data and space constraint data of the driving system in the electric scooter into the two-stage optimization model to calculate the heat dissipation structure and obtain the heat dissipation channel parameters;

[0029] It is understandable that the execution subject of the present invention may be a heat dissipation device of the electric scooter driving system, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0030] Specifically, the driving system of the electric scooter is divided into three-dimensional grids, and the entire heat dissipation space is divided into several grid units to generate spatial discretization data, which is convenient for the subsequent accurate identification and calculation of the heat source position and heat flow path. The mutual connection between the grid units and the heat conduction path constitute the basis of the system heat dissipation analysis. The heat source is identified on the spatial discretization data to determine the key heat source position inside the motor. The heat source includes the stator winding, rotor and bearing of the motor. By accurately locating the heat source position of the key components, the specific heat source node data is obtained. The heat source node data is input into the first stage of the two-stage optimization model to calculate the number of radiators. In the calculation process, the thermal conductivity of the heat dissipation material and the spatial volume constraint are comprehensively considered to determine the optimal number of radiators. In this stage, the thermal conductivity of the material plays a key role in determining the heat dissipation efficiency, while the volume constraint needs to consider the feasibility of installing the radiator in the limited internal space of the scooter. Based on the balance of these two factors, the number of radiators suitable for the system is obtained. Based on the number of radiators, the length, width and height of the radiator are optimized. The heat sink size is adjusted to maximize the heat dissipation performance while meeting the constraints of space and heat conduction. In the size optimization process, the heat sink material characteristics, space geometric limitations and installation feasibility are considered to obtain the heat sink size parameters. The heat sink size parameters are input into the second stage of the two-stage optimization model to calculate the heat sink spacing. Based on the hierarchical constraint generation algorithm, the spacing range between adjacent heat sinks is determined to avoid mutual influence between heat sinks and ensure the optimal heat transfer efficiency. The heat sink arrangement parameters are obtained by optimizing the heat sink arrangement. This parameter determines the arrangement method between heat sinks to maximize the heat dissipation surface area and heat flow, thereby improving the overall heat dissipation performance. The heat conduction path analysis is performed on the heat sink arrangement parameters, and the heat flow transfer path of the entire system is calculated based on the heat conduction equation. The application of the heat conduction equation can accurately describe the heat transfer method in each grid unit and generate heat flow conduction data. Based on the heat flow conduction data, the heat dissipation channel is optimized to determine the cross-sectional area and inclination angle of the heat dissipation channel. In the process of determining the parameters, the heat flow law and fluid mechanics characteristics in the heat dissipation channel are taken into account to ensure that the heat can be effectively dissipated through the channel in the shortest possible time. The cross-sectional area and inclination angle of the heat dissipation channel need to adapt to the constraints of the physical space and ensure that the heat flow can pass smoothly, minimize the flow resistance, and reduce the possibility of local overheating. Through this optimization process, the channel structure data is obtained to ensure that the design of the heat dissipation channel is optimal in terms of mechanics and thermodynamics. The flow field distribution of the channel structure data is calculated to obtain the heat dissipation channel parameters.Through flow field distribution calculation, the flow state of heat flux in the heat dissipation channel is predicted, potential hot spots and inefficient areas are identified, and finally the heat dissipation channel parameters are obtained, including the geometry and size of the channel, as well as the distribution characteristics of the internal heat flux, so as to ensure that the heat dissipation process of the entire electric scooter drive system can be carried out stably and efficiently to achieve the ideal heat dissipation effect.

[0031] The heat flux conduction data is matrix decomposed. By converting the heat flux density data into an eigenvector matrix and an eigenvalue matrix, the complexity of the heat flux data is simplified, and the original heat flux distribution information is disassembled into a mathematical form that is easier to analyze and process, and characteristic data representing the heat flux characteristics are obtained. The heat flux characteristic data is input into the adaptive cutting plane algorithm, and the direction of the initial plane is determined by analyzing the spatial posture of the cutting plane in the main direction of the heat flux. The main direction of the heat flux distribution, that is, the main path direction of heat transfer, is found according to the eigenvector. In the process of main direction analysis through the cutting plane, the direction information of the eigenvector is used to determine the initial spatial posture of the cutting plane, and the plane direction data is obtained to ensure that the optimized design of the heat dissipation channel can conform to the direction of the natural flow of heat, reduce thermal resistance, and thus improve heat dissipation efficiency. Constraint condition data is constructed according to the plane direction data to better regulate the movement of the cutting plane. Based on the constraint condition data, the cutting plane is iteratively divided along the main direction of the heat flow to generate multiple sub-plane unit data. The cutting process of the cutting plane along the main direction is an iterative optimization process, which makes the geometric shape of the heat dissipation channel more consistent with the characteristic distribution of the heat flow through continuous decomposition and adjustment. Each sub-plane unit represents a part of the heat dissipation channel, and the segmentation operation captures the slight changes in heat flow more finely. The thermal resistance of the sub-plane unit data is calculated to understand the thermal resistance distribution of each part in the heat dissipation channel. Through the thermal resistance calculation, the thermal resistance distribution data of the system is obtained to help identify the high thermal resistance areas in the heat dissipation path so that further measures can be taken to reduce the thermal resistance of these areas. Based on the thermal resistance distribution data, the position of the cutting plane is adaptively adjusted to obtain the cross-sectional optimization data. The adaptive adjustment process ensures that the position of the cutting plane can be dynamically adjusted as the heat flow characteristics change, so that the cross-sectional area of ​​the channel can reduce the thermal resistance as much as possible at each position to achieve the best heat dissipation effect. Based on the cross-sectional optimization data, the natural convection effect at different inclination angles is calculated to obtain the inclination angle optimization data. The effect of natural convection depends on the inclination angle of the channel. The appropriate inclination angle can enhance convection and improve the heat dissipation efficiency. The calculation of the inclination angle optimization data needs to fully consider the characteristics of air flow and the driving force of the temperature difference between the inside and outside of the channel on the flow, so as to find the optimal inclination angle that can maximize the natural convection effect under geometric constraints. The inclination angle optimization data and the cross-section optimization data are combined to generate the spatial configuration of the heat dissipation channel and obtain the complete channel structure data. The optimization results of the cross-sectional area and the inclination angle are combined to form a three-dimensional, overall heat dissipation channel design. The flow field distribution of the generated channel structure data is calculated to obtain the heat dissipation channel parameters. The goal of the flow field distribution calculation is to simulate and analyze the flow state of air or other cooling media in the heat dissipation channel to ensure that the designed heat dissipation channel can effectively guide heat to the outside during actual operation.In the flow field distribution calculation, the fluid movement in the channel is numerically simulated to identify the existing flow dead corners or eddy current areas, and the geometry of the channel is fine-tuned to address these problems to ensure that the efficiency of the entire heat dissipation process is optimized.

[0032] 102. Based on the heat dissipation channel parameters, the heat source temperature and the heat sink temperature are sampled and synchronized in real time to obtain the heat flow distribution characteristics;

[0033] Specifically, the heat source of the drive system is layered based on the heat dissipation channel parameters to obtain multi-layer heat source data. The heat source in the entire drive system is divided into multiple independent heat source layers, such as the motor stator winding layer, the heat sink layer and the controller layer. Each layer represents a different heat dissipation unit. This layered method can more accurately analyze the heat generation and dissipation of each component. Hot spots are located based on the multi-layer heat source data, and the spatial coordinates of the temperature measurement points in the system are accurately analyzed to obtain the temperature measurement point distribution data. The temperature measurement point distribution data is input into the three-layer temperature monitoring model. The three-layer temperature monitoring model performs partition temperature sampling for the stator winding layer, the heat sink layer and the controller layer respectively to obtain the inter-layer temperature data. In the process of partition temperature sampling, the temperature change of each layer is recorded independently to obtain a more refined heat source temperature distribution. These data help capture the temperature changes of each heat dissipation layer under different working conditions. The temperature data of each layer is aligned, and the sampled data of each layer is synchronously calibrated according to the timestamp to obtain multi-layer temperature field data, ensuring that the temperature data of each layer can be compared and analyzed under the same time reference, thereby eliminating the error caused by the sampling time difference. Based on the multi-layer temperature field data, the thermal flow field is reconstructed and the temperature gradient is spatially mapped. The thermal flow field reconstruction process analyzes the temperature change law between each layer and integrates these data into an overall thermal field distribution map, thereby intuitively displaying the heat flow transfer state inside the system. At the same time, the spatial mapping of the temperature gradient helps to identify the temperature differences between various regions in the system, helps to determine the main path of the heat flow and the potential thermal resistance area, and obtains the thermal field reconstruction data. The thermal field reconstruction data is associated with the real-time speed data of the motor, and the mapping relationship between the temperature field and the motor speed is established to obtain the speed correlation data. During the operation of the electric scooter, the speed of the motor is one of the important factors affecting the heat generation. By associating the temperature field data with the speed data, we can better understand the impact of different speeds on the heat generation and dissipation inside the system. The speed correlation data is coupled with the load torque data to establish the dynamic response relationship between the temperature field and the load, and obtain the load response data. Load torque is another key factor that directly affects the heat generation. Different load states will cause different heat to be generated inside the motor, so the change of the temperature field will also be affected by the load accordingly. By coupling the speed and load data, the dynamic response characteristics of the motor temperature field under various working conditions are comprehensively analyzed. Based on the load response data, the temperature field is dynamically decomposed and reorganized to finally obtain the heat flux distribution characteristics. The process of dynamic feature decomposition is to decompose the temperature field data into multiple independent feature patterns, which reflect the heat transfer characteristics of the system at different times and under different working conditions. By reorganizing these features, a heat flux distribution feature map is constructed.

[0034] 103. Perform heat conduction model calculation and multi-region temperature classification on the heat flow distribution characteristics to obtain temperature dynamic distribution data, and input the temperature dynamic distribution data into the temperature field prediction model to perform heat flow prediction calculation to obtain heat flow prediction data;

[0035] Specifically, based on the heat flow distribution characteristics, the position of the radiator and the distribution of the heat source are analyzed, and the corresponding heat conduction differential equations are constructed to obtain the basic heat conduction model used to describe the heat transfer law in the system. The construction of the basic heat conduction model is based on the classical heat conduction equation. By introducing the boundary conditions of each part of the system, the thermal conductivity of the material, and the power of the heat source, a mathematical model that can fully reflect the heat conduction mechanism in the system is established. Based on the basic heat conduction model, signal decomposition and multi-scale analysis are performed to decompose the complex heat transfer process into behaviors at different time and space scales to obtain multi-scale temperature data, which is easier to capture the slight changes in heat transfer in the system. Multi-point temperature classification is performed on the multi-scale temperature data to accurately divide the temperature of different areas in the system. Through multi-point temperature classification, temperature classification data of each area is obtained, which represents the temperature change characteristics of different positions in the system. Feature dimensionality reduction operation is performed on the temperature classification data to obtain temperature principal component data. The dimension of the temperature data is reduced by principal component analysis, thereby simplifying the data structure, making subsequent analysis and prediction more efficient and easy to implement. In this process, the original high-dimensional temperature data is compressed into a few principal components, which effectively represent the main change trend of the entire temperature distribution, thereby reducing the computational complexity while maintaining the core information of the temperature distribution. The obtained temperature principal component data is mapped in time and space, and the change law of the temperature field over time and space is converted into a state transfer equation to obtain the temperature dynamic distribution data. The temperature distribution data is converted from a static temperature value to a dynamic description that changes with time and space, so as to better understand the change law of temperature in different regions and time periods. By constructing a state transfer equation to describe the process of temperature change in the system over time and space, the evolution law of the entire temperature field is made clearer. The temperature dynamic distribution data is state predicted to obtain temperature prediction data. State prediction is based on the state transfer equation, and uses past temperature distribution data to infer future temperature changes to ensure that measures can be taken in advance in actual operation to prevent overheating. In order to improve the accuracy of the prediction, the temperature prediction data is compensated for the minimum temperature deviation to obtain more accurate temperature correction data. The purpose of the minimum temperature deviation compensation is to correct the error during the prediction process, so that the result of the temperature prediction is closer to the true value and the reliability of the prediction model is improved. The temperature correction data is input into the temperature field prediction model to calculate the heat transfer to obtain the final heat flow prediction data. In the temperature field prediction model, the heat flow distribution state of the entire system in the future time is obtained by simulating and calculating the heat transfer process.

[0036] 104. Perform thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters;

[0037] Specifically, the heat flow prediction data is segmented and the interference pattern is identified. The heat flow prediction data is segmented according to a specific temperature change law to form several time intervals, each of which represents the distribution state of the heat flow in a specific time period. Through interval segmentation, the abnormal change pattern in the heat flow can be better identified, especially those temperature mutations caused by interference factors. Interference pattern recognition is performed on the segmented data to effectively extract the characteristic data of temperature interference. These temperature interference data reflect the impact of some uncontrollable factors inside or outside the system on the temperature field, such as the instantaneous increase of load, the power fluctuation of the motor, etc. The time domain feature extraction of the temperature interference data is performed to obtain the time domain feature data. By analyzing the change law of the temperature interference data on the time axis, the frequency, amplitude and duration of the interference are captured. The time domain feature data is input into the temperature interference analysis model to establish a mapping relationship between the interference source and the temperature fluctuation, so as to identify the location and intensity of the target interference source and obtain the interference source data. Through this mapping relationship, the specific source of the temperature fluctuation is determined, such as the overheating of a motor winding, the sudden change of the external environment temperature or the abnormal heat flow caused by the power change. The interference source data not only includes the location of the interference, but also includes the specific impact intensity of the interference on the system temperature fluctuation. The spatial distribution analysis operation of the interference source data is performed to understand how these interferences propagate and attenuate within the system. By constructing the propagation path and attenuation law of the interference, the interference transmission data is obtained to describe how the interference spreads in space and its energy attenuation during the transmission process. The analysis of interference propagation can help identify which areas in the system are susceptible to interference and help determine the effective compensation position. Based on the interference transmission data, the interference characteristics on different transmission paths are classified and combined to obtain thermal interference characteristic data. The compensation control law is constructed for the thermal interference characteristic data to determine how to effectively compensate for these interferences and obtain compensation control data. The construction of the compensation control law is based on the thermal interference characteristics in the system, combined with the response capability of the heat dissipation system, and through the method of mathematical modeling, a strategy for real-time regulation of the interference source is formed. The compensation control data describes how to adjust the power of the radiator, the speed of the fan, or other active heat dissipation measures to effectively offset the temperature fluctuations caused by the interference. According to the compensation control data, a compensation control model of the heat dissipation system is established to calculate the optimal compensation amount and obtain the heat dissipation compensation parameters, which describe the specific values ​​of the compensation measures that the system needs to take under different interference conditions, including the strength, duration, and specific location of the compensation. These parameters are used to adjust the heat dissipation strategy of the system to ensure that the temperature of the electric scooter drive system can be kept within a reasonable range under the influence of various interferences.

[0038] 105. According to the heat dissipation compensation parameters, the heat load of the driving condition is divided and the heat dissipation power is optimized to obtain the heat dissipation control parameters.

[0039] Specifically, based on the heat dissipation compensation parameters, the corresponding relationship between the motor load rate and the heat dissipation compensation amount is constructed to generate the working condition characteristic data. By linking the running load of the motor with the corresponding heat dissipation demand, the heat dissipation requirements under different loads can be clearly defined. The working condition characteristic data represents the changes in the heat dissipation requirements of the system under various load conditions. The partition boundary calculation is performed on the working condition characteristic data to obtain the working condition boundary data. Different working conditions are divided into several independent areas, each area corresponds to different loads and heat dissipation requirements, so that the heat dissipation requirements of different areas can be accurately controlled in actual operation. Based on the working condition boundary data, the working condition membership function is established, and the entire working condition space is smoothly divided to obtain the working condition division data. The function of the working condition membership function is to divide the working condition space into multiple membership areas, so that each working condition state can be smoothly mapped to a specific area. The smooth division method can avoid the sudden change of heat dissipation control due to the drastic change of the working condition, so that the heat dissipation regulation of the system when dealing with different loads is more stable and continuous. Through further processing of the working condition division data, the heat load is calculated for each divided area to obtain the heat load partition data. The heat load partition data describes the specific heat generation in each operating area. Based on the heat load partition data, the difference between the maximum heat dissipation of the radiator and the current heat load is calculated to determine the power demand of the system and obtain the power demand data. By analyzing the matching between the heat dissipation capacity of the radiator under the current conditions and the heat load, it is determined whether the heat dissipation power needs to be increased or whether the working state of the radiator needs to be adjusted. The power demand data reflects the additional heat dissipation power required by the system under the current working conditions. The power demand data is mapped with control parameters to obtain control mapping data. The process of control parameter mapping is to convert the power demand into specific control instructions, such as adjusting the speed of the fan, changing the cooling medium flow of the radiator, etc., to ensure that the system can be regulated according to the expected heat dissipation requirements. Based on the control mapping data, multi-objective optimization is performed to obtain optimized control data. Under the premise of meeting the heat dissipation requirements, the comprehensive optimization of multiple objectives such as minimizing energy consumption, minimizing noise and maximizing system reliability is achieved. In the process of multi-objective optimization, multiple influencing factors such as heat dissipation efficiency, system stability, energy consumption, etc. are comprehensively considered, and finally a compromise solution is found to make the heat dissipation control achieve the best state as a whole. Through multi-objective optimization, the optimized control data obtained ensures that the system can manage heat dissipation in the best way under various working conditions. Based on the optimized control data, a partition control rule library and a switching mechanism are constructed to obtain heat dissipation control parameters. The optimized control data under different working conditions are converted into a set of executable rules, which describe how the system should adjust the heat dissipation parameters in each working area. The switching mechanism is used to switch between different control rules when the system working conditions change to ensure the continuity and stability of heat dissipation control.By combining the partition control rule base and the switching mechanism, dynamic control of the entire cooling system is achieved, allowing the system to quickly adapt to and maintain the optimal cooling state under any operating conditions.

[0040] In the embodiment of the present invention, through the combination of a two-stage optimization model and a hierarchical constraint generation algorithm, the precise design of the heat dissipation structure is achieved, the optimal heat dissipation channel layout is obtained in a limited space, and the problem of low space utilization of traditional heat dissipation solutions is solved; a multi-layer temperature monitoring and real-time synchronization strategy is adopted to establish a complete heat flux distribution feature extraction system, accurately grasp the dynamic change law of the temperature field of the system, and based on the dual guarantee of heat conduction model solution and temperature field prediction model, significantly improve the system's prediction accuracy of temperature changes, so that the heat dissipation control has the ability to respond in advance; through thermal disturbance analysis and adaptive compensation mechanism, a complete interference suppression system is established, which effectively reduces the impact of external environmental changes on the system temperature; the working condition adaptive heat load division method is adopted to achieve accurate distribution of heat dissipation power, which greatly improves the heat dissipation efficiency of the system under different working conditions; a complete thermal management control system is constructed, and through multi-objective optimization and dynamic adjustment, the system has intelligent heat dissipation control capabilities.

[0041] In a specific embodiment, the process of executing step 101 may specifically include the following steps:

[0042] (1) Perform three-dimensional meshing on the drive system of the electric scooter and divide the heat dissipation space into several grid cells to obtain spatial discretization data;

[0043] (2) Perform heat source identification on the spatial discretized data to determine the heat source locations of the motor stator winding, rotor, and bearings, obtain heat source node data, and input the heat source node data into the first stage of the two-stage optimization model to calculate the number of heat sinks. The heat sink number parameters are obtained based on the thermal conductivity and volume constraints of the material.

[0044] (3) Based on the number of heat sinks, the length, width and height of the heat sink are optimized to obtain the heat sink size parameters;

[0045] (4) Inputting the radiator size parameters into the second stage of the two-stage optimization model to calculate the radiator spacing, determining the spacing range of adjacent radiators based on the hierarchical constraint generation algorithm, and obtaining the radiator arrangement parameters;

[0046] (5) Perform heat conduction path analysis on the heat sink arrangement parameters, calculate the heat flow transfer path based on the heat conduction equation, and obtain heat flow conduction data;

[0047] (6) Optimize the heat conduction data to determine the cross-sectional area and inclination angle of the heat dissipation channel, obtain the channel structure data, and calculate the flow field distribution of the channel structure data to obtain the heat dissipation channel parameters.

[0048] Specifically, the geometric structure of the electric scooter drive system is divided into three-dimensional grids, and the heat dissipation space is divided into a number of grid units to obtain spatial discretization data. For example, using finite element analysis software such as ANSYS or COMSOL, the electric scooter drive system model is divided into a number of volume units to obtain a three-dimensional grid. These grid units are recorded as ,in is the number of the grid unit. The spatial coordinates of each grid unit are , which transforms the complex heat dissipation problem into a local calculation for each small unit, so as to better describe the spatial distribution of heat flow. The heat source is identified on the spatial discretized data, the location of the main heat sources such as the stator winding, rotor and bearing in the motor is determined, and the heat source node data is obtained. The temperature change of the heat source is crucial to the heat dissipation design of the entire system. The temperature of the heat source node is recorded as , where h represents the type of heat source (such as stator, rotor, bearing), For each heat source node, identify its location in the entire grid and its power density , with this information, the heat source distribution model of the drive system is established. The heat source node data is input into the first stage of the two-stage optimization model to calculate the number of heat sinks. Consider the thermal conductivity of the material And the volume constraints of the heat dissipation space , to determine the optimal number of heat sink parameters The optimization model uses the following objective function to solve the optimal number of radiators:

[0049] ;

[0050] in, is the total heat, For the effective heat dissipation area of ​​each radiator, is a trade-off parameter used to balance heat dissipation efficiency and space constraints. By optimizing the above objective function, the optimal heat sink quantity parameter is obtained. Based on the number of radiators, the geometric dimensions of the radiator (including length ,width ,high ) to obtain the dimensional parameters of the heat sink. By optimizing these dimensions, the surface area of ​​the heat sink is maximized and the heat dissipation efficiency is improved. The optimization process is expressed by the following formula:

[0051] ;

[0052] At the same time, the volume constraints need to be met:

[0053] ;

[0054] By solving these equations, the optimal geometric dimensions that meet the heat dissipation requirements are obtained. The heat sink size parameters are input into the second stage of the two-stage optimization model to calculate the heat sink spacing. The spacing range of adjacent heat sinks is determined through the hierarchical constraint generation algorithm to obtain the heat sink arrangement parameters. The spacing between heat sinks is crucial to improving heat dissipation efficiency. Too small a spacing will result in poor air flow between heat sinks, which will reduce heat dissipation efficiency, while too large a spacing will increase the volume of the system. Through the hierarchical constraint generation algorithm, the spacing of each heat sink is calculated to ensure that its spacing is within the optimal range. This range is determined by the following relationship:

[0055] ;

[0056] in, is the thermal resistance, is the cost function of air flow. Through optimization, the arrangement of the radiator is ensured to achieve the best balance between system volume and heat dissipation efficiency. The heat conduction path of the radiator arrangement parameters is analyzed, and the heat flow transfer path is calculated based on the heat conduction equation to obtain the heat flow conduction data. In this process, the Fourier heat conduction equation is used:

[0057] ;

[0058] in, is the temperature field, is the thermal diffusion coefficient, is the heat source term, is the material density, is the specific heat capacity of the material. By solving the above partial differential equation, the heat transfer path in and around the heat sink is obtained, so as to conduct a detailed heat transfer analysis. Channel optimization is performed on the heat flow conduction data to determine the cross-sectional area of ​​the heat dissipation channel. and tilt angle , and obtain the channel structure data. The goal of channel optimization is to maximize the heat flow in the channel, thereby improving the heat dissipation effect. The determination of is mainly based on the flow resistance and thermal conductivity of the heat flow:

[0059] ;

[0060] in, is the flow resistance, is the channel length, is the thermal conductivity. Similarly, the tilt angle The optimization of is determined by optimizing the natural convection effect:

[0061] ;

[0062] in, is the convective heat transfer coefficient, is the temperature difference. Through the above optimization, the optimal cross-sectional area and inclination angle of the heat dissipation channel are determined to improve the heat dissipation efficiency. The flow field distribution of the obtained channel structure data is calculated to obtain the heat dissipation channel parameters. In the flow field distribution calculation, the flow characteristics of air or cooling medium are considered, and the Navier-Stokes equation is used to describe the movement of the fluid in the heat dissipation channel:

[0063] ;

[0064] in, is the velocity field, is the fluid density, For pressure, is the dynamic viscosity, is the volume force. By solving these equations, the flow field distribution in the heat dissipation channel is obtained, thereby determining the optimal heat dissipation channel parameters.

[0065] In a specific embodiment, the execution step performs channel optimization on the heat flow conduction data, determines the cross-sectional area and the inclination angle of the heat dissipation channel, obtains the channel structure data, and performs flow field distribution calculation on the channel structure data to obtain the heat dissipation channel parameters. The process may specifically include the following steps:

[0066] (1) Perform matrix decomposition on the heat flux conduction data, convert the heat flux density data into eigenvector matrix and eigenvalue matrix, and obtain heat flux characteristic data;

[0067] (2) Input the heat flow characteristic data into the cutting plane in the adaptive cutting plane algorithm for main direction analysis, determine the spatial posture of the initial cutting plane according to the direction of the characteristic vector, and obtain the plane direction data;

[0068] (3) Construct constraint data according to the plane direction data, and iteratively divide the cutting plane along the main direction of the heat flow based on the constraint data to obtain multiple sub-plane unit data;

[0069] (4) Calculate the thermal resistance of the sub-plane unit data to obtain thermal resistance distribution data, and adaptively adjust the position of the cutting plane based on the thermal resistance distribution data to obtain cross-sectional optimization data;

[0070] (5) Based on the cross-sectional optimization data, the natural convection effect at different inclination angles is calculated to obtain the inclination angle optimization data;

[0071] (6) The inclination angle optimization data and the cross-section optimization data are combined to generate the spatial configuration of the heat dissipation channel, obtain the channel structure data, and calculate the flow field distribution of the channel structure data to obtain the heat dissipation channel parameters.

[0072] Specifically, the heat flux data is matrix decomposed to convert the heat flux data into eigenvector matrix and eigenvalue matrix to obtain heat flux characteristic data. In heat conduction analysis, the heat flux data is a complex multi-dimensional data set, which is simplified and features are extracted through matrix decomposition technology. Assume that the heat flux data is a matrix , whose elements Represents the heat flux density at different spatial positions. Perform singular value decomposition on the matrix, namely:

[0073] ;

[0074] in, is the left eigenvector matrix, is a diagonal matrix whose diagonal elements are singular values, is the right eigenvector matrix. Through this decomposition, the heat flux density data is converted into an eigenvector matrix and an eigenvalue matrix. The eigenvector represents the main direction of the heat flow, and the eigenvalue represents the heat flow intensity in these directions. This information is used as heat flow characteristic data for optimizing the heat dissipation design. Based on the heat flow characteristic data, it is input into the adaptive cutting plane algorithm for main direction analysis. In the adaptive cutting plane algorithm, the cutting plane is used to identify the main transfer path of the heat flow. Using the eigenvector matrix , determines the main direction of heat flow, that is, the main eigenvector in the matrix This eigenvector represents the main flow direction of heat flow in space. By analyzing this eigenvector, the spatial posture of the initial cutting plane is determined and the plane direction data is obtained. Let the plane direction be vector n, then , represents the specific direction of the cutting plane in three-dimensional space. According to the plane direction data, the constraint condition data is constructed to ensure that the cutting plane can be reasonably divided along the main direction of the heat flow. The construction of the constraint condition data takes into account the physical boundary conditions of heat flow transfer and the geometric characteristics of the system. Based on these constraints, the cutting plane is iteratively divided along the main direction of the heat flow to obtain multiple sub-plane unit data. Let the initial cutting plane be , by following the main direction By step length Perform iterative segmentation to obtain multiple sub-planes ,in The thermal resistance is calculated for each sub-plane unit data to obtain the thermal resistance distribution data. The thermal resistance is calculated based on the following formula:

[0075] ;

[0076] in, Indicates The thermal resistance of each sub-plane, is the thickness of the subplane, is the thermal conductivity of the material, is the area of ​​the sub-plane. By calculating the thermal resistance of each sub-plane, the thermal resistance distribution data of the entire system is obtained, which reflects the difficulty of heat transfer on different paths. Based on the thermal resistance distribution data, the position of the cutting plane is adaptively adjusted to optimize the heat transfer performance of the entire heat dissipation channel. If the thermal resistance of a sub-plane is too large, the position of the cutting plane needs to be adjusted to reduce the thermal resistance on the path and obtain the cross-sectional optimization data. Based on the cross-sectional optimization data, the natural convection effect at different inclination angles is calculated to obtain the inclination angle optimization data. The efficiency of natural convection depends on the inclination angle of the channel. The appropriate inclination angle can maximize the heat dissipation. Assume the inclination angle of the channel is , calculate the natural convection heat transfer coefficient for each angle :

[0077] ;

[0078] in, and is a constant related to the flow state, is the acceleration due to gravity, is the coefficient of thermal expansion, is the temperature difference, is the characteristic length, is the kinematic viscosity, is the thermal diffusion coefficient. By calculating the natural convection effect at different angles, an optimal angle is found to maximize the natural convection effect, and the inclination optimization data is obtained. The inclination optimization data and the cross-section optimization data are combined to generate the spatial configuration of the heat dissipation channel and obtain the channel structure data. The cross-section optimization data determines the cross-sectional shape and size of the channel, while the inclination optimization data determines the overall direction and angle of the channel. The combination of the two determines the final three-dimensional configuration of the heat dissipation channel. For example, the cross-sectional area of ​​the channel is , whose inclination angle is the optimal angle , then the entire channel is represented as a channel with a cross-sectional area Inclined with the horizontal plane The geometric structure of the channel structure data is used to calculate the flow field distribution to obtain the heat dissipation channel parameters. In the flow field distribution calculation, the flow characteristics of air or cooling medium in the channel are considered, and the Navier-Stokes equation is used to describe the motion state of the fluid in the channel:

[0079] ;

[0080] in, is the fluid density, is the velocity field, For pressure, is the dynamic viscosity, is the external force term. By solving the Navier-Stokes equations, the flow field distribution in the heat dissipation channel is obtained, thereby determining the speed, pressure and other parameters of the air flow in the channel. These heat dissipation channel parameters are used to evaluate the heat dissipation effect of the channel. For example, the higher the flow rate, the faster the heat dissipation, and the pressure distribution helps to determine whether there are problems such as flow dead spots.

[0081] In a specific embodiment, the process of executing step 102 may specifically include the following steps:

[0082] (1) Based on the heat dissipation channel parameters, the drive system is stratified to obtain multi-layer heat source data. Hot spots are located and the spatial coordinates of the temperature measurement points are analyzed to obtain the temperature measurement point distribution data.

[0083] (2) Input the temperature measurement point distribution data into the three-layer temperature monitoring model, perform partition temperature sampling on the stator winding layer, heat sink layer, and controller layer, and obtain the inter-layer temperature data;

[0084] (3) Align the inter-layer temperature data, synchronize and calibrate the sampling data of each layer according to the timestamp, obtain multi-layer temperature field data, and reconstruct the thermal flow field and spatially map the temperature gradient based on the multi-layer temperature field data to obtain thermal field reconstruction data;

[0085] (4) Correlate the thermal field reconstruction data with the motor speed data collected in real time, establish a mapping relationship between the temperature field and the speed, and obtain the speed correlation data;

[0086] (5) The speed correlation data and the load torque data are coupled and calculated to establish the dynamic response relationship between the temperature field and the load, and the load response data is obtained. Based on the load response data, the temperature field is dynamically decomposed and reorganized to obtain the heat flux distribution characteristics.

[0087] Specifically, the heat source of the drive system is layered based on the heat dissipation channel parameters, and they are divided into several layers according to the distribution of different heat sources and their respective thermal characteristics, so as to facilitate accurate temperature monitoring and analysis. In this embodiment, the heat sources of the drive system are mainly concentrated in the stator windings, heat sinks and controller parts of the motor. Assume that through the heat source layering, data of multiple heat source layers are obtained, and each layer is regarded as an independent heat source area, and its corresponding thermal power is expressed as Indicates that Represent different layers, such as stator winding layer, heat sink layer and controller layer. Perform hotspot location and temperature measurement point spatial coordinate analysis on multi-layer heat source data, determine the area with the highest temperature in each layer of heat source, that is, the hotspot, and arrange temperature measurement points around the hotspot. Assume that the temperature distribution in each layer is represented by the temperature field function Indicates that is the three-dimensional spatial coordinate. By performing gradient analysis on the temperature field of each layer, the location with the largest temperature gradient is found, which is the hotspot of the system. Several temperature measurement points are arranged around the hotspot to obtain the spatial coordinates of the temperature measurement points. These coordinate information constitutes the temperature measurement point distribution data, which is used for subsequent real-time temperature monitoring to ensure that the system can respond and control the temperature changes in the high-temperature area in a timely manner. The temperature measurement point distribution data is input into the three-layer temperature monitoring model, and the temperature of different heat source layers is sampled through the model. The temperature monitoring model is set to three layers of monitoring, corresponding to the motor stator winding layer, heat sink layer and controller layer. For each temperature measurement point, record its temperature changes over time. ,in is the number of the temperature measurement point, is time. By sampling, we can obtain the temperature variation data of each layer over time, i.e., the interlayer temperature data. Assume that at a certain moment The sampled temperatures of the stator winding layer, heat sink layer, and controller layer are , and After obtaining the interlayer temperature data, data alignment is performed. The data of different layers are synchronized and calibrated according to the timestamp to ensure that the temperature data of each layer are compared and analyzed under the same time reference. Assume that the time interval of temperature sampling is , for each time point , the temperature data of all layers need to be aligned, that is , , Must be in the same The aligned data are compared and processed at the same time. These aligned data form multi-layer temperature field data, describing the change of temperature inside the drive system over time and space. Based on the multi-layer temperature field data, the thermal flow field is reconstructed and the temperature gradient space is mapped. The thermal flow field reconstruction calculates the heat flow distribution inside the system through the temperature field data. The heat flux density is calculated using Fourier's law:

[0088] ;

[0089] in, is the heat flux vector, is the thermal conductivity, is the temperature gradient. Through this formula, the heat flow field inside the entire system is reconstructed according to the temperature field data to obtain the thermal field reconstruction data. The spatial mapping of the temperature gradient helps identify the areas with the fastest temperature changes by calculating the rate of change of temperature in space. These areas are usually areas that require special attention in order to optimize the heat dissipation design. The thermal field reconstruction data is associated with the real-time collected motor speed data to establish a mapping relationship between the temperature field and the speed. The speed of the motor The temperature inside the system is significantly affected. The higher the speed, the more heat is generated, which leads to a temperature rise. By correlating the temperature field data with the motor speed data, the temperature field is established. and speed The mapping relationship between them is usually described by a linear regression model:

[0090] ;

[0091] in, and is the regression coefficient, which reflects the influence of the speed on the temperature field. In this way, the speed-related data is obtained to describe the change law of the temperature field under different speeds. The speed-related data is coupled with the load torque data to establish the dynamic response relationship between the temperature field and the load. Load torque It is one of the important factors affecting temperature. When the load increases, the motor needs to output more power, which leads to higher heat generation. By coupling the speed correlation data with the load torque data, the dynamic response relationship between the temperature field and the load is obtained. The coupling relationship is described by the following formula:

[0092] ;

[0093] in, is the coefficient describing the effect of load on temperature, , is the aforementioned regression coefficient. Through this coupling model, the dynamic response of the temperature field under different load and speed conditions is described to obtain the load response data. Based on the load response data, the dynamic characteristics of the temperature field are decomposed and reorganized to obtain the heat flow distribution characteristics. The time series data of the temperature field is decomposed into several independent characteristic modes, and the temperature data is represented as a set of orthogonal principal components through dimensionality reduction methods such as principal component analysis:

[0094] ;

[0095] in, is the time-dependent coefficient, is the spatially dependent basis function, The number of main components. Through this decomposition method, the complex temperature field data is simplified into a few characteristic modes, which is convenient for analysis and control. These characteristic modes are reorganized to generate a comprehensive heat flow distribution characteristic diagram, which describes the main transfer paths and distribution characteristics of the heat flow inside the system.

[0096] In a specific embodiment, the process of executing step 103 may specifically include the following steps:

[0097] (1) Based on the heat flux distribution characteristics, the heat conduction differential equations are constructed for the heat sink position and heat source distribution to obtain the basic heat conduction model. Signal decomposition and multi-scale analysis are performed based on the basic heat conduction model to obtain multi-scale temperature data.

[0098] (2) Perform multi-point temperature classification on the multi-scale temperature data to obtain temperature classification data, and perform feature dimension reduction on the temperature classification data to obtain temperature principal component data;

[0099] (3) Perform time-space mapping on the temperature principal component data, and transform the variation law of the temperature field over time and space into a state transfer equation to obtain the temperature dynamic distribution data;

[0100] (4) Perform state prediction on the temperature dynamic distribution data to obtain temperature prediction data, perform minimum temperature deviation compensation on the temperature prediction data to obtain temperature correction data, and input the temperature correction data into the temperature field prediction model to perform heat transfer calculation to obtain heat flow prediction data.

[0101] Specifically, based on the heat flux distribution characteristics, the heat conduction differential equations are constructed for the position of the radiator and the distribution of the heat source to obtain the basic heat conduction model. The heat flux distribution characteristics include the position, size and direction of each heat source in the system. Using these characteristics, a heat conduction equation group that reflects the heat transfer in the system is constructed. The Fourier heat conduction equation is the core mathematical tool that describes how heat is transferred in space:

[0102] ;

[0103] in, Indicates at location and time The temperature at is the thermal diffusion coefficient, is the thermal conductivity, is the density of the material, is the specific heat capacity, is the heat source term, which describes the heat generation at different locations. The heat conduction equations are the basic model of heat conduction, which describes the generation and diffusion of heat in the drive system. Signal decomposition and multi-scale analysis are performed on the basis of the basic model of heat conduction to obtain multi-scale temperature data. The heat transfer behavior of the system at different time and space scales is captured to more fully understand the law of temperature change. The temperature signal is decomposed using wavelet transform. Wavelet transform is a tool suitable for non-stationary signal analysis, which decomposes complex temperature signals into multiple components of different frequencies and scales. For example, assuming the temperature signal is , through wavelet transform, it can be expressed as:

[0104] ;

[0105] in, is the wavelet coefficient, is the wavelet basis function, and Represent the index of scale and position respectively. Through multi-scale decomposition, the temperature signal components at different scales are obtained, that is, multi-scale temperature data. These data help to identify the temperature change characteristics at different frequencies in the system. Multi-point temperature classification is performed on multi-scale temperature data. The temperature data at different locations and times are classified to find the common characteristics of temperature distribution. Suppose the multi-scale temperature data is , indicating the Scale, location The temperature at . By performing cluster analysis on these data, the temperature data are classified into several groups, each representing an area with a similar temperature change pattern. The obtained temperature classification data helps to simplify the subsequent analysis process. Feature dimensionality reduction is performed on the temperature classification data to obtain the temperature principal component data. The purpose of feature dimensionality reduction is to reduce the dimension of the data so that the high-dimensional temperature data can be represented by fewer variables while maintaining the main features of the data. Assume that the temperature classification data is is a The matrix of represents the number of samples, Represents the feature number. Through principal component analysis, the original data is reduced to The principal components ( :

[0106] ;

[0107] in, is a left singular matrix, is a diagonal matrix whose diagonal elements are singular values, is a right singular matrix. singular values ​​and corresponding eigenvectors to obtain the temperature principal component data. The temperature principal component data represents the main temperature change mode in the system. The temperature principal component data is mapped in time and space, and the change law of the temperature field over time and space is converted into a state transfer equation to obtain the temperature dynamic distribution data. The purpose of time and space mapping is to establish a mathematical model to describe the temperature change process, which is used to predict the temperature change of the system in the future. Assume that the temperature principal component is , and its evolution over time is described by the state transition equation:

[0108] ;

[0109] in, is the state transfer matrix, which describes the intrinsic dynamic characteristics of the system. is the input matrix, is an external input (such as load change). Through the state equation, the temperature dynamic distribution data is obtained to describe the temperature change trend of the system under different working conditions. The state prediction of the temperature dynamic distribution data is performed to obtain the temperature prediction data. In the state prediction, the state transition equation is used to predict the future temperature, and the temperature change is understood in advance so that corresponding heat dissipation measures can be taken. Assume that at time The temperature state is , then in the future time The temperature state of is predicted by the equation of state:

[0110] ;

[0111] Through this formula, the temperature prediction data for the future time is obtained. The temperature prediction data is compensated for the minimum temperature deviation to obtain the temperature correction data. The prediction error caused by model error or external interference is corrected. Assume that the predicted temperature is The actual measured temperature is , then the temperature correction data after compensation is:

[0112] ;

[0113] in, The compensation coefficient is adjusted to minimize the deviation between the predicted value and the actual value. The temperature correction data is input into the temperature field prediction model to calculate the heat transfer to obtain the final heat flow prediction data. By combining the corrected temperature data, the temperature field prediction model more accurately calculates the heat transfer in the system and obtains the heat flow prediction data. These data are used to evaluate the heat load of each part of the system to ensure that the radiator and other cooling equipment can meet future heat dissipation needs.

[0114] In a specific embodiment, the process of executing step 104 may specifically include the following steps:

[0115] (1) Perform interval segmentation and interference pattern recognition on the heat flow prediction data to obtain temperature interference data, and perform time domain feature extraction on the temperature interference data to obtain time domain feature data;

[0116] (2) Input the time domain feature data into the temperature interference analysis model, and by establishing a mapping relationship between the interference source and the temperature fluctuation, identify the location and intensity of the target interference source and obtain the interference source data;

[0117] (3) Perform spatial distribution analysis on the interference source data, obtain interference transmission data by constructing the interference propagation path and attenuation law, and classify and combine the interference features on different transmission paths based on the interference transmission data to obtain thermal interference feature data;

[0118] (4) Construct a compensation control law for the thermal interference characteristic data to obtain compensation control data. Based on the compensation control data, establish a compensation control model for the heat dissipation system to calculate the optimal compensation amount and obtain the heat dissipation compensation parameters.

[0119] Specifically, the heat flow prediction data is segmented and the interference pattern is identified to obtain temperature interference data. The heat flow prediction data describes the heat flow distribution at different times and spaces within the system. These data contain temperature disturbances caused by factors such as load changes and environmental fluctuations. In order to identify these disturbances, the heat flow prediction data is segmented, that is, the entire data set is divided into several small intervals according to a certain time period or space segment. Assume that the heat flow data is a function , then the time The axis is divided into several intervals, each interval is recorded as . In each interval, the data is analyzed to identify abnormal patterns in temperature, which are interference patterns, and the temperature interference data is obtained. The interference pattern is identified by calculating the rate of change of temperature. If the rate of change exceeds a certain threshold, it is considered that there is interference in the interval. The time domain feature extraction of the temperature interference data is performed to obtain the time domain feature data. Describe the time variation characteristics of the interference in order to better understand the nature of the interference. Suppose the interference data is By analyzing the characteristics of the signal, a series of time domain features are extracted, including the average value , Standard Deviation , Peak and rise time Etc. Through these time domain feature data, the intensity, fluctuation amplitude and frequency of the interference are quantitatively described. The time domain feature data is input into the temperature interference analysis model, and the location and intensity of the target interference source are identified by establishing a mapping relationship between the interference source and the temperature fluctuation, and the interference source data is obtained. The core of the temperature interference analysis model is to establish a causal relationship between the interference source and the system temperature fluctuation, which is achieved through regression analysis or machine learning models. Assume that the location of the interference source is , the interference intensity is By fitting the relationship between the time domain characteristic data and the potential heat source in the system, the specific location and strength of the interference source can be obtained. For example, a linear regression model is used to estimate the intensity of the interference source:

[0120] ;

[0121] in, and is the regression coefficient, is the number of features. After identifying the interference source data, perform spatial distribution analysis on the interference source data to understand the propagation of the interference in the system. By constructing the propagation path and attenuation law of the interference, the interference transmission data is obtained. The propagation path of the interference is simulated by the heat conduction equation:

[0122] ;

[0123] in, For location and time The temperature at is the thermal diffusion coefficient, is the heat source term, which represents the heat input of the interference source. By numerically solving this equation, we can obtain the propagation of interference in space and its attenuation law at different positions. Based on the interference transmission data, the interference characteristics on different transmission paths are classified and combined to obtain thermal interference characteristic data. The compensation control law is constructed for the thermal interference characteristic data to obtain the compensation control data. Determine how to effectively offset or reduce the impact of the interference source on the system temperature. Let the interference characteristic be , the compensation control law is expressed by designing a control function:

[0124] ;

[0125] in, is the control input, is the reference temperature, , , are proportional, differential, and integral control gains. By adjusting these gain coefficients, rapid response and compensation to disturbances can be achieved to maintain the stable temperature of the system. Based on the compensation control data, a compensation control model for the heat dissipation system is established to calculate the optimal compensation amount and obtain the heat dissipation compensation parameters. The compensation control model finds the best control input through optimization calculation to minimize the system temperature fluctuation. Assume that the heat dissipation power of the radiator is , the optimal compensation amount is determined by optimizing the problem:

[0126] ;

[0127] Subject to system physical constraints, such as maximum heat dissipation power limit and minimum power limit By solving the optimization problem, the best heat dissipation power control strategy is obtained, thereby determining the heat dissipation compensation parameters. .

[0128] In a specific embodiment, the process of executing step 105 may specifically include the following steps:

[0129] (1) Based on the heat dissipation compensation parameters, the corresponding relationship between the motor load rate and the heat dissipation compensation amount is constructed to obtain the operating condition characteristic data, and the partition boundary calculation is performed on the operating condition characteristic data to obtain the operating condition boundary data;

[0130] (2) Based on the working condition boundary data, the working condition membership function is established to smoothly divide the working condition space to obtain the working condition division data, and the heat load is calculated on the working condition division data to obtain the heat load partition data;

[0131] (3) Based on the heat load partition data, the difference between the maximum heat dissipation of the radiator and the current heat load is calculated to determine the power demand, obtain the power demand data, and perform control parameter mapping on the power demand data to obtain control mapping data;

[0132] (4) Perform multi-objective optimization based on the control mapping data to obtain optimized control data, and build a partition control rule library and switching mechanism based on the optimized control data to obtain the heat dissipation control parameters.

[0133] Specifically, the heat dissipation compensation parameter is used to describe the heat dissipation required by the system under different working conditions. Assume that the load rate of the motor is , refers to the ratio of the motor's current load to its maximum load, usually varying in the interval [0,1], and the heat dissipation compensation is expressed as , that is, the amount of heat that needs to be compensated to maintain the temperature stability of the system under different loads. In order to establish the corresponding relationship between the load rate and the amount of heat dissipation compensation, a functional relationship is obtained by fitting the experimental data as shown below:

[0134] ;

[0135] in, and The fitting coefficient is obtained through experimental data regression, which describes the nonlinear relationship between the load rate and the heat dissipation compensation. Based on this relationship, the system's operating characteristic data under different load conditions is obtained. The operating characteristic data describes the required heat dissipation under different load rates. The partition boundary calculation is performed on the operating characteristic data to obtain the operating boundary data. The operating space is divided into several independent areas, each area represents different load conditions and corresponding heat dissipation requirements. Assuming the load rate The range of variation is from 0 to 1, dividing the entire load space into intervals, such as , , and so on. Each interval corresponds to a different heat dissipation requirement, and the boundaries of these intervals are expressed as , called the working condition boundary data. The boundary is determined by clustering the load rate so that the working condition characteristics in each interval have similar heat dissipation requirements. Based on the obtained working condition boundary data, the working condition membership function is established to smoothly divide the entire working condition space to obtain the working condition division data. The working condition membership function is a mathematical tool used to describe the possibility that a specific load rate belongs to a certain interval. The membership function is established using fuzzy logic methods. For example, for a certain interval , load factor The membership function of is expressed as:

[0136] ;

[0137] The membership function smoothly divides the entire working condition space to obtain working condition division data. The working condition division data describes which working condition area the system load rate belongs to at a certain moment, and the load conditions in each area are relatively smooth and continuous, which helps to reduce the jump phenomenon in the control process and ensure the stable operation of the system. The heat load is calculated for the working condition division data to obtain the heat load partition data. Determine the heat generation in each working condition interval, the heat load Calculated by the power consumption of the motor. Assume that the input power of the motor is , the efficiency of the motor is , the heat load is expressed as:

[0138] ;

[0139] For each operating area, calculate the corresponding heat load and obtain the heat load partition data. The heat load partition data reflects the heat generation of the system under different load rate conditions. Based on the heat load partition data, calculate the difference between the maximum heat dissipation of the radiator and the current heat load to determine the power demand of the system and obtain the power demand data. Assume that the maximum heat dissipation of the radiator is , the power demand is expressed as:

[0140] ;

[0141] if , it means that the current heat load exceeds the maximum capacity of the radiator and additional heat dissipation power is required. , the radiator capacity is sufficient to cope with the current heat load. Through this calculation, the power demand data of each working condition area is obtained. The power demand data is mapped with control parameters to obtain control mapping data. The power demand data is converted into specific control instructions, such as adjusting the speed of the fan or the cooling medium flow of the heat sink. Assume that the speed of the fan is , the control parameter mapping is performed through the following linear relationship:

[0142] ;

[0143] in, To control the gain, is the basic speed of the fan. Through this mapping, the heat dissipation demand is directly converted into the actual control input, thereby realizing the heat dissipation management of the system. Based on the control mapping data, multi-objective optimization is performed to obtain the optimized control data. Under the premise of meeting the heat dissipation demand, multiple goals such as minimizing energy consumption, minimizing noise, and maximizing system reliability are achieved. Set a comprehensive optimization objective function, for example:

[0144] ;

[0145] in, is the energy consumption of the fan, is the noise term, is the reference temperature, is the actual temperature, is the weight coefficient used to balance various optimization objectives. By solving the objective function, the optimized control data is obtained. Based on the optimized control data, a partition control rule base and a switching mechanism are constructed to obtain the heat dissipation control parameters. The construction of the partition control rule base is based on the optimized control data, and the optimization results of each operating area are stored as a control rule. These rules describe how to adjust the parameters of equipment such as radiators and fans under different load conditions. According to the current operating condition division data, the corresponding control strategy is selected from the rule base. Therefore, when the operating condition of the system changes, the control strategy can be switched smoothly to avoid control jumps. Assume that the current operating condition area is , the next time step enters the working condition area , the switching mechanism ensures that the control parameters such as fan speed smoothly transition from the current value to the new set value. This is achieved through interpolation functions, for example:

[0146] ;

[0147] In this way, smooth control switching is performed between different operating conditions to obtain the final heat dissipation control parameters.

[0148] The heat dissipation method of the electric scooter driving system in the embodiment of the present invention is described above. The heat dissipation device of the electric scooter driving system in the embodiment of the present invention is described below. Figure 2 In one embodiment of the present invention, an embodiment of the heat dissipation device of the electric scooter driving system includes:

[0149] The calculation module 201 is used to input the heat source distribution data and the space constraint data of the driving system in the electric scooter into the two-stage optimization model to calculate the heat dissipation structure and obtain the heat dissipation channel parameters;

[0150] The sampling module 202 is used to perform partition sampling and real-time synchronization of the heat source temperature and the heat sink temperature based on the heat dissipation channel parameters to obtain heat flow distribution characteristics;

[0151] Prediction module 203, used to solve the heat conduction model and classify the multi-region temperature of the heat flow distribution characteristics to obtain temperature dynamic distribution data, and input the temperature dynamic distribution data into the temperature field prediction model to perform heat flow prediction calculation to obtain heat flow prediction data;

[0152] The compensation module 204 is used to perform thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters;

[0153] The optimization module 205 is used to divide the heat load and optimize the heat dissipation power of the driving condition according to the heat dissipation compensation parameters to obtain the heat dissipation control parameters.

[0154] Through the collaborative cooperation of the above-mentioned components, through the combination of a two-stage optimization model and a hierarchical constraint generation algorithm, the precise design of the heat dissipation structure is achieved, the optimal heat dissipation channel layout is obtained in a limited space, and the problem of low space utilization of traditional heat dissipation solutions is solved; by adopting multi-layer temperature monitoring and real-time synchronization strategies, a complete heat flux distribution feature extraction system is established, and the dynamic change law of the system's temperature field is accurately grasped. Based on the dual guarantees of heat conduction model solution and temperature field prediction model, the system's prediction accuracy of temperature changes is significantly improved, so that the heat dissipation control has the ability to respond in advance; through thermal disturbance analysis and adaptive compensation mechanism, a complete interference suppression system is established, which effectively reduces the impact of external environmental changes on system temperature; by adopting a condition-adaptive heat load division method, the precise allocation of heat dissipation power is achieved, which greatly improves the heat dissipation efficiency of the system under different working conditions; a complete thermal management control system is constructed, and through multi-objective optimization and dynamic adjustment, the system has intelligent heat dissipation control capabilities.

[0155] above Figure 2 The heat dissipation device of the electric scooter driving system in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The heat dissipation device of the electric scooter driving system in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0156] Figure 3 3 is a schematic diagram of the structure of a heat dissipation device for a driving system of an electric scooter provided in an embodiment of the present invention. The heat dissipation device 300 for a driving system of an electric scooter may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage device terminals) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 may be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the heat dissipation device 300 for the driving system of an electric scooter. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the heat dissipation device 300 for the driving system of an electric scooter to implement the steps of the heat dissipation method for the driving system of an electric scooter.

[0157] The electric scooter driving system heat dissipation device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that Figure 3 The illustrated structure of the heat dissipation device for the electric scooter driving system does not constitute a limitation on the heat dissipation device for the electric scooter driving system provided by the present invention, and may include more or fewer components than illustrated, or a combination of certain components, or a different arrangement of components.

[0158] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the electric scooter drive system heat dissipation method.

[0159] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0160] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0161] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A heat dissipation method for an electric scooter driving system, characterized in that: The method comprises: The heat source distribution data and space constraint data of the driving system in the electric scooter are input into the two-stage optimization model to calculate the heat dissipation structure and obtain the heat dissipation channel parameters; specifically, the driving system in the electric scooter is divided into three-dimensional grids, and the heat dissipation space is divided into a number of grid units to obtain spatial discretization data; the heat source is identified on the spatial discretization data to determine the heat source positions of the motor stator winding, rotor and bearing to obtain the heat source node data, and the heat source node data is input into the first stage of the two-stage optimization model to calculate the number of radiators, and the number of radiators is obtained according to the thermal conductivity and volume constraints of the material; based on the number of radiators parameters, optimizing the geometric dimensions of the length, width and height of the heat sink to obtain the dimensional parameters of the heat sink; inputting the dimensional parameters of the heat sink into the second stage of the two-stage optimization model to calculate the spacing between the heat sinks, determining the spacing range of adjacent heat sinks based on the hierarchical constraint generation algorithm, and obtaining the radiator arrangement parameters; performing a heat conduction path analysis on the radiator arrangement parameters, calculating the heat flow transfer path based on the heat conduction equation, and obtaining the heat flow conduction data; performing channel optimization on the heat flow conduction data, determining the cross-sectional area and inclination angle of the heat dissipation channel, and obtaining the channel structure data; performing a flow field distribution calculation on the channel structure data, and obtaining the heat dissipation channel parameters; Based on the heat dissipation channel parameters, the heat source temperature and the heat sink temperature are sampled and synchronized in real time to obtain heat flow distribution characteristics; Performing heat conduction model calculation and multi-region temperature classification on the heat flow distribution characteristics to obtain temperature dynamic distribution data, and inputting the temperature dynamic distribution data into a temperature field prediction model to perform heat flow prediction calculation to obtain heat flow prediction data; Performing thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters; The heat load is divided and the heat dissipation power is optimized for the driving condition according to the heat dissipation compensation parameter to obtain the heat dissipation control parameter.

2. The heat dissipation method of the electric scooter driving system according to claim 1, characterized in that: The heat conduction data is subjected to channel optimization calculation to determine the cross-sectional area and inclination angle of the heat dissipation channel to obtain channel structure data, and the channel structure data is subjected to flow field distribution calculation to obtain heat dissipation channel parameters, including: Performing matrix decomposition on the heat flux conduction data, converting the heat flux density data into an eigenvector matrix and an eigenvalue matrix, and obtaining heat flux characteristic data; Inputting the heat flow characteristic data into the cutting plane in the adaptive cutting plane algorithm to perform main direction analysis, determining the spatial posture of the initial cutting plane according to the direction of the characteristic vector, and obtaining plane direction data; Constructing constraint condition data according to the plane direction data, and iteratively dividing the cutting plane along the main direction of heat flow based on the constraint condition data to obtain a plurality of sub-plane unit data; Performing thermal resistance calculation on the sub-plane unit data to obtain thermal resistance distribution data, and adaptively adjusting the position of the cutting plane based on the thermal resistance distribution data to obtain cross-section optimization data; Based on the cross-section optimization data, natural convection effects at different inclination angles are calculated to obtain inclination optimization data; The inclination angle optimization data and the cross-section optimization data are combined to generate a spatial configuration of a heat dissipation channel to obtain channel structure data, and flow field distribution calculation is performed on the channel structure data to obtain heat dissipation channel parameters.

3. The heat dissipation method of the electric scooter driving system according to claim 2, characterized in that: Based on the heat dissipation channel parameters, the heat source temperature and the heat sink temperature are sampled and synchronized in real time to obtain heat flow distribution characteristics, including: Based on the heat dissipation channel parameters, the drive system is subjected to heat source stratification to obtain multi-layer heat source data, and hot spot positioning and temperature measurement point spatial coordinate analysis are performed on the multi-layer heat source data to obtain temperature measurement point distribution data; The temperature measurement point distribution data is input into a three-layer temperature monitoring model, and temperature sampling is performed on the stator winding layer, the heat sink layer, and the controller layer to obtain interlayer temperature data; Performing data alignment on the interlayer temperature data, synchronously calibrating the sampling data of each layer according to the timestamp to obtain multi-layer temperature field data, and performing thermal flow field reconstruction and temperature gradient space mapping based on the multi-layer temperature field data to obtain thermal field reconstruction data; Correlating the thermal field reconstruction data with the motor speed data collected in real time, establishing a mapping relationship between the temperature field and the speed, and obtaining speed correlation data; The speed-related data and the load torque data are coupled and calculated to establish a dynamic response relationship between the temperature field and the load to obtain the load response data. Based on the load response data, the temperature field is dynamically decomposed and reorganized to obtain the heat flow distribution characteristics.

4. The heat dissipation method of the electric scooter driving system according to claim 3, characterized in that: The heat flow distribution characteristics are solved by a heat conduction model and multi-region temperature classification to obtain temperature dynamic distribution data, and the temperature dynamic distribution data is input into a temperature field prediction model to perform heat flow prediction calculation to obtain heat flow prediction data, including: Based on the heat flow distribution characteristics, a heat conduction differential equation group is constructed for the heat sink position and the heat source distribution to obtain a heat conduction basic model, and signal decomposition and multi-scale analysis are performed based on the heat conduction basic model to obtain multi-scale temperature data; Performing multi-point temperature classification on the multi-scale temperature data to obtain temperature classification data, and performing feature dimension reduction on the temperature classification data to obtain temperature principal component data; Performing time-space mapping on the temperature principal component data, and converting the variation law of the temperature field over time and space into a state transfer equation to obtain temperature dynamic distribution data; The temperature dynamic distribution data is subjected to state prediction to obtain temperature prediction data, the temperature prediction data is subjected to minimum temperature deviation compensation to obtain temperature correction data, and the temperature correction data is input into a temperature field prediction model to perform heat transfer calculation to obtain heat flow prediction data.

5. The heat dissipation method of the electric scooter driving system according to claim 4, characterized in that: The step of performing thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters includes: Performing interval segmentation and interference pattern recognition on the heat flow prediction data to obtain temperature interference data, and performing time domain feature extraction on the temperature interference data to obtain time domain feature data; Inputting the time domain characteristic data into the temperature interference analysis model, identifying the location and intensity of the target interference source by establishing a mapping relationship between the interference source and the temperature fluctuation, and obtaining the interference source data; Performing spatial distribution analysis on the interference source data, obtaining interference transmission data by constructing the propagation path and attenuation law of the interference, and classifying and combining the interference features on different transmission paths based on the interference transmission data to obtain thermal interference feature data; A compensation control law is constructed for the thermal interference characteristic data to obtain compensation control data, and a compensation control model of the heat dissipation system is established based on the compensation control data to calculate the optimal compensation amount and obtain the heat dissipation compensation parameters.

6. The heat dissipation method of the electric scooter driving system according to claim 5, characterized in that: The heat load division and heat dissipation power optimization of the driving condition according to the heat dissipation compensation parameter to obtain the heat dissipation control parameter includes: Based on the heat dissipation compensation parameter, a corresponding relationship between the motor load rate and the heat dissipation compensation amount is constructed to obtain operating condition characteristic data, and a partition boundary calculation is performed on the operating condition characteristic data to obtain operating condition boundary data; According to the working condition boundary data, a working condition membership function is established to smoothly divide the working condition space to obtain working condition division data, and a heat load is calculated on the working condition division data to obtain heat load partition data; Based on the heat load partition data, the difference between the maximum heat dissipation of the radiator and the current heat load is calculated to determine the power demand, to obtain power demand data, and control parameter mapping is performed on the power demand data to obtain control mapping data; Based on the control mapping data, multi-objective optimization is performed to obtain optimized control data, and based on the optimized control data, a partition control rule library and a switching mechanism are constructed to obtain heat dissipation control parameters.

7. A heat dissipation device for an electric scooter driving system, characterized in that: The device is used to perform the heat dissipation method of the electric scooter driving system according to any one of claims 1 to 6, and comprises: A calculation module is used to input the heat source distribution data and space constraint data of the driving system in the electric scooter into a two-stage optimization model to calculate the heat dissipation structure and obtain the heat dissipation channel parameters; specifically comprising: performing three-dimensional grid division on the driving system in the electric scooter, and dividing the heat dissipation space into a number of grid units to obtain spatial discretization data; performing heat source identification on the spatial discretization data, determining the heat source positions of the motor stator winding, rotor and bearing, obtaining heat source node data, and inputting the heat source node data into the first stage of the two-stage optimization model to calculate the number of radiators, and obtaining the number parameters of radiators according to the thermal conductivity and volume constraints of the material; based on the heat dissipation The number of heat sinks is determined by the number of heat sinks, and the length, width and height of the heat sink are geometrically optimized to obtain the heat sink size parameters; the heat sink size parameters are input into the second stage of the two-stage optimization model to calculate the heat sink spacing, and the spacing range of adjacent heat sinks is determined based on the hierarchical constraint generation algorithm to obtain the heat sink arrangement parameters; the heat conduction path analysis is performed on the heat sink arrangement parameters, and the heat flow transfer path is calculated based on the heat conduction equation to obtain the heat flow conduction data; the channel optimization is performed on the heat flow conduction data to determine the cross-sectional area and inclination angle of the heat dissipation channel to obtain the channel structure data, and the flow field distribution calculation is performed on the channel structure data to obtain the heat dissipation channel parameters; A sampling module, used to perform partition sampling and real-time synchronization of the heat source temperature and the heat sink temperature based on the heat dissipation channel parameters to obtain heat flow distribution characteristics; A prediction module is used to perform heat conduction model calculation and multi-region temperature classification on the heat flow distribution characteristics to obtain temperature dynamic distribution data, and input the temperature dynamic distribution data into a temperature field prediction model to perform heat flow prediction calculation to obtain heat flow prediction data; A compensation module, used for performing thermal disturbance analysis and adaptive compensation on the heat flow prediction data to obtain heat dissipation compensation parameters; The optimization module is used to divide the heat load and optimize the heat dissipation power of the driving condition according to the heat dissipation compensation parameters to obtain the heat dissipation control parameters.

8. A heat dissipation device for an electric scooter driving system, characterized in that: The heat dissipation device of the electric scooter driving system comprises: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the electric scooter drive system heat dissipation device to execute the electric scooter drive system heat dissipation method according to any one of claims 1-6.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the heat dissipation method of the electric scooter driving system according to any one of claims 1-6 is implemented.

Images