Self-adaptive charging power supply control method and system

Abstract

The invention provides a self-adaptive charging power supply control method and system, and relates to the field of power supply control, and the method comprises the steps: capturing a voltage and temperature time sequence signal of a battery in real time through a high-precision sensor, carrying out the multi-scale time sequence modeling of the dynamic characteristics of the battery through a time convolution network in combination with the charging behavior data of a user, and obtaining a multi-scale time sequence model; and extracting high-frequency nonlinear fluctuation characteristics of the battery state. Meanwhile, semantic association coding is performed on the user behavior data, and a mapping relation between the user demand and the battery health state is established. Then, a cross-modal feature interaction mechanism is introduced, dynamic coupling analysis of the battery state and the user requirement is achieved, finally, a charging voltage regulation and control instruction which is optimally matched with the current battery state and meets the user scene requirement is generated through a decoder, charging parameters are dynamically adjusted, and collaborative optimization of the charging efficiency, the battery life and the personalized scene is achieved.

Description

Technical Field

[0001] The present application relates to the field of power supply control, and more particularly to an adaptive charging power supply control method and system in an embodiment of the present application. Background Art

[0002] With the widespread application of lithium-ion batteries in electric vehicles, portable electronic devices, energy storage systems and other fields, the demand for intelligent battery charging management has become increasingly prominent. The traditional constant current / constant voltage (CC / CV) charging strategy realizes charging control by presetting fixed parameters, and its core principle relies on the static model of the battery. However, the battery exhibits highly dynamic characteristics in actual use: its internal electrochemical state (such as polarization effect, capacity attenuation) and external operating conditions (such as ambient temperature, load fluctuation) will continue to change with the charge and discharge cycle. Fixed parameter strategies cannot perceive the real-time state of the battery, which can easily lead to overcharging, undercharging or local overheating, aggravating degradation phenomena such as lithium dendrite growth and SEI film thickening, and significantly shortening battery life. In addition, users' demands for charging speed, safety and scene adaptability are becoming more and more diverse (such as emergency fast charging, slow charging at night, and low temperature protection). The rigid parameter configuration of traditional methods is difficult to meet the dynamic optimization needs in personalized scenarios.

[0003] Therefore, an adaptive charging power control scheme is desired. Summary of the invention

[0004] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides an adaptive charging power control method and system, which uses a high-precision sensor to capture the voltage and temperature timing signals of the battery in real time, combines the user's charging behavior data, and uses a time convolution network to perform multi-scale timing modeling on the dynamic characteristics of the battery, and extracts the high-frequency nonlinear fluctuation characteristics of the battery state. At the same time, the user behavior data is semantically associated and encoded to establish a mapping relationship between user needs and battery health status. Next, a cross-modal feature interaction mechanism is introduced to realize the dynamic coupling analysis of battery status and user needs. Finally, the decoder generates a charging voltage control instruction that optimally matches the current battery state and meets the user scenario requirements, and dynamically adjusts the charging parameters to achieve coordinated optimization of charging efficiency, battery life and personalized scenarios.

[0005] According to one aspect of the present application, there is provided an adaptive charging power supply control method, which includes: Monitor the battery status data of the charged power source in real time through a battery sensor, wherein the battery status data includes voltage data and temperature data; Acquire user charging behavior data, wherein the user charging behavior data includes charging time, charging frequency, charging duration, charging mode, and charging target power; Performing time-series fusion encoding on the battery status data according to the time dimension to obtain a multi-dimensional fusion representation of the battery status; Performing mapping association coding on the user charging behavior data based on the charging information to obtain an association coding representation of the user charging behavior; Performing battery state-charging behavior cross-modal feature independent restriction interactive analysis on the multi-dimensional fusion representation of the battery state and the associated coding representation of the user charging behavior to obtain a battery state-charging behavior interactive coding representation; Based on the battery state-charging behavior interactive coding representation, a target charging voltage recommended decoding value is determined to control the charged power source to charge.

[0006] According to another aspect of the present application, an adaptive charging power supply control system is provided, which includes: A battery status data real-time monitoring module, used to monitor the battery status data of the charged power source in real time through a battery sensor, wherein the battery status data includes voltage data and temperature data; A user charging behavior data acquisition module, used to acquire user charging behavior data, wherein the user charging behavior data includes charging time, charging frequency, charging duration, charging mode and charging target power; A battery state time series fusion coding module, used for performing time series fusion coding on the battery state data according to the time dimension to obtain a multi-dimensional fusion representation of the battery state; A user charging behavior association coding module, used for performing mapping association coding on the user charging behavior data based on charging information to obtain a user charging behavior association coding representation; A battery status-charging behavior interactive coding module, used to perform battery status-charging behavior cross-modal feature independent restriction interactive analysis on the multi-dimensional fusion representation of the battery status and the associated coding representation of the user charging behavior to obtain a battery status-charging behavior interactive coding representation; The target charging voltage recommended decoding value determination module is used to determine the target charging voltage recommended decoding value based on the battery state-charging behavior interactive coding representation to control the charged power source to charge.

[0007] Compared with the prior art, the present application provides an adaptive charging power control method and system, which uses a high-precision sensor to capture the voltage and temperature time series signals of the battery in real time, combines the user's charging behavior data, and uses a time convolution network to perform multi-scale time series modeling on the dynamic characteristics of the battery, and extracts the high-frequency nonlinear fluctuation characteristics of the battery state. At the same time, the user behavior data is semantically associated and encoded to establish a mapping relationship between user needs and battery health status. Then, a cross-modal feature interaction mechanism is introduced to achieve dynamic coupling analysis of battery status and user needs. Finally, the decoder generates a charging voltage control instruction that optimally matches the current battery state and meets the user's scenario requirements, dynamically adjusts the charging parameters, and achieves coordinated optimization of charging efficiency, battery life, and personalized scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0009] Figure 1 Flow chart of an adaptive charging power control method according to an embodiment of the present application.

[0010] Figure 2 Schematic diagram of data flow of an adaptive charging power control method according to an embodiment of the present application.

[0011] Figure 3 The present invention is a flow chart of performing time-series fusion encoding on the battery status data according to the time dimension in the adaptive charging power supply control method according to an embodiment of the present application to obtain a multi-dimensional fusion representation of the battery status.

[0012] Figure 4 A flowchart of performing independent restricted interactive analysis of the battery state-charging behavior cross-modal features on the multi-dimensional fusion representation of the battery state and the associated coding representation of the user charging behavior in the adaptive charging power control method according to an embodiment of the present application to obtain a battery state-charging behavior interactive coding representation.

[0013] Figure 5 is a system block diagram of an adaptive charging power control system according to an embodiment of the present application. DETAILED DESCRIPTION

[0014] Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0015] With the popularization of lithium-ion batteries in electric vehicles, mobile electronic devices, and energy storage systems, the demand for intelligent battery charging management has become increasingly important. The traditional constant current / constant voltage (CC / CV) charging method relies on preset fixed parameters and is controlled based on a static battery model. However, in actual operation, the battery exhibits very dynamic characteristics: its internal electrochemical state (such as polarization, capacity decay) and external use conditions (such as ambient temperature, load changes) will continue to change with the charging and discharging process. Since these fixed parameters cannot reflect the real-time status of the battery, the use of this strategy is prone to cause problems such as overcharging, undercharging, or local overheating, which accelerates the degradation process such as lithium dendrite formation and solid electrolyte interface film thickening, thereby greatly shortening the battery life. At the same time, users have increasingly diverse requirements for charging speed, safety, and adaptation to specific scenarios (such as emergency fast charging, slow charging at night, low temperature protection, etc.), and the fixed parameter settings of traditional methods are difficult to meet the optimization needs in different scenarios.

[0016] In recent years, adaptive charging technology has become a research hotspot. Its core goal is to balance charging efficiency and battery health by sensing the battery status in real time and dynamically adjusting charging parameters. Existing methods mainly rely on single-dimensional optimization: some studies build state observers based on equivalent circuit models or electrochemical impedance spectroscopy (EIS), but the model is highly dependent and the computational complexity is high; other solutions use neural networks to predict the battery state of health (SOH), but they are not deeply integrated with user behavior characteristics, resulting in insufficient strategy flexibility. In addition, battery status data shows corresponding time series changes over time, and traditional feature extraction methods are difficult to efficiently capture the high-frequency nonlinear time series fluctuation characteristics of battery status such as battery voltage / temperature.

[0017] In view of the static parameter limitations and insufficient adaptability of traditional charging strategies to user scenarios, an adaptive charging control scheme based on collaborative analysis of multimodal data is constructed in the technical solution of this application. It uses high-precision sensors to capture the voltage and temperature time series signals of the battery in real time, combines user charging behavior data (such as charging period, mode preference, etc.), and uses a time convolution network to perform multi-scale time series modeling on the dynamic characteristics of the battery, extracting the high-frequency nonlinear fluctuation characteristics of the battery state; at the same time, the user behavior data is semantically associated and encoded through a fully connected network to establish a mapping relationship between user needs and battery health status. On this basis, a cross-modal feature interaction mechanism is introduced to achieve dynamic coupling analysis of battery status and user needs while retaining the independence of battery physical characteristics and user behavior semantics. Finally, the decoder generates charging voltage control instructions that best match the current battery state and meet the needs of the user scenario, dynamically adjusts the charging parameters, and achieves collaborative optimization of charging efficiency, battery life, and personalized scenarios.

[0018] The present application proposes an adaptive charging power supply control method. Figure 1 Flow chart of an adaptive charging power control method according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the adaptive charging power supply control method according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the embodiment of the present application, the adaptive charging power control method includes: S110, monitoring the battery status data of the charged power source in real time through a battery sensor, the battery status data including voltage data and temperature data; S120, acquiring user charging behavior data, the user charging behavior data including charging time, charging frequency, charging duration, charging mode and charging target power; S130, performing temporal fusion encoding on the battery status data according to the time dimension to obtain a multi-dimensional fusion representation of the battery status; S140, performing mapping association encoding on the user charging behavior data based on charging information to obtain a user charging behavior association encoding representation; S150, performing battery status-charging behavior cross-modal feature independent restriction interactive analysis on the multi-dimensional fusion representation of the battery status and the user charging behavior association encoding representation to obtain a battery status-charging behavior interactive encoding representation; S160, based on the battery status-charging behavior interactive encoding representation, determining a target charging voltage recommended decoding value to control the charging of the charged power source.

[0019] In the above-mentioned adaptive charging power supply control method, the step S110 monitors the battery status data of the charged power supply in real time through the battery sensor, and the battery status data includes voltage data and temperature data. It should be understood that with the increasing application of lithium-ion batteries in electric vehicles, mobile electronic devices, energy storage systems and other fields, the demand for intelligent battery charging management has become more and more important. The traditional constant current / constant voltage (CC / CV) charging method relies on preset fixed parameters and is controlled based on the static model of the battery. However, in actual operation, the battery exhibits very dynamic characteristics: its internal electrochemical state (such as polarization phenomenon, capacity decay) and external use conditions (such as ambient temperature, load changes) will continue to change with the charging and discharging process. Since these fixed parameters cannot reflect the real-time status of the battery, the use of this strategy is prone to cause problems such as overcharging, undercharging or local overheating, which accelerates the degradation process such as lithium dendrite formation and solid electrolyte interface film thickening, thereby greatly shortening the battery life. Therefore, it is necessary to monitor the battery status data of the charged power supply in real time through the battery sensor, especially the voltage data and temperature data. Specifically, the state of the battery, especially the voltage and temperature, are important indicators for evaluating the health of the battery. Voltage directly reflects the equilibrium state of the electrochemical reaction inside the battery, while temperature is closely related to the heat generation and heat dissipation inside the battery. By accurately measuring these two parameters, the working state of the battery can be effectively monitored, abnormal conditions can be detected in time, and corresponding measures can be taken to prevent potential safety risks. For example, when the battery voltage drops rapidly or the temperature rises sharply, this may be a sign of a short circuit or other fault inside the battery. At this time, charging should be stopped immediately to avoid more serious damage. In order to achieve real-time monitoring of battery voltage and temperature, high-precision sensors are usually required to be integrated into the battery management system. These sensors can collect voltage signals at both ends of the battery and temperature information on the surface or inside of the battery at a high frequency, and transmit these data to the control system for processing.

[0020] In the above-mentioned adaptive charging power supply control method, the step S120 obtains the user's charging behavior data, and the user's charging behavior data includes charging time, charging frequency, charging duration, charging mode and charging target power. It should be understood that different time periods often correspond to different power demand and environmental conditions. For example, the user may choose to charge at night, when the grid load is low and the power supply is relatively stable; while during the day, especially during peak hours, the grid load is large, which may affect the charging efficiency. In addition, some areas implement time-of-use electricity price policies during specific time periods to encourage users to charge during low periods to save costs. By analyzing the user's charging time, their preferences can be identified and the charging strategy can be adjusted accordingly to meet user needs while taking into account economy and environmental protection. In addition, frequent charging operations will accelerate the aging process of the battery and shorten its service life. Therefore, understanding the user's charging frequency helps to predict the battery's decay rate and take corresponding maintenance measures. For example, if it is found that a user's charging frequency is significantly higher than the average level, the user can be advised to adopt a more reasonable charging method, such as reducing unnecessary partial charging times to avoid the battery being in a high charge state for a long time, thereby slowing down the battery aging rate. In addition, the charging time is directly related to charging efficiency and safety. Generally speaking, although fast charging can replenish a large amount of energy for the device in a short time, this high-intensity charging process will cause great pressure on the battery, increase the risk of overheating, and may even cause safety accidents. On the contrary, although slow charging takes a long time, it puts less pressure on the battery and is conducive to extending the battery life. By recording and analyzing the user's charging time, it can help to determine whether the current charging rate is appropriate, and dynamically adjust the charging parameters according to the actual situation to ensure that it can meet the needs of the user and ensure the safe operation of the battery. In addition, the choice of charging mode also has a profound impact on battery performance. At present, the common charging modes on the market mainly include constant current (CC) and constant voltage (CV) in two basic forms, in addition to fast charging, slow charging and other variants. Under different modes, the electrochemical reaction mechanism inside the battery is different, which in turn affects the battery's charging efficiency, heat generation and aging rate. By collecting the user's charging mode preferences, it is possible to better match the charging plan that best suits the user. For example, fast charging services are provided when the user is in urgent need of using the device, while in other cases, a more gentle slow charging method is recommended to achieve the best overall benefit. Finally, the charging target power refers to the target capacity percentage that the user wants to charge the battery to. This value is usually closely related to the user's specific application scenario. For example, long-distance travelers may want to charge the battery to 100% to ensure that they will not run out of power on the way; while daily commuters may only need to keep it at a lower level to meet their needs for the day. By monitoring the charging target power, the system can more accurately grasp the user's actual needs and optimize the charging process accordingly.For example, when the target power set by the user is low, the charging rate can be appropriately reduced to reduce the pressure on the battery; and when full charging is required, the charging speed can be increased as much as possible within a safe range. It is worth noting that all the above types of user charging behavior data have certain randomness and individual differences. This means that even the same device may show completely different usage patterns in the hands of different users. Therefore, it is obviously not enough to manage the battery by relying solely on preset fixed parameters. Only by fully considering and utilizing these personalized charging behavior data can we truly realize customized services for each user, maximize the user experience, and ensure the safety and durability of the battery. Specifically, for parameters such as charging time, charging frequency, and charging duration, the charging and discharging status of the battery can be monitored in real time through built-in current sensors and voltage sensors; and the choice of charging mode often depends on the user's settings or operation records on the device; as for the charging target power, it is usually set by the user or determined based on the system's default values. All of this information needs to be accurately recorded and transmitted to the corresponding processing unit for further analysis.

[0021] Figure 3 The flowchart is a method for performing temporal fusion encoding of the battery status data according to the time dimension in the adaptive charging power supply control method according to the embodiment of the present application to obtain a multi-dimensional fusion representation of the battery status. Figure 3 As shown, in an embodiment of the present application, the step S130, performs time series fusion encoding on the battery status data according to the time dimension to obtain a multi-dimensional fusion representation of the battery status, including: S131, after integrating the battery status data into a voltage data time series vector and a temperature data time series vector according to the time dimension, the voltage data time series vector and the temperature data time series vector are respectively passed through a time encoder based on a time convolutional network to obtain a voltage data time series encoding vector and a temperature data time series encoding vector; S132, feature-level fusion of the voltage data time series encoding vector and the temperature data time series encoding vector to obtain a battery status multi-dimensional fusion representation vector as the battery status multi-dimensional fusion representation.

[0022] Specifically, in step S131, after the battery status data is integrated into a voltage data time series vector and a temperature data time series vector according to the time dimension, the voltage data time series vector and the temperature data time series vector are respectively passed through a time series encoder based on a time convolution network to obtain a voltage data time series encoding vector and a temperature data time series encoding vector. It should be understood that, considering that the traditional charging strategy relies on a static parameter model, it is difficult to capture the dynamic time series characteristics of the battery voltage / temperature during the charging and discharging process, and the degradation phenomena such as the internal polarization effect and capacity decay of the battery are often reflected in the voltage / temperature time series signal through high-frequency nonlinear fluctuations. The existing time series modeling and feature extraction methods are limited in their feature extraction capabilities for long sequence data due to the gradient attenuation problem, and it is difficult to effectively analyze the multi-scale correlation features in the transient response of the battery. To this end, in the technical solution of the present application, after the battery status data is integrated into a voltage data time series vector and a temperature data time series vector according to the time dimension, the voltage data time series vector and the temperature data time series vector are respectively passed through a time series encoder based on a time convolution network to obtain a voltage data time series encoding vector and a temperature data time series encoding vector. It is worth mentioning that through the causal dilated convolution structure of the temporal convolutional network, local features of different time scales are extracted layer by layer, and its parallel computing advantage is used to overcome the bottleneck of time series modeling of traditional recurrent neural networks. This step aims to establish a high-precision representation of the dynamic evolution of the battery state, and capture short-term mutations (such as voltage jumps at the moment of charging) and long-term trends (such as temperature rise drift caused by cycle aging) in the voltage / temperature sequence through the dilated convolution kernel of the temporal convolutional network, thereby revealing the implicit correlation between the real-time health status of the battery and the external working conditions. In this way, it helps to improve the expressive ability of battery state characteristics, so that subsequent fusion and decision modules can dynamically optimize charging parameters based on finer-grained time series information, which not only avoids the control deviation caused by overfitting static models, but also provides a physically interpretable feature basis for cross-modal interactive analysis, and finally realizes the synchronous adaptation of charging strategies to battery degradation dynamics and user scenario requirements.

[0023] Specifically, in step S132, the voltage data time series encoding vector and the temperature data time series encoding vector are feature-level fused to obtain a multi-dimensional fusion representation vector of the battery state as the multi-dimensional fusion representation of the battery state. It should be understood that, considering that voltage changes directly reflect the dynamic balance of electrochemical reactions inside the battery (such as lithium ion insertion / extraction rate), and temperature fluctuations are closely related to ohmic heat, polarization heat and environmental heat exchange, the two together constitute a multi-dimensional observation window for the battery health state. However, traditional methods often cause potential correlations between features (such as voltage drops caused by increased polarization at high temperatures) to be severed when processing the time series correlation of voltage and temperature data independently, making it difficult to construct a global state representation. To this end, in the technical solution of the present application, the voltage data time series encoding vector and the temperature data time series encoding vector are feature-level fused to obtain a multi-dimensional fusion representation vector of the battery state. The time-series encoding vectors of voltage and temperature are jointly modeled in latent space through feature-level fusion, aiming to break through the limitations of single signal analysis and use the complementarity of voltage and temperature to reveal the coupling relationship between battery degradation mechanisms (such as the temperature rise hysteresis effect associated with SEI film thickening) and external working conditions (such as the coordinated temperature-pressure oscillation caused by fast charging). This step nonlinearly associates the transient response features in the voltage sequence (such as the voltage inflection point at the end of charging) with the thermal inertia features in the temperature sequence (such as the temperature accumulation caused by heat dissipation delay) through tensor splicing or attention weighting mechanisms, thereby constructing a fusion representation covering multiple dimensions of electricity, heat, and aging. This design effectively enhances the model's ability to identify complex implicit failure modes (such as voltage anomalies caused by local overheating), enabling subsequent cross-modal interactive analysis to generate charging strategies based on more complete battery status information, avoiding control risks caused by misjudgment of a single signal (such as relying solely on voltage may ignore potential thermal runaway hazards), and accurately balancing the charging rate and temperature rise constraints through electrothermal coupling analysis, ultimately achieving an upgrade in battery health management from "single-dimensional threshold control" to "multi-physical field collaborative optimization."

[0024] In an embodiment of the present application, the step S140, mapping and associating the user charging behavior data based on charging information to obtain a user charging behavior associating coding representation, includes: passing the user charging behavior data through a charging information mapping and associating encoder based on a fully connected layer to obtain a user charging behavior associating coding vector as the user charging behavior associating coding representation. It should be understood that since the user charging behavior data (such as charging period preference, target power setting) contains personalized needs and scenario constraint information, the traditional charging strategy lacks semantic analysis of user behavior and is difficult to convert discrete charging habits (such as frequent fast charging, low power maintenance at night) into quantifiable control parameters. Although existing solutions attempt to introduce user preferences, they often directly use raw data splicing or simple rule mapping, resulting in the implicit association between user intention and battery status not being deeply mined (for example, the 80% target power set by the user may correspond to different health protection priorities, depending on its charging frequency and usage scenario). To this end, in the technical solution of the present application, the user charging behavior data is passed through a charging information mapping and associating encoder based on a fully connected layer to obtain a user charging behavior associating coding vector. The charging information mapping association encoder is constructed through the fully connected layer to map the multi-dimensional and heterogeneous user behavior data (including time series attributes such as charging time and statistical features such as charging frequency) to a unified semantic space, aiming to deconstruct the potential decision logic behind user behavior (such as the "emergency fast charging" mode implies a strong demand for charging speed, while the "low temperature protection" mode gives priority to temperature rise constraints). This step captures the complex interactions between user behavior characteristics through nonlinear transformations (such as high-frequency short-time charging is often negatively correlated with high target power requirements), and generates a low-dimensional encoding vector with semantic coherence, so that user needs can be embedded in the subsequent cross-modal interaction process in a parsable form. In actual applications, this design breaks through the rigid mapping limitations of traditional rule engines, allowing the charging strategy to not only recognize explicit instructions (such as charging mode selection) but also infer implicit needs (such as predicting users' future usage scenarios based on historical charging period distribution), and ultimately achieve dynamic alignment of battery health management strategies with user behavior habits - for example, when it is recognized that the user has been using the "night slow charging" mode for a long time, the charging curve is automatically optimized to use the valley power period to extend the trickle charging stage, thereby satisfying the user's scenario preferences while minimizing the battery capacity decay.

[0025] Figure 4 A flowchart of performing battery state-charging behavior cross-modal feature independent restriction interactive analysis on the multi-dimensional fusion representation of the battery state and the associated coding representation of the user charging behavior in the adaptive charging power control method according to an embodiment of the present application to obtain a battery state-charging behavior interactive coding representation. Figure 4As shown, in an embodiment of the present application, the step S150 performs battery state-charging behavior cross-modal feature independence restriction interactive analysis on the multidimensional fusion representation of the battery state and the user charging behavior associated coding representation to obtain a battery state-charging behavior interactive coding representation, including: S151, transforming the multidimensional fusion representation vector of the battery state to obtain a set of battery state principal component linear transformation feature coding vectors; S152, performing inter-modal independence constraint analysis on the set of battery state principal component linear transformation feature coding vectors and the user charging behavior associated coding vector to determine a set of battery state-charging behavior principal component inter-modal independence soft constraint factors; S153, based on the set of battery state-charging behavior principal component inter-modal independence soft constraint factors, dynamically and adaptively aggregate the set of battery state principal component linear transformation feature coding vectors and the user charging behavior associated coding vector to obtain a battery state-charging behavior interactive coding vector as the battery state-charging behavior interactive coding representation. It should be understandable that the multi-dimensional fusion representation vector of the battery state and the user charging behavior associated coding vector respectively carry two types of heterogeneous modal information: physical characteristics and semantic needs: the former characterizes the dynamic health status of the battery through electro-thermal coupling analysis, and the latter reveals the user's scenario-based charging intention through behavior mapping. However, traditional cross-modal interaction methods often simply splice or weightedly fuse, ignoring the essential differences between the two types of data in information density (high-frequency sampling of battery time series data and sparse event records of user behavior) and semantic levels (electrochemical degradation mechanisms and user subjective preferences), resulting in redundant features interfering with effective information transmission during the interaction process (such as the charging mode set by the user may mask the actual temperature rise pressure that the battery is currently under). Therefore, in the technical solution of the present application, the multi-dimensional fusion representation vector of the battery state and the user charging behavior associated coding vector are further subjected to feature interaction analysis of independent constraint representation between modalities to obtain the battery state-charging behavior interaction coding vector. Specifically, the process of feature interaction analysis of independent constraint representation between modalities is to decouple the multi-dimensional fusion representation vector of battery state through principal component analysis (PCA), remove minor variation components (such as random noise) and extract core electro-thermal coupling principal components (such as the correlation pattern between voltage inflection point and temperature rise rate), generate low-dimensional orthogonal feature subspace, and eliminate internal redundancy; then align the numerical scale of battery state principal components and user behavior encoding through linear transformation (for example, normalize the voltage fluctuation amplitude to a range close to the user's charging frequency) to avoid imbalance of modal contribution due to magnitude difference. In other words, by constructing an interactive feature space of battery state and user behavior, the independence relationship between modalities is explicitly modeled at the same time. The complementarity and independence boundaries of the two types of modalities in the long-range correlation global structure are captured through the intrinsic normative potential representation of the topological order matrix (for example, the dynamic balance relationship between the user's fast charging demand and the physical constraint of the maximum allowable charging current of the battery).The dynamic calculation mechanism of the soft constraint factor further realizes the adaptive adjustment of the fusion weight: when the battery health status is highly conflicting with the user's needs (such as the user starts fast charging in a high temperature environment), the independence coding matrix suppresses high-risk interaction paths through stability optimization (reduces the contribution weight of the fast charging mode to the voltage boost instruction), while strengthening the effective feature association within the safety boundary (improving charging efficiency within the allowable temperature rise range). In practical effect, this design breaks through the rigid limitations of traditional hard constraint fusion and enables the charging strategy to have multi-objective dynamic trade-off capabilities. For example, when the user selects the "low temperature protection" mode, the temperature rise rate feature in the main component of the battery status enhances its influence on the decision through independent soft constraints, dynamically suppressing the charging current fluctuations that may cause lithium precipitation in a low temperature environment; in the "emergency fast charging" scenario, the semantic weight of the user behavior encoding is increased, driving the decoder to maximize the charging power within the battery safety threshold. The final generated battery state-charging behavior interaction encoding vector explores the potential optimization space through fine-grained response interaction (such as identifying conditions for extending the trickle phase during night charging to utilize natural heat dissipation), which not only ensures battery health but also accurately adapts to user scenario needs, realizing a paradigm upgrade from "single-modal dominance" to "multi-modal game optimization".

[0026] In an embodiment of the present application, the step S151, transforming the battery status multidimensional fusion representation vector to obtain a set of battery status principal component linear transformation feature coding vectors, includes: S1511, performing principal component analysis on the battery status multidimensional fusion representation vector to obtain a set of battery status principal component feature coding vectors; S1512, linearly transforming each battery status principal component feature coding vector in the set of battery status principal component feature coding vectors to obtain a set of battery status principal component linear transformation feature coding vectors, wherein each battery status principal component linear transformation feature coding vector in the set of battery status principal component linear transformation feature coding vectors has the same characteristic scale as the user charging behavior associated coding vector.

[0027] Specifically, in step S1511, principal component analysis is performed on the multi-dimensional fusion representation vector of the battery state to obtain a set of principal component feature encoding vectors of the battery state, which is expressed as follows using the principal component analysis formula of the battery state: in, is the multi-dimensional fusion representation vector of the battery state, For principal component analysis, is the covariance matrix of the multi-dimensional fusion representation vector of the battery state, is the orthogonal matrix of the multidimensional characteristic principal components of the battery state, , and are the first, second and third principal component feature encoding vectors of the battery status. The multi-dimensional feature principal component encoding vector of the battery state, is the multi-dimensional feature diagonal matrix of battery status, , and They are , and The corresponding eigenvalues ​​are for The transposed matrix of The number of eigenvalues ​​in the multi-dimensional fusion representation vector of the battery state, Represents the extraction of diagonal elements. It should be understood that the multidimensional fusion representation vector of the battery state is a high-dimensional feature set of voltage and temperature data generated by time series fusion encoding, which naturally has information redundancy and multicollinearity problems. Specifically, the physical parameters such as voltage and temperature will form a complex nonlinear coupling relationship during the dynamic charging and discharging process, resulting in a large number of inefficient or repeated correlation dimensions in the original feature space. This high-dimensional redundant feature not only increases the computational complexity, but also may mask the key variation characteristics of the true degradation mode of the battery, thereby affecting the accuracy of subsequent cross-modal interaction analysis. Principal component analysis can mathematically remove secondary noise interference by means of orthogonal transformation, and project the multidimensional fusion representation vector of the battery state to the principal component space composed of orthogonal basis vectors, thereby extracting a low-dimensional feature set that characterizes the core dynamic characteristics of the battery state, providing a de-redundant, high-information-density input basis for subsequent cross-modal interaction. In this way, a compact representation form of battery status features is constructed, and the dimension reduction reconstruction and information purification of the feature space are achieved through principal component analysis to eliminate the linear correlation between multi-dimensional time series features such as voltage and temperature, and map the original high-dimensional fusion vector to a set of orthogonal principal component axes, so that a few head principal components can focus on reflecting the main variation modes in the dynamic characteristics of the battery (such as polarization effect, temperature rise rate change, etc.). This process not only optimizes the dimensional structure of the feature space, but more importantly, it selects the key principal components that are sensitive to the battery health state through the principle of maximizing information entropy, so that the subsequent cross-modal interaction mechanism can focus on the core variation direction of the battery physical characteristics, avoiding the interference of invalid features on the semantic analysis of user behavior. After implementing principal component analysis, the multi-dimensional fusion representation vector of the battery state is converted into a set of low-dimensional orthogonal principal component feature encoding vectors. This operation significantly reduces the feature dimension, alleviates the "dimensionality disaster" problem, and improves feature independence by eliminating multicollinearity. The set of battery state principal component feature encoding vectors can more efficiently characterize the core modes of dynamic behaviors such as battery voltage fluctuations and temperature drift by retaining the components with the largest variance contribution in the original data.

[0028] Specifically, in step S1512, each battery state principal component feature coding vector in the set of battery state principal component feature coding vectors is linearly transformed to obtain a set of battery state principal component linear transformation feature coding vectors, wherein each battery state principal component linear transformation feature coding vector in the set of battery state principal component linear transformation feature coding vectors has the same characteristic scale as the user charging behavior associated coding vector, and is expressed as a battery state linear transformation formula: in, Express The principal component feature encoding vectors of each battery state in are linearly transformed. and are the first, second, and third in the set of linear transformation feature encoding vectors of the battery status principal component. and The battery state principal component linear transformation feature encoding vector, is the set of linear transformation feature encoding vectors of the battery state principal component. It should be understood that the battery state principal component feature encoding vector and the user charging behavior associated encoding vector are derived from two types of heterogeneous modal data, physical signals and user behavior, respectively, and there are significant differences in the scale of their original feature distribution. The main component features of the battery (such as the voltage fluctuation principal component and the temperature rise trend principal component) are usually generated by dimensionality reduction of time series data with high sampling rate, and the numerical range is affected by the physical dimension of the electrochemical parameters; while the user behavior encoding vector (such as charging frequency and target power) is generated based on the statistics of discrete behavioral events, and the numerical distribution is constrained by semantic mapping rules. If cross-modal interaction is performed directly, the scale difference will cause the model to be overly sensitive to high-level modalities (such as the voltage amplitude principal component), while the contribution of low-level modalities (such as charging mode encoding) will be suppressed, causing information bias and decision bias in the feature interaction process. Through scale normalization and spatial mapping, the dimensional barrier between the main component features of the battery and the user behavior encoding can be eliminated, and the fairness foundation of cross-modal interaction can be constructed. Specifically, by applying a linear transformation matrix to the battery principal component feature encoding vector, it is projected to the feature scale space that matches the user behavior encoding vector, so that the numerical range and distribution variance of the two types of modes tend to be consistent. This operation not only achieves alignment at the numerical level, but also preliminarily establishes an association mapping channel between battery physical characteristics and user behavior semantics through geometric structure adjustment of the feature space, providing compatibility guarantee for the collaborative fusion of multi-dimensional features in subsequent cross-modal interaction analysis, and avoiding the problem of modal contribution imbalance caused by scale differences. After linear transformation, the set of battery principal component feature encoding vectors and user behavior association encoding vectors are strictly aligned in the feature scale. The numerical range of the battery principal component features (such as low-order principal components reflecting polarization effects and high-order principal components representing temperature rise inertia) is compressed or expanded to the same order of magnitude as the user behavior encoding (such as charging period preference encoding and target power quantization value), ensuring that the weight distribution of the two types of features in the cross-modal interaction process is not disturbed by the original dimension. The scaled feature space enables the interactive model to fairly evaluate the game relationship between battery health constraints and user scenario requirements. For example, when the fast charging mode is activated, the main component characteristics of the battery temperature rise and the user charging time encoding can dynamically balance the charging rate and thermal runaway risk at the same scale, thereby improving the robustness and interpretability of the charging strategy in multi-objective optimization scenarios.

[0029] In an embodiment of the present application, the step S152, performing an inter-modal independence constraint analysis on the set of the battery state principal component linear transformation feature coding vectors and the user charging behavior associated coding vectors to determine a set of battery state-charging behavior principal component inter-modal independence soft constraint factors, includes: S1521, inputting each battery state principal component linear transformation feature coding vector in the set of the user charging behavior associated coding vector and the battery state principal component linear transformation feature coding vector into an inter-modal independence modeling unit to obtain a set of battery state-charging behavior principal component inter-modal independence coding matrices; S1522, Each battery state-charging behavior principal component modal independence coding matrix in the set of behavior principal component modal independence coding matrices is subjected to canonical stability optimization to obtain a set of battery state-charging behavior principal component modal independence canonical optimization coding matrices, wherein the canonical stability optimization process is determined by a battery state-charging behavior eigenvector composed of each eigenvalue of the battery state-charging behavior principal component modal independence coding matrix; S1523, based on the set of battery state-charging behavior principal component modal independence canonical optimization coding matrices, calculate the set of battery state-charging behavior principal component modal independence soft constraint factors.

[0030] Specifically, in step S1521, each battery state principal component linear transformation feature coding vector in the set of the user charging behavior associated coding vector and the battery state principal component linear transformation feature coding vector is input into the inter-modal independence modeling unit to obtain a set of battery state-charging behavior principal component inter-modal independence coding matrices, which is expressed as the inter-modal independence modeling formula: in, and is a feature mapping function, such as a linear mapping or a nonlinear kernel function, Associating a coding vector with the user's charging behavior, For the The length of the linear transformation feature encoding vector of the battery state principal component, is the first in the set of independence encoding matrices between the main component modes of battery state-charging behavior The main component of battery state-charging behavior modal independence coding matrix. It should be understood that, considering that the linear transformation feature coding vector of the main component of battery state (characterizing electrochemical dynamic characteristics) and the user charging behavior association coding vector (reflecting user scenario requirements) have been scale-aligned, the two are essentially physical modal and behavioral modal, and their deep correlation is highly nonlinear and asymmetric. Traditional cross-modal fusion methods rely on implicit learning of modal relationships and are easily disturbed by redundant features (such as the unnecessary correlation between the battery temperature rise trend and the user's charging time preference), resulting in the key complementary information being masked by noise during the interaction process. By explicitly modeling modal independence, the inherent properties and interactive dependencies of the two types of modalities can be separated, avoiding decision-making bias caused by information redundancy during feature coupling (for example, misjudging the user's fast charging demand as the battery's high-rate tolerance). In this way, a measurable modal independence constraint space is constructed, and the complementary boundary between the battery's physical state and user behavior requirements is dynamically parsed through a learnable coding matrix. Specifically, the inter-modal independence modeling unit does not simply calculate the statistical independence index, but maps the interaction between the battery principal component characteristics (such as the voltage fluctuation principal component, the temperature rise rate principal component) and the user behavior encoding (such as charging mode preference, target power setting) to a low-dimensional orthogonal subspace by parameterizing the generation of the inter-modal independence encoding matrix of the battery state-charging behavior principal component. This explicit modeling mechanism can quantify the dynamic balance between inter-modal independence and correlation (for example, identifying the positive synergistic effect of user fast charging demand and battery polarization voltage recovery ability), provide directional guidance for subsequent cross-modal aggregation, and ensure that the interaction process focuses on feature combinations with physical meaning and scene adaptability. After inter-modal independence modeling, the set of inter-modal independence encoding matrices of the battery state-charging behavior principal component is output, which is essentially a dynamic weight map describing the interaction intensity of the two types of modal features. Each battery state-charging behavior principal component modal independence coding matrix corresponds to the independence quantification result of a specific principal component and user behavior coding (such as the strength of the negative correlation between a temperature rise principal component and the user's charging frequency), and interactive stability control is achieved by constraining the matrix spectral radius or eigenvalue distribution.

[0031] Specifically, in step S1522, each battery state-charging behavior principal component modal independence coding matrix in the set of the battery state-charging behavior principal component modal independence coding matrix is ​​subjected to canonical stability optimization to obtain a set of battery state-charging behavior principal component modal independence canonical optimization coding matrices, wherein the canonical stability optimization process is determined by the battery state-charging behavior eigenvector composed of each eigenvalue of the battery state-charging behavior principal component modal independence coding matrix, and is expressed as a canonical stability optimization formula: in, For the said Each eigenvalue of the inter-modal independence encoding matrix of the battery state-charging behavior principal component, is the battery state-charging behavior eigenvector composed of the above eigenvalues, is matrix multiplication, is the battery state-charging behavior canonical association encoding vector, The first set of encoding matrices in the set of battery state-charging behavior principal component modal independence norm optimization The battery state-charging behavior main component modal independence norm optimization coding matrix. It should be understood that the battery state-charging behavior main component modal independence coding matrix, as a topological order matrix, has a potential risk of instability in the modal interaction structure due to its asymmetry and long-range correlation characteristics. The eigenvalue distribution of the original battery state-charging behavior main component modal independence coding matrix may cause the eigenvector phase accumulation offset due to local disturbances (such as instantaneous fluctuations in battery voltage or sudden changes in user charging mode), destroying the global topological consistency of modal interaction. For example, the strong asymmetric coupling between fast charging demand and temperature rise principal component may cause the battery state-charging behavior main component modal independence coding matrix to have eigenvalue divergence in dynamic scenarios, resulting in violent oscillation of cross-modal feature aggregation weights, affecting the stability and continuity of the charging strategy. This step reconstructs the geometric structure of the battery state-charging behavior main component modal independence coding matrix to enhance its anti-interference ability by introducing topological invariant constraints based on the eigennormal potential. Specifically, the battery state-charging behavior eigenvector composed of the eigenvalues ​​is used as the mathematical basis of the canonical potential to map the global topological association of the asymmetric matrix to the canonical association coding vector space. The phase accumulation is calibrated by the first-order canonical potential term to eliminate the interference of the high-order perturbation term on the long-range correlation structure, and the structural stability optimization of the battery state-charging behavior main component modal independence coding matrix under asymmetric conditions is achieved. In this way, it is ensured that local characteristic disturbances (such as temporary adjustment of the charging target by the user) during the modal interaction process will not destroy the global balance relationship between the battery health constraint and the user's needs, and maintain the smooth transition of the cross-modal fusion weights. After the canonical stability optimization, the set of the battery state-charging behavior main component modal independence canonical optimization coding matrices obtained constrains the spectral distribution of its autocorrelation matrix through the eigennormal potential, so that the eigenvalue modulus and phase angle of the matrix meet the stability boundary conditions (such as limiting the real part of the eigenvalue to be negative to ensure the convergence of the dynamic system). This structural stability ensures the robustness of the cross-modal interaction mechanism, enabling the charging strategy to maintain a dynamic balance between battery health and user needs under complex working conditions, and avoiding control command oscillations or strategy failures caused by local data fluctuations.

[0032] Specifically, the step S1523, based on the set of battery state-charging behavior main component modal independence specification optimization coding matrices, calculates the set of battery state-charging behavior main component modal independence soft constraint factors, and the battery state-charging behavior main component modal independence soft constraint factor calculation formula is expressed as: in, represents the square of the norm of the matrix F, Represents the set of soft constraint factors for the independence of the main components of the battery state-charging behavior. It should be understood that although the normative optimization coding matrix of the independence of the main components of the battery state-charging behavior after normative optimization ensures the structural stability of the modal interaction through topological constraints, its essence is still a static mathematical representation, which is difficult to directly adapt to the real-time changes of the dynamic characteristics of the battery and the user's behavior. For example, when the battery health state suddenly changes (such as a sharp increase in the temperature rise rate) or the user temporarily switches the charging mode (such as from conventional charging to fast charging), the modal independence constraint with fixed weights may cause decision lag or rigidity, and cannot accurately capture the impact of the scene dynamics on the interaction intensity. Therefore, it is necessary to convert the modal independence information contained in the normative optimization coding matrix of the independence of the main components of the battery state-charging behavior after normative optimization into dynamically adjustable constraint parameters to cope with the nonlinear evolution of multi-modal game relations under complex working conditions. Specifically, through the quantitative extraction of the soft constraint factors of the independence of the main components of the battery state-charging behavior, an adaptive mapping mechanism between modal independence constraints and dynamic feature interactions can be established. Among them, based on the intrinsic normative potential distribution of the normative optimization coding matrix of the independence between the main component modes of the battery state-charging behavior after normative optimization (such as the stability margin reflected by the real part of the eigenvalue and the phase accumulation rate corresponding to the imaginary part), a differentiable soft constraint factor calculation function of the independence between the main component modes of the battery state-charging behavior is designed to convert the topological invariance characteristics of the normative optimization coding matrix of the independence between the main component modes of the battery state-charging behavior (such as long-range correlation strength and local disturbance suppression ability) into dynamic weight coefficients that can act on the feature aggregation process. Its core goal is to interpret the static topological structure of the normative optimization coding matrix of the independence between the main component modes of the battery state-charging behavior into flexible constraint rules, so that cross-modal interaction can dynamically balance the game relationship between independence maintenance and correlation utilization according to real-time scenario requirements (such as battery polarization degree and user charging urgency), and avoid suboptimal decision-making problems caused by hard threshold segmentation. After the battery state-charging behavior main component mode independence soft constraint factor, each standard optimized battery state-charging behavior main component mode independence standard optimization coding matrix is ​​mapped to the dynamic constraint strength parameter between the corresponding principal component and the user behavior mode. For example, when the independence of a battery temperature rise principal component and the user's fast charging demand coding is high (manifested as a negative shift in the real part of the encoding matrix eigenvalue), the battery state-charging behavior main component mode independence soft constraint factor automatically enhances the independence weight of the principal component, suppressing the excessive demand for charging current in the fast charging mode; on the contrary, when the independence of the two decreases (eigenvalue approaches neutrality), the battery state-charging behavior main component mode independence soft constraint factor moderately relaxes the constraint, allowing user demand to dominate the charging strategy within a safe threshold.This dynamic adjustment mechanism enables the charging control system to flexibly adjust the decision boundary of multimodal fusion according to the coupling strength between the real-time degradation state of the battery and the user's scenario intention. It not only prevents the control risk caused by overfitting a single mode (such as blindly following user demand and causing overcharging), but also makes full use of modal complementarity to improve charging efficiency and scenario adaptability, and ultimately achieves adaptive convergence of the global optimization goal.

[0033] In an embodiment of the present application, the step S153, based on the set of soft constraint factors for independence between the battery state-charging behavior principal component modes, dynamically and adaptively aggregates the set of battery state principal component linear transformation feature coding vectors and the user charging behavior associated coding vector to obtain a battery state-charging behavior interaction coding vector as the battery state-charging behavior interaction coding representation, including: S1531, inputting each battery state principal component linear transformation feature coding vector in the set of the user charging behavior associated coding vector and the battery state principal component linear transformation feature coding vector into a feature interaction response unit to obtain a set of fine-grained response interaction coding vectors between the battery state-charging behavior principal component modes; S1532, based on the set of soft constraint factors for independence between the battery state-charging behavior principal component modes, dynamically and adaptively aggregates the set of fine-grained response interaction coding vectors between the battery state-charging behavior principal component modes to obtain the battery state-charging behavior interaction coding vector.

[0034] Specifically, in step S1531, each battery state principal component linear transformation feature coding vector in the set of the user charging behavior associated coding vector and the battery state principal component linear transformation feature coding vector is input into a feature interaction response unit to obtain a set of fine-grained response interaction coding vectors between the battery state and the charging behavior principal component modalities, which is expressed as a feature interaction response formula: in, Indicates cascade processing, , and represents positional dot multiplication, positional summation, and positional dot division. and represents the trainable weight matrix and the trainable bias vector, The first in the set of fine-grained response interaction encoding vectors representing the battery state-charging behavior principal component modalities The fine-grained response interaction encoding vector between the main component modes of battery state-charging behavior. It should be understood that traditional coarse-grained fusion methods (such as weighted summation or simple splicing) are difficult to capture the implicit coupling mechanism between the main components of the battery (such as polarization voltage recovery rate, temperature rise trend) and user behavior (such as charging period preference, target power setting), such as the dynamic game relationship between the transient response of the specific main component of the battery when the fast charging mode is activated and the user's charging time requirement. Direct shallow interaction is prone to the loss of key complementary information, and it is impossible to accurately quantify the differential impact weight of user scenario requirements on charging strategies. The feature interaction response unit realizes the fine-grained semantic association analysis of cross-modal features, and constructs a deep interaction network of battery physical characteristics and user behavior intentions from a microscopic dimension. Specifically, the feature interaction response unit mines the dependency and complementarity of the main component features of the battery (such as the temporal pattern of the main component of voltage fluctuation) and the user behavior encoding (such as the statistical characteristics of charging frequency) in the subspace layer by layer through nonlinear mapping functions (such as cross-attention mechanism, tensor product operation), and generates a fine-grained response interaction encoding vector between the main component modes of battery state-charging behavior. In this way, interpreting user scenario requirements (such as emergency fast charging) into dynamic constraints of battery principal components (such as the maximum allowable charging current slope) can establish interpretable feature-level interaction rules, providing a decision-making basis for adaptive charging strategies that is both physically reasonable and scenario-adaptive. After being processed by the feature interaction response unit, the output set of fine-grained response interaction coding vectors between the battery state-charging behavior principal component modalities can accurately characterize the dynamic correlation strength between user behavior and battery principal components. For example, when the user selects the "low temperature protection" mode, the fine-grained response interaction coding vector between the battery state-charging behavior principal component modalities can strengthen the inhibitory effect of the temperature rise principal component on the charging power and weaken the voltage principal component weight related to the charging speed; in the "efficient energy replenishment" scenario, the synergistic effect of the battery's high rate tolerance principal component and the user's target power is preferentially activated. This fine-grained interaction mechanism enables the charging strategy to dynamically allocate the game priority between battery health constraints and user needs in the feature dimension according to real-time operating conditions. It not only avoids the strategy rigidity caused by global hard rules, but also explores potential optimization space through local feature interaction (such as using the battery polarization relaxation characteristics to moderately increase the trickle stage voltage), ultimately achieving an adaptive balance between charging efficiency, battery life and scenario requirements.

[0035] Specifically, in step S1532, based on the set of independence soft constraint factors between the battery state-charging behavior principal component modes, the set of fine-grained response interaction coding vectors between the battery state-charging behavior principal component modes is dynamically adaptively aggregated to obtain the battery state-charging behavior interaction coding vector, which is expressed as a dynamic adaptive aggregation formula: in, for function, is the number of vectors in the set of fine-grained response interaction encoding vectors between the battery state-charging behavior principal component modes, It is the battery state-charging behavior interaction coding vector. It should be understood that by driving the dynamic aggregation mechanism through the soft constraint factor of independence between the main component modes of battery state-charging behavior, the selective fusion of fine-grained interaction features and the optimization integration of global information can be achieved. Specifically, based on the set of soft constraint factors of independence between the main component modes of battery state-charging behavior (reflecting the strength of independence between each main component and user behavior encoding), the fine-grained response interaction coding vector between the main component modes of battery state-charging behavior is weightedly aggregated, and the contribution weights of different interaction paths are dynamically adjusted to retain cross-modal complementary information while eliminating redundant noise through soft constraints, thereby generating a battery state-charging behavior interaction coding vector with both high information density and low redundancy, providing the decoder with an accurate basis for optimizing the charging strategy.

[0036] In an embodiment of the present application, the step S160, based on the battery state-charging behavior interaction coding representation, determines the target charging voltage recommended decoding value to control the charging of the charged power source, including: S161, passing the battery state-charging behavior interaction coding vector through a decoder-based charging voltage controller to obtain a target charging voltage recommended decoding value; S162, based on the target charging voltage recommended decoding value, controlling the charged power source to charge.

[0037] Specifically, in step S161, the battery state-charging behavior interaction coding vector is passed through a decoder-based charging voltage controller to obtain a target charging voltage recommended decoding value. It should be understood that since the battery state-charging behavior interaction coding vector is an abstract representation of multimodal deep fusion, although it contains a dynamic game relationship between battery health constraints and user scenario requirements, its high-dimensional nonlinear characteristics cannot be directly mapped into executable charging voltage control instructions. Traditional control strategies rely on threshold triggering or linear regression, and it is difficult to parse the multi-objective optimization logic implied in such complex interactive features (such as the compromise relationship between fast charging requirements and temperature rise constraints). To this end, in the technical solution of the present application, the battery state-charging behavior interaction coding vector is further passed through a decoder-based charging voltage controller to obtain a target charging voltage recommended decoding value. Specifically, the decoder-based charging voltage controller interprets the compressed complementary information (such as the correlation pattern between the user's fast charging intention and the battery polarization voltage recovery rate) in the battery state-charging behavior interaction coding vector into a continuously adjustable charging voltage trajectory through a multi-layer deconvolution or fully connected network, thereby obtaining a target charging voltage recommended decoding value.

[0038] Specifically, in step S162, based on the target charging voltage recommended decoded value, the charged power source is controlled to charge. It should be understood that after the target charging voltage recommended decoded value is determined, the charged power source can be precisely controlled accordingly. This means that during the entire charging cycle, the charging parameters will be dynamically adjusted according to the target charging voltage recommended decoded value to adapt to changes in the battery state. For example, if it is detected that the battery temperature rises too fast, the system may temporarily reduce the charging current and gradually restore it after the temperature stabilizes; if it is found that the battery voltage is close to the full charge threshold, it will automatically switch to the trickle charging mode to prevent damage caused by overcharging. This real-time response mechanism ensures that the most suitable charging conditions can be obtained regardless of the battery state. Compared with the traditional fixed parameter charging method, the advantage of this method lies in its flexibility and adaptability. It can respond to slight changes in the battery state in real time, adjust the charging strategy in time, and avoid problems caused by improper parameter settings. At the same time, since the charging voltage is dynamically generated according to the actual needs of the battery, it can minimize unnecessary energy loss and improve the overall charging efficiency. In addition, the charging control method based on the target charging voltage recommended decoded value helps to extend the service life of the battery. A reasonable charging strategy can not only avoid damage to the battery caused by overcharging or discharging, but also effectively alleviate various adverse phenomena during battery aging, such as lithium dendrite growth and thickening of the solid electrolyte interface film. By accurately controlling the charging voltage, the electrochemical balance inside the battery can be maintained to the greatest extent, slowing down the rate of battery performance degradation. This is especially important for applications that require long-term stable operation, such as electric vehicles and energy storage power stations.

[0039] In summary, the adaptive charging power control method based on the embodiment of the present application is explained, which uses a high-precision sensor to capture the voltage and temperature time series signals of the battery in real time, combines the user's charging behavior data, and uses a time convolution network to perform multi-scale time series modeling on the dynamic characteristics of the battery, and extracts the high-frequency nonlinear fluctuation characteristics of the battery state. At the same time, the user behavior data is semantically associated and encoded to establish a mapping relationship between user needs and battery health status. Next, a cross-modal feature interaction mechanism is introduced to realize the dynamic coupling analysis of battery status and user needs. Finally, the decoder generates a charging voltage control instruction that optimally matches the current battery state and meets the user scenario requirements, and dynamically adjusts the charging parameters to achieve coordinated optimization of charging efficiency, battery life, and personalized scenarios.

[0040] Figure 5 : is a system block diagram of an adaptive charging power supply control system according to an embodiment of the present application. Figure 5As shown, according to the adaptive charging power supply control system 100 of the embodiment of the present application, it includes: a battery status data real-time monitoring module 110, which is used to monitor the battery status data of the charged power supply in real time through a battery sensor, and the battery status data includes voltage data and temperature data; a user charging behavior data acquisition module 120, which is used to acquire user charging behavior data, and the user charging behavior data includes charging time, charging frequency, charging duration, charging mode and charging target power; a battery status time series fusion encoding module 130, which is used to perform time series fusion encoding on the battery status data according to the time dimension to obtain a multi-dimensional fusion representation of the battery status; a user The charging behavior association coding module 140 is used to perform mapping association coding on the user charging behavior data based on the charging information to obtain the user charging behavior association coding representation; the battery state-charging behavior interaction coding module 150 is used to perform battery state-charging behavior cross-modal feature independent restriction interaction analysis on the battery state multi-dimensional fusion representation and the user charging behavior association coding representation to obtain the battery state-charging behavior interaction coding representation; the target charging voltage recommended decoding value determination module 160 is used to determine the target charging voltage recommended decoding value based on the battery state-charging behavior interaction coding representation to control the charged power source to charge.

[0041] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned adaptive charging power supply control system have been referred to above. Figures 1 to 4 The adaptive charging power control method has been described in detail in the description of the adaptive charging power control method, and therefore, its repeated description will be omitted.

[0042] In summary, an adaptive charging power supply control system based on an embodiment of the present application is explained, which uses a high-precision sensor to capture the voltage and temperature timing signals of the battery in real time, combines the user's charging behavior data, and uses a time convolution network to perform multi-scale timing modeling on the dynamic characteristics of the battery, and extracts the high-frequency nonlinear fluctuation characteristics of the battery state. At the same time, the user behavior data is semantically associated and encoded to establish a mapping relationship between user needs and the battery health status. Next, a cross-modal feature interaction mechanism is introduced to achieve dynamic coupling analysis of battery status and user needs. Finally, a decoder is used to generate a charging voltage control instruction that optimally matches the current battery state and meets the user's scenario requirements, and the charging parameters are dynamically adjusted to achieve coordinated optimization of charging efficiency, battery life, and personalized scenarios.

Claims

1. An adaptive charging power supply control method, characterized in that: include: Monitor the battery status data of the charged power source in real time through a battery sensor, wherein the battery status data includes voltage data and temperature data; Acquire user charging behavior data, wherein the user charging behavior data includes charging time, charging frequency, charging duration, charging mode, and charging target power; Performing time-series fusion encoding on the battery status data according to the time dimension to obtain a multi-dimensional fusion representation of the battery status; Performing mapping association coding on the user charging behavior data based on the charging information to obtain an association coding representation of the user charging behavior; Performing battery state-charging behavior cross-modal feature independent restriction interactive analysis on the multi-dimensional fusion representation of the battery state and the associated coding representation of the user charging behavior to obtain a battery state-charging behavior interactive coding representation; Based on the battery state-charging behavior interactive coding representation, a target charging voltage recommended decoding value is determined to control the charged power source to charge.

2. The adaptive charging power supply control method according to claim 1, characterized in that: The battery status data is subjected to time-series fusion encoding according to the time dimension to obtain a multi-dimensional fusion representation of the battery status, including: After integrating the battery status data into a voltage data time series vector and a temperature data time series vector according to the time dimension, the voltage data time series vector and the temperature data time series vector are respectively passed through a time encoder based on a time convolutional network to obtain a voltage data time series encoding vector and a temperature data time series encoding vector; The voltage data time series encoding vector and the temperature data time series encoding vector are fused at feature level to obtain a battery state multi-dimensional fusion representation vector as the battery state multi-dimensional fusion representation.

3. The adaptive charging power supply control method according to claim 2, characterized in that: The user charging behavior data is subjected to mapping association coding based on charging information to obtain a user charging behavior association coding representation, including: passing the user charging behavior data through a charging information mapping association encoder based on a fully connected layer to obtain a user charging behavior association coding vector as the user charging behavior association coding representation.

4. The adaptive charging power supply control method according to claim 3, characterized in that: The multi-dimensional fusion representation of the battery state and the associated coding representation of the user charging behavior are subjected to a battery state-charging behavior cross-modal feature independent restriction interactive analysis to obtain a battery state-charging behavior interactive coding representation, including: Transforming the multi-dimensional fusion representation vector of the battery state to obtain a set of linear transformation feature encoding vectors of the main components of the battery state; Performing inter-modal independence constraint analysis on the set of linear transformation feature coding vectors of the battery state principal component and the user charging behavior associated coding vector to determine a set of battery state-charging behavior principal component modal independence soft constraint factors; Based on the set of independence soft constraint factors between the battery state-charging behavior principal component modes, the set of linear transformation feature coding vectors of the battery state principal components and the user charging behavior associated coding vector are dynamically and adaptively aggregated to obtain a battery state-charging behavior interaction coding vector as the battery state-charging behavior interaction coding representation.

5. The adaptive charging power supply control method according to claim 4, characterized in that: The multi-dimensional fusion characterization vector of the battery state is transformed to obtain a set of linear transformation feature encoding vectors of the main components of the battery state, including: Performing principal component analysis on the multi-dimensional fusion representation vector of the battery state to obtain a set of principal component feature encoding vectors of the battery state; Performing a linear transformation on each battery status principal component feature coding vector in the set of battery status principal component feature coding vectors to obtain a set of battery status principal component linear transformation feature coding vectors, wherein each battery status principal component linear transformation feature coding vector in the set of battery status principal component linear transformation feature coding vectors has the same characteristic scale as the user charging behavior associated coding vector.

6. The adaptive charging power supply control method according to claim 5, characterized in that: Performing inter-modal independence constraint analysis on the set of linear transformation feature coding vectors of the battery state principal component and the user charging behavior associated coding vector to determine a set of soft constraint factors for inter-modal independence of the battery state-charging behavior principal component, including: Inputting each battery state principal component linear transformation feature coding vector in the set of the user charging behavior association coding vector and the battery state principal component linear transformation feature coding vector into an inter-modality independence modeling unit to obtain a set of battery state-charging behavior principal component inter-modality independence coding matrices; Performing canonical stability optimization on each battery state-charging behavior principal component modal independence coding matrix in the set of battery state-charging behavior principal component modal independence coding matrices to obtain a set of battery state-charging behavior principal component modal independence canonical optimization coding matrices, wherein the canonical stability optimization process is determined by a battery state-charging behavior eigenvector composed of each eigenvalue of the battery state-charging behavior principal component modal independence coding matrix; Based on the set of battery state-charging behavior principal component modal independence norm optimization coding matrices, a set of battery state-charging behavior principal component modal independence soft constraint factors is calculated.

7. The adaptive charging power supply control method according to claim 6, characterized in that: Based on the set of the independence soft constraint factors between the battery state and charging behavior principal component modes, the set of the battery state principal component linear transformation feature coding vectors and the user charging behavior associated coding vector are dynamically and adaptively aggregated to obtain a battery state-charging behavior interaction coding vector as the battery state-charging behavior interaction coding representation, including: Inputting each battery state principal component linear transformation feature coding vector in the set of the user charging behavior association coding vector and the battery state principal component linear transformation feature coding vector into a feature interaction response unit to obtain a set of fine-grained response interaction coding vectors between battery state and charging behavior principal component modes; Based on the set of independence soft constraint factors between the battery state-charging behavior principal component modes, the set of fine-grained response interaction coding vectors between the battery state-charging behavior principal component modes is dynamically and adaptively aggregated to obtain the battery state-charging behavior interaction coding vector.

8. The adaptive charging power supply control method according to claim 7, characterized in that: Based on the battery state-charging behavior interactive coding representation, determining a target charging voltage recommended decoding value to control the charged power source to charge, including: Passing the battery state-charging behavior interaction coding vector through a decoder-based charging voltage controller to obtain a target charging voltage recommended decoding value; Based on the target charging voltage recommendation decoded value, the charged power source is controlled to perform charging.

9. An adaptive charging power supply control system, characterized in that: include: A battery status data real-time monitoring module, used to monitor the battery status data of the charged power source in real time through a battery sensor, wherein the battery status data includes voltage data and temperature data; A user charging behavior data acquisition module, used to acquire user charging behavior data, wherein the user charging behavior data includes charging time, charging frequency, charging duration, charging mode and charging target power; A battery state time series fusion coding module, used for performing time series fusion coding on the battery state data according to the time dimension to obtain a multi-dimensional fusion representation of the battery state; A user charging behavior association coding module, used to perform mapping association coding on the user charging behavior data based on charging information to obtain a user charging behavior association coding representation; A battery status-charging behavior interactive coding module, used to perform battery status-charging behavior cross-modal feature independent restriction interactive analysis on the multi-dimensional fusion representation of the battery status and the associated coding representation of the user charging behavior to obtain a battery status-charging behavior interactive coding representation; The target charging voltage recommended decoding value determination module is used to determine the target charging voltage recommended decoding value based on the battery state-charging behavior interactive coding representation to control the charged power source to charge.

10. The adaptive charging power supply control system according to claim 9, characterized in that: The battery status timing fusion encoding module is used to: After integrating the battery status data into a voltage data time series vector and a temperature data time series vector according to the time dimension, the voltage data time series vector and the temperature data time series vector are respectively passed through a time encoder based on a time convolutional network to obtain a voltage data time series encoding vector and a temperature data time series encoding vector; The voltage data time series encoding vector and the temperature data time series encoding vector are fused at feature level to obtain a battery state multi-dimensional fusion representation vector as the battery state multi-dimensional fusion representation.

Images