A cloud platform-based startup battery safety monitoring and early warning method and system

Abstract

This invention proposes a cloud-based method and system for monitoring and providing early warning of the safety of parked batteries. It belongs to the field of battery safety monitoring technology. The method includes: deploying multi-dimensional data acquisition nodes for the parked battery of a truck to generate a full lifecycle monitoring dataset; constructing a cloud platform communication link based on the full lifecycle monitoring dataset to form a distributed battery monitoring network architecture; and building a cloud-based parked battery safety monitoring system to achieve real-time dynamic acquisition of battery voltage, current, temperature, and location data, significantly improving the comprehensiveness and accuracy of battery health status monitoring.

Description

Technical Field

[0001] This invention proposes a cloud-based method and system for monitoring and providing early warnings for battery safety, belonging to the field of battery safety monitoring technology. Background Technology

[0002] In truck operation scenarios, the start-up battery serves as a critical power source, and its operational stability directly impacts vehicle safety and operational efficiency. Traditional battery management systems rely heavily on local sensors for single-parameter monitoring, which suffers from limitations such as limited data acquisition dimensions and delayed anomaly response. Especially in low-temperature environments or under high-load conditions, sudden changes in battery performance are difficult to predict in advance, leading to vehicle start-up failures or even safety accidents. While existing technologies can achieve basic data transmission, they lack the ability to deeply analyze the health status of the battery throughout its entire lifecycle. They cannot establish dynamic correlation models between multiple parameters such as voltage, current, and temperature, resulting in a fault warning accuracy rate of less than 60% and difficulty in locating specific faulty battery cells.

[0003] In low-temperature winter environments, the reduced activity of lithium batteries leads to particularly severe start-up difficulties. Existing preheating solutions often require manual intervention and cannot intelligently adjust preheating power based on real-time battery status. Furthermore, traditional systems use a fixed sampling frequency, generating redundant data during periods of stable battery performance, while missing crucial features due to excessively long sampling intervals during periods of sudden performance changes, resulting in inaccurate early warning timing. More critically, current technologies lack a cloud-based collaborative analysis framework, creating information silos between vehicle battery data. This hinders the discovery of potential common risks through big data mining, impeding the transformation of battery maintenance from reactive repair to proactive prevention. Therefore, there is an urgent need for a battery safety monitoring system that can integrate multi-dimensional data and achieve intelligent early warning and remote control. Summary of the Invention

[0004] This invention provides a cloud platform-based method and system for monitoring the safety of resident batteries and providing early warnings, in order to solve the problems mentioned in the background section above: This invention proposes a cloud-based method for monitoring and providing early warning of battery safety during startup, the method comprising: S1. Deploy multi-dimensional data acquisition nodes for truck start-up batteries to generate a full life cycle monitoring dataset for start-up batteries; construct cloud platform communication links based on the full life cycle monitoring dataset for start-up batteries to form a distributed battery monitoring network architecture. S2. Based on the distributed battery monitoring network architecture, dynamically adjust the data sampling frequency and synchronously collect battery voltage, current, temperature and location data to generate the original battery operating status data stream; perform outlier removal and data smoothing on the original battery operating status data stream to generate standardized battery health characteristic data. S3. Perform multimodal parameter correlation analysis based on standardized battery health characteristic data to obtain battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data respectively; perform risk assessment by coupling battery capacity decay trend data and internal resistance abnormal fluctuation data to generate comprehensive battery health score data. S4. Use the comprehensive battery health score data to trigger the AI ​​prediction model to generate a battery failure probability prediction curve; combine the location data to expand the spatial dimension of the failure probability prediction curve to generate a three-dimensional risk heat map; use the three-dimensional risk heat map to locate the abnormal state of the battery and generate accurate fault warning coordinate data. S5. Based on accurate fault warning coordinate data, perform graded warning threshold matching to generate multi-level warning signals; based on the multi-level warning signals, activate differentiated response strategies, and when the highest level warning is reached, automatically trigger the battery power cut-off command, while simultaneously pushing emergency handling solutions through mobile terminals.

[0005] The present invention proposes a system for implementing the cloud platform-based method for monitoring and providing early warning of battery safety, as described above. The system comprises: Link construction module: Deploys multi-dimensional data acquisition nodes for truck valet batteries to generate a full lifecycle monitoring dataset for valet batteries; constructs cloud platform communication links based on the full lifecycle monitoring dataset for valet batteries to form a distributed battery monitoring network architecture; Data acquisition module: Based on a distributed battery monitoring network architecture, it dynamically adjusts the data sampling frequency and synchronously collects battery voltage, current, temperature and location data to generate a raw battery operating status data stream; it performs outlier removal and data smoothing on the raw battery operating status data stream to generate standardized battery health characteristic data; The comprehensive scoring module performs multimodal parameter correlation analysis based on standardized battery health characteristic data to obtain battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data. It then performs a risk assessment by coupling the battery capacity decay trend data and internal resistance abnormal fluctuation data to generate a comprehensive battery health score. Fault prediction module: It uses comprehensive battery health score data to trigger AI prediction model calculation to generate battery fault probability prediction curve; it combines location data to expand the spatial dimension of the fault probability prediction curve to generate a three-dimensional risk heat map; it uses the three-dimensional risk heat map to locate abnormal battery conditions and generate accurate fault warning coordinate data. Multi-level early warning module: Based on accurate fault early warning coordinate data, it performs hierarchical early warning threshold matching to generate multi-level early warning signals; based on the multi-level early warning signals, it initiates differentiated response strategies, and when the highest level of early warning is reached, it automatically triggers a battery power cut-off command, while simultaneously pushing emergency handling solutions through mobile terminals.

[0006] The beneficial effects of this invention are as follows: By constructing a cloud-based battery safety monitoring system, real-time dynamic acquisition of battery voltage, current, temperature, and location data is achieved, significantly improving the comprehensiveness and accuracy of battery health status monitoring. The system adopts adaptive sampling frequency technology, reducing data transmission volume during stable battery operation and encrypting the acquisition of key features during periods of sudden performance changes, effectively reducing redundant data transmission and lowering cloud storage pressure. The cloud-based AI analysis model, through multi-parameter correlation calculations, can predict battery capacity degradation trends 48 hours in advance, enhancing the ability to predict abnormal states and avoiding vehicle breakdowns caused by sudden battery failures. In low-temperature winter environments, the system can automatically adjust the preheating power according to the real-time battery temperature, which can quickly restore battery activity to ensure a high start-up success rate and reduce ineffective energy consumption. When a serious fault is detected, the system can accurately locate the faulty battery cell and automatically cut off the power supply, avoiding chain reactions that could lead to overall battery pack damage. It can also push emergency plans containing location information to mobile terminals, significantly improving maintenance efficiency and vehicle uptime. Attached Figure Description

[0007] Figure 1 This is a diagram illustrating the steps of the method described in this invention; Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0008] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0009] One embodiment of the present invention, such as Figure 1 As shown, a method for monitoring and providing early warning of battery safety based on a cloud platform is disclosed, the method comprising: S1. Deploy multi-dimensional data acquisition nodes for truck start-up batteries to generate a full life cycle monitoring dataset for start-up batteries; construct cloud platform communication links based on the full life cycle monitoring dataset for start-up batteries to form a distributed battery monitoring network architecture. S2. Based on the distributed battery monitoring network architecture, dynamically adjust the data sampling frequency and synchronously collect battery voltage, current, temperature and location data to generate the original battery operating status data stream; perform outlier removal and data smoothing on the original battery operating status data stream to generate standardized battery health characteristic data. S3. Perform multimodal parameter correlation analysis based on standardized battery health characteristic data to obtain battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data respectively; perform risk assessment by coupling battery capacity decay trend data and internal resistance abnormal fluctuation data to generate comprehensive battery health score data. S4. Use the comprehensive battery health score data to trigger the AI ​​prediction model to generate a battery failure probability prediction curve; combine the location data to expand the spatial dimension of the failure probability prediction curve to generate a three-dimensional risk heat map; use the three-dimensional risk heat map to locate the abnormal state of the battery and generate accurate fault warning coordinate data. S5. Based on the accurate fault warning coordinate data, perform graded warning threshold matching to generate multi-level warning signals; based on the multi-level warning signals, activate the differentiated response strategy. When the highest level warning is reached, automatically trigger the battery power cut-off command and push the emergency handling plan through the mobile terminal. The emergency handling plan includes fault location information.

[0010] The working principle and effects of the above technical solution are as follows: Through multi-dimensional data collection and cloud platform network architecture, the completeness of battery lifecycle data monitoring is improved, reducing data omissions caused by single collection methods. Dynamically adjusting the sampling frequency and optimizing the data processing flow enhances the accuracy of battery health characteristic data and reduces the interference of outliers on analysis results. Multi-modal parameter correlation analysis and coupled risk assessment improve the scientific rigor of battery health scoring and avoid misjudgments caused by single parameter judgments. The combination of AI prediction models and 3D risk heat maps improves the timeliness and accuracy of fault warnings and reduces the time cost of fault troubleshooting. Tiered warning and differentiated response strategies can quickly trigger corresponding processing procedures and automatically cut off power supply in high-risk situations, preventing safety accidents caused by battery failures. Simultaneously, the accurate delivery of emergency response plans enhances the efficiency of fault handling, reduces the risk of increased losses due to information lag, and overall improves the safety and reliability of truck battery operation.

[0011] In one embodiment of the present invention, S1 includes: S11. Analyze the truck's operating route and the environmental characteristics of the battery installation area to generate a data acquisition node location scheme; S12. Deploy multi-dimensional data acquisition equipment according to the site selection plan and complete equipment debugging and calibration to generate a full life cycle monitoring dataset for the start-up battery. S13. Optimize the transmission protocol of the monitoring dataset to form a unified data interaction standard; S14. Establish cloud platform communication links according to data interaction standards and construct a distributed battery monitoring network architecture.

[0012] The working principle and effects of the above technical solution are as follows: By analyzing the truck's operating route and the environmental characteristics of the battery installation area, a site selection scheme for the data acquisition nodes is formulated, improving the rationality of node deployment and reducing the adverse effects of environmental obstruction and signal interference on data acquisition. Deploying multi-dimensional equipment and completing debugging and calibration enhances the completeness and accuracy of the full lifecycle monitoring dataset, avoiding data distortion caused by equipment errors. Optimizing the transmission protocol for the monitoring dataset to form a unified data interaction standard reduces the risk of data transmission compatibility between different devices and improves the smoothness of data interaction. Building a cloud platform communication link according to a unified standard and constructing a distributed monitoring network architecture ensures the stability of data transmission, expands the coverage of battery monitoring, avoids monitoring interruptions caused by single link failures, provides reliable basic support for subsequent battery safety monitoring, and overall improves the quality of the monitoring system's initial deployment and data transmission assurance capabilities.

[0013] In one embodiment of the present invention, S11 includes: Collect data on the types of roads, frequency of travel, and duration of stay for trucks during daily operations, and simultaneously collect data on temperature, humidity, vibration intensity, and electromagnetic interference in the battery installation area to generate a basic information set; Extract key influencing factors from the basic information set, analyze the data transmission stability requirements under different road sections and environmental conditions, and generate feature analysis results; Based on the results of feature analysis, the coverage and density requirements for data collection are clarified, potential areas with smooth signal transmission and convenient installation are screened out, and a list of candidate sites is generated. On-site surveys and verifications were conducted on the candidate sites to eliminate adverse factors such as physical obstruction and environmental erosion, and a site selection plan for data acquisition nodes was generated.

[0014] The working principle and effects of the above technical solution are as follows: By comprehensively collecting truck operation data and environmental information of the battery installation area, the completeness of basic information is improved, avoiding site selection bias caused by incomplete data. Extracting key influencing factors and analyzing data transmission stability requirements enhances the targeting of site selection and reduces the risk of signal interruption after subsequent equipment deployment. Combining the analysis results to screen candidate areas ensures both smooth signal transmission and ease of installation, reducing deployment and maintenance costs. On-site investigation and verification of candidate areas eliminates adverse factors such as physical obstruction and environmental erosion, improving the feasibility and reliability of the site selection scheme. This avoids equipment failure or data acquisition failure due to environmental issues after node deployment, laying a solid foundation for stable subsequent battery data acquisition and comprehensively improving the scientific nature and efficiency of data acquisition node deployment.

[0015] In one embodiment of the present invention, step S14 includes: Analyze the transmission rate, data format and compatibility requirements in the unified data interaction standard, plan the connection path and communication mode between the cloud platform and the data acquisition node, and generate a link construction plan. Deploy gateway devices, transmission modules, and network adapter components according to the planning scheme, build a basic communication link framework, and generate the initial link connection structure; Perform connectivity tests and signal strength checks on the initial link connection structure to identify link interruptions, excessive latency, and data packet loss issues, and generate link optimization and adjustment parameters. Based on the optimized and adjusted parameters, the basic communication link framework was configured and iterated to form a stable data transmission channel between each acquisition node and the cloud platform, thus constructing a distributed battery monitoring network architecture.

[0016] The working principle and effects of the above technical solution are as follows: By analyzing the core requirements of the unified data interaction standard and planning connection paths and communication modes, the targeted nature of link construction is improved, avoiding resource waste or compatibility issues caused by blind deployment. Deploying relevant equipment and components according to the planned scheme and building a basic communication link framework enhances the rationality of the link structure and reduces transmission obstacles caused by components. Connectivity testing and signal detection are performed on the initial link to promptly identify interruptions, delays, and packet loss issues, improving the timeliness of fault detection and reducing the risk of sudden failures during subsequent operation. Iterative configuration of the link framework based on optimized parameters forms a stable transmission channel, ensuring data transmission efficiency between each acquisition node and the cloud platform, strengthening the overall stability of the distributed network, avoiding monitoring disconnection caused by a single link failure, providing reliable support for real-time battery data upload, and comprehensively improving the construction quality and operational assurance capabilities of the distributed monitoring network architecture.

[0017] In one embodiment of the present invention, S2 includes: S21. Collect communication delay data of each node in the distributed battery monitoring network architecture and generate network transmission efficiency evaluation results. S22. Adjust the dynamic data sampling frequency according to the evaluation results, and synchronously collect battery voltage, current, temperature and positioning data to generate the original battery operating status data stream. S23. Use statistical analysis methods to identify abnormal values ​​in the raw data stream, perform screening and removal processes, and generate preliminary cleaned data. S24. Implement a data smoothing algorithm on the preliminary purification data to eliminate data fluctuation interference and generate standardized battery health characteristic data.

[0018] The working principle and effects of the above technical solution are as follows: By collecting communication delay data from each node to assess network transmission efficiency, the dynamic sampling frequency is adjusted accordingly, ensuring real-time data acquisition while avoiding problems such as excessive network load or insufficient data collection. Simultaneous acquisition of multi-dimensional data such as battery voltage and current enhances the comprehensiveness of the raw data stream and reduces the analytical limitations caused by single-dimensional data. Statistical methods are used to remove outliers, reducing the interference of extreme data on the results. Data smoothing further eliminates fluctuations, improving the accuracy of standardized health characteristic data. The entire process ensures that the data closely reflects the actual operating state and has good usability, avoiding deviations in battery health assessment caused by data distortion. This lays a solid data foundation for subsequent risk analysis and fault early warning, comprehensively improving the scientific rigor and reliability of battery status monitoring.

[0019] In one embodiment of the present invention, S3 includes: S31. Extract multimodal parameters from standardized battery health characteristic data and establish a parameter correlation analysis model; S32. Analyze the battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data through model calculations. S33. Perform risk factor superposition calculation on capacity decay trend data and internal resistance abnormal fluctuation data to generate coupled risk assessment results. S34. Combine temperature gradient change data to correct and optimize the coupling risk assessment results, and generate comprehensive battery health score data.

[0020] The working principle and effects of the above technical solution are as follows: By extracting multimodal parameters from standardized data to establish a correlation analysis model, the correlation between parameters is enhanced, avoiding the limitations of assessment caused by single-parameter analysis. The model is used to analyze three core data types: capacity decay, internal resistance fluctuation, and temperature gradient, improving the targeting of key health indicators and reducing the omission of important risk information. Risk factors are superimposed on capacity decay and internal resistance fluctuation data, taking into account the synergistic impact of the two core risks and enhancing the comprehensiveness of coupled risk assessment, avoiding one-sided conclusions caused by individual judgments. The assessment results are corrected and optimized by combining temperature gradient data, further improving the accuracy of the comprehensive battery health score and avoiding assessment bias caused by ignoring temperature factors. The entire process makes the health assessment more closely reflect the actual operating state of the battery, providing a reliable basis for subsequent fault warnings and improving the scientific rigor and effectiveness of battery safety monitoring.

[0021] In one embodiment of the present invention, S33 includes: S331. Extract key risk factors from capacity decay trend data, such as decay rate and decay period. Simultaneously extract core influencing factors from internal resistance abnormal fluctuation data, such as fluctuation amplitude and fluctuation frequency, to generate a two-dimensional risk factor set. S332. Analyze the impact of each risk factor on the safe operation of the battery, quantify the weight allocation of the factors in the two-dimensional risk factor set, and generate a weighted risk factor matrix. S333. Perform superposition calculation on the factors in the weighted risk factor matrix according to the weight ratio, integrate the associated risks of capacity decay and internal resistance fluctuation, and generate preliminary coupled risk data. S334. Perform extreme value screening and logical verification on the preliminary coupling risk data, eliminate unreasonable values ​​generated during the calculation process, and generate coupling risk assessment results.

[0022] The working principle and effects of the above technical solution are as follows: By extracting dual-dimensional risk factors of capacity decay and internal resistance fluctuation, the comprehensiveness of risk information is enhanced, avoiding the omission of key risks caused by single-dimensional analysis. Analyzing the impact of each factor on battery safety and quantifying its weight improves the rationality of factor importance differentiation and reduces assessment bias caused by treating all factors equally. The weighted superposition calculation and fusion of related risks not only highlights the synergistic effect of the two core risks but also enhances the overall integrity of the coupled risk assessment, avoiding one-sided conclusions from individual judgments. Extreme value screening and logical verification are performed on the preliminary coupled risk data to eliminate unreasonable values, reducing the interference of calculation errors on the results and improving the accuracy of the coupled risk assessment results. The entire process makes the risk assessment more closely reflect the actual safety state of the battery, providing reliable support for subsequent health scoring and improving the overall scientific rigor and effectiveness of battery risk identification.

[0023] In one embodiment of the present invention, S333 includes: Separate the capacity decay factor and internal resistance fluctuation factor in the weighted risk factor matrix to generate two independent risk factor subsets; The factors within each subset of risk factors are summed according to their weight proportions to obtain the cumulative value of capacity decay risk and the cumulative value of internal resistance volatility risk, respectively. The correlation mechanism between the two types of risk cumulative values ​​was analyzed, and the synergistic risk value of the two was calculated by a nonlinear fusion algorithm. By integrating the original features of the collaborative risk values ​​and the cumulative values ​​of the two types of risks, preliminary coupled risk data is generated.

[0024] The working principle and effects of the above technical solution are as follows: By splitting capacity decay and internal resistance fluctuation into two risk factors to generate independent subsets, the targeting of single-type risk calculation is improved, and the calculation deviation caused by the confusion of different types of factors is avoided. The risk accumulation value is obtained by weighting and summing each subset, enhancing the overall representation of single-type risks and reducing the fragmentation problem caused by analyzing individual factors separately. A nonlinear fusion algorithm is used to analyze the correlation mechanism between the two types of risks, which can accurately capture their synergistic effect while avoiding complex interactive influences that linear superposition cannot reflect, thus improving the accuracy of the synergistic risk values. Integrating the synergistic values ​​with the original features to generate preliminary coupled data retains core risk information without losing the original characteristics of the data, reducing the evaluation distortion caused by information simplification, and making the coupled risk data more consistent with the actual risk state of the battery. This provides a high-quality foundation for subsequent verification and optimization, and overall improves the scientificity and accuracy of risk fusion calculation.

[0025] In one embodiment of the present invention, step S4 includes: S41. Input the comprehensive battery health score data into the AI ​​prediction model, start the model iterative calculation, and generate the battery failure probability prediction curve. S42. Extract spatial coordinate information from the positioning data and perform data fusion processing with the fault probability prediction curve to generate spatially correlated fault prediction data. S43. Perform three-dimensional coordinate mapping calculations on spatially correlated fault prediction data to generate a three-dimensional risk heat map. S44. Analyze the abnormal concentration areas in the three-dimensional risk thermal distribution map, determine the location of the fault, and generate accurate fault early warning coordinate data.

[0026] The working principle and effects of the above technical solution are as follows: The comprehensive battery health score is input into the AI ​​prediction model for iterative calculation, generating a fault probability prediction curve. This improves the accuracy of fault trend judgment and avoids subjective biases caused by traditional experience-based judgments. Integrating spatial coordinates from location data with the prediction curve allows fault prediction to move beyond probability values, enhancing the spatial correlation of data and reducing the difficulty in locating fault positions. Generating a risk heat map through 3D coordinate mapping makes risk distribution more intuitive and easier to understand, improving the efficiency of abnormal area identification and avoiding the problem of difficulty in quickly capturing core risks using pure data. Analyzing abnormal concentration areas in the heat map to determine fault locations improves the accuracy of early warning coordinates, reduces the time cost of fault investigation, and avoids delays in handling due to ambiguous location. The entire process ensures both the scientific nature of fault prediction and the practicality of location, providing reliable support for subsequent tiered early warning systems and improving the timeliness and effectiveness of battery fault early warnings overall.

[0027] In one embodiment of the present invention, S43 includes: Extract the fault probability values ​​and corresponding spatial coordinate information from the spatial correlation fault prediction data, filter valid data points and remove duplicate records to generate a three-dimensional mapping basic dataset. Analyze the spatial distribution range and probability value interval of the basic dataset, define the axis range and scale accuracy of the three-dimensional coordinate system, and generate a standardized three-dimensional coordinate system; Each data point in the 3D mapping base dataset is embedded into a standardized 3D coordinate system according to the coordinate correspondence, so as to complete the precise binding of probability value and spatial position and generate a 3D discrete data point set. Spatial interpolation is performed on a set of three-dimensional discrete data points to fill the gaps between data points and smooth the transition of probability distribution, thereby generating a continuous three-dimensional risk distribution grid. Based on the probability values, the corresponding thermal color gradients are matched, and the three-dimensional risk distribution grid is fused and rendered with the color gradients to generate a three-dimensional risk thermal distribution map.

[0028] The working principle and effects of the above technical solution are as follows: By extracting fault probability values ​​and spatial coordinate information and filtering for duplicates, the validity of the basic data for 3D mapping is ensured, reducing the interference of duplicate or invalid records on subsequent calculations. A standardized 3D coordinate system is defined, improving the uniformity and scale accuracy of coordinate mapping and avoiding spatial position deviations caused by chaotic axis domains. Data points are embedded into the coordinate system according to their correspondence, achieving precise binding between probability values ​​and spatial positions, enhancing their correlation, and reducing the problem of location being disconnected from risk probability. Spatial interpolation fills data gaps, making the risk distribution more continuous and smooth, avoiding the blurring of risk areas caused by discrete data points. Combining probability values ​​with thermal color gradient rendering makes risk levels intuitively identifiable, improving the efficiency of abnormal area identification, avoiding the difficulty of quickly capturing core risks in pure data form, and overall improving the accuracy and practicality of the 3D risk heat map, providing clear and reliable visualization support for subsequent fault location.

[0029] In one embodiment of the present invention, step S5 includes: S51. Analyze the risk level parameters corresponding to the accurate fault early warning coordinate data and formulate early warning threshold matching standards; S52. According to the matching standard, the fault risk level is compared with the threshold to generate a multi-level early warning signal; S53. Formulate differentiated response strategies based on multi-level early warning signals and initiate corresponding processing procedures at different levels; S54. When the warning signal reaches the highest level, the battery power supply cut-off mechanism is automatically activated, and detailed information on the fault location is collected simultaneously. S55. Integrate fault location information to develop an emergency response plan, and push the emergency response plan to relevant personnel via mobile terminals.

[0030] The working principle and effects of the above technical solution are as follows: By sorting out risk level parameters and formulating early warning threshold matching standards, the rationality of threshold division is improved, avoiding the problem of early warning distortion caused by confusion between different risk levels. Multi-level early warning signals are generated according to the standards, and differentiated response strategies are formulated and handled in stages based on the signals. This not only accurately matches the handling needs of different risks, reducing resource waste caused by over-response, but also avoids the imbalance of ignoring low-risk situations and delaying high-risk situations. The highest-level early warning automatically activates the power supply cut-off mechanism, which can quickly block the path of fault deterioration and avoid safety accidents caused by battery failure. Simultaneously collecting fault information and compiling and pushing emergency plans improves the pertinence and timeliness of the handling, reduces delays caused by personnel lacking information, makes the emergency process more orderly and efficient, and comprehensively improves the safety and handling efficiency of battery failure emergency response.

[0031] In one embodiment of the present invention, S53 includes: Extract the level identifier, risk spread rate and impact range information from the multi-level early warning signals, and generate a signal classification feature dataset after classification and organization. Analyze the security threat level corresponding to different levels of signals in the dataset, divide the processing priority sequence, clarify the type and quantity of response resources required for each level, and generate a resource configuration list; By combining priority sequences and resource allocation lists, a hierarchical processing flow was developed, which includes real-time data reporting, notification of relevant personnel, and adjustment of equipment status, thus forming a process execution standard. The tiered processing procedure execution specifications are associated with the warning signals of each level, triggering a signal level matching mechanism to automatically start the corresponding level of processing procedure.

[0032] The working principle and effects of the above technical solution are as follows: By extracting the level identifiers, risk spread rates, and impact ranges of multi-level early warning signals to generate a feature dataset, the integrity of signal information is enhanced, avoiding response judgment biases caused by missing key parameters. Analyzing the security threat level of different levels of signals, prioritizing processing, and clarifying resource allocation requirements improves the rationality of resource scheduling and reduces inefficiencies caused by resource mismatch or insufficiency. Establishing a hierarchical processing flow and execution standard that includes data reporting, personnel notification, and equipment adjustment makes response operations more systematic and avoids chaos and disorder during the handling process. Associating the process standard with early warning signals triggers an automatic matching and activation mechanism, improving the timeliness of the response and avoiding delays caused by manual intervention. It can quickly concentrate resources to deal with high-level risks and efficiently handle low-level hidden dangers without wasting resources, thus improving the overall accuracy and orderliness of battery fault response.

[0033] One embodiment of the present invention, such as Figure 2As shown, the present invention proposes a system for implementing the cloud platform-based method for monitoring and providing early warning of battery safety, the system comprising: Link construction module: Deploys multi-dimensional data acquisition nodes for truck valet batteries to generate a full lifecycle monitoring dataset for valet batteries; constructs cloud platform communication links based on the full lifecycle monitoring dataset for valet batteries to form a distributed battery monitoring network architecture; Data acquisition module: Based on a distributed battery monitoring network architecture, it dynamically adjusts the data sampling frequency and synchronously collects battery voltage, current, temperature and location data to generate a raw battery operating status data stream; it performs outlier removal and data smoothing on the raw battery operating status data stream to generate standardized battery health characteristic data; The comprehensive scoring module performs multimodal parameter correlation analysis based on standardized battery health characteristic data to obtain battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data. It then performs a risk assessment by coupling the battery capacity decay trend data and internal resistance abnormal fluctuation data to generate a comprehensive battery health score. Fault prediction module: It uses comprehensive battery health score data to trigger AI prediction model calculation to generate battery fault probability prediction curve; it combines location data to expand the spatial dimension of the fault probability prediction curve to generate a three-dimensional risk heat map; it uses the three-dimensional risk heat map to locate abnormal battery conditions and generate accurate fault warning coordinate data. Multi-level early warning module: Based on accurate fault early warning coordinate data, it performs graded early warning threshold matching to generate multi-level early warning signals; based on the multi-level early warning signals, it initiates a differentiated response strategy. When the highest level early warning is reached, it automatically triggers a battery power cut-off command and pushes an emergency handling plan through the mobile terminal. The emergency handling plan includes fault location information.

[0034] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for monitoring and providing early warning of battery safety based on a cloud platform, characterized in that, The method includes: S1. Deploy multi-dimensional data acquisition nodes for truck start-up batteries to generate a full life cycle monitoring dataset for start-up batteries; construct cloud platform communication links based on the full life cycle monitoring dataset for start-up batteries to form a distributed battery monitoring network architecture. S2. Based on the distributed battery monitoring network architecture, dynamically adjust the data sampling frequency and synchronously collect battery voltage, current, temperature and location data to generate the original battery operating status data stream; perform outlier removal and data smoothing on the original battery operating status data stream to generate standardized battery health characteristic data. S3. Perform multimodal parameter correlation analysis based on standardized battery health characteristic data to obtain battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data respectively; perform risk assessment by coupling battery capacity decay trend data and internal resistance abnormal fluctuation data to generate comprehensive battery health score data. S4. Use the comprehensive battery health score data to trigger the AI ​​prediction model to generate a battery failure probability prediction curve; combine the location data to expand the spatial dimension of the failure probability prediction curve to generate a three-dimensional risk heat map; use the three-dimensional risk heat map to locate the abnormal state of the battery and generate accurate fault warning coordinate data. S5. Based on accurate fault warning coordinate data, perform graded warning threshold matching to generate multi-level warning signals; based on the multi-level warning signals, activate differentiated response strategies, and when the highest level warning is reached, automatically trigger the battery power cut-off command, while simultaneously pushing emergency handling solutions through mobile terminals.

2. The cloud-based method for monitoring and providing early warning of battery safety during startup, as described in claim 1, is characterized in that... S1 includes: S11. Analyze the truck's operating route and the environmental characteristics of the battery installation area to generate a data acquisition node location scheme; S12. Deploy multi-dimensional data acquisition equipment according to the site selection plan and complete equipment debugging and calibration to generate a full life cycle monitoring dataset for the start-up battery. S13. Optimize the transmission protocol of the monitoring dataset to form a unified data interaction standard; S14. Establish cloud platform communication links according to data interaction standards and construct a distributed battery monitoring network architecture.

3. The cloud-based method for monitoring and providing early warning of battery safety during startup, as described in claim 1, is characterized in that... S2 includes: S21. Collect communication delay data of each node in the distributed battery monitoring network architecture and generate network transmission efficiency evaluation results. S22. Adjust the dynamic data sampling frequency according to the evaluation results, and synchronously collect battery voltage, current, temperature and positioning data to generate the original battery operating status data stream. S23. Use statistical analysis methods to identify abnormal values ​​in the raw data stream, perform screening and removal processes, and generate preliminary cleaned data. S24. Implement a data smoothing algorithm on the preliminary purification data to eliminate data fluctuation interference and generate standardized battery health characteristic data.

4. The cloud-based battery safety monitoring and early warning method according to claim 1, characterized in that, The S3 includes: S31. Extract multimodal parameters from standardized battery health characteristic data and establish a parameter correlation analysis model; S32. Analyze the battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data through model calculations. S33. Perform risk factor superposition calculation on capacity decay trend data and internal resistance abnormal fluctuation data to generate coupled risk assessment results. S34. Combine temperature gradient change data to correct and optimize the coupling risk assessment results, and generate comprehensive battery health score data.

5. The cloud-based method for monitoring and providing early warning of battery safety as described in claim 4, characterized in that, S33 includes: S331. Extract key risk factors from capacity decay trend data and simultaneously extract core influencing factors from internal resistance abnormal fluctuation data to generate a two-dimensional risk factor set. S332. Analyze the impact of each risk factor on the safe operation of the battery, quantify the weight allocation of the factors in the two-dimensional risk factor set, and generate a weighted risk factor matrix. S333. Perform superposition calculation on the factors in the weighted risk factor matrix according to the weight ratio, integrate the associated risks of capacity decay and internal resistance fluctuation, and generate preliminary coupled risk data. S334. Perform extreme value screening and logical verification on the preliminary coupling risk data, eliminate unreasonable values ​​generated during the calculation process, and generate coupling risk assessment results.

6. The cloud-based method for monitoring and providing early warning of battery safety during startup, as described in claim 1, is characterized in that... The S4 includes: S41. Input the comprehensive battery health score data into the AI ​​prediction model, start the model iterative calculation, and generate the battery failure probability prediction curve. S42. Extract spatial coordinate information from the positioning data and perform data fusion processing with the fault probability prediction curve to generate spatially correlated fault prediction data. S43. Perform three-dimensional coordinate mapping calculations on spatially correlated fault prediction data to generate a three-dimensional risk heat map. S44. Analyze the abnormal concentration areas in the three-dimensional risk thermal distribution map, determine the location of the fault, and generate accurate fault early warning coordinate data.

7. The cloud-based method for monitoring and providing early warning of battery safety as described in claim 6, characterized in that, S43 includes: Extract the fault probability values ​​and corresponding spatial coordinate information from the spatial correlation fault prediction data, filter valid data points and remove duplicate records to generate a three-dimensional mapping basic dataset. Analyze the spatial distribution range and probability value interval of the basic dataset, define the axis range and scale accuracy of the three-dimensional coordinate system, and generate a standardized three-dimensional coordinate system; Each data point in the 3D mapping base dataset is embedded into a standardized 3D coordinate system according to the coordinate correspondence, so as to complete the precise binding of probability value and spatial position and generate a 3D discrete data point set. Spatial interpolation is performed on a set of three-dimensional discrete data points to fill the gaps between data points and smooth the transition of probability distribution, thereby generating a continuous three-dimensional risk distribution grid. Based on the probability values, the corresponding thermal color gradients are matched, and the three-dimensional risk distribution grid is fused and rendered with the color gradients to generate a three-dimensional risk thermal distribution map.

8. The cloud-based method for monitoring and providing early warning of battery safety during startup, as described in claim 1, is characterized in that... The S5 includes: S51. Analyze the risk level parameters corresponding to the accurate fault early warning coordinate data and formulate early warning threshold matching standards; S52. According to the matching standard, the fault risk level is compared with the threshold to generate a multi-level early warning signal; S53. Formulate differentiated response strategies based on multi-level early warning signals and initiate corresponding processing procedures at different levels; S54. When the warning signal reaches the highest level, the battery power supply cut-off mechanism is automatically activated, and detailed information on the fault location is collected simultaneously. S55. Integrate fault location information to develop an emergency response plan, and push the emergency response plan to relevant personnel via mobile terminals.

9. The cloud-based method for monitoring and providing early warning of battery safety as described in claim 8, characterized in that, S53 includes: Extract the level identifier, risk spread rate and impact range information from the multi-level early warning signals, and generate a signal classification feature dataset after classification and organization. Analyze the security threat level corresponding to different levels of signals in the dataset, divide the processing priority sequence, clarify the type and quantity of response resources required for each level, and generate a resource configuration list; By combining priority sequences and resource allocation lists, a hierarchical processing flow was developed, which includes real-time data reporting, notification of relevant personnel, and adjustment of equipment status, thus forming a process execution standard. The tiered processing procedure execution specifications are associated with the warning signals of each level, triggering a signal level matching mechanism to automatically start the corresponding level of processing procedure.

10. A system for implementing the cloud platform-based method for monitoring and providing early warning of battery safety as described in claim 1, characterized in that, The system includes: Link construction module: Deploys multi-dimensional data acquisition nodes for truck valet batteries to generate a full lifecycle monitoring dataset for valet batteries; constructs cloud platform communication links based on the full lifecycle monitoring dataset for valet batteries to form a distributed battery monitoring network architecture; Data acquisition module: Based on a distributed battery monitoring network architecture, it dynamically adjusts the data sampling frequency and synchronously collects battery voltage, current, temperature and location data to generate a raw battery operating status data stream; it performs outlier removal and data smoothing on the raw battery operating status data stream to generate standardized battery health characteristic data; The comprehensive scoring module performs multimodal parameter correlation analysis based on standardized battery health characteristic data to obtain battery capacity decay trend data, internal resistance abnormal fluctuation data, and temperature gradient change data. It then performs a risk assessment by coupling the battery capacity decay trend data and internal resistance abnormal fluctuation data to generate a comprehensive battery health score. Fault prediction module: It uses comprehensive battery health score data to trigger AI prediction model calculation to generate battery fault probability prediction curve; it combines location data to expand the spatial dimension of the fault probability prediction curve to generate a three-dimensional risk heat map; it uses the three-dimensional risk heat map to locate abnormal battery conditions and generate accurate fault warning coordinate data. Multi-level early warning module: Based on accurate fault early warning coordinate data, it performs hierarchical early warning threshold matching to generate multi-level early warning signals; based on the multi-level early warning signals, it initiates differentiated response strategies, and when the highest level of early warning is reached, it automatically triggers a battery power cut-off command, while simultaneously pushing emergency handling solutions through mobile terminals.

Images