Intelligent optimization management method and system for battery efficiency of unmanned aerial vehicle

Abstract

The invention relates to an intelligent optimization management method and system for unmanned aerial vehicle battery efficiency, and the method comprises the following steps: carrying out the dynamic threshold division of a battery health state evaluation matrix through a recursive partition optimization algorithm, and obtaining a battery work prediction state; obtaining a target task of the unmanned aerial vehicle, and performing adaptive matching calculation on the target task based on the battery working prediction state to obtain a task power consumption optimization configuration scheme; performing intelligent discharge strategy planning on the unmanned aerial vehicle battery based on the task power consumption optimal configuration scheme to obtain a battery discharge strategy control parameter set; through a heterogeneous computing architecture, the unmanned aerial vehicle is regulated and controlled in real time to complete the target task based on the battery discharge strategy control parameter set, and the problems that a traditional battery management method is often based on a fixed charge and discharge strategy, actual performance changes of a battery in different use scenes are not fully considered, and the service life of the battery is affected are solved. The service life of the battery is shortened; and the energy utilization efficiency is low.

Description

Technical Field

[0001] The present invention relates to the technical field of drone batteries, and in particular to an intelligent optimization management method and system for drone battery performance. Background Art

[0002] With the rapid development of drone technology, drones are increasingly being used in both civilian and military applications, ranging from aerial photography and agricultural monitoring to logistics and distribution and reconnaissance missions in complex environments. However, the operating time and performance of drones are highly dependent on battery efficiency, making battery management a key factor affecting a drone's overall performance. Traditional battery management methods often rely on fixed charging and discharging strategies that fail to fully account for the actual performance variations of batteries in different usage scenarios, resulting in shortened battery life and inefficient energy utilization.

[0003] Faced with these challenges, researchers have begun exploring intelligent approaches to optimize drone battery management to improve efficiency and lifespan. Despite this, practical applications still present numerous challenges. For one thing, existing research focuses primarily on static assessments of battery health, lacking effective monitoring and prediction of battery state changes under dynamic operating conditions. Furthermore, given the diverse nature of drone missions and the significant variations in battery energy consumption requirements, existing battery management strategies struggle to adapt to specific mission needs. These challenges limit drones' ability to sustain continuous operation and maintain autonomy during complex missions.

[0004] Furthermore, the working environment faced by drones during missions is complex and changeable, including changes in temperature, humidity, and other external conditions, which can significantly affect battery performance. Currently, research on how to precisely control the operating state of batteries in unstable environments is not yet in-depth, especially in terms of optimizing energy efficiency in conjunction with specific target missions. Therefore, it is particularly important to develop a method that can respond to changes in the external environment in real time and intelligently adjust the battery discharge strategy based on the requirements of the target mission. This will not only help improve the overall operating efficiency of drones, but also extend battery life and reduce operating costs. Summary of the Invention

[0005] The main purpose of the present invention is to provide an intelligent optimization management method and system for drone battery efficiency, which solves the technical problem that traditional battery management methods are often based on fixed charging and discharging strategies and fail to fully consider the actual performance changes of batteries in different usage scenarios, resulting in shortened battery life and low energy utilization efficiency.

[0006] To achieve the above objectives, the present invention provides an intelligent optimization management method for the battery efficiency of a drone, which is applied to a drone equipped with a drone battery, and comprises the following steps: Performing multi-dimensional feature extraction on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set; Performing a battery health status hierarchical cluster analysis on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix; Performing dynamic threshold partitioning on the battery health status assessment matrix through a recursive partitioning optimization algorithm to obtain a battery operating prediction state; Obtaining a target mission of the UAV, and performing adaptive matching calculation on the target mission based on the predicted battery operating state to obtain a mission power consumption optimization configuration plan; Performing intelligent discharge strategy planning on the UAV battery based on the mission power consumption optimization configuration scheme to obtain a battery discharge strategy control parameter set; Through a heterogeneous computing architecture, the UAV is controlled in real time to complete the target task based on the battery discharge strategy control parameter set.

[0007] Furthermore, the multi-dimensional feature extraction is performed on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set, including: Performing multi-scale decomposition on the discharge characteristic curve of the UAV battery to obtain a voltage fluctuation feature sequence, and performing time-frequency domain transformation on the voltage fluctuation feature sequence to obtain a voltage spectrum feature matrix, which includes voltage decay rate, voltage fluctuation amplitude, and voltage stability range; Performing nonlinear dynamic response analysis on the voltage spectrum characteristic matrix to obtain a battery load response characteristic spectrum, and performing topological feature decomposition on the battery load response characteristic spectrum to obtain a battery working state feature set; A multi-parameter correlation analysis is performed on the battery operating state feature set to obtain a battery performance degradation trend sequence, and feature space mapping is performed on the battery performance degradation trend sequence to obtain a battery performance feature vector set; wherein the battery performance feature vector set includes a performance decay rate, a remaining life prediction value, and a discharge efficiency coefficient.

[0008] Furthermore, the battery health status hierarchical cluster analysis is performed on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix, including: Performing high-dimensional spatial transformation and sparse representation on the battery performance feature vector set to obtain a battery feature sparse representation matrix, and performing multi-kernel function similarity calculation on the battery feature sparse representation matrix to obtain a battery state affinity spectrum; wherein the battery state affinity spectrum includes capacity decay trajectory points, voltage recovery characteristic index, and load response gradient map; A battery state diffusion mapping network is constructed based on the battery state affinity spectrum, and topological feature extraction and dynamic evolution analysis are performed on the battery state diffusion mapping network to obtain a battery state topological structure diagram. The battery state topological structure diagram is then subjected to spectral decomposition and hierarchical cutting to obtain a battery state multi-level classification structure; wherein the battery state multi-level classification structure includes a discharge efficiency threshold range, an internal resistance change rate distribution, and a temperature sensitivity classification boundary; Performing density peak extraction and boundary optimization on the multi-level classification structure of battery status to obtain a battery health status cluster center set, and performing adaptive radius expansion on the battery health status cluster center set to obtain a battery health status classification area map; Based on the battery health status classification area map, the state feature mapping and conversion of the UAV battery are performed to obtain a battery state quantitative evaluation vector, and the battery state quantitative evaluation vector is orthogonalized and matrix reconstructed to obtain a battery health status evaluation matrix.

[0009] Furthermore, the recursive partitioning optimization algorithm is used to dynamically partition the battery health status assessment matrix to obtain the battery operating prediction state, including: Performing phase space reconstruction and dimensionality reduction processing on the battery health status assessment matrix to obtain a battery state feature space, and recursively bisectioning the battery state feature space using a recursive partitioning optimization algorithm to obtain a battery state initial partition set, wherein the battery state initial partition set includes a capacity degradation partition point, an internal resistance change boundary value, and a voltage response critical threshold; Constructing a battery state decision tree based on the initial battery state partition set, and performing information entropy optimization pruning on the battery state decision tree to obtain an optimal partition boundary set; wherein the optimal partition boundary set includes a health state dividing line, a working performance cluster center, and an abnormal state isolation interval; Dynamic parameter adjustment and nonlinear mapping are performed on the optimal partition boundary set to obtain a dynamic partition map of the battery state, and multi-dimensional threshold optimization is performed on the dynamic partition map of the battery state to obtain an optimized battery state threshold set; wherein the optimized battery state threshold set includes a capacity decay rate threshold, a power output limit, and a temperature sensitive interval delimiter; Based on the battery state optimization threshold set, a multi-scenario deduction analysis is performed on the UAV battery to obtain a battery operation prediction state; wherein the battery operation prediction state includes a battery remaining capacity prediction value, power output reliability, and discharge safety boundary parameters.

[0010] Furthermore, the adaptive matching calculation is performed on the target task based on the predicted battery working state to obtain a task power consumption optimization configuration scheme, including: Performing segmented deconstruction analysis on the target task to obtain a task load characteristic sequence, and performing time-domain power decomposition on the task load characteristic sequence to obtain a task power consumption distribution map, wherein the task power consumption distribution map includes a flight power demand curve, a load power consumption fluctuation range, and a task execution timing table; Based on the task power consumption distribution map, the predicted battery working state is mapped and transformed to obtain a battery power output matching sequence, and the battery power output matching sequence is cross-validated in multiple dimensions to obtain a task battery matching matrix, wherein the task battery matching matrix includes a power matching coefficient, a discharge depth threshold, and an operating temperature boundary; Performing hierarchical optimization processing on the task battery matching matrix to obtain a task power consumption scheduling strategy set, and performing dynamic constraint solving on the task power consumption scheduling strategy set to obtain a power consumption scheduling execution plan; Constructing a task energy flow graph based on the power consumption scheduling execution scheme, and performing multi-path optimization calculation on the task energy flow graph to obtain a task energy allocation sequence; Multi-objective collaborative optimization is performed on the task energy allocation sequence to obtain a task power consumption optimization configuration scheme.

[0011] Furthermore, the intelligent discharge strategy planning of the UAV battery is performed based on the mission power consumption optimization configuration scheme to obtain a battery discharge strategy control parameter set, including: Performing power flow topology decomposition on the task power consumption optimization configuration scheme to obtain a discharge power flow distribution sequence, and performing multi-scenario energy link tracing on the discharge power flow distribution sequence to obtain a discharge strategy topology map, wherein the discharge strategy topology map includes power flow path points, energy allocation weight coefficients, and a discharge depth control curve; Constructing a discharge state transition network based on the discharge strategy topology map, performing state probability transition analysis on the discharge state transition network to obtain a discharge state migration matrix, and performing spectral clustering boundary extraction on the discharge state migration matrix to obtain a discharge control threshold set; Performing multi-dimensional constraint mapping on the discharge control threshold set to obtain a discharge control constraint vector group, and performing nonlinear adaptive adjustment on the discharge control constraint vector group to obtain a discharge strategy optimization sequence; Constructing a discharge efficiency feature map based on the discharge strategy optimization sequence, and performing multi-level feature decoupling on the discharge efficiency feature map to obtain a discharge efficiency energy data set; A multi-objective collaborative optimization process is performed on the discharge efficiency energy data set to obtain a battery discharge strategy control parameter set.

[0012] Furthermore, the UAV is provided with a battery management unit and a flight control system interaction interface. The UAV is controlled in real time to complete the target task based on the battery discharge strategy control parameter set through a heterogeneous computing architecture, including: Performing resource allocation mapping on the battery discharge strategy control parameter set to obtain a computing load distribution matrix, and performing parallel task segmentation on the computing load distribution matrix using a preset heterogeneous computing architecture to obtain a heterogeneous computing execution unit sequence, wherein the heterogeneous computing execution unit sequence includes a power control instruction set, a voltage regulation execution sequence, and a temperature compensation control point; Constructing a hierarchical execution priority network based on the heterogeneous computing execution unit sequence, obtaining a task execution graph based on the hierarchical execution priority network, and performing critical path extraction on the task execution graph to obtain a sequential execution pipeline strategy; Performing resource conflict detection and resolution on the sequential execution pipeline strategy to obtain an optimized execution strategy set, and concurrently scheduling and arranging the optimized execution strategy set to obtain a heterogeneous execution instruction stream; Based on the heterogeneous execution instruction stream, the interaction interface between the battery management unit and the flight control system is mapped to obtain a battery flight control collaborative control sequence, and the battery flight control collaborative control sequence is dynamically optimized and embedded deployed to obtain a UAV execution control instruction set. Based on the UAV execution control instruction set, the UAV is controlled in real time to complete the target task.

[0013] The present invention also provides an intelligent optimization management system for the battery efficiency of a drone, which is applied to a drone equipped with a drone battery, and includes: An extraction module, configured to perform multi-dimensional feature extraction on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set; An analysis module, configured to perform a battery health status hierarchical cluster analysis on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix; A partitioning module is used to perform dynamic threshold partitioning on the battery health status assessment matrix through a recursive partitioning optimization algorithm to obtain a battery operating prediction state; A calculation module is used to obtain a target task of the UAV and perform adaptive matching calculation on the target task based on the predicted working state of the battery to obtain a task power consumption optimization configuration plan; A planning module, configured to perform intelligent discharge strategy planning for the UAV battery based on the mission power consumption optimization configuration scheme, and obtain a battery discharge strategy control parameter set; An execution module is used to control the UAV in real time to complete the target task based on the battery discharge strategy control parameter set through a heterogeneous computing architecture.

[0014] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0015] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0016] The present invention provides an intelligent optimization management method for the battery efficiency of a drone, comprising the following steps: performing multidimensional feature extraction on the discharge characteristic curve of the drone battery to obtain a battery performance feature vector set; performing a hierarchical clustering analysis of the battery health status of the drone battery based on the battery performance feature vector set to obtain a battery health status assessment matrix; performing dynamic threshold partitioning on the battery health status assessment matrix using a recursive partitioning optimization algorithm to obtain a battery operating prediction state; obtaining a target mission for the drone, and performing adaptive matching calculations on the target mission based on the battery operating prediction state to obtain a mission power consumption optimization configuration scheme; performing intelligent discharge strategy planning on the drone battery based on the mission power consumption optimization configuration scheme to obtain a battery discharge strategy control parameter set; and, using a heterogeneous computing architecture, controlling the drone in real time to complete the target mission based on the battery discharge strategy control parameter set. This method solves the technical problem that traditional battery management methods are often based on fixed charge and discharge strategies and fail to fully consider the actual performance changes of batteries in different usage scenarios, resulting in shortened battery life and low energy utilization efficiency. By obtaining the specific target mission of the drone and performing adaptive matching calculations based on the battery operating prediction state, a mission power consumption optimization configuration scheme that best suits the current mission requirements can be formulated. This approach ensures that energy consumption is optimized when performing specific tasks, which has the beneficial effect of improving the working efficiency and endurance of the drone. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 1 is a schematic diagram of the steps of an intelligent optimization management method for drone battery efficiency according to one embodiment of the present invention; Figure 2 This is a structural block diagram of an intelligent optimization management device for drone battery efficiency according to one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0018] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] like Figure 1 As shown, Figure 1 This is a schematic diagram of the steps of a method for intelligent optimization management of drone battery efficiency in one embodiment of the present invention; In one embodiment of the present invention, a method for intelligent optimization management of drone battery performance is provided, which is applied to a drone equipped with a drone battery, and includes the following steps: Step S1: performing multi-dimensional feature extraction on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set.

[0021] Specifically, to extract multidimensional features from the discharge characteristic curves of drone batteries to generate a set of battery performance feature vectors, it is first necessary to collect battery discharge data under different operating conditions. This data includes, but is not limited to, parameters such as voltage, current, temperature, and discharge time. By collecting data on the time-varying changes of these parameters to form discharge characteristic curves, this process is a key step in understanding the actual battery performance. Next, advanced data analysis techniques, such as machine learning algorithms or statistical methods, are used to extract multidimensional feature information from these discharge characteristic curves, such as voltage drop rate, maximum discharge current, and temperature change rate, thereby constructing a set of feature vectors that comprehensively reflects battery performance. This step not only helps accurately describe the battery's current operating state but also lays the foundation for further analysis of its health. For example, in agricultural monitoring missions, drones must fly for long periods of time to cover large areas of farmland. In this case, battery performance directly determines the successful completion of the mission. By extracting multidimensional features from the discharge characteristic curves of drone batteries before and after such missions, we can obtain key information such as the rate at which voltage drops with increasing load and the temperature trend during discharge. Suppose, for example, that during a specific mission cycle, a battery's voltage drop rate is found to be significantly higher than that of other similar batteries. This suggests the possibility of increased internal resistance, which in turn affects the battery's overall health assessment. Based on these eigenvectors, hierarchical cluster analysis can be used to further refine the battery's health classification and implement appropriate maintenance measures to ensure the drone's efficient operation in future missions. Therefore, in-depth analysis and feature extraction of discharge characteristic curves not only improves the intelligence level of the battery management system, but also effectively extends battery life and enhances the drone's overall mission execution efficiency. This process represents a complete chain from data collection to feature extraction and final decision support, and is of great significance for optimizing drone battery management.

[0022] Step S2: performing a battery health status hierarchical cluster analysis on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix.

[0023] Specifically, the process of performing a hierarchical clustering analysis of the battery health status of drone batteries based on a battery performance feature vector set to generate a battery health assessment matrix first relies on multidimensional feature data extracted from discharge characteristic curves. These feature vectors contain key information reflecting battery health, such as voltage drop rate, temperature change rate, and maximum discharge current. Next, a hierarchical clustering algorithm is used to classify the data in these feature vector sets. By calculating the similarity or distance between different batteries, a tree-like structure diagram reflecting the battery health status, namely a hierarchical clustering tree, is constructed. This process not only identifies the commonalities and differences between batteries but also classifies batteries into different health levels based on predefined criteria, forming a battery health assessment matrix. For example, in an agricultural monitoring mission, suppose we have a fleet of drones used for long-term flight monitoring, each equipped with a specific battery model. By analyzing the discharge characteristic curves of each battery before and after the mission, we obtain the respective battery performance feature vector sets. Then, by processing these feature vector sets using a hierarchical clustering algorithm, we can find that some batteries have significantly accelerated voltage drop rates and more severe temperature changes due to long-term use or improper maintenance. By constructing a hierarchical clustering tree, we can clearly see which batteries are in good health and which are showing early signs of aging. This allows us to develop a personalized maintenance plan or replacement strategy for each battery based on the battery health assessment matrix. For example, batteries assessed as healthy can continue to be scheduled for high-load missions; while batteries showing early aging are recommended for further inspection or early replacement, thereby ensuring the efficient operation of the entire drone fleet and the successful completion of the mission. This process not only improves the accuracy of battery management, but also provides important decision-making support for drone operations, helping to optimize resource allocation and extend the service life of equipment. In this way, a complete closed-loop management system is achieved, from data collection to health status assessment to specific application guidance.

[0024] Step S3: Dynamically partition the battery health status evaluation matrix using a recursive partitioning optimization algorithm to obtain a predicted battery operating state.

[0025] Specifically, the process of dynamically partitioning the battery health status assessment matrix using a recursive partitioning optimization algorithm to obtain predicted battery operating states relies first on the battery health status assessment matrix constructed in the previous step. This matrix contains multidimensional characteristic information for each battery and uses hierarchical cluster analysis to determine the health status of different batteries. The core of the recursive partitioning optimization algorithm lies in automatically finding the optimal partition boundaries based on this health status data, thereby achieving refined battery health status classification and dynamic threshold setting. Specifically, the algorithm gradually partitions the entire dataset into different subsets, each representing a specific health status interval, and sets a corresponding threshold range for each interval. This dynamic thresholding can more accurately reflect the performance changes of batteries in actual use, thereby providing more accurate operating status predictions. For example, in an agricultural monitoring mission, suppose that the health status assessment matrix of a batch of drone batteries has been obtained through hierarchical cluster analysis. Next, when applying the recursive partitioning optimization algorithm to process this data, the algorithm first performs a preliminary partitioning of the entire dataset to identify battery groups that exhibit significant health differences. For example, some batteries may exhibit characteristics such as rapid voltage drop and large temperature fluctuations, while others are more stable. As the algorithm iterates recursively, the health status intervals for each battery group are gradually refined, ultimately forming a series of subsets with clear threshold ranges. These thresholds not only consider the current data distribution but also incorporate historical data and expected future trends to ensure prediction accuracy. For example, for batteries classified as "early aging," their dynamic thresholds may be set more stringently to provide timely warnings of potential problems; while for batteries in good health, relatively loose thresholds can be set to allow greater operational flexibility. This method provides a predicted operating status for each battery, providing an important basis for subsequent mission planning. For example, when formulating flight plans, the predicted status of each battery can be used to appropriately arrange mission loads, avoiding the risk of mission failure due to battery performance degradation and ensuring that the drone fleet remains efficient during agricultural monitoring missions. This process embodies a comprehensive pipeline from data processing to health status prediction to practical application guidance, significantly enhancing the intelligence of battery management systems.

[0026] Step S4: obtaining the target mission of the UAV, and performing adaptive matching calculation on the target mission based on the predicted battery working state to obtain a mission power consumption optimization configuration solution.

[0027] Specifically, the process of obtaining the drone's target mission and adaptively matching it to the predicted battery operating status to arrive at an optimized power consumption configuration for the mission requires first clarifying the drone's specific mission requirements and objectives. These missions may include long-term agricultural monitoring, logistics distribution, or reconnaissance in complex environments, each with distinct battery energy consumption requirements. Once the target mission is defined, the system combines the predicted battery operating status obtained in the previous step with an adaptive matching algorithm to analyze the compatibility between the current battery's actual performance and the mission requirements. This process not only considers battery health but also factors such as flight path, payload weight, and weather conditions to ensure the optimal power consumption configuration for the mission. For example, in an agricultural monitoring mission, suppose a drone is required to cover a large area of ​​farmland and capture high-resolution imagery. The system first obtains the specific mission requirements, such as flight time, coverage area size, and required image resolution. Then, based on the predicted battery operating status obtained in the previous step, the system comprehensively evaluates each battery's remaining charge, voltage stability, temperature trends, and other factors. For example, if a battery is predicted to be in an early stage of aging, with a rapid voltage drop and large temperature fluctuations, the system may recommend reducing the flight time and coverage area of ​​the drone powered by that battery, or adjusting the flight altitude to reduce energy consumption. Meanwhile, for batteries in good health, the system can schedule them for longer-distance or higher-load missions. Through this adaptive matching calculation, the system dynamically adjusts each drone's mission parameters, such as flight speed and aerial photography frequency, to develop an optimized mission power allocation that comprehensively considers battery performance and mission requirements. Ultimately, this not only ensures successful mission completion but also maximizes battery resources, extending the drone's overall operational time and improving mission efficiency and success rate. For example, in actual operation, the system may automatically plan the drone's flight route based on the predicted battery operating status, prioritizing batteries with sufficient charge and stable performance for critical mission segments, ensuring efficient and accurate completion of the entire monitoring mission. This process demonstrates a complete chain from battery health prediction to optimized mission configuration and practical application guidance, demonstrating the power and flexibility of intelligent battery management systems.

[0028] Step S5: performing intelligent discharge strategy planning on the UAV battery based on the mission power consumption optimization configuration scheme to obtain a battery discharge strategy control parameter set.

[0029] Specifically, the process of intelligently planning a discharge strategy for drone batteries based on the mission power optimization configuration to obtain a set of battery discharge strategy control parameters first relies on the mission power optimization configuration obtained in the previous step. This configuration details the energy consumption pattern, flight path, payload weight, and other relevant factors required for each drone to perform a specific mission. Next, the system uses this information to develop a refined intelligent discharge strategy to ensure that the battery can efficiently and safely provide the required power to the drone. Specifically, intelligent discharge strategy planning comprehensively considers factors such as the battery's current health status, remaining charge, and mission requirements. An algorithm is used to calculate the optimal discharge rate and time distribution, thereby generating a set of battery discharge strategy control parameters. For example, in an agricultural monitoring mission, assume that the mission power optimization configuration for each drone has been determined, including details such as flight path, coverage area size, and shooting frequency. Based on this information, the system begins intelligent discharge strategy planning for the drone batteries. For example, for drones assigned to long-duration missions, the system may design a progressive discharge strategy, adopting a more conservative discharge rate at the beginning of the mission to ensure that the battery has sufficient charge to cope with unexpected situations later in the mission. Furthermore, for drones expected to experience higher temperatures, the system may adjust its discharge strategy, reducing periods of high power output to prevent a sharp drop in battery performance due to excessive temperatures. Furthermore, the system dynamically adjusts the discharge strategy based on the battery's health status. If a battery shows early signs of aging, its discharge strategy may become more cautious, setting lower thresholds to prevent premature depletion. Ultimately, these strategies are translated into a set of specific control parameters, such as maximum discharge current, lower voltage limit, and upper temperature limit, forming the battery discharge strategy control parameter set. These parameters not only guide the actual battery discharge process but also provide the drone with essential protection mechanisms, ensuring stable battery operation throughout the mission, extending its service life and increasing the success rate of mission completion. In this way, the system achieves a complete closed-loop management system from mission requirement analysis to battery health management and intelligent discharge strategy formulation, significantly improving the reliability and efficiency of the drone system. This process demonstrates the ability of an intelligent battery management system to flexibly adapt its strategy to meet diverse mission requirements in complex and changing mission environments.

[0030] Step S6: Using a heterogeneous computing architecture, the UAV is controlled in real time based on the battery discharge strategy control parameter set to complete the target mission.

[0031] Specifically, a heterogeneous computing architecture is used to control the drone's mission completion process in real time based on a battery discharge strategy control parameter set. This relies primarily on the detailed battery discharge strategy control parameter set generated in the previous step. These parameters include key indicators such as maximum discharge current, lower voltage limit, and upper temperature limit, ensuring efficient and safe battery operation during the mission. The heterogeneous computing architecture leverages the strengths of diverse computing resources (such as CPUs, GPUs, and FPGAs) to enable real-time processing and dynamic adjustment of these parameters. Specifically, the system utilizes parallel processing and rapid response across the different computing units within the heterogeneous computing architecture based on the current mission requirements and environmental conditions, taking into account the actual battery status, thereby achieving precise control of the drone's operations. For example, in an agricultural monitoring mission, suppose a detailed battery discharge strategy control parameter set has been defined for each drone and they are preparing to begin a long-term farmland monitoring mission. During the mission, the drone will need to capture high-resolution images according to a pre-set flight path and capture frequency. This is where the heterogeneous computing architecture comes into play: the CPU processes sensor data, such as battery voltage, current, temperature, and the drone's location. The GPU handles complex image processing tasks, ensuring the drone can analyze and store captured images in real time during flight. Furthermore, the FPGA can be used to accelerate specific computing tasks, such as rapidly adjusting the discharge rate to respond to emergencies or optimize power consumption. For example, when a drone enters a high-temperature area, the system can quickly detect the rising battery temperature through the heterogeneous computing architecture and dynamically adjust the discharge strategy based on pre-set temperature limits, reducing output power to prevent battery overheating. Furthermore, if strong winds or other adverse weather conditions are encountered during a mission, the system can replan the flight path or adjust the flight speed based on real-time flight data, ensuring the drone can continue its mission within safe limits. All of these adjustments are made based on the battery discharge strategy control parameter set, ensuring that every operation is performed under optimal conditions. In this way, the heterogeneous computing architecture not only improves system responsiveness and processing power, but also ensures that the drone can operate stably and efficiently in complex and changing mission environments, ultimately successfully completing its intended mission. This process demonstrates the complete chain from battery health management to intelligent discharge strategy to real-time regulation, reflecting the powerful functions and flexibility of intelligent drone management systems in practical applications.

[0032] In a specific embodiment, the multi-dimensional feature extraction is performed on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set, including: Performing multi-scale decomposition on the discharge characteristic curve of the UAV battery to obtain a voltage fluctuation feature sequence, and performing time-frequency domain transformation on the voltage fluctuation feature sequence to obtain a voltage spectrum feature matrix, which includes voltage decay rate, voltage fluctuation amplitude, and voltage stability range; Performing nonlinear dynamic response analysis on the voltage spectrum characteristic matrix to obtain a battery load response characteristic spectrum, and performing topological feature decomposition on the battery load response characteristic spectrum to obtain a battery working state feature set; A multi-parameter correlation analysis is performed on the battery operating state feature set to obtain a battery performance degradation trend sequence, and feature space mapping is performed on the battery performance degradation trend sequence to obtain a battery performance feature vector set; wherein the battery performance feature vector set includes a performance decay rate, a remaining life prediction value, and a discharge efficiency coefficient.

[0033] Specifically, when extracting multidimensional features from the discharge characteristic curve of a drone battery to obtain a set of battery performance feature vectors, the discharge characteristic curve must first be decomposed at multiple scales to obtain a voltage fluctuation feature sequence. This process captures voltage variation information at different time scales, revealing the dynamic behavior of the battery during actual operation. Next, these voltage fluctuation feature sequences are transformed into the time-frequency domain to generate a voltage spectrum feature matrix, which includes key indicators such as voltage decay rate, voltage fluctuation amplitude, and voltage stability range. This step not only provides the battery's response characteristics at different frequencies but also helps identify potential failure modes or performance degradation trends. Subsequently, based on the voltage spectrum feature matrix, the system performs nonlinear dynamic response analysis to generate a battery load response feature map. This map reflects the battery's response under different load conditions, and topological feature decomposition is further used to extract a set of battery operating state features. For example, in an agricultural monitoring mission, if a battery exhibits large voltage fluctuations during high-load flight, nonlinear dynamic response analysis can reveal abnormal response under specific load conditions. Next, by performing multi-parameter correlation analysis on the battery operating status feature set, the system can identify specific trends in battery performance degradation, generating a battery performance degradation trend sequence. This sequence not only contains the performance degradation rate but also provides important information such as the remaining life prediction value and discharge efficiency coefficient. Finally, the battery performance degradation trend sequence is mapped into a feature space to generate a final set of battery performance feature vectors. These feature vectors include a detailed description of the battery health status, such as the performance degradation rate, remaining life prediction value, and discharge efficiency coefficient, providing a solid foundation for subsequent health management. For example, in an agricultural monitoring mission, if a performance feature vector set for a battery is obtained, indicating a high performance degradation rate and a short remaining life, the system can use this information to preemptively schedule maintenance or replacement, ensuring that drones are not interrupted by battery issues during long monitoring missions. In this way, the entire process, from extracting the voltage fluctuation feature sequence to generating the battery performance feature vector set, not only improves the understanding and management capabilities of battery health status, but also provides important decision support for drone operations, ensuring smooth mission completion and improving overall equipment reliability. This process demonstrates the complete chain from data collection, feature extraction to health status assessment, and reflects the powerful functions and flexibility of the intelligent battery management system in complex application scenarios.

[0034] In a specific embodiment, the battery health status hierarchical cluster analysis is performed on the drone battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix, including: Performing high-dimensional spatial transformation and sparse representation on the battery performance feature vector set to obtain a battery feature sparse representation matrix, and performing multi-kernel function similarity calculation on the battery feature sparse representation matrix to obtain a battery state affinity spectrum; wherein the battery state affinity spectrum includes capacity decay trajectory points, voltage recovery characteristic index, and load response gradient map; A battery state diffusion mapping network is constructed based on the battery state affinity spectrum, and topological feature extraction and dynamic evolution analysis are performed on the battery state diffusion mapping network to obtain a battery state topological structure diagram. The battery state topological structure diagram is then subjected to spectral decomposition and hierarchical cutting to obtain a battery state multi-level classification structure; wherein the battery state multi-level classification structure includes a discharge efficiency threshold range, an internal resistance change rate distribution, and a temperature sensitivity classification boundary; Performing density peak extraction and boundary optimization on the multi-level classification structure of battery status to obtain a battery health status cluster center set, and performing adaptive radius expansion on the battery health status cluster center set to obtain a battery health status classification area map; Based on the battery health status classification area map, the state feature mapping and conversion of the UAV battery are performed to obtain a battery state quantitative evaluation vector, and the battery state quantitative evaluation vector is orthogonalized and matrix reconstructed to obtain a battery health status evaluation matrix.

[0035] Specifically, when performing a hierarchical clustering analysis of the battery health status of drone batteries based on a set of battery performance feature vectors to obtain a battery health status assessment matrix, the battery performance feature vectors must first undergo a high-dimensional spatial transformation and sparse representation. This process transforms the original multidimensional feature vectors into a higher-dimensional spatial representation, making the underlying structure in the data more clearly visible. Specifically, by performing a sparse representation of the battery performance feature vectors, redundant information is removed while retaining key features, generating a battery feature sparse representation matrix. Next, the battery feature sparse representation matrix is ​​processed using a multi-kernel function similarity calculation method to obtain a battery state affinity spectrum. The battery state affinity spectrum includes important indicators such as capacity decay trajectory points, voltage recovery characteristic index, and load response gradient map. These indicators can comprehensively reflect the battery's health status under different operating conditions. For example, in an agricultural monitoring mission, suppose we have obtained a performance feature vector set for a drone battery, which contains information such as voltage fluctuation, discharge efficiency, and remaining life prediction under different load conditions. By performing high-dimensional spatial transformation and sparse representation on these feature vectors, the system identifies the features that most significantly influence the battery's state of health (SOH) and generates a sparse representation matrix of the battery features. This matrix is ​​then processed using a multi-kernel function similarity calculation method to generate a battery state affinity spectrum. For example, a battery exhibiting significant capacity decay trajectory points during long-duration flight indicates rapid aging under high-load conditions. Simultaneously, a low voltage recovery characteristic index indicates that the battery is unable to quickly recover to its optimal state after recharging. These are all important indicators of the battery's state of health. Based on the battery state affinity spectrum, the system constructs a battery state diffusion mapping network. This network reveals the inherent connections between battery states of health through topological feature extraction and dynamic evolution analysis. Specifically, by analyzing the battery state diffusion mapping network, a battery state topology map is obtained. This map not only demonstrates the similarities and differences in the health states of different batteries but also provides a foundation for subsequent hierarchical segmentation. For example, in an agricultural monitoring mission, suppose we have a fleet of drones used for long-duration flight monitoring, each equipped with a specific battery model. By constructing a battery state diffusion mapping network and performing topological feature extraction, we can find that the health status of some batteries is significantly different from that of other batteries due to long-term use or improper maintenance. Furthermore, by performing spectral decomposition and hierarchical cutting on the battery state topological structure diagram, the system can divide these batteries into multiple hierarchical classification structures, each level corresponding to a different health state interval. These hierarchical classification structures include discharge efficiency threshold intervals, internal resistance change rate distribution, and temperature sensitivity classification boundaries, which provide a basis for subsequent refined management. Next, the system performs density peak extraction and boundary optimization on the multi-level classification structure of the battery state to generate a cluster center set of the battery health state.This process identifies density peaks within each classification structure and optimizes their boundaries to ensure that each cluster center accurately represents the health status of a specific battery type. For example, in the aforementioned agricultural monitoring task, suppose several batteries are classified as "early aging," characterized by low discharge efficiency and high internal resistance. By extracting density peaks and optimizing boundaries for these battery status data, the system can determine the specific cluster centers for these batteries and assign appropriate health status ranges. The system then adaptively expands the radius of these cluster centers to generate a battery health status classification region map. This step not only considers the current data distribution but also incorporates historical data and expected future trends to ensure the accuracy and stability of the classification results. Finally, based on the battery health status classification region map, the system maps and transforms the state features of the drone batteries to generate a quantitative battery health assessment vector. This process accurately describes the battery health status by mapping the actual state of each battery into the corresponding classification region and converting it into a quantifiable assessment vector. For example, in an agricultural monitoring mission, if a battery is classified as "good health" after the above steps, its corresponding battery health quantitative assessment vector might include high discharge efficiency, low internal resistance change rate, and good temperature sensitivity. To further improve the accuracy of the assessment results, the system performs orthogonalization and matrix reconstruction on the battery health quantitative assessment vector, ultimately generating a battery health assessment matrix. This matrix not only contains detailed battery health information but also provides an important reference for subsequent mission planning and maintenance strategies. In this way, the entire process, from generating a set of battery performance feature vectors to forming a battery health assessment matrix, not only improves the understanding and management of battery health, but also provides important decision support for drone operations. For example, in an agricultural monitoring mission, assuming that the health assessment matrix for each drone battery is obtained, the system can use this information to rationally arrange mission assignments and maintenance plans for each drone. Batteries in good health can continue to perform high-load missions, while batteries showing early signs of degradation are recommended for further inspection or early replacement, ensuring that the drone fleet remains efficient during agricultural monitoring missions. This process demonstrates the complete chain from data collection, feature extraction, health status assessment to practical application guidance, and demonstrates the powerful functionality and flexibility of intelligent battery management systems in complex application scenarios. This approach not only improves the overall operating efficiency of drones, but also extends battery life, reduces operating costs, and provides reliable protection for various application scenarios.

[0036] In a specific embodiment, the dynamic threshold partitioning of the battery health status assessment matrix by a recursive partitioning optimization algorithm to obtain the battery operating prediction state includes: Performing phase space reconstruction and dimensionality reduction processing on the battery health status assessment matrix to obtain a battery state feature space, and recursively bisectioning the battery state feature space using a recursive partitioning optimization algorithm to obtain a battery state initial partition set, wherein the battery state initial partition set includes a capacity degradation partition point, an internal resistance change boundary value, and a voltage response critical threshold; Constructing a battery state decision tree based on the initial battery state partition set, and performing information entropy optimization pruning on the battery state decision tree to obtain an optimal partition boundary set; wherein the optimal partition boundary set includes a health state dividing line, a working performance cluster center, and an abnormal state isolation interval; Dynamic parameter adjustment and nonlinear mapping are performed on the optimal partition boundary set to obtain a dynamic partition map of the battery state, and multi-dimensional threshold optimization is performed on the dynamic partition map of the battery state to obtain an optimized battery state threshold set; wherein the optimized battery state threshold set includes a capacity decay rate threshold, a power output limit, and a temperature sensitive interval delimiter; Based on the battery state optimization threshold set, a multi-scenario deduction analysis is performed on the UAV battery to obtain a battery operation prediction state; wherein the battery operation prediction state includes a battery remaining capacity prediction value, power output reliability, and discharge safety boundary parameters.

[0037] Specifically, in the process of assessing the health status of drone batteries, a recursive partitioning optimization algorithm is used to dynamically partition the battery health status assessment matrix to further refine and optimize its predicted operating state. This process begins with phase space reconstruction and dimensionality reduction of the battery health status assessment matrix. This method extracts key state features from high-dimensional data, forming a simplified battery state feature space that fully represents the battery health status. For example, in an agricultural monitoring mission, suppose we have obtained several drone battery health status assessment matrices. These matrices contain information on battery capacity, internal resistance, voltage response, and other aspects. By performing phase space reconstruction and dimensionality reduction on these matrices, we can effectively compress the complex information into a few core dimensions, such as discharge efficiency and temperature change rate, to facilitate subsequent analysis. Next, a recursive partitioning optimization algorithm is used to recursively bisection this battery state feature space to generate an initial set of battery state partitions. This step aims to identify boundary points that clearly distinguish different battery health states by continuously partitioning the dataset. In this process, the initial set of battery state partitions includes important indicators such as capacity degradation partition points, internal resistance change boundary values, and voltage response critical thresholds. For example, consider drones used in agricultural monitoring. Suppose we discover that some batteries experience a significant capacity drop after a certain number of charge-discharge cycles, while others exhibit a high rate of change in internal resistance or abnormal voltage response. Using a recursive partitioning optimization algorithm, the system identifies these differences and establishes an initial set of battery state partitions. This step provides the foundation for subsequent decision tree construction. Based on this initial set of battery state partitions, the system constructs a battery state decision tree and performs entropy-optimized pruning to obtain the optimal set of partition boundaries. Decision trees are powerful classification tools that automatically learn how to classify data based on input features. In this example, the battery state decision tree not only considers factors such as capacity degradation, internal resistance change, and voltage response, but also incorporates other parameters that may affect battery health. Entropy-optimized pruning removes branches from the decision tree that contribute little to classification or are prone to overfitting, thereby ensuring the model's generalization. For example, in an agricultural monitoring mission, suppose we construct a decision tree encompassing multiple battery health states. After information entropy optimization and pruning, we ultimately determine key parameters such as the health state dividing line, performance cluster center, and abnormal state isolation interval. These parameters together constitute the optimal partition boundary set. The system then dynamically adjusts the parameters and nonlinearly maps the optimal partition boundary set to generate a dynamic battery state partition map. This process allows for dynamic adjustment of partition boundaries based on real-time data, making battery health state assessment more accurate and flexible. For example, in agricultural monitoring missions, the battery's operating environment and load conditions may change with seasonal changes or flight missions.Through dynamic parameter adjustment, the system updates partition boundaries based on the latest environmental and operating conditions, generating a dynamic battery status partition map reflecting the current battery health. Furthermore, the system performs multi-dimensional threshold optimization on the dynamic battery status partition map to obtain an optimized battery status threshold set. These optimized thresholds include capacity decay rate thresholds, power output limits, and temperature sensitivity range delimiters, providing quantitative criteria for subsequent scenario simulations. Finally, based on the optimized battery status threshold set, the system conducts multi-scenario analysis of the drone's batteries to obtain a predicted battery operating status. For example, in an agricultural monitoring mission, considering an upcoming long-duration flight, the system can pre-evaluate each battery's suitability for the mission based on information such as the predicted remaining capacity, power output reliability, and discharge safety margin parameters. Specifically, if a battery's predicted remaining capacity is low and its power output reliability does not meet mission requirements, the system may recommend replacing it. Conversely, for batteries that demonstrate good health, the system will confirm their ability to support the mission. The entire process, from processing the battery health assessment matrix to ultimately deriving a predicted battery operating status, not only improves understanding and management of battery health status but also provides crucial decision support for drone operations. This approach not only improves overall drone efficiency but also extends battery life, reduces operating costs, and provides reliable support for a variety of application scenarios. This process demonstrates a complete chain from data collection, feature extraction, health status assessment, to practical application guidance, demonstrating the powerful functionality and flexibility of intelligent battery management systems in complex application scenarios.

[0038] In a specific embodiment, the adaptive matching calculation is performed on the target task based on the predicted battery operating state to obtain a task power consumption optimization configuration scheme, including: Performing segmented deconstruction analysis on the target task to obtain a task load characteristic sequence, and performing time-domain power decomposition on the task load characteristic sequence to obtain a task power consumption distribution map, wherein the task power consumption distribution map includes a flight power demand curve, a load power consumption fluctuation range, and a task execution timing table; Based on the task power consumption distribution map, the predicted battery working state is mapped and transformed to obtain a battery power output matching sequence, and the battery power output matching sequence is cross-validated in multiple dimensions to obtain a task battery matching matrix, wherein the task battery matching matrix includes a power matching coefficient, a discharge depth threshold, and an operating temperature boundary; Performing hierarchical optimization processing on the task battery matching matrix to obtain a task power consumption scheduling strategy set, and performing dynamic constraint solving on the task power consumption scheduling strategy set to obtain a power consumption scheduling execution plan; Constructing a task energy flow graph based on the power consumption scheduling execution scheme, and performing multi-path optimization calculation on the task energy flow graph to obtain a task energy allocation sequence; Multi-objective collaborative optimization is performed on the task energy allocation sequence to obtain a task power consumption optimization configuration scheme.

[0039] Specifically, the process of adaptively matching the target task based on the predicted battery operating state to obtain an optimized power consumption configuration requires first performing a segmented decomposition analysis of the target task. This process aims to decompose complex tasks into multiple manageable subtasks and extract key task load characteristic sequences from these subtasks. For example, in an agricultural monitoring mission, suppose a drone is required to cover a large area of ​​farmland and capture high-resolution imagery. The mission includes multiple phases, such as takeoff, cruise, image capture, and return. By performing a detailed segmented decomposition analysis of these phases, the system can identify the specific load requirements of each phase, such as the high power demand during takeoff, stable power consumption during cruise, and intermittent high loads during image capture. Next, the task load characteristic sequence is decomposed in the time domain to generate a task power consumption distribution map. This map not only includes the flight power demand curve, but also displays the payload power consumption fluctuation range and the task execution schedule, comprehensively reflecting the power consumption requirements of each phase during the mission. Next, based on the task power consumption distribution map, the system performs a mapping transformation on the predicted battery operating state to generate a battery power output matching sequence. This step combines the mission's power requirements with the battery's health status and predicted performance to ensure that each battery can support the mission under optimal conditions. For example, in an agricultural monitoring mission, if a battery has a high predicted remaining capacity and a wide depth of discharge threshold, the system will assign it to mission phases requiring higher power output, such as takeoff or filming. Batteries with less favorable health conditions will be assigned to more stable cruise missions. To ensure the accuracy and reliability of this matching, the system also performs multi-dimensional cross-validation on the battery power output matching sequence to generate a mission battery matching matrix. This matrix includes not only power matching coefficients but also key parameters such as depth of discharge thresholds and operating temperature limits, ensuring that each battery can operate efficiently within its safe range. The system then performs a hierarchical optimization process on the mission battery matching matrix to generate a set of mission power scheduling strategies. This process comprehensively considers factors such as battery health status, mission requirements, and environmental conditions to develop a detailed power scheduling strategy. For example, in an agricultural monitoring mission, assuming we've determined which batteries are suitable for which mission phases, we need to further refine the specific operation of each battery, such as adjusting the power output rate during takeoff and optimizing the flight speed during cruise. Then, we dynamically solve the constraints of the mission power scheduling strategy set to generate a power scheduling execution plan. This plan not only takes into account the current mission requirements but also incorporates real-time environmental changes and battery health status to ensure that each operation is performed under optimal conditions. For example, in the event of sudden weather changes, the system can dynamically adjust the flight path and power output strategy based on the latest meteorological data to ensure the smooth completion of the mission.Based on the power consumption scheduling execution plan generated above, the system constructs a task energy flow graph and performs multi-path optimization on it to generate a task energy allocation sequence. The task energy flow graph details the energy flow and distribution during the task, while the multi-path optimization calculation further improves the overall efficiency of the system by finding the most efficient energy transmission path. For example, in an agricultural monitoring task, suppose a group of drones need to collaborate to complete a large-scale farmland monitoring mission. The system can optimize the energy allocation paths in the energy flow graph to ensure that each drone can perform the mission under optimal conditions. Specifically, for drones far from charging stations, the system may prioritize batteries with good health and sufficient remaining capacity to reduce the need for mid-flight battery replacement. For drones closer to charging stations, the system can flexibly adjust their task order to ensure efficient completion of the overall mission. Finally, the system performs multi-objective collaborative optimization on the task energy allocation sequence to generate a task power optimization configuration. Multi-objective collaborative optimization aims to balance multiple conflicting objectives, such as maximizing task completion rate, minimizing energy consumption, and extending battery life. For example, in an agricultural monitoring task, suppose we need to cover the largest possible area within a limited time and ensure that all drones can return safely to base. Through multi-objective collaborative optimization, the system can generate the optimal mission power consumption configuration plan while meeting these objectives. Specifically, the system may recommend that certain drones lower their flight altitudes during specific time periods to reduce energy consumption, or adjust their shooting frequency to conserve power, while ensuring high-quality completion of the entire monitoring mission. The entire process, from extracting the mission payload feature sequence to generating the final mission power optimization configuration plan, not only improves the understanding and management capabilities of mission requirements but also provides important decision-making support for drone operations. This approach not only improves the overall efficiency of drones, but also extends battery life, reduces operating costs, and provides reliable protection for various application scenarios. This process demonstrates a complete chain from data collection, feature extraction, health status assessment, to practical application guidance, and demonstrates the powerful functionality and flexibility of intelligent battery management systems in complex application scenarios.

[0040] In a specific embodiment, the intelligent discharge strategy planning of the UAV battery based on the mission power consumption optimization configuration scheme is performed to obtain a battery discharge strategy control parameter set, including: Performing power flow topology decomposition on the task power consumption optimization configuration scheme to obtain a discharge power flow distribution sequence, and performing multi-scenario energy link tracing on the discharge power flow distribution sequence to obtain a discharge strategy topology map, wherein the discharge strategy topology map includes power flow path points, energy allocation weight coefficients, and a discharge depth control curve; Constructing a discharge state transition network based on the discharge strategy topology map, performing state probability transition analysis on the discharge state transition network to obtain a discharge state migration matrix, and performing spectral clustering boundary extraction on the discharge state migration matrix to obtain a discharge control threshold set; Performing multi-dimensional constraint mapping on the discharge control threshold set to obtain a discharge control constraint vector group, and performing nonlinear adaptive adjustment on the discharge control constraint vector group to obtain a discharge strategy optimization sequence; Constructing a discharge efficiency feature map based on the discharge strategy optimization sequence, and performing multi-level feature decoupling on the discharge efficiency feature map to obtain a discharge efficiency energy data set; A multi-objective collaborative optimization process is performed on the discharge efficiency energy data set to obtain a battery discharge strategy control parameter set.

[0041] Specifically, when planning an intelligent discharge strategy for a drone battery based on a mission-based power optimization configuration to obtain a set of battery discharge strategy control parameters, the mission-based power optimization configuration must first undergo a power flow topological decomposition. This process aims to decompose complex mission power requirements into multiple specific discharge power flow distribution sequences, thereby better understanding the energy flow during different mission phases. For example, in an agricultural monitoring mission, suppose a drone is required to cover a large area of ​​farmland and capture high-resolution imagery. The mission includes multiple phases, such as takeoff, cruise, image capture, and return. By performing a power flow topological decomposition of these phases, the system can identify the specific power requirements of each phase, such as high power output during takeoff, stable power consumption during cruise, and intermittent high loads during image capture. Next, multi-scenario energy link tracing is performed on the discharge power flow distribution sequence to generate a discharge strategy topological map. This map not only includes power flow path points but also displays energy allocation weight coefficients and discharge depth control curves, comprehensively reflecting the energy allocation and discharge strategy at each phase of the mission. Next, based on the discharge strategy topological map, the system constructs a discharge state transition network and performs state probability transition analysis on it to generate a discharge state transition matrix. This step analyzes the state transition probabilities of batteries at different discharge states, helping the system understand the battery's behavior under different operating conditions. For example, in an agricultural monitoring mission, if a battery has a high power output requirement during takeoff, the system can use the discharge state transition network to predict its discharge state changes during the subsequent cruise and photography phases. Spectral clustering is then performed on the discharge state transition matrix to extract boundaries and generate a set of discharge control thresholds. These thresholds not only take into account the current mission requirements but also incorporate the battery's health status and environmental conditions, ensuring that each operation is performed under optimal conditions. For example, in high-temperature environments, the system can dynamically adjust the discharge control thresholds based on the latest temperature data to prevent battery overheating. The system then performs multi-dimensional constraint mapping on the discharge control threshold set to generate a set of discharge control constraint vectors. This process comprehensively considers factors such as battery health, mission requirements, and environmental conditions to develop a detailed set of discharge control constraint rules. For example, in an agricultural monitoring mission, assuming that we have determined which batteries are suitable for which mission phases, the next step is to further refine the specific operation of each battery, such as adjusting the power output rate during takeoff or optimizing the flight speed during cruise. The system then performs nonlinear adaptive adjustments to the set of discharge control constraint vectors to generate an optimized discharge strategy sequence. This sequence not only considers the current mission requirements but also incorporates real-time environmental changes and battery health, ensuring that each operation is performed under optimal conditions. For example, in the event of unexpected weather changes, the system can dynamically adjust the flight path and power output strategy based on the latest meteorological data to ensure successful mission completion.Based on the generated discharge strategy optimization sequence, the system constructs a discharge efficiency feature map and performs multi-level feature decoupling on it to generate a discharge efficiency energy dataset. The discharge efficiency feature map details the battery's performance under different discharge strategies. Multi-level feature decoupling further improves the system's overall efficiency by isolating different influencing factors. For example, in an agricultural monitoring mission, suppose a group of drones need to collaborate to complete a large-scale farmland monitoring mission. The system can optimize the energy allocation path in the discharge efficiency feature map to ensure that each drone can perform the mission under optimal conditions. Specifically, for drones farther from charging stations, the system may prioritize batteries with better health and sufficient remaining capacity to reduce the need for mid-flight battery replacement. For drones closer to charging stations, the system can flexibly adjust their mission order to ensure efficient overall mission completion. Finally, the system performs multi-objective collaborative optimization on the discharge efficiency energy dataset to generate a set of battery discharge strategy control parameters. Multi-objective collaborative optimization aims to balance multiple conflicting objectives, such as maximizing mission completion rate, minimizing energy consumption, and extending battery life. For example, in an agricultural monitoring mission, suppose we need to cover the largest possible area within a limited timeframe and ensure all drones return safely to base. Through multi-objective collaborative optimization, the system can generate the optimal set of battery discharge strategy control parameters while meeting these objectives. Specifically, the system might recommend that certain drones lower their flight altitudes during specific time periods to reduce energy consumption, or adjust their capture frequency to conserve power, while ensuring high-quality completion of the overall monitoring mission. The entire process, from generating mission power optimization configurations to finalizing the battery discharge strategy control parameter set, not only improves understanding and management capabilities of mission requirements but also provides important decision-making support for drone operations. For example, in an actual agricultural monitoring mission, suppose we have obtained mission power optimization configurations for each drone battery. The system first performs power flow topological decomposition on these configurations, identifying the specific power requirements for takeoff, cruising, capture, and return, and generating a discharge strategy topology map. Next, the system constructs a discharge state transition network, analyzes the state transition probabilities of the batteries under different discharge states, generates a discharge state transition matrix, and extracts a set of discharge control thresholds from this matrix. These thresholds are dynamically adjusted based on battery health and environmental conditions to ensure that each battery operates efficiently within its safe range. The system then applies multi-dimensional constraint mapping to the set of discharge control thresholds, generating a set of discharge control constraint vectors. Through nonlinear adaptive adjustment, it generates an optimized discharge strategy sequence. Based on this, the system constructs a discharge efficiency feature map, performs multi-level feature decoupling, and generates a discharge efficiency energy dataset.Finally, the system performs multi-objective collaborative optimization on this data to generate a final set of battery discharge strategy control parameters, ensuring that each battery discharges according to the optimal strategy during the mission, thereby improving mission success and efficiency. This entire process demonstrates a complete chain from data collection, feature extraction, health status assessment, to practical application guidance, and demonstrates the powerful functionality and flexibility of intelligent battery management systems in complex application scenarios. This approach not only improves the overall efficiency of drones, but also extends battery life, reduces operating costs, and provides reliable support for a variety of application scenarios.

[0042] In a specific embodiment, the UAV is provided with a battery management unit and a flight control system interaction interface. The UAV is controlled in real time to complete the target task based on the battery discharge strategy control parameter set through a heterogeneous computing architecture, including: Performing resource allocation mapping on the battery discharge strategy control parameter set to obtain a computing load distribution matrix, and performing parallel task segmentation on the computing load distribution matrix using a preset heterogeneous computing architecture to obtain a heterogeneous computing execution unit sequence, wherein the heterogeneous computing execution unit sequence includes a power control instruction set, a voltage regulation execution sequence, and a temperature compensation control point; Constructing a hierarchical execution priority network based on the heterogeneous computing execution unit sequence, obtaining a task execution graph based on the hierarchical execution priority network, and performing critical path extraction on the task execution graph to obtain a sequential execution pipeline strategy; Performing resource conflict detection and resolution on the sequential execution pipeline strategy to obtain an optimized execution strategy set, and concurrently scheduling and arranging the optimized execution strategy set to obtain a heterogeneous execution instruction stream; Based on the heterogeneous execution instruction stream, the interaction interface between the battery management unit and the flight control system is mapped to obtain a battery flight control collaborative control sequence, and the battery flight control collaborative control sequence is dynamically optimized and embedded deployed to obtain a UAV execution control instruction set. Based on the UAV execution control instruction set, the UAV is controlled in real time to complete the target task.

[0043] Specifically, when using a heterogeneous computing architecture to control a drone in real time to complete a target mission based on a battery discharge strategy control parameter set, resource allocation mapping of the battery discharge strategy control parameter set is first required to generate a computational load distribution matrix. This process aims to decompose complex mission requirements into multiple specific computational tasks and allocate them appropriately based on system resource availability. For example, in an agricultural monitoring mission, suppose a drone is required to cover a large area of ​​farmland and capture high-resolution imagery. The mission includes multiple phases such as takeoff, cruise, capture, and return. By mapping these phases, the system can identify the specific computational requirements of each phase, such as high power output control during takeoff, stable power regulation during cruise, and intermittent high load management during capture. Next, the computational load distribution matrix is ​​partitioned into parallel tasks using the pre-defined heterogeneous computing architecture to generate a sequence of heterogeneous compute execution units. This sequence includes not only the power control instruction set, but also the voltage regulation execution sequence and temperature compensation control points, comprehensively reflecting the energy management and control requirements of each phase of the mission. Next, based on the heterogeneous compute execution unit sequence, the system constructs a hierarchical execution priority network and generates a task execution graph based on this network. This step helps the system develop a detailed execution plan by analyzing the dependencies and priorities between different computational tasks. For example, in an agricultural monitoring task, if a battery has a high power output requirement during takeoff, the system can use a hierarchical execution priority network to predict the order of its computational tasks during the subsequent cruise and photography phases. It then extracts critical paths from the task execution graph and generates sequential execution pipeline strategies. These strategies not only consider the current task requirements but also incorporate the battery's health status and environmental conditions to ensure that each operation is performed under optimal conditions. For example, in high-temperature environments, the system can dynamically adjust the sequential execution pipeline strategy based on the latest temperature data to prevent battery overheating. The system then detects and resolves resource conflicts within the sequential execution pipeline strategy to generate a set of optimized execution strategies. This process comprehensively considers factors such as system computing resources and communication bandwidth to ensure efficient parallel execution of various tasks. For example, in an agricultural monitoring task, assuming that we have determined which computational tasks are suitable for parallel execution, we need to further refine the specific execution methods for each task, such as adjusting the power output rate during takeoff or optimizing the flight speed during cruise. The optimized execution strategy set is then concurrently scheduled and orchestrated to generate a heterogeneous execution instruction stream. This instruction stream not only takes into account the current mission requirements but also incorporates real-time environmental changes and battery health to ensure that each operation is performed under optimal conditions. For example, in the event of sudden weather changes, the system can dynamically adjust the flight path and power output strategy based on the latest meteorological data to ensure successful mission completion. Based on this generated heterogeneous execution instruction stream, the system maps the interaction interface between the battery management unit and the flight control system to generate a battery-flight-control coordinated control sequence.This process combines the control instructions of the battery management unit with the operational instructions of the flight control system to ensure that the drone can perform its mission under optimal conditions. Specifically, in an agricultural monitoring mission, if a battery requires a higher power output during takeoff, the system will generate a corresponding power control instruction set and map it to the flight control system's operational instructions, ensuring that the drone receives sufficient power support during takeoff. The battery-flight control coordinated control sequence is then dynamically optimized and embedded, generating the drone's execution control instruction set. This instruction set not only takes into account the current mission requirements but also incorporates real-time environmental changes and battery health to ensure that each operation is performed under optimal conditions. For example, in the event of sudden weather changes, the system can dynamically adjust the flight path and power output strategy based on the latest meteorological data to ensure successful mission completion. Finally, based on the drone's execution control instruction set, the system controls the drone in real time to achieve the desired mission. Throughout this process, the system leverages a heterogeneous computing architecture to achieve efficient parallel processing and resource allocation, ensuring that the drone can operate efficiently in complex and changing mission environments. For example, in an agricultural monitoring mission, suppose a group of drones need to collaborate to monitor a large area of ​​farmland. The system can optimize the energy allocation path within the battery and flight control coordinated control sequence to ensure that each drone can execute the mission under optimal conditions. Specifically, for drones farther from charging stations, the system may prioritize batteries with better health and sufficient remaining capacity to reduce the need for mid-flight battery replacements. For drones closer to charging stations, the system can flexibly adjust the task order to ensure efficient completion of the overall mission. For example, in an actual agricultural monitoring mission, suppose we have a set of discharge strategy control parameters for each drone battery. The system first maps these parameter sets to resource allocation, generates a computational load distribution matrix, and partitions them into parallel tasks using a pre-defined heterogeneous computing architecture, resulting in a sequence of heterogeneous computing execution units. Next, a hierarchical execution priority network is constructed based on this sequence, generating a task execution graph and extracting sequential execution pipeline strategies from it. The system then detects and resolves resource conflicts within the sequential execution pipeline strategies, generating an optimized set of execution strategies, and orchestrates these strategies concurrently to generate a heterogeneous execution instruction stream. Based on this, the system maps the interface between the battery management unit and the flight control system, generating a coordinated control sequence for the battery and flight control. Dynamic parameter optimization and embedded deployment are then performed to generate the drone's execution control instruction set. Ultimately, based on the drone's execution control instruction set, the system controls the drone in real time to complete the desired mission. These control sequences, after dynamic parameter optimization and embedded deployment, are ultimately converted into the drone's execution control instruction set. For example, when a drone needs to return home from a low-battery mission, the system optimizes the battery output power and flight path in real time to ensure a safe landing while maximizing the utilization of the remaining battery power.The entire solution tightly integrates battery efficiency management with flight control through a heterogeneous computing architecture, achieving intelligent management across the entire process, from parameter perception to command execution. Through multi-layered optimization strategies and collaborative control mechanisms, the system ensures mission completion while efficiently utilizing battery energy. This collaborative optimization approach is particularly suitable for demanding UAV applications, such as long-duration reconnaissance, emergency rescue, and precision agriculture, where both battery efficiency and mission performance are critical performance indicators. Furthermore, the entire process, from generating the battery discharge strategy control parameter set to finalizing the UAV execution control command set, not only enhances understanding and management capabilities for mission requirements but also provides crucial decision-making support for UAV operations. This approach not only improves the overall efficiency of UAVs, but also extends battery life, reduces operating costs, and provides reliable support for a variety of application scenarios. This process demonstrates a complete chain from data acquisition, feature extraction, health status assessment, to practical application guidance, demonstrating the powerful functionality and flexibility of intelligent battery management systems in complex application scenarios. This approach ensures that UAVs discharge and control according to the optimal strategy during mission execution, thereby improving mission success and efficiency.

[0044] The above describes the intelligent optimization management method of the drone battery efficiency in the embodiment of the present invention. The following describes the intelligent optimization management system of the drone battery efficiency in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, an intelligent optimization management system for drone battery efficiency includes: An extraction module 21 is used to perform multi-dimensional feature extraction on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set; An analysis module 22 is configured to perform a battery health status hierarchical cluster analysis on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix; a partitioning module 23 for performing dynamic threshold partitioning on the battery health status assessment matrix using a recursive partitioning optimization algorithm to obtain a predicted battery operating state; The calculation module 24 is used to obtain the target mission of the UAV and perform adaptive matching calculation on the target mission based on the predicted battery working state to obtain a task power consumption optimization configuration plan; A planning module 25 is configured to perform intelligent discharge strategy planning for the UAV battery based on the mission power consumption optimization configuration scheme to obtain a battery discharge strategy control parameter set; The execution module 26 is used to control the UAV in real time to complete the target task based on the battery discharge strategy control parameter set through a heterogeneous computing architecture.

[0045] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0046] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, memory, display screen, input device, network interface and database connected via a system bus. The processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the above method is implemented.

[0047] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0048] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the above-described method when executed by a processor. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0049] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware using a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media provided herein and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM.

[0050] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0051] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An intelligent optimization management method for drone battery performance, characterized in that: The method is applied to a drone equipped with a drone battery, and includes the following steps: Performing multi-dimensional feature extraction on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set; Performing a battery health status hierarchical cluster analysis on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix; Performing dynamic threshold partitioning on the battery health status assessment matrix through a recursive partitioning optimization algorithm to obtain a battery operating prediction state; Obtaining a target mission of the UAV, and performing adaptive matching calculation on the target mission based on the predicted battery operating state to obtain a mission power consumption optimization configuration plan; Performing intelligent discharge strategy planning on the UAV battery based on the mission power consumption optimization configuration scheme to obtain a battery discharge strategy control parameter set; Through a heterogeneous computing architecture, the UAV is controlled in real time to complete the target task based on the battery discharge strategy control parameter set.

2. The intelligent optimization management method for drone battery performance according to claim 1, characterized in that: The multi-dimensional feature extraction is performed on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set, including: Performing multi-scale decomposition on the discharge characteristic curve of the UAV battery to obtain a voltage fluctuation feature sequence, and performing time-frequency domain transformation on the voltage fluctuation feature sequence to obtain a voltage spectrum feature matrix, which includes voltage decay rate, voltage fluctuation amplitude, and voltage stability range; Performing nonlinear dynamic response analysis on the voltage spectrum characteristic matrix to obtain a battery load response characteristic spectrum, and performing topological feature decomposition on the battery load response characteristic spectrum to obtain a battery working state feature set; A multi-parameter correlation analysis is performed on the battery operating state feature set to obtain a battery performance degradation trend sequence, and feature space mapping is performed on the battery performance degradation trend sequence to obtain a battery performance feature vector set; wherein the battery performance feature vector set includes a performance decay rate, a remaining life prediction value, and a discharge efficiency coefficient.

3. The intelligent optimization management method for drone battery performance according to claim 1, characterized in that: The battery health status hierarchical cluster analysis is performed on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix, including: Performing high-dimensional spatial transformation and sparse representation on the battery performance feature vector set to obtain a battery feature sparse representation matrix, and performing multi-kernel function similarity calculation on the battery feature sparse representation matrix to obtain a battery state affinity spectrum; wherein the battery state affinity spectrum includes capacity decay trajectory points, voltage recovery characteristic index, and load response gradient map; A battery state diffusion mapping network is constructed based on the battery state affinity spectrum, and topological feature extraction and dynamic evolution analysis are performed on the battery state diffusion mapping network to obtain a battery state topological structure diagram. The battery state topological structure diagram is then subjected to spectral decomposition and hierarchical cutting to obtain a battery state multi-level classification structure; wherein the battery state multi-level classification structure includes a discharge efficiency threshold range, an internal resistance change rate distribution, and a temperature sensitivity classification boundary; Performing density peak extraction and boundary optimization on the multi-level classification structure of battery status to obtain a battery health status cluster center set, and performing adaptive radius expansion on the battery health status cluster center set to obtain a battery health status classification area map; Based on the battery health status classification area map, the state feature mapping and conversion of the UAV battery are performed to obtain a battery state quantitative evaluation vector, and the battery state quantitative evaluation vector is orthogonalized and matrix reconstructed to obtain a battery health status evaluation matrix.

4. The intelligent optimization management method for drone battery performance according to claim 1, characterized in that: The method of performing dynamic threshold partitioning on the battery health status assessment matrix by a recursive partitioning optimization algorithm to obtain a battery operating prediction state includes: Performing phase space reconstruction and dimensionality reduction processing on the battery health status assessment matrix to obtain a battery state feature space, and recursively bisectioning the battery state feature space using a recursive partitioning optimization algorithm to obtain a battery state initial partition set, wherein the battery state initial partition set includes a capacity degradation partition point, an internal resistance change boundary value, and a voltage response critical threshold; Constructing a battery state decision tree based on the initial battery state partition set, and performing information entropy optimization pruning on the battery state decision tree to obtain an optimal partition boundary set; wherein the optimal partition boundary set includes a health state dividing line, a working performance cluster center, and an abnormal state isolation interval; Dynamic parameter adjustment and nonlinear mapping are performed on the optimal partition boundary set to obtain a dynamic partition map of the battery state, and multi-dimensional threshold optimization is performed on the dynamic partition map of the battery state to obtain an optimized battery state threshold set; wherein the optimized battery state threshold set includes a capacity decay rate threshold, a power output limit, and a temperature sensitive interval delimiter; Based on the battery state optimization threshold set, a multi-scenario deduction analysis is performed on the UAV battery to obtain a battery operation prediction state; wherein the battery operation prediction state includes a battery remaining capacity prediction value, power output reliability, and discharge safety boundary parameters.

5. The intelligent optimization management method for drone battery performance according to claim 1, characterized in that: The adaptive matching calculation is performed on the target task based on the predicted battery working state to obtain a task power consumption optimization configuration scheme, including: Performing segmented deconstruction analysis on the target task to obtain a task load characteristic sequence, and performing time-domain power decomposition on the task load characteristic sequence to obtain a task power consumption distribution map, wherein the task power consumption distribution map includes a flight power demand curve, a load power consumption fluctuation range, and a task execution timing table; Based on the task power consumption distribution map, the predicted battery working state is mapped and transformed to obtain a battery power output matching sequence, and the battery power output matching sequence is cross-validated in multiple dimensions to obtain a task battery matching matrix, wherein the task battery matching matrix includes a power matching coefficient, a discharge depth threshold, and an operating temperature boundary; Performing hierarchical optimization processing on the task battery matching matrix to obtain a task power consumption scheduling strategy set, and performing dynamic constraint solving on the task power consumption scheduling strategy set to obtain a power consumption scheduling execution plan; Constructing a task energy flow graph based on the power consumption scheduling execution scheme, and performing multi-path optimization calculation on the task energy flow graph to obtain a task energy allocation sequence; Multi-objective collaborative optimization is performed on the task energy allocation sequence to obtain a task power consumption optimization configuration scheme.

6. The intelligent optimization management method for drone battery performance according to claim 1, characterized in that: The intelligent discharge strategy planning of the UAV battery based on the mission power consumption optimization configuration scheme is performed to obtain a battery discharge strategy control parameter set, including: Performing power flow topology decomposition on the task power consumption optimization configuration scheme to obtain a discharge power flow distribution sequence, and performing multi-scenario energy link tracing on the discharge power flow distribution sequence to obtain a discharge strategy topology map, wherein the discharge strategy topology map includes power flow path points, energy allocation weight coefficients, and a discharge depth control curve; Constructing a discharge state transition network based on the discharge strategy topology map, performing state probability transition analysis on the discharge state transition network to obtain a discharge state migration matrix, and performing spectral clustering boundary extraction on the discharge state migration matrix to obtain a discharge control threshold set; Performing multi-dimensional constraint mapping on the discharge control threshold set to obtain a discharge control constraint vector group, and performing nonlinear adaptive adjustment on the discharge control constraint vector group to obtain a discharge strategy optimization sequence; Constructing a discharge efficiency feature map based on the discharge strategy optimization sequence, and performing multi-level feature decoupling on the discharge efficiency feature map to obtain a discharge efficiency energy data set; A multi-objective collaborative optimization process is performed on the discharge efficiency energy data set to obtain a battery discharge strategy control parameter set.

7. The intelligent optimization management method for drone battery performance according to claim 1, characterized in that: The UAV is provided with a battery management unit and a flight control system interaction interface. The UAV is controlled in real time to complete the target mission based on the battery discharge strategy control parameter set through a heterogeneous computing architecture, including: Performing resource allocation mapping on the battery discharge strategy control parameter set to obtain a computing load distribution matrix, and performing parallel task segmentation on the computing load distribution matrix using a preset heterogeneous computing architecture to obtain a heterogeneous computing execution unit sequence, wherein the heterogeneous computing execution unit sequence includes a power control instruction set, a voltage regulation execution sequence, and a temperature compensation control point; Constructing a hierarchical execution priority network based on the heterogeneous computing execution unit sequence, obtaining a task execution graph based on the hierarchical execution priority network, and performing critical path extraction on the task execution graph to obtain a sequential execution pipeline strategy; Performing resource conflict detection and resolution on the sequential execution pipeline strategy to obtain an optimized execution strategy set, and concurrently scheduling and arranging the optimized execution strategy set to obtain a heterogeneous execution instruction stream; Based on the heterogeneous execution instruction stream, the interaction interface between the battery management unit and the flight control system is mapped to obtain a battery flight control collaborative control sequence, and the battery flight control collaborative control sequence is dynamically optimized and embedded deployed to obtain a UAV execution control instruction set. Based on the UAV execution control instruction set, the UAV is controlled in real time to complete the target task.

8. An intelligent optimization management system for drone battery efficiency, characterized by: Applied to a drone, wherein the drone is equipped with a drone battery, including: An extraction module, configured to perform multi-dimensional feature extraction on the discharge characteristic curve of the UAV battery to obtain a battery performance feature vector set; An analysis module, configured to perform a battery health status hierarchical cluster analysis on the UAV battery based on the battery performance feature vector set to obtain a battery health status evaluation matrix; A partitioning module is used to perform dynamic threshold partitioning on the battery health status assessment matrix through a recursive partitioning optimization algorithm to obtain a battery operating prediction state; A calculation module is used to obtain a target task of the UAV and perform adaptive matching calculation on the target task based on the predicted working state of the battery to obtain a task power consumption optimization configuration plan; A planning module, configured to perform intelligent discharge strategy planning for the UAV battery based on the mission power consumption optimization configuration scheme, and obtain a battery discharge strategy control parameter set; An execution module is used to control the UAV in real time to complete the target task based on the battery discharge strategy control parameter set through a heterogeneous computing architecture.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images