Multi-module lithium battery pack system based on cloud computing management

Abstract

The invention discloses a multi-module lithium battery pack system based on cloud computing management, relates to the technical field of cloud computing management, and is used for solving the problem of poor management precision of a multi-module lithium battery pack. Comprising a lithium battery data acquisition module, a battery degradation analysis module, a degradation acceleration point determination module and a strategy management module. According to the invention, data acquisition and preprocessing are carried out on a multi-module lithium battery pack through an Internet of Things sensor and an edge computing device, the data are transmitted to a cloud platform, Fourier transform is used to extract features, a nonlinear regression model is combined to evaluate the health state of the lithium battery, and a degradation speed is determined and a degradation acceleration phenomenon is identified through a segmented nonlinear regression model. Kalman filtering and nonlinear regression are combined, a degradation acceleration point is identified, load distribution is optimized through an intelligent scheduling algorithm after identification, the load of a degradation module is reduced, a scheduling signal is generated by analyzing load adjustment information, a system is guided to perform strategy scheduling management, and a scheduling strategy and energy management efficiency are optimized.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing management technology, and more specifically, to a multi-module lithium battery pack system based on cloud computing management. Background Art

[0002] With the widespread application of electric vehicles, energy storage systems and other equipment, the demand for lithium battery packs as the main energy storage components is increasing. The multi-module lithium battery pack system is composed of multiple lithium battery modules, and each module operates independently in power supply, energy storage and management. This design improves the redundancy and scalability of the system. Through cloud computing technology, unified management and remote monitoring of multiple lithium battery modules can be achieved, and various data of lithium battery packs, such as voltage, current, temperature, charge and discharge status, can be collected in real time. Then, the health status and remaining life of lithium batteries can be predicted through big data analysis and machine learning technology. In addition, the cloud computing platform can also dynamically schedule different lithium battery modules according to real-time needs to ensure efficient operation of the system.

[0003] Deficiencies in the existing technology: During the use of the lithium battery modules of a multi-module lithium battery pack, there is an accelerated degradation of lithium battery performance. That is, due to the uneven distribution of voltage, temperature and load among the modules, some lithium battery modules will be under greater pressure during frequent charging and discharging, resulting in intensified internal chemical reactions and accelerated aging of the modules. This degradation will not only shorten the life of individual modules, but also affect the performance consistency of the entire lithium battery system, resulting in reduced energy utilization, and increased maintenance costs and risks. The inability to monitor and adjust this unbalanced state in a timely manner has a significant impact on the long-term operating efficiency and stability of the lithium battery pack. Summary of the invention

[0004] In order to overcome the above defects of the prior art, there is a solution as follows to solve the problem of poor management accuracy of multi-module lithium battery packs in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A multi-module lithium battery pack system based on cloud computing management includes a lithium battery data acquisition module, a battery degradation analysis module, a degradation acceleration point determination module, and a strategy management module, and each module is connected through signals; Lithium battery data acquisition module, used to collect multi-module lithium battery pack data through IoT sensors, pre-process the data using edge computing devices, and transmit the pre-processed data to the cloud computing platform; The battery degradation analysis module is used to extract features from preprocessed data using Fourier transform, build a nonlinear regression model to evaluate the health status of lithium batteries, and use piecewise nonlinear regression to determine the degradation rate of lithium batteries and the accelerated degradation phenomenon of lithium battery modules; Degradation acceleration point determination module, which is used to identify the degradation acceleration point of lithium batteries in real time through Kalman filtering combined with nonlinear regression model, and optimize load distribution through intelligent scheduling algorithm after identifying the degradation acceleration point; The strategy management module is used to analyze the load scheduling of the intelligent scheduling algorithm, obtain the load regulation information in the scheduling optimization process, analyze the load regulation information and generate different signals, and perform strategy scheduling management based on the generated different signals.

[0006] In a preferred embodiment, for collecting multi-module lithium battery pack data through an Internet of Things sensor, using an edge computing device to pre-process the data, and transmitting the pre-processed data to a cloud computing platform, the specific steps are as follows: Install IoT sensors on lithium battery modules to collect voltage, current, temperature and internal impedance data; Edge computing devices compress and encode data, filter out abnormal data, suppress noise in collected data, prioritize sensor data based on its importance and time sensitivity, and upload data of different priorities to the cloud computing platform in a hierarchical manner.

[0007] In a preferred embodiment, the preprocessed data is used to extract features using Fourier transform, a nonlinear regression model is constructed to evaluate the health status of the lithium battery, and a segmented nonlinear regression is used to determine the degradation rate of the lithium battery and the accelerated degradation phenomenon of the lithium battery module. The specific steps include: Perform Fourier transform on the preprocessed data to extract feature data, and then perform a denoising process based on wavelet transform on the feature data to ensure that high-frequency interference components in the measurement are removed; Combining the voltage, current, and temperature change rates in historical data, nonlinear feature variables are constructed, and various features are input into the nonlinear regression model in the form of multidimensional feature vectors to determine the estimated health status of the lithium battery module; Based on the results of nonlinear mapping, a piecewise nonlinear regression analysis is performed, and the health status is iteratively updated in combination with the current and historical status of the lithium battery module. The changes in the lithium battery during the degradation process are determined based on the iterative update results. The degradation rate is adjusted nonlinearly, an adjustment factor of the degradation rate is introduced, the sudden fluctuations in the health status of the lithium battery are determined, and the degradation rate is re-estimated in the multi-module lithium battery pack according to the change of the adjustment factor.

[0008] In a preferred embodiment, the Kalman filter is combined with a nonlinear regression model to identify the accelerated degradation point of the lithium battery in real time, and after identifying the accelerated degradation point, the load distribution is optimized by an intelligent scheduling algorithm. The specific steps are as follows: Define the initial health status estimate and covariance matrix of the lithium battery module, and establish the initial health status estimate through the historical data of the lithium battery and the preliminary prediction results of the nonlinear regression model; The output results of the nonlinear regression model for evaluating the lithium battery module are used as the observation data of the Kalman filter to predict the health status of the lithium battery; By monitoring the rate of change of the health status estimate in real time, the time point of accelerated degradation can be identified; When the rate of change of the state estimation value of the Kalman filter exceeds a preset threshold within a time, it is determined that the current lithium battery module is in an accelerated degradation state, and the time is marked as a degradation acceleration point; When the accelerated degradation point of the lithium battery module is identified, the cloud computing platform triggers the intelligent scheduling algorithm to optimize the use strategy of the lithium battery module.

[0009] In a preferred embodiment, the method for analyzing the load scheduling of the intelligent scheduling algorithm and obtaining the load regulation information in the scheduling optimization process includes the following steps: Obtaining load regulation information during the dispatch optimization process, the load regulation information including regulation balance information and dispatch response information; The fluctuation trend information includes the load regulation balance index, and the load response information includes the response efficiency dispatch index; The obtained load regulation balance index and response efficiency dispatch index are combined to generate a dispatch management coefficient; The load regulation balance index and response efficiency dispatch index are directly proportional to the dispatch management coefficient.

[0010] In a preferred embodiment, analyzing the load regulation information and generating different signals, and performing strategic scheduling management according to the generated different signals, includes the following steps: Compare the generated forecast adjustment factor with the set management assessment threshold; If the forecast adjustment coefficient is greater than or equal to the management evaluation threshold, a strategy scheduling stability signal is generated, indicating that the scheduling strategy is running stably and no additional adjustment is required; If the predicted adjustment coefficient is less than the management evaluation threshold, a scheduling abnormal fluctuation signal is generated, indicating that the scheduling strategy of the optimization scheduling algorithm should be adjusted in terms of load regulation, balance or response efficiency.

[0011] The technical effects and advantages of the multi-module lithium battery pack system based on cloud computing management of the present invention are as follows: The present invention uses IoT sensors combined with edge computing devices to perform real-time data collection and preprocessing of multi-module lithium battery packs, ensuring high quality and high efficiency when data is transmitted to the cloud computing platform. The frequency spectrum characteristics of the battery module are extracted by Fourier transform, and the health status of the lithium battery is evaluated in combination with a nonlinear regression model, which can more accurately capture the degradation trend of the lithium battery. The piecewise nonlinear regression model is used to determine the degradation speed of the battery module, especially to identify the accelerated degradation phenomenon, thereby providing early warning of the risk of imminent failure of the lithium battery. The Kalman filter is combined with nonlinear regression to identify the accelerated degradation point of the lithium battery in real time, so that the system can respond quickly and adjust the load distribution.

[0012] After identifying the degradation acceleration point, the intelligent scheduling algorithm optimizes the load distribution of each battery module, reduces the load of the degraded module, and slows down its degradation rate. By analyzing the load adjustment information, the present invention generates different scheduling signals to guide the system to perform strategic scheduling management. This innovative method significantly improves the stability and life of the lithium battery pack, ensures the efficient operation of the battery system in load scheduling and energy management, and optimizes the accuracy of the scheduling strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the structure of a multi-module lithium battery pack system based on cloud computing management of the present invention. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] In order to achieve the above objectives, Figure 1 A schematic diagram of the structure of a multi-module lithium battery pack system based on cloud computing management of the present invention is given, which specifically includes a lithium battery data acquisition module, a battery degradation analysis module, a degradation acceleration point determination module and a strategy management module, and each module is connected by a signal; It includes a lithium battery data acquisition module, a battery degradation analysis module, a degradation acceleration point determination module, and a strategy management module, and each module is connected through signals; The lithium battery data acquisition module is used to collect multi-module lithium battery pack data through IoT sensors, pre-process the data using edge computing devices, and transmit the pre-processed data to the cloud computing platform; The battery degradation analysis module is used to extract features from preprocessed data using Fourier transform, build a nonlinear regression model to evaluate the health status of lithium batteries, and use piecewise nonlinear regression to determine the degradation rate of lithium batteries and the accelerated degradation phenomenon of lithium battery modules; Degradation acceleration point determination module, which is used to identify the degradation acceleration point of lithium batteries in real time through Kalman filtering combined with nonlinear regression model, and optimize load distribution through intelligent scheduling algorithm after identifying the degradation acceleration point; The strategy management module is used to analyze the load scheduling of the intelligent scheduling algorithm, obtain the load regulation information in the scheduling optimization process, analyze the load regulation information and generate different signals, and perform strategy scheduling management based on the generated different signals.

[0016] Install high-precision IoT sensors on each lithium battery module to measure key physical parameters such as voltage, current, temperature and internal impedance; In a multi-module lithium battery pack system, real-time monitoring of lithium battery module performance is crucial, especially in order to deal with imbalance and degradation between modules. Through IoT sensors and edge computing devices, all-round monitoring of each lithium battery module can more efficiently collect, process and analyze data, and optimize the operation of the entire system. Each lithium battery module in a multi-module lithium battery pack is monitored through IoT sensors (such as voltage sensors, current sensors, and temperature sensors) and edge computing devices. The multi-dimensional data collected in real time includes voltage, current, temperature, charge and discharge times, internal impedance, etc., reflecting the operating status and health of the lithium battery module. The real-time collection and preprocessing of data are performed by edge devices and finally transmitted to the relevant cloud computing platform for large-scale analysis and storage; The voltage sensor monitors the voltage changes of each lithium battery module, captures the voltage fluctuations during the charging and discharging process, and detects voltage imbalance; the current sensor tracks the charging and discharging current of each module, ensures the stability of the current output, and identifies the current anomaly; the temperature sensor monitors the temperature changes of the lithium battery module, captures the temperature rise that may lead to thermal runaway, and ensures the timeliness of heat dissipation management; the internal impedance sensor evaluates the internal impedance changes of each lithium battery module, identifies the signs of aging and the early signs of performance degradation; The data will be transmitted to the cloud computing platform in real time through wireless communication technology (such as LoRa, Wi-Fi or 4G / 5G) for storage and analysis. Before transmitting the data, the edge computing device compresses and encodes the data to reduce communication bandwidth consumption and improve transmission efficiency. According to the importance and time sensitivity of the sensor data, data with different priorities will be uploaded in layers. For example, data with high temperature or abnormal voltage will be immediately transmitted to the cloud computing platform for rapid processing and decision-making; The packaged, encrypted and verified data is uploaded to the cloud computing platform in real time through the selected wireless communication protocol to ensure that the status of the lithium battery pack can be monitored and processed in real time. Each data packet contains accurate timestamp information to ensure that multi-module data can be accurately analyzed in time series on the cloud computing platform. According to system requirements, the upload priority of different data types is set. For example, key data such as overtemperature and abnormal voltage are uploaded in real time, while routine data such as the health status of lithium batteries can be uploaded regularly. After receiving the data, the cloud computing platform verifies it to confirm the integrity of the data. Once the verification is passed, the confirmation information is returned to the edge device. If the verification fails, the retransmission process is automatically triggered. In the framework of cloud computing, preliminary data processing by edge computing devices can reduce the burden on the cloud and ensure the quality and reliability of uploaded data. Invalid or obviously abnormal data can be filtered out locally by edge computing devices. For example, if a sensor reading deviates from the normal range, edge computing will perform secondary confirmation to avoid transmitting erroneous data. Physical parameters are smoothed by edge algorithms to suppress noise in collected data and ensure data accuracy, especially in the case of small fluctuations in voltage or current; The edge computing device automatically adjusts the sampling frequency according to the operating status of the lithium battery module. For example, when the lithium battery module shows signs of accelerated degradation such as high temperature and rapid voltage drop, the device will increase the data sampling frequency to capture more details so that the cloud computing platform can perform more accurate analysis.

[0017] Based on the big data analysis capabilities of the cloud computing platform, a health status prediction model for lithium battery modules is established. The model uses the collected real-time and historical data to predict the health status and future degradation trends of lithium battery modules through machine learning algorithms, with a particular focus on signs of accelerated degradation of lithium batteries. The pre-processed data will maintain high quality when transmitted to the cloud computing platform. The pre-processed data will be Fourier transformed to extract feature data, and the feature data will be subjected to a denoising process based on wavelet transform to ensure that high-frequency interference components in the measurement are removed and important low-frequency information in the signal is retained, thereby enhancing the identification of health change trends of lithium batteries. Feature extraction requires multi-dimensional analysis in the time domain and frequency domain. First, the lithium battery voltage data is Fourier transformed to extract spectrum features to identify periodic changes and nonlinear degradation trends. Secondly, the voltage, current, and temperature change rates in historical data are combined to construct nonlinear feature variables, and various features are input into the prediction model in the form of multi-dimensional feature vectors, such as the voltage fluctuation feature extraction formula: , where V(t) represents the real-time voltage, Represents the historical reference voltage, is the time decay factor, is the voltage fluctuation characteristic, T represents the total duration, and this formula is used to capture the abnormal characteristic changes in voltage fluctuations to identify the degradation trend of the health status; Based on the preprocessed feature data, a nonlinear regression model is used to evaluate the health status of lithium batteries. In a multi-module lithium battery pack, the health status of lithium battery modules has complex nonlinear change characteristics, especially in the charge and discharge cycle, the performance degradation of lithium batteries is not linear. Therefore, a nonlinear regression model is used for prediction, that is, the health status of lithium battery modules is evaluated by a nonlinear regression model, and the nonlinear degradation process of lithium batteries is captured based on multi-dimensional lithium battery operation data (such as voltage, current, temperature, etc.): A nonlinear mapping function f is constructed to map the input data to capture complex dynamic relationships. In order to avoid the limitations of simple linear combinations, a nonlinear activation function is used to process the input. The mapping formula is: , represents the estimated health status of the lithium battery module at the current time t, is the weight matrix, which represents the influence weight of each input variable on the health status; is the input vector, which contains various real-time data of the lithium battery module (voltage, current, temperature, etc.), b is the bias vector, which adjusts the offset of the mapping, and f is the nonlinear activation function (such as ReLU, Tanh, etc.), which is used to capture the nonlinear characteristics of the lithium battery state change; Based on the results of nonlinear mapping, it is necessary to further estimate the current health status of the lithium battery. This model takes into account multiple dynamic parameters of the lithium battery, and based on the characteristics of the multi-module lithium battery pack, comprehensively analyzes the health status of each module to ensure that the independent changes and mutual influences of each module are captured. The segmented nonlinear regression method is used to combine the current and historical states of the lithium battery module, and the recursive formula is used to iteratively update the health status to describe the rapid changes of the lithium battery during the degradation process. The health status estimation formula is: , where is the estimated health status at the current time t, is the health status of the previous moment, which serves as the basis for the current status. is the attenuation coefficient, which controls the rate of change of the lithium battery state; is the voltage value of the lithium battery at the current moment, is the threshold voltage of the lithium battery, beyond which the health status will show accelerated degradation; is the influence coefficient of the ith physical parameter (temperature, current, etc.) on the health status, It is the function of the influence of temperature and current on the health status of lithium batteries, reflecting the nonlinear relationship of accelerated lithium battery degradation under high temperature or high current conditions; The influence function expression is: , which indicates that the temperature Exceeding the threshold After that, the health status of lithium batteries will be affected more severely, showing a nonlinear growth relationship; at the same time, the square term of current reflects the nonlinear effect of current on lithium batteries; In the context of multi-module lithium battery packs, the health status of different lithium battery modules may differ, and the differences may gradually widen as the usage time increases. When estimating the health status, it is necessary to dynamically adjust the status of each module and conduct a comprehensive assessment of the health status of all modules. A dynamic weight adjustment mechanism is introduced. When the health status of different modules differs too much, a higher weight is given to the module with poor health status to ensure that the system can detect possible faults early and comprehensively assess the health status. formula: ,in, is the health status of the i-th lithium battery module at time t, is the weight of the i-th lithium battery module. As the health status degrades, the weight will increase dynamically, giving higher attention to the severely degraded modules; The degradation rate is adjusted nonlinearly. Considering that the degradation process of lithium battery modules will show nonlinear changes over time, the adjustment factor of degradation rate is introduced to capture the sudden fluctuations in the health status of lithium batteries. In a multi-module lithium battery pack, some lithium battery modules may experience accelerated degradation in different time periods. The nonlinear regression model should dynamically adjust the estimation of the degradation rate according to real-time data, that is, the dynamic degradation rate is estimated according to different adjustment factors. The adaptive optimization formula is: ,in, is the degradation rate of the next time step; is the change in health status at the current time step; and is the nonlinear adjustment coefficient, which controls the change of degradation speed; It is the minimum value of the health status of the lithium battery, reflecting the degradation limit. When the lithium battery is close to failure, the degradation rate will accelerate; In actual operation, the model will be adjusted according to the status differences and overall health level of different modules to ensure the accuracy and robustness of the model. The optimization mechanism based on the gradient descent method is used to update the model parameters (such as attenuation coefficient, temperature and current influence coefficient, etc.) in real time. In summary, the real-time monitoring data (such as voltage, current, temperature, etc.) of each lithium battery module is input into the nonlinear mapping layer to generate a preliminary estimate of the health status of the lithium battery. The health status of the lithium battery is updated through a nonlinear regression model, and combined with historical data to calculate the comprehensive health status at the current moment. According to the real-time change in the health status, the degradation speed is dynamically adjusted to ensure that the phenomenon of accelerated degradation is captured. Based on system feedback, the gradient descent method is used to adjust the model parameters to ensure the long-term stability of the model in practical applications and improve the safety and operation efficiency of the system.

[0018] In a multi-module lithium battery pack, the degradation acceleration point is an important early signal that the lithium battery module is about to enter the failure stage. Therefore, accurately identifying these acceleration points is crucial for the stable operation and preventive maintenance of the lithium battery system. The state estimation method based on Kalman filtering, combined with the output results of the lithium battery module evaluation through the nonlinear regression model, can further accurately identify the degradation acceleration points of lithium batteries; By combining the output data of the nonlinear regression model, the Kalman filter will be used to smooth and suppress the changes in the health status of the lithium battery, estimate the degradation acceleration point in real time, and define the initial health status estimate and covariance matrix of the lithium battery module. And through the historical data of lithium batteries and the preliminary prediction results of the nonlinear regression model, the initial health state estimate is established, and the Kalman filter state vector formula is: ,in, is the health status estimate at the current time t; F is the state transfer matrix, which controls the state at the previous time The impact on the current state; B is the control matrix, for control inputs (such as current or temperature changes); is process noise, reflecting the unobservable random fluctuations in the system; The output results of the lithium battery module evaluation through the nonlinear regression model are used as the observation data of the Kalman filter to predict the health status of the lithium battery, so that the final health status estimate is smoother and the degradation acceleration point can be identified. When the health status of the lithium battery changes sharply, the degradation acceleration point is accurately identified. The degradation acceleration point is often manifested as a sudden drop in the health status. Therefore, by detecting the change rate of the health status estimate in real time, the time point of accelerated degradation can be identified; When the change rate of the health state estimate value of the Kalman filter exceeds the preset threshold in a short period of time, the current lithium battery module is determined to be in an accelerated degradation state, and the moment is marked as the accelerated degradation point. The identification formula is: ,if , then mark the moment is the degradation acceleration point, where is the threshold of the degradation acceleration point. By setting a suitable threshold, the normal degradation process and the degradation acceleration point can be effectively distinguished. After identifying the degradation acceleration point, the Kalman filter needs to dynamically adjust the state covariance matrix And the observation noise covariance matrix R is used to cope with the rapid changes in the lithium battery state. In the accelerated stage of lithium battery degradation, the system uncertainty increases, so it is necessary to increase the weight of the covariance matrix to strengthen the suppression of noise.

[0019] When the degradation speed of the lithium battery module is detected to be accelerated (degradation acceleration point), the cloud computing platform will trigger the intelligent scheduling algorithm to optimize the use strategy of the lithium battery module. By adjusting the charge and discharge load, the workload of the module with accelerated degradation is reduced to delay its degradation. The specific steps are as follows: After identifying the accelerated degradation points of some module lithium battery packs, firstly, all modules in the lithium battery pack need to be prioritized. Based on the results of the health status prediction model, the lithium battery modules are divided into different priorities according to their health status. Modules with better health status should bear greater loads, while modules with accelerated degradation are lowered in priority and have their loads reduced to avoid further accelerated degradation. That is, the modules are prioritized according to the current health status value and degradation rate of each lithium battery module: ,in is the health status of the i-th module, is the degradation rate of the i-th module, and the module with better health status and slower degradation rate has higher priority; Assign higher priority modules to high-load tasks, taking on the main charging and discharging tasks first, while lower priority modules have less burden and remain at a lower charging and discharging cycle; After priority classification, the intelligent scheduling algorithm ensures that the use of lithium battery modules is dynamically adjusted by optimizing the load distribution of each module in real time. This process is based on the changes in the voltage, current and health status of each module, combined with the real-time power demand of the lithium battery group and the external load conditions. The specific steps are as follows: Monitor the voltage, current and temperature of each lithium battery module in real time, and make load scheduling decisions based on the total load demand of the current system. The load distribution formula is: ;in, is the load distribution amount of the i-th module; It is a dynamic weight, based on health status and degradation acceleration indicator (When the degradation acceleration point is identified, ); is the total load demand of the current system; For lithium battery modules with identified degradation acceleration points, the intelligent scheduling algorithm will implement a load reduction mechanism to reduce its workload to slow down the degradation rate. By controlling the number and amplitude of its charge and discharge, the frequency of chemical reactions inside the lithium battery is reduced, thereby slowing down its performance degradation. For lithium battery modules in good health, the scheduling algorithm will assign more charging and discharging tasks to increase their workload to compensate for the load reduction of modules with accelerated degradation. By appropriately increasing the frequency of use of modules in good health, the overall efficiency of the system can be improved and the stability of power supply can be ensured. Analyze the load regulation and balance, response efficiency and scheduling effectiveness of the intelligent scheduling algorithm, that is, analyze the load scheduling of the intelligent scheduling algorithm, determine the response effect of load adjustment, whether the health status of the lithium battery module has been successfully improved, and obtain the load regulation information in the scheduling optimization process. The load regulation information includes regulation balance information and scheduling response information; The regulation balance information includes the load regulation balance index and is calibrated as LBRI, and the dispatch response information includes the response efficiency dispatch index and is calibrated as RSEI; The load regulation balance index is used to measure the performance of the intelligent scheduling algorithm in regulating the load of lithium battery modules and maintaining load balance. It reflects whether the scheduling algorithm can achieve reasonable load distribution between modules with different health states when adjusting the workload of lithium battery modules, and whether such distribution reaches a balanced state. By comprehensively measuring the intensity of load regulation and the balance of load distribution, the load management effect of the intelligent scheduling algorithm on multi-module lithium battery packs is evaluated, and the strength of the intelligent scheduling algorithm to adjust the load distribution when facing lithium battery modules with different health states is determined. The higher the load regulation balance index, the stronger the load adjustment, the faster the degradation of the module load is greatly reduced, and the healthier the module load is correspondingly increased; A high load regulation balance index indicates that the system can properly regulate the load while maintaining load balance, which helps to extend the overall service life of the lithium battery pack and improve the operating efficiency of the system; if the load regulation balance index is low, it indicates that the scheduling strategy needs to be further optimized, especially in the case of uneven load distribution or insufficient regulation. At this time, the load should be further reduced for the modules with poor health status, or the balance of load distribution should be improved: The logic for obtaining the load regulation balance index is as follows: Get the load of the i-th lithium battery module after load adjustment and load before adjustment , get the health status of the i-th module , calculate the regulated load response value: , get the change of module i load in the historical period , get the current module load and the sum of the current loads of all modules , calculate the dynamic adjustment value: ; Get the actual load of the i-th lithium battery module after the intelligent scheduling algorithm is executed , calculate the average load of all modules: , where n is the total number of lithium battery modules, and the load regulation balance index is calculated as: .

[0020] It should be noted that the health status of the module is the output result of the lithium battery module evaluation through the nonlinear regression model as the observation data of the Kalman filter, and the health status of the lithium battery is predicted to obtain the estimated health status value; the setting of the historical period is determined according to actual needs, and the range of the historical period can be a historical day, a historical month, etc.

[0021] The response efficiency dispatch index (RSEI) is used to measure the response effect of the lithium battery module after the load adjustment of the intelligent dispatch algorithm and the overall effectiveness of the dispatch strategy. It combines the health status response degree of each lithium battery module to the load adjustment and the success rate of the dispatch strategy, reflects the system's operating performance after the load adjustment, and determines the changes in the health status of each lithium battery module after the load adjustment, that is, whether the dispatch algorithm can effectively improve the health status of the lithium battery. By quantifying the changes in the health status after the load adjustment, the response efficiency dispatch index can determine whether the lithium battery module has responded well to the load adjustment; The response efficiency scheduling index directly reflects the success of the scheduling strategy. If the index is high, it means that the system operates well after load adjustment and the health status of the lithium battery module has improved. This index can be used as an important reference for whether the scheduling strategy needs to be optimized; if the response efficiency scheduling index is low, it means that there may be problems with the current scheduling strategy, and the improvement in health status after load adjustment is not significant. System managers can optimize the scheduling algorithm and adjust the load distribution ratio of each module based on the changes in this index; The response efficiency dispatch index can quantify the positive or negative impact of load adjustment on the health of lithium battery modules, helping the system better control the degradation rate of lithium batteries. Through this index, managers can judge the effect of load adjustment on extending the life of lithium battery packs, thereby optimizing lithium battery management solutions. The logic for obtaining the response efficiency scheduling index is as follows: Get the maximum load value of the lithium battery module , get the health status value of the jth module after load adjustment and the health status value before load adjustment , calculate the change in the health status of the jth module after load adjustment, and the calculation formula is: , the module response efficiency value is calculated: , where is the current load of the jth module, and m is the total number of modules; Get the load value of the jth module after load adjustment The load value before load adjustment , calculate the load change: , calculate the effective value of scheduling: , the response efficiency scheduling index is calculated based on the scheduling effective value and the module response efficiency value, and the calculation expression is: .

[0022] It should be noted that the health status value is obtained by monitoring the voltage, internal resistance, temperature and other key parameters of the lithium battery module, and calculating the health status of the module after data processing. This parameter reflects the remaining life or performance level of the lithium battery; the load value is calculated by monitoring the output power or input power of each lithium battery module, combined with the current, voltage and load conditions, and the load value represents the workload of each module; by calculating the change in the health status of the lithium battery module before and after the load adjustment, the quantitative result of the improvement or deterioration of the health status is obtained, that is, the change in health status.

[0023] The adjustment balance information and the dispatch response information are combined to generate the dispatch management coefficient, that is, the obtained load adjustment balance index and the response efficiency dispatch index are combined to generate the dispatch management coefficient. The expression is: , where , is the preset proportional coefficient of the load regulation balance index and the response efficiency dispatch index, and , Both are greater than 0.

[0024] The specific method of jointly generating the scheduling management coefficient may involve multiple algorithms and models, which depends on the actual situation and application requirements. In this embodiment, the load regulation balance index and the response efficiency scheduling index can be combined by weighted summation to generate a comprehensive scheduling management coefficient. This scheduling management coefficient can be used as an input for the effect of using an intelligent scheduling algorithm for lithium batteries, and is used to determine the load optimization of the lithium battery by the scheduling algorithm.

[0025] It should be noted that the size of the preset proportional coefficient is a specific value obtained by quantifying each parameter. In order to facilitate subsequent comparison, the size of the coefficient depends on the amount of sample data and the preliminary setting of the corresponding preset proportional coefficient for each group of sample data by technical personnel in this field. It is not unique, as long as it does not affect the proportional relationship between the parameter and the quantized value. For example, the load regulation balance index is proportional to the scheduling management coefficient. The load regulation balance index and the response efficiency scheduling index are normalized to have the same dimension and range. This can be achieved by subtracting the mean from the original data and dividing it by the standard deviation, or mapping the data to the range of [0, 1].

[0026] The larger the load regulation balance index and the larger the response efficiency scheduling index, the larger the jointly generated scheduling management coefficient, indicating that the scheduling strategy performs well in load regulation and health status response. The health status of the lithium battery module after load adjustment is significantly improved. The scheduling strategy has a significant effect on delaying the degradation of lithium batteries and improving the overall operating efficiency. This shows that the load regulation not only maintains the balance, but also reasonably distributes the load to the lithium battery modules with different health status, preventing some modules from accelerating degradation due to excessive load, and extending the overall service life of the lithium battery group. The system achieves efficient operation under intelligent scheduling, and the lithium batteries with poor health status are properly protected, while the lithium battery modules with good health status bear more loads, effectively allocating system resources and improving the operating efficiency and safety of the entire lithium battery group. The smaller the load regulation balance index and the response efficiency dispatch index, the smaller the dispatch management coefficient generated by the combination, indicating that the intelligent dispatch strategy has deficiencies in load regulation intensity, load distribution balance, and response efficiency, and the system has failed to reach an optimized state. This may lead to accelerated degradation of lithium battery modules, low system operation efficiency, and insufficient resource utilization, affecting the overall performance and service life of the lithium battery pack; Compare the generated scheduling management coefficient with the preset management evaluation threshold to generate a strategy scheduling stability signal and a scheduling abnormal fluctuation signal; After obtaining the scheduling management coefficient, the scheduling management coefficient is compared with the management evaluation threshold; If the scheduling management coefficient is greater than or equal to the management evaluation threshold, a strategy scheduling stability signal is generated, indicating that the current scheduling strategy of the system is running stably, and the load regulation, balance and module response efficiency are good. The system can work at a higher efficiency and successfully reduce the load of the degradation acceleration module. At the same time, the utilization rate of the modules in good health is improved. The health of the entire system is being effectively improved, and the load distribution between the modules is reasonable, avoiding the situation where the load is too heavy and concentrated in individual modules. Modules with large differences in health status have been properly adjusted, and lithium battery pack resources can be more fully utilized; If the scheduling management coefficient is less than the management evaluation threshold, an abnormal scheduling fluctuation signal is generated, indicating that there are problems with the scheduling strategy in terms of load regulation, balance or response efficiency. The system fails to maintain stable load distribution and efficient operation. The execution of the scheduling strategy is not ideal, especially the modules with accelerated degradation may still bear a large load and fail to be effectively protected, which may accelerate the degradation of the lithium battery pack. The scheduling algorithm needs to be further optimized to deal with this abnormal situation.

[0027] When an abnormal scheduling fluctuation signal is generated, it means that the current scheduling strategy has problems in load regulation, balance or response efficiency. In order to restore the stability of the system and optimize the load distribution of lithium batteries, the scheduling strategy must be adjusted. The following are the specific adjustment steps: Use real-time monitoring data to analyze the source of abnormal scheduling fluctuation signals, determine whether there is insufficient or excessive load regulation, and some modules may not receive sufficient load adjustment, resulting in overload or insufficient load. Analyze whether the load distribution among the lithium battery modules is unbalanced. Some modules may be overloaded while other modules are underloaded. Re-evaluate the response of the lithium battery modules to load adjustment and confirm whether the health status improvement effect after scheduling is significant. At the same time, the anomaly detection algorithm or historical data comparison can be used to quickly locate the source of the anomaly. For example, the anomaly point can be detected through trend analysis, and the module or parameter that causes the abnormal fluctuation can be accurately determined in combination with the real-time lithium battery load changes. After each load adjustment, the health status of the lithium battery module is monitored in real time, feedback data of the adjusted load is obtained, the effect of the adjustment is evaluated in time, a feedback loop is introduced, and real-time data is input into the scheduling algorithm for dynamic adjustment to ensure that the load regulation strategy can adapt to real-time changes; After completing the adjustment of the dispatch parameters, continue to monitor the dispatch management coefficient (SMC). If the SMC continues to rise and stabilizes at a high level, it means that the adjustment of the system dispatch strategy is effective; if it still fluctuates, further adjustment is required. When the dispatch fluctuates abnormally, the system can generate dynamic warnings to remind the operation and maintenance personnel to intervene in time. Through the automated abnormal detection and warning mechanism, it ensures that the system can respond quickly and take corresponding adjustment measures when fluctuations occur; When an abnormal scheduling fluctuation signal is generated, it is necessary to quickly identify the source of the problem and dynamically adjust the scheduling strategy. By adjusting the load regulation weight, optimizing balance, introducing feedback mechanism, optimizing response efficiency and other steps, the stability of the system can be restored, and the long-term effectiveness of the intelligent scheduling strategy can be ensured through continuous monitoring and feedback mechanism.

[0028] It should be noted that the threshold information in this embodiment is pre-set by professionals and will not be explained in detail here. Some parameter English letters in the embodiments have the same situation, but different meanings are explained when used, which will not be explained one by one here.

[0029] The present invention uses IoT sensors combined with edge computing devices to perform real-time data collection and preprocessing on multi-module lithium battery packs, ensuring high quality and high efficiency when data is transmitted to the cloud computing platform. The spectrum characteristics of the battery module are extracted through Fourier transform, and the health status of the lithium battery is evaluated in combination with a nonlinear regression model, which can more accurately capture the degradation trend of the lithium battery. The piecewise nonlinear regression model is used to determine the degradation rate of the battery module, especially to identify the accelerated degradation phenomenon, so as to provide early warning of the risk of imminent failure of the lithium battery. The Kalman filter is combined with nonlinear regression to identify the accelerated degradation point of the lithium battery in real time, so that the system can respond quickly and adjust the load distribution.

[0030] After identifying the degradation acceleration point, the intelligent scheduling algorithm optimizes the load distribution of each battery module, reduces the load of the degraded module, and slows down its degradation rate. By analyzing the load regulation information, the present invention generates different scheduling signals to guide the system to perform strategic scheduling management. This innovative method significantly improves the stability and life of the lithium battery pack, ensures the efficient operation of the battery system in load scheduling and energy management, and optimizes the accuracy of the scheduling strategy.

[0031] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0032] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0033] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0034] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0035] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0036] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-module lithium battery pack system based on cloud computing management, characterized by: It includes a lithium battery data acquisition module, a battery degradation analysis module, a degradation acceleration point determination module, and a strategy management module, and each module is connected through signals; The lithium battery data acquisition module is used to collect multi-module lithium battery pack data through IoT sensors, pre-process the data using edge computing devices, and transmit the pre-processed data to the cloud computing platform; The battery degradation analysis module is used to extract features from preprocessed data using Fourier transform, build a nonlinear regression model to evaluate the health status of lithium batteries, and use piecewise nonlinear regression to determine the degradation rate of lithium batteries and the accelerated degradation phenomenon of lithium battery modules; Degradation acceleration point determination module, which is used to identify the degradation acceleration point of lithium batteries in real time through Kalman filtering combined with nonlinear regression model, and optimize load distribution through intelligent scheduling algorithm after identifying the degradation acceleration point; The strategy management module is used to analyze the load scheduling of the intelligent scheduling algorithm, obtain the load regulation information in the scheduling optimization process, analyze the load regulation information and generate different signals, and perform strategy scheduling management according to the generated different signals; The degradation acceleration point determination module is used to identify the lithium battery degradation acceleration point in real time through Kalman filtering combined with a nonlinear regression model, and optimize the load distribution through an intelligent scheduling algorithm after identifying the degradation acceleration point. The specific steps are as follows: Define the initial health status estimate and covariance matrix of the lithium battery module, and establish the initial health status estimate through the historical data of the lithium battery and the preliminary prediction results of the nonlinear regression model; The output results of the nonlinear regression model for evaluating the lithium battery module are used as the observation data of the Kalman filter to predict the health status of the lithium battery; By monitoring the rate of change of the health status estimate in real time, the time point at which degradation accelerates can be identified; When the rate of change of the state estimation value of the Kalman filter exceeds a preset threshold within a time, it is determined that the current lithium battery module is in an accelerated degradation state, and the time is marked as a degradation acceleration point; When the accelerated degradation point of the lithium battery module is identified, the cloud computing platform triggers the intelligent scheduling algorithm to optimize the use strategy of the lithium battery module.

2. The multi-module lithium battery pack system based on cloud computing management according to claim 1, characterized in that: It is used to collect multi-module lithium battery pack data through IoT sensors, pre-process the data using edge computing devices, and transmit the pre-processed data to the cloud computing platform. The specific steps are as follows: Install IoT sensors on lithium battery modules to collect voltage, current, temperature and internal impedance data; Edge computing devices compress and encode data, filter out abnormal data, suppress noise in collected data, prioritize sensor data based on its importance and time sensitivity, and upload data of different priorities to the cloud computing platform in a hierarchical manner.

3. The multi-module lithium battery pack system based on cloud computing management according to claim 2, characterized in that: It is used to extract features from preprocessed data using Fourier transform, build a nonlinear regression model to evaluate the health status of lithium batteries, and use piecewise nonlinear regression to determine the degradation rate of lithium batteries and the accelerated degradation phenomenon of lithium battery modules. The specific steps include: Perform Fourier transform on the preprocessed data to extract feature data, and then perform a denoising process based on wavelet transform on the feature data to ensure that high-frequency interference components in the measurement are removed; Combining the voltage, current, and temperature change rates in historical data, nonlinear feature variables are constructed, and various features are input into the nonlinear regression model in the form of multidimensional feature vectors to determine the estimated health status of the lithium battery module; Based on the results of nonlinear mapping, a piecewise nonlinear regression analysis is performed, and the health status is iteratively updated based on the current and historical status of the lithium battery module. The changes in the lithium battery during the degradation process are determined based on the iterative update results. The degradation rate is adjusted nonlinearly, an adjustment factor of the degradation rate is introduced, the sudden fluctuations in the health status of the lithium battery are determined, and the degradation rate is re-estimated in the multi-module lithium battery pack according to the change of the adjustment factor.

4. The multi-module lithium battery pack system based on cloud computing management according to claim 3, characterized in that: It is used to analyze the load dispatching situation of the intelligent dispatching algorithm and obtain the load regulation information in the dispatching optimization process, including the following steps: Obtaining load regulation information during the dispatch optimization process, the load regulation information including regulation balance information and dispatch response information; The fluctuation trend information includes the load regulation balance index, and the load response information includes the response efficiency dispatch index; The obtained load regulation balance index and response efficiency dispatch index are combined to generate a dispatch management coefficient; The load regulation balance index and response efficiency dispatch index are directly proportional to the dispatch management coefficient.

5. The multi-module lithium battery pack system based on cloud computing management according to claim 4, characterized in that: Analyze the load regulation information and generate different signals, and perform strategic scheduling management according to the generated different signals, including the following steps: Compare the generated forecast adjustment factor with the set management assessment threshold; If the forecast adjustment coefficient is greater than or equal to the management evaluation threshold, a strategy scheduling stability signal is generated, indicating that the scheduling strategy is running stably and no additional adjustment is required; If the predicted adjustment coefficient is less than the management evaluation threshold, a scheduling abnormal fluctuation signal is generated, indicating that the scheduling strategy of the optimization scheduling algorithm should be adjusted in terms of load regulation, balance or response efficiency.

Images