Retired battery active equalization method and online detection device based on direct current converter and time division multiplexing

Abstract

The invention relates to the technical field of battery management, in particular to a decommissioned battery active equalization method and online detection device based on a direct current converter and time division multiplexing, and the method comprises the following steps: S1, collecting the health state, capacity and historical use data of a main control chip and a sensor; s2, classifying the battery pack through a preset classification algorithm; s3, dynamically adjusting voltage output; s4, adjusting parameters of the switching network based on the voltage in the S3; s5, short pulse current is applied to the battery pack, and current flowing in the equalization process is controlled; s6, automatically adjusting the working parameters of the DC converter and the configuration of the switching network; and S7, automatically adjusting the parameters in the steps S3 to S6 according to an evaluation result. According to the invention, by implementing the active pulse modulation technology and real-time battery health monitoring, the energy efficiency and safety of the retired battery in secondary utilization are remarkably improved, and meanwhile, the battery maintenance is optimized and the service life of the battery is prolonged through the intelligent management system.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management, and in particular to a retired battery active balancing method and an online detection device based on a DC converter and time-sharing multiplexing. Background Art

[0002] With the rapid development of the electric vehicle industry, a large number of batteries are retired after reaching their design life. However, these retired batteries are often not completely failed and still have available remaining capacity. Therefore, their secondary use has become an important issue. The reuse of retired batteries can not only reduce environmental pollution, but also improve resource utilization efficiency. However, these batteries often face problems such as capacity attenuation and uneven performance during use. These problems have brought significant challenges to the effective management and reuse of batteries.

[0003] Existing technologies have multiple limitations in handling the equalization and health monitoring of retired batteries. First, traditional battery management systems usually focus on new batteries and are not sufficient to handle the complex situation of aging batteries caused by disrepair and irregular use. Second, existing battery equalization technologies, such as passive equalization or simple active equalization methods, are inefficient and cannot meet the needs of fast and efficient equalization of retired batteries during secondary utilization. In addition, the monitoring of the real-time health status of the battery pack often lacks accuracy and timeliness, which limits the safety and reliability of battery use.

[0004] Therefore, there is an urgent need for an advanced battery management system that can achieve efficient energy transfer, accurate health monitoring, and automatically adjust the balancing strategy to adapt to changes in battery status. Such a system will provide greater flexibility and higher economic benefits for the reuse of retired batteries. Summary of the invention

[0005] Based on the above objectives, the present invention provides a retired battery active balancing method and an online detection device based on a DC converter and time-sharing multiplexing.

[0006] A method for active balancing of retired batteries based on DC converter and time-sharing multiplexing comprises the following steps:

[0007] S1: Automatically detect retired battery packs connected to the system through the main control chip and sensors, and collect their health status, capacity and historical usage data;

[0008] S2: Classify the battery packs according to the data collected in step S1 using a preset classification algorithm, and analyze their energy requirements and balancing priorities;

[0009] S3: using a preset buck-boost isolated DC converter, dynamically adjusting the voltage output according to the analysis result of step S2 to meet the energy transfer requirements between different battery groups;

[0010] S4: Control the battery pack to access the input or output of the DC converter through a preset switch network, and adjust the parameters of the switch network based on the voltage of S3 to maximize the energy transmission efficiency;

[0011] S5: applying active pulse modulation technology to apply short pulse current to the battery pack according to the switch network configuration of step S4 to control the current flow during the balancing process;

[0012] S6: Integrated temperature sensor to continuously monitor the operating temperature of the battery pack and automatically adjust the operating parameters of the DC converter and the configuration of the switch network according to the pulse modulation results of S5 to prevent the battery from overheating;

[0013] S7: Use the main control chip to inject sine waves of different frequencies into the battery, collect response signals, and combine cloud data analysis to evaluate the battery status in real time according to the preset health assessment algorithm. Automatically adjust the parameters of steps S3 to S6 according to the assessment results to optimize the balancing strategy.

[0014] Furthermore, the S1 specifically includes:

[0015] S11: Configure multiple sensors in the system, including voltage sensors, current sensors, and temperature sensors. Each sensor is calibrated and initialized before being connected to the system to ensure the accuracy and reliability of data collection;

[0016] S12: using the main control chip to automatically detect the battery pack connected to the system through the interface circuit, and identifying the model and manufacturing information of each battery pack by reading the unique identifier of the battery pack;

[0017] S13: The main control chip instructs each sensor to collect real-time data of the battery pack. The voltage sensor is used to measure the voltage of each battery cell, the current sensor is used to record the charge and discharge current, and the temperature sensor is used to monitor the battery temperature.

[0018] S14: The main control chip accesses the historical usage data of the battery pack through built-in or external storage, including the number of charge and discharge cycles, historical fault records and maintenance history;

[0019] S15: Based on the real-time and historical data collected in steps S13 and S14, the main control chip uses a preset algorithm to evaluate the health status. The preset algorithm includes state estimation, capacity decay analysis and life prediction. The specific calculation formula used for the evaluation is: Among them, SOH represents the percentage of battery health status, C current is the current measured capacity of the battery, and C ratedis the rated capacity of the battery; based on historical failure data and usage frequency, the expected life reduction of the battery is calculated. The specific calculation formula is: EOL = C initial ·e -k·n , where EOI represents the expected service life of the battery, C initial is the initial capacity, k is the decay constant, and n represents the number of charge and discharge cycles.

[0020] Furthermore, the S2 specifically includes:

[0021] S21: Standardize all battery data collected in S1, including removing outliers, filling missing data, and normalizing data;

[0022] S22: Apply a preset clustering algorithm to classify the battery packs. Specifically, the K-means clustering algorithm is used to classify the batteries into different categories based on the health status, historical usage data, and battery capacity characteristics of the batteries. The calculation formula used in the clustering process is: Among them, C i represents the i-th category; x represents the characteristic data point of a single battery pack; μ i Category C i The center point of ; k is the number of preset categories;

[0023] S23: Based on the classification results of the battery packs, the typical energy requirements of each category of battery packs are analyzed. Specifically, the average capacity loss rate and typical load response characteristics of each category of battery packs are calculated using a statistical method. The calculation formula is: Among them, E avg represents the average energy requirement; E j represents the actual energy demand of the jth battery pack in the category; n is the number of battery packs in the category;

[0024] S24: Determine the balancing priority of each battery group based on the health status and energy demand analysis results of the battery group. Specifically, a weighted scoring system is used to combine the health status and energy demand into a balancing priority score. The calculation formula is: P = w1×SOH+w2×E need , where P is the balancing priority score; SOH is the health status percentage of the battery pack; E need is the energy demand of the battery pack; w1 and w2 are the weight coefficients of health status and energy demand, respectively.

[0025] Furthermore, the S3 specifically includes:

[0026] S31: Based on the classification of each battery pack and the energy demand analysis obtained in S2, the preset control system will calculate the ideal voltage level required for each battery pack;

[0027] S32: The buck-boost isolated DC converter automatically selects a boost or buck mode based on the comparison between the current voltage of each battery pack and the ideal voltage calculated in step S31. The specific selection is based on the following judgment logic:

[0028] Where Vcurrent is the current voltage of the battery pack, V 理想 It is the ideal voltage calculated based on energy requirements;

[0029] S33: The DC converter dynamically adjusts its output voltage according to the mode selected in S32 to ensure that the voltage of each battery pack gradually approaches or reaches its ideal voltage;

[0030] S34: During the entire adjustment process, the voltage and current status of each battery pack is continuously monitored, and a feedback control algorithm is used to optimize the voltage regulation process to cope with the impact of battery status changes or external load fluctuations. The specific formula for the voltage regulation is:

[0031] Among them, V 调整 Represents the output of the voltage adjustment control signal; K p is the proportional gain; K i is the integral gain; K d is the differential gain; e(t) is the error or deviation, which represents the difference between the target voltage and the actual voltage at the instant of time t; ∫e(t)dt represents the sum of the deviation e(t) integrated over time; Represents the rate of change of the deviation e(t).

[0032] Furthermore, the S4 specifically includes:

[0033] S41: using a preset switch network, the network consisting of electronic switches, including transistors, relays or solid-state switches, each switch being configured to control the connection between the battery pack and the DC converter, so that the battery pack can be connected to the input or output of the converter as needed;

[0034] S42: Based on the voltage regulation parameters in step S3, the control logic of the switch network will determine whether each battery pack should be connected to the input or output of the DC converter, making the decision based on the following rules:

[0035] When the target voltage of the battery pack is higher than the current voltage, the battery pack is connected to the output terminal of the converter;

[0036] When the target voltage of the battery pack is lower than the current voltage, the battery pack is connected to the input terminal of the converter;

[0037] S43: The switch network optimizes the energy flow by controlling the access points of each battery pack, including adjusting the connection time and sequence to ensure that the electric energy flows from the high-energy battery pack to the low-energy battery pack, thereby maximizing the energy conversion efficiency of the overall system;

[0038] S44: After the switch network configuration is implemented, the energy flow and voltage changes between the battery packs will be continuously monitored, and the switch state will be dynamically adjusted based on real-time data to adapt to changes in battery performance or changes in external load conditions.

[0039] Furthermore, the S5 specifically includes:

[0040] S51: according to the specific configuration of each battery pack connected to the DC converter obtained in step S4, set the parameters of the active pulse modulation technology, including pulse width, pulse amplitude and pulse frequency;

[0041] S52: using an electronic control unit to generate a customized short pulse current signal, the short pulse current signal will match the parameters set in step S51, ensuring that each pulse can effectively promote or control the flow of current to achieve a balancing goal;

[0042] S53: applying the generated pulse signal to the corresponding battery pack through the switch network, and controlling the precise flow of current between the battery packs by adjusting the pulse application time and duration, thereby achieving effective energy balance;

[0043] S54: continuously monitoring the current and voltage response of each battery group after receiving the pulse, and the monitoring data is used to evaluate the pulse modulation effect and provide necessary feedback information for subsequent pulse adjustment;

[0044] S55: Based on the real-time feedback data collected in step S54, the parameters of pulse width, amplitude and frequency are dynamically adjusted.

[0045] Furthermore, the S6 specifically includes:

[0046] S61: A temperature sensor is integrated into each battery pack to monitor the battery's operating temperature in real time. The temperature sensor is placed at the hot spot of the battery pack to ensure the accuracy and representativeness of the temperature reading;

[0047] S62: continuously collects temperature data of the battery pack from the temperature sensor, and uses a high-precision analog-to-digital converter to read the analog signal of the sensor and convert the analog signal into digital data for easy processing and analysis;

[0048] S63: Analyze the collected temperature data in real time and compare it with the preset temperature threshold. The set threshold is determined based on the thermal stability and safety standards of the battery material. The specific calculation formula is: T max =Tambient +ΔT, where T max is the maximum safe operating temperature of the battery, T ambient is the ambient temperature, ΔT is a predetermined temperature increment based on the battery type and configuration;

[0049] S64: When the temperature of a battery pack approaches or exceeds the threshold value T max When the pulse modulation parameters set in step S5 are automatically adjusted, including reducing the pulse amplitude or frequency to reduce heat generation and prevent the battery from overheating;

[0050] S65: According to the temperature data feedback, adjust the operating parameters and switch network configuration of the DC converter, including adjusting the output voltage or changing the connection configuration of the battery pack to control the current flow direction and intensity.

[0051] Furthermore, the S7 specifically includes:

[0052] S71: The main control chip controls the signal generator to generate sine waves of different frequencies, and the sine waves are accurately injected into each battery pack. The frequency range is set according to the typical response characteristics of the battery;

[0053] S72: The response of the battery pack to the injected sine wave is recorded by the sensor, and the response signal is converted into a digital form and transmitted to the main control system in real time for analysis;

[0054] S73: The main control chip uses a preset signal processing algorithm to analyze the collected response signal and extract features including impedance, phase angle and amplitude. The feature data is analyzed by advanced algorithms to evaluate the health status of the battery. Specifically, the impedance change of the battery is calculated using the following formula: Where Z(f) is the impedance at frequency f, V(f) and I(f) are the voltage and current responses at the injected frequency, respectively;

[0055] S74: The collected characteristic data is uploaded to the cloud, where the data analysis platform uses a machine learning algorithm to evaluate the battery status, specifically a neural network algorithm to predict the health of the battery and calculate its remaining service life;

[0056] S75: Based on the analysis results from the cloud, the main control chip receives an evaluation report on the battery health status and adjusts the parameters in S3 to S6 in real time, including adjusting the voltage output of the DC converter, modifying the configuration of the switch network, and adjusting the parameters of the pulse modulation.

[0057] Furthermore, the S74 specifically includes:

[0058] S741: Before training and predicting the neural network model, preprocess the battery response data collected from S73, including normalization, to standardize the voltage, current and impedance data of the battery to a predetermined range, so as to optimize the performance and prediction accuracy of the neural network;

[0059] S742: Selecting features of the battery health status from the preprocessed data, including the frequency response of the impedance, the number of charge and discharge cycles, and the historical temperature change, and using the features as inputs to the neural network;

[0060] S743: Configure a multi-layer perceptron neural network, including an input layer, several hidden layers and an output layer. Nodes between each layer transmit information through weighted connections. The specific formula of the neural network is: y=f(W·x+b), where x is the input feature vector, W is the weight matrix, b is the bias vector, and f is the activation function;

[0061] S744: Use historical data sets to train the neural network, including battery data with known health status as training samples. The training process optimizes weights and biases through the back propagation algorithm to minimize the difference between the predicted value and the actual value. The loss function used is the mean square error. The specific formula is: Among them, L is the loss function, y i is the actual value of the sample, is the predicted value, n is the number of samples;

[0062] S745: Use the trained neural network to predict the health status of the newly collected battery data. The prediction result is used to estimate the remaining service life of the battery. By combining the predicted health status with the typical life curve of the battery, the expected life reduction is calculated. The specific algorithm is: RUL = L typical ×(1-SOH predicted ), where RUL is the remaining useful life, L typical is the typical life of the battery type, SOH predicted is the health status ratio predicted by the neural network.

[0063] An online detection device for an active equalization method for retired batteries, used to implement the above-mentioned active equalization method for retired batteries based on a DC converter and time-sharing multiplexing, includes the following modules:

[0064] Data acquisition module: used to collect key parameters of the connected battery pack in real time, including voltage, current, and temperature data;

[0065] Signal generation and injection module: generates and injects sine waves of different frequencies into each battery pack for subsequent health status assessment;

[0066] Signal analysis module: receives the battery response signal and analyzes the signal to evaluate the health of the battery;

[0067] Data analysis and diagnosis module: Analyzes the data received from the signal analysis module, uses a preset health assessment algorithm to determine the battery status, and predicts the remaining service life;

[0068] Communication control module: adjusts the battery balancing strategy and the operating parameters of the DC converter based on the diagnosis results;

[0069] User Interface Module: Used to provide a user interface for operators to monitor battery status and adjust operating parameters, while displaying real-time diagnostic and prediction results.

[0070] Beneficial effects of the present invention:

[0071] The present invention, through active pulse modulation technology and intelligent balancing strategy, can effectively improve the energy conversion efficiency of retired batteries during the reuse process. It also reduces energy loss and ensures the maximum energy utilization of batteries during the balancing process by precisely controlling the energy flow between battery packs. This feature is of great significance for improving the economic value and environmental sustainability of retired batteries.

[0072] The present invention evaluates the health status of the battery in real time through an integrated online monitoring device, and combines cloud data analysis to provide accurate health diagnosis and life prediction. This real-time monitoring and prediction capability enables the battery management system to adjust the balancing strategy in a timely manner, thereby preventing potential risks such as battery over-discharge and overheating, and significantly improving the safety and reliability of battery use.

[0073] The present invention, through an automated feedback mechanism, the system not only reduces the workload of daily maintenance, but also ensures that the battery pack maintains optimal performance throughout its entire life cycle. In addition, the application of the system helps to reduce operating costs caused by reduced battery performance and improve overall economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0075] Figure 1 A schematic diagram of an active balancing method for retired batteries according to an embodiment of the present invention;

[0076] Figure 2 Schematic diagram of an online detection device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0078] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. The words "include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0079] like Figure 1 As shown, a retired battery active balancing method based on DC converter and time-sharing multiplexing includes the following steps:

[0080] S1: Automatically detect retired battery packs connected to the system through the main control chip and sensors, and collect their health status, capacity and historical usage data;

[0081] S2: Classify the battery packs according to the data collected in step S1 using a preset classification algorithm, and analyze their energy requirements and balancing priorities;

[0082] S3: using a preset buck-boost isolated DC converter, dynamically adjusting the voltage output according to the analysis result of step S2 to meet the energy transfer requirements between different battery groups;

[0083] S4: Control the battery pack to access the input or output of the DC converter through a preset switch network, and adjust the parameters of the switch network based on the voltage of S3 to maximize the energy transmission efficiency;

[0084] S5: applying active pulse modulation technology to apply short pulse current to the battery pack according to the switch network configuration of step S4 to control the current flow during the balancing process;

[0085] S6: Integrated temperature sensor to continuously monitor the operating temperature of the battery pack and automatically adjust the operating parameters of the DC converter and the configuration of the switch network according to the pulse modulation results of S5 to prevent the battery from overheating;

[0086] S7: Use the main control chip to inject sine waves of different frequencies into the battery, collect response signals, and combine cloud data analysis to evaluate the battery status in real time according to the preset health assessment algorithm. Automatically adjust the parameters of steps S3 to S6 according to the assessment results to optimize the balancing strategy.

[0087] S1 specifically includes:

[0088] S11: Configure multiple sensors in the system, including voltage sensors, current sensors, and temperature sensors. Each sensor is calibrated and initialized before being connected to the system to ensure the accuracy and reliability of data collection;

[0089] S12: using the main control chip to automatically detect the battery pack connected to the system through the interface circuit, and identifying the model and manufacturing information of each battery pack by reading the unique identifier of the battery pack (such as RFID or barcode);

[0090] S13: The main control chip instructs each sensor to collect real-time data of the battery pack. The voltage sensor is used to measure the voltage of each battery cell, the current sensor is used to record the charge and discharge current, and the temperature sensor is used to monitor the battery temperature. The collected data is used to evaluate the instant working status of the battery pack.

[0091] S14: The main control chip accesses the historical usage data of the battery pack through built-in or external storage, including the number of charge and discharge cycles, historical fault records and maintenance history;

[0092] S15: Based on the real-time and historical data collected in steps S13 and S14, the main control chip uses a preset algorithm to evaluate the health status. The preset algorithm includes state estimation, capacity decay analysis and life prediction. The specific calculation formula used for the evaluation is: Among them, SOH represents the percentage of battery health status, C current is the current measured capacity of the battery, and C rated is the rated capacity of the battery; in addition, based on historical failure data and usage frequency, the expected life reduction of the battery is calculated. The specific calculation formula is: EOL = C initial ·e -k·n , where EOL represents the expected service life of the battery, C initial is the initial capacity, k is the decay constant, and n represents the number of charge and discharge cycles.

[0093] S2 specifically includes:

[0094] S21: Standardize all battery data (voltage, current, temperature and historical usage data) collected in S1, including removing outliers, filling missing data and normalizing data for effective classification and analysis;

[0095] S22: Apply a preset clustering algorithm to classify the battery pack. Specifically, the K-means clustering algorithm is used to classify the batteries into different categories based on the battery health status (SOH), historical usage data, and battery capacity characteristics. The calculation formula used in the clustering process is: Among them, C irepresents the i-th category, which is the result of dividing the battery pack into different groups; x represents the characteristic data points of a single battery pack, including multi-dimensional data such as health status (SOH), historical usage data, and battery capacity; μ i Category C i The center point is the mean of all battery pack features in this category; k is the number of preset categories, which is determined in advance based on the total number of battery packs and feature differences. The clustering algorithm, such as Kmeans, determines which category each battery pack should belong to by minimizing the sum of intra-class distances in the feature space;

[0096] S23: Based on the classification results of the battery packs, the typical energy requirements of each category of battery packs are analyzed. Specifically, the average capacity loss rate and typical load response characteristics of each category of battery packs are calculated using a statistical method. The calculation formula is: Among them, E avg The average energy demand is calculated for all battery packs in a certain category to estimate the energy consumption of the battery packs in this category under typical conditions. j represents the actual energy demand of the jth battery pack in the category, which is usually evaluated by historical energy consumption data; n is the number of battery packs in the category, which is used to calculate the average value;

[0097] S24: Determine the balancing priority of each battery group based on the health status and energy demand analysis results of the battery group. Specifically, a weighted scoring system is used to combine the health status and energy demand into a balancing priority score. The calculation formula is: P = w1×SOH+w2×E need , where P is the balancing priority score. A priority is set for each battery pack based on the comprehensive health status and energy demand. The battery pack with a high priority will be given priority for energy balancing; SOH is the health status percentage of the battery pack, which is used to reflect the ratio of the current capacity of the battery to the original capacity; E need is the energy demand of the battery pack, obtained from step S23; w1 and w2 are weight coefficients of health status and energy demand, respectively, which are preset according to the importance of the actual balancing strategy.

[0098] S3 specifically includes:

[0099] S31: Based on the classification of each battery group and the energy demand analysis obtained in S2, the preset control system will calculate the ideal voltage level required for each battery group, and the ideal voltage level ensures the effective transfer of energy from the battery group with higher power to the battery group with lower power;

[0100] S32: The buck-boost isolated DC converter automatically selects a boost or buck mode based on the comparison between the current voltage of each battery pack and the ideal voltage calculated in step S31. The specific selection is based on the following judgment logic:

[0101] Where Vcurrent is the current voltage of the battery pack, V 理想 It is the ideal voltage calculated based on energy requirements;

[0102] S33: The DC converter dynamically adjusts its output voltage according to the mode selected in S32 to ensure that the voltage of each battery pack gradually approaches or reaches its ideal voltage. The adjustment specifically uses a closed-loop control system to provide real-time feedback of voltage readings and accurately control output to maintain the accuracy and stability of voltage regulation;

[0103] S34: During the entire adjustment process, the voltage and current status of each battery pack is continuously monitored, and a feedback control algorithm (such as a PID controller) is used to optimize the voltage regulation process to cope with the impact of battery status changes or external load fluctuations. The specific formula for the voltage regulation is:

[0104] Among them, V 调整 represents the output of the voltage regulation control signal, which is the actual control input that the DC converter applies to the battery pack to achieve the desired voltage; K p K is the proportional gain, a parameter of the PID controller, which is used to adjust the direct impact of the deviation e(t) on the control output. The proportional gain determines the speed and intensity of the response to the current error. i K is the integral gain, another parameter of the PID controller, which is used to accumulate the deviation e(t) to eliminate the steady-state error. The integral action helps the system cope with long-term continuous deviations. d is the differential gain, which is the third parameter of the PID controller. It is used to predict the future trend of the deviation e(t) and to suppress the oscillation and overshoot of the system by adjusting the control output; e(t) is the error or deviation, which means the difference between the target voltage and the actual voltage at the instant of time t. It is the key parameter for adjusting the target in the feedback control system; ∫e(t)dt represents the sum of the integral of the deviation e(t) over time, which reflects the cumulative error and is used to calculate the integral control action; It represents the rate of change of the deviation e(t), that is, the derivative of the deviation with time, which is used to calculate the differential control action.

[0105] S4 specifically includes:

[0106] S41: using a preset switch network, the network consisting of electronic switches, including transistors, relays or solid-state switches, each switch being configured to control the connection between the battery pack and the DC converter, so that the battery pack can be connected to the input or output of the converter as needed;

[0107] S42: Based on the voltage regulation parameters in step S3, the control logic of the switch network will determine whether each battery pack should be connected to the input or output of the DC converter, making the decision based on the following rules:

[0108] When the target voltage of the battery pack is higher than the current voltage, the battery pack is connected to the output terminal of the converter;

[0109] When the target voltage of the battery pack is lower than the current voltage, the battery pack is connected to the input terminal of the converter;

[0110] S43: The switch network optimizes the energy flow by controlling the access points of each battery pack, including adjusting the connection time and sequence to ensure that the electric energy flows from the high-energy battery pack to the low-energy battery pack, thereby maximizing the energy conversion efficiency of the overall system;

[0111] S44: After the switch network configuration is implemented, the energy flow and voltage changes between the battery packs will be continuously monitored, and the switch state will be dynamically adjusted based on real-time data to adapt to changes in battery performance or changes in external load conditions.

[0112] S5 specifically includes:

[0113] S51: according to the specific configuration of each battery pack connected to the DC converter obtained in step S4, the parameters of the active pulse modulation technology are set, including pulse width, pulse amplitude and pulse frequency, and these parameters are optimized according to the current state of the battery pack and the required rapid response degree;

[0114] S52: using an electronic control unit to generate a customized short pulse current signal, the short pulse current signal will match the parameters set in step S51, ensuring that each pulse can effectively promote or control the flow of current to achieve a balancing goal;

[0115] S53: applying the generated pulse signal to the corresponding battery pack through the switch network, and controlling the precise flow of current between the battery packs by adjusting the pulse application time and duration, thereby achieving effective energy balance;

[0116] S54: continuously monitoring the current and voltage response of each battery group after receiving the pulse, and the monitoring data is used to evaluate the pulse modulation effect and provide necessary feedback information for subsequent pulse adjustment;

[0117] S55: Based on the real-time feedback data collected in step S54, the parameters of pulse width, amplitude and frequency are dynamically adjusted to optimize the energy balancing effect of the battery pack and ensure that all battery packs reach a consistent state of charge.

[0118] S6 specifically includes:

[0119] S61: A temperature sensor is integrated into each battery pack to monitor the battery's operating temperature in real time. The temperature sensor is placed at the hot spot of the battery pack to ensure the accuracy and representativeness of the temperature reading;

[0120] S62: continuously collect temperature data of the battery pack from the temperature sensor, and use a high-precision analog-to-digital converter (ADC) to read the analog signal of the sensor and convert the analog signal into digital data for easy processing and analysis;

[0121] S63: Analyze the collected temperature data in real time and compare it with the preset temperature threshold. The set threshold is determined based on the thermal stability and safety standards of the battery material. The specific calculation formula is: T max =T ambient +ΔT, where T max is the maximum safe operating temperature of the battery, T ambient is the ambient temperature, ΔT is a predetermined temperature increment based on the battery type and configuration;

[0122] S64: When the temperature of a battery pack approaches or exceeds the threshold value T max When the pulse modulation parameters set in step S5 are automatically adjusted, including reducing the pulse amplitude or frequency to reduce heat generation and prevent the battery from overheating;

[0123] S65: Based on the temperature data feedback, adjust the operating parameters and switch network configuration of the DC converter, including adjusting the output voltage or changing the connection configuration of the battery pack to control the current flow direction and intensity to further manage heat generation.

[0124] S7 specifically includes:

[0125] S71: The main control chip controls the signal generator to generate sine waves of different frequencies, and the sine waves are accurately injected into each battery pack. The frequency range is set according to the typical response characteristics of the battery in order to capture the dynamic behavior of the battery;

[0126] S72: The response of the battery pack to the injected sine wave is recorded by a sensor (such as a voltage and current sensor), and the response signal is converted into a digital form and transmitted to the main control system in real time for analysis;

[0127] S73: The main control chip uses a preset signal processing algorithm to analyze the collected response signal and extract features including impedance, phase angle and amplitude. The feature data is analyzed by advanced algorithms to evaluate the health status of the battery. Specifically, the impedance change of the battery is calculated using the following formula: Where Z(f) is the impedance at frequency f, V(f) and I(f) are the voltage and current responses at the injected frequency, respectively;

[0128] S74: The collected characteristic data is uploaded to the cloud, where the data analysis platform uses a machine learning algorithm to evaluate the battery status, specifically a neural network algorithm to predict the health of the battery and calculate its remaining service life;

[0129] S75: Based on the analysis results from the cloud, the main control chip receives an evaluation report on the battery health status and adjusts the parameters in S3 to S6 in real time, including adjusting the voltage output of the DC converter, modifying the configuration of the switching network, and adjusting the parameters of the pulse modulation. These adjustments are aimed at optimizing the balancing strategy to ensure the long-term stability and efficiency of the battery pack.

[0130] S74 specifically includes:

[0131] S741: Before training and predicting the neural network model, preprocess the battery response data collected from S73, including normalization, to standardize the voltage, current and impedance data of the battery to a predetermined range (usually 0 to 1) to optimize the performance and prediction accuracy of the neural network;

[0132] S742: Selecting features that are helpful for predicting the health status of the battery from the preprocessed data, including the frequency response of the impedance, the number of charge and discharge cycles, and the historical temperature change, and the features are used as inputs of the neural network;

[0133] S743: Configure a multi-layer perceptron (MLP) neural network, including an input layer, several hidden layers, and an output layer. The nodes (neurons) between each layer transmit information through weighted connections. The weights and biases are parameters learned during the network training process. The specific formula of the neural network is: y = f(W x + b), where x is the input feature vector, W is the weight matrix, b is the bias vector, and f is the activation function. ReLU or Sigmoid function is usually selected to increase the nonlinearity of the model;

[0134] S744: Use historical data sets to train the neural network, including battery data with known health status as training samples. The training process optimizes weights and biases through the back propagation algorithm to minimize the difference between the predicted value and the actual value. The loss function used is the mean square error. The specific formula is: Among them, L is the loss function, y i is the actual value of the sample, is the predicted value, n is the number of samples;

[0135] S745: Use the trained neural network to predict the health status of the newly collected battery data. The prediction result is used to estimate the remaining service life of the battery. By combining the predicted health status with the typical life curve of the battery, the expected life reduction is calculated. The specific algorithm is: RUL = L typical ×(1-SOHpredicted ), where RUL is the remaining useful life, L typical is the typical life of the battery type, SOH predicted is the health status ratio predicted by the neural network.

[0136] like Figure 2 As shown, an online detection device for an active equalization method for retired batteries is used to implement the above-mentioned active equalization method for retired batteries based on DC converter and time-sharing multiplexing, including the following modules:

[0137] Data acquisition module: used to collect key parameters of the connected battery pack in real time, including voltage, current, and temperature data;

[0138] Signal generation and injection module: generates and injects sine waves of different frequencies into each battery pack for subsequent health status assessment;

[0139] Signal analysis module: receives the battery response signal and analyzes the signal to evaluate the health of the battery;

[0140] Data analysis and diagnosis module: Analyzes the data received from the signal analysis module, uses a preset health assessment algorithm to determine the battery status, and predicts the remaining service life;

[0141] Communication control module: adjusts the battery balancing strategy and the operating parameters of the DC converter based on the diagnosis results;

[0142] User Interface Module: Used to provide a user interface for operators to monitor battery status and adjust operating parameters, while displaying real-time diagnostic and prediction results.

[0143] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the above aspects of the present invention, which are not provided in detail for the sake of simplicity.

[0144] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the claims. Therefore, any omissions, 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 method for active balancing of retired batteries based on DC converter and time-sharing multiplexing, characterized in that: The following steps are involved: S1: Automatically detect retired battery packs connected to the system through the main control chip and sensors, and collect their health status, capacity and historical usage data; S2: Classify the battery packs according to the data collected in step S1 using a preset classification algorithm, and analyze their energy requirements and balancing priorities; S3: using a preset buck-boost isolated DC converter, dynamically adjusting the voltage output according to the analysis result of step S2 to meet the energy transfer requirements between different battery groups; S4: Control the battery pack to access the input or output of the DC converter through a preset switch network, and adjust the parameters of the switch network based on the voltage of S3 to maximize the energy transmission efficiency; S5: applying active pulse modulation technology to apply short pulse current to the battery pack according to the switch network configuration of step S4 to control the current flow during the balancing process; S6: Integrated temperature sensor to continuously monitor the operating temperature of the battery pack and automatically adjust the operating parameters of the DC converter and the configuration of the switch network according to the pulse modulation results of S5 to prevent the battery from overheating; S7: Use the main control chip to inject sine waves of different frequencies into the battery, collect response signals, and combine cloud data analysis to evaluate the battery status in real time according to the preset health assessment algorithm. Automatically adjust the parameters of steps S3 to S6 according to the assessment results to optimize the balancing strategy.

2. The method for active balancing of retired batteries based on DC converter and time-sharing multiplexing according to claim 1 is characterized in that: The S1 specifically includes: S11: Configure multiple sensors in the system, including voltage sensors, current sensors, and temperature sensors. Each sensor is calibrated and initialized before being connected to the system to ensure the accuracy and reliability of data collection; S12: using the main control chip to automatically detect the battery pack connected to the system through the interface circuit, and identifying the model and manufacturing information of each battery pack by reading the unique identifier of the battery pack; S13: The main control chip instructs each sensor to collect real-time data of the battery pack. The voltage sensor is used to measure the voltage of each battery cell, the current sensor is used to record the charge and discharge current, and the temperature sensor is used to monitor the battery temperature. S14: The main control chip accesses the historical usage data of the battery pack through built-in or external storage, including the number of charge and discharge cycles, historical fault records and maintenance history; S15: Based on the real-time and historical data collected in steps S13 and S14, the main control chip uses a preset algorithm to evaluate the health status. The preset algorithm includes state estimation, capacity decay analysis and life prediction. The specific calculation formula used for the evaluation is: Among them, SOH represents the percentage of battery health status, C current is the current measured capacity of the battery, and C rated is the rated capacity of the battery; based on historical failure data and usage frequency, the expected life reduction of the battery is calculated. The specific calculation formula is: EOL = C initial ·e -k·n , where EOL represents the expected service life of the battery, C initial is the initial capacity, j is the decay constant, and n represents the number of charge and discharge cycles.

3. The method for active balancing of retired batteries based on DC converter and time-sharing multiplexing according to claim 2 is characterized in that: The S2 specifically includes: S21: Standardize all battery data collected in S1, including removing outliers, filling missing data, and normalizing data; S22: Apply a preset clustering algorithm to classify the battery packs. Specifically, the K-means clustering algorithm is used to classify the batteries into different categories based on the health status, historical usage data, and battery capacity characteristics of the batteries. The calculation formula used in the clustering process is: Among them, C i represents the i-th category; x represents the characteristic data point of a single battery pack; μ i Category C i The center point of ; k is the number of preset categories; S23: Based on the classification results of the battery packs, the typical energy requirements of each category of battery packs are analyzed. Specifically, the average capacity loss rate and typical load response characteristics of each category of battery packs are calculated using a statistical method. The calculation formula is: Among them, E avg represents the average energy requirement; E j represents the actual energy demand of the jth battery pack in the category; n is the number of battery packs in the category; S24: Determine the balancing priority of each battery group based on the health status and energy demand analysis results of the battery group. Specifically, a weighted scoring system is used to combine the health status and energy demand into a balancing priority score. The calculation formula is: P = w1×SOH+w2×E need , where P is the balancing priority score; SOH is the health status percentage of the battery pack; E need is the energy demand of the battery pack; w1 and w2 are the weight coefficients of health status and energy demand, respectively.

4. The method for active balancing of retired batteries based on DC converter and time-sharing multiplexing according to claim 3 is characterized in that: The S3 specifically includes: S31: Based on the classification of each battery pack and the energy demand analysis obtained in S2, the preset control system will calculate the ideal voltage level required for each battery pack; S32: The buck-boost isolated DC converter automatically selects a boost or buck mode based on the comparison between the current voltage of each battery pack and the ideal voltage calculated in step S31. The specific selection is based on the following judgment logic: Where Vcurrent is the current voltage of the battery pack, V 理想 It is the ideal voltage calculated based on energy requirements; S33: The DC converter dynamically adjusts its output voltage according to the mode selected in S32 to ensure that the voltage of each battery pack gradually approaches or reaches its ideal voltage; S34: During the entire adjustment process, the voltage and current status of each battery pack is continuously monitored, and a feedback control algorithm is used to optimize the voltage regulation process to cope with the impact of battery status changes or external load fluctuations. The specific formula for the voltage regulation is: Among them, V 调整 Represents the output of the voltage adjustment control signal; K p is the proportional gain; K i is the integral gain; K d is the differential gain; e(t) is the error or deviation, which represents the difference between the target voltage and the actual voltage at the instant of time t; ∫e(t)dt represents the sum of the deviation e(t) integrated over time; Represents the rate of change of the deviation e(t).

5. The method for active equalization of retired batteries based on DC converter and time-sharing multiplexing according to claim 4 is characterized in that: The S4 specifically includes: S41: using a preset switch network, the network consisting of electronic switches, including transistors, relays or solid-state switches, each switch being configured to control the connection between the battery pack and the DC converter, so that the battery pack can be connected to the input or output of the converter as needed; S42: Based on the voltage regulation parameters in step S3, the control logic of the switch network will determine whether each battery pack should be connected to the input or output of the DC converter, making the decision based on the following rules: When the target voltage of the battery pack is higher than the current voltage, the battery pack is connected to the output terminal of the converter; When the target voltage of the battery pack is lower than the current voltage, the battery pack is connected to the input terminal of the converter; S43: The switch network optimizes the energy flow by controlling the access points of each battery pack, including adjusting the connection time and sequence to ensure that the electric energy flows from the high-energy battery pack to the low-energy battery pack, thereby maximizing the energy conversion efficiency of the overall system; S44: After the switch network configuration is implemented, the energy flow and voltage changes between the battery packs will be continuously monitored, and the switch state will be dynamically adjusted based on real-time data to adapt to changes in battery performance or changes in external load conditions.

6. The method for active balancing of retired batteries based on DC converter and time-sharing multiplexing according to claim 5, characterized in that: The S5 specifically includes: S51: according to the specific configuration of each battery pack connected to the DC converter obtained in step S4, set the parameters of the active pulse modulation technology, including pulse width, pulse amplitude and pulse frequency; S52: using an electronic control unit to generate a customized short pulse current signal, the short pulse current signal will match the parameters set in step S51, ensuring that each pulse can effectively promote or control the flow of current to achieve a balancing goal; S53: applying the generated pulse signal to the corresponding battery pack through the switch network, and controlling the precise flow of current between the battery packs by adjusting the pulse application time and duration, thereby achieving effective energy balance; S54: continuously monitoring the current and voltage response of each battery group after receiving the pulse, and the monitoring data is used to evaluate the pulse modulation effect and provide necessary feedback information for subsequent pulse adjustment; S55: Based on the real-time feedback data collected in step S54, the parameters of pulse width, amplitude and frequency are dynamically adjusted.

7. The method for active balancing of retired batteries based on DC converter and time-sharing multiplexing according to claim 6, characterized in that: The S6 specifically includes: S61: A temperature sensor is integrated into each battery pack to monitor the battery's operating temperature in real time. The temperature sensor is placed at the hot spot of the battery pack to ensure the accuracy and representativeness of the temperature reading; S62: continuously collects temperature data of the battery pack from the temperature sensor, and uses a high-precision analog-to-digital converter to read the analog signal of the sensor and convert the analog signal into digital data for easy processing and analysis; S63: Analyze the collected temperature data in real time and compare it with the preset temperature threshold. The set threshold is determined based on the thermal stability and safety standards of the battery material. The specific calculation formula is: T max =T ambient +ΔT, where T max is the maximum safe operating temperature of the battery, T ambient is the ambient temperature, ΔT is a predetermined temperature increment based on the battery type and configuration; S64: When the temperature of a battery pack approaches or exceeds the threshold value T max When the pulse modulation parameters set in step S5 are automatically adjusted, including reducing the pulse amplitude or frequency to reduce heat generation and prevent the battery from overheating; S65: According to the temperature data feedback, adjust the operating parameters and switch network configuration of the DC converter, including adjusting the output voltage or changing the connection configuration of the battery pack to control the current flow direction and intensity.

8. The method for active balancing of retired batteries based on DC converter and time-sharing multiplexing according to claim 7, characterized in that: The S7 specifically includes: S71: The main control chip controls the signal generator to generate sine waves of different frequencies, and the sine waves are accurately injected into each battery pack. The frequency range is set according to the typical response characteristics of the battery; S72: The response of the battery pack to the injected sine wave is recorded by the sensor, and the response signal is converted into a digital form and transmitted to the main control system in real time for analysis; S73: The main control chip uses a preset signal processing algorithm to analyze the collected response signal and extract features including impedance, phase angle and amplitude. The feature data is analyzed by advanced algorithms to evaluate the health status of the battery. Specifically, the impedance change of the battery is calculated using the following formula: Where Z(f) is the impedance at frequency f, V(f) and I(f) are the voltage and current responses at the injected frequency, respectively; S74: The collected characteristic data is uploaded to the cloud, where the data analysis platform uses a machine learning algorithm to evaluate the battery status, specifically a neural network algorithm to predict the health of the battery and calculate its remaining service life; S75: Based on the analysis results from the cloud, the main control chip receives an evaluation report on the battery health status and adjusts the parameters in S3 to S6 in real time, including adjusting the voltage output of the DC converter, modifying the configuration of the switch network, and adjusting the parameters of the pulse modulation.

9. The method for active balancing of retired batteries based on DC converter and time-sharing multiplexing according to claim 8, characterized in that: The S74 specifically includes: S741: Before training and predicting the neural network model, preprocess the battery response data collected from S73, including normalization, to standardize the voltage, current and impedance data of the battery to a predetermined range, so as to optimize the performance and prediction accuracy of the neural network; S742: Selecting features of the battery health status from the preprocessed data, including the frequency response of the impedance, the number of charge and discharge cycles, and the historical temperature change, and using the features as inputs to the neural network; S743: Configure a multi-layer perceptron neural network, including an input layer, several hidden layers and an output layer. Nodes between each layer transmit information through weighted connections. The specific formula of the neural network is: y=f(W·x+b), where x is the input feature vector, W is the weight matrix, b is the bias vector, and f is the activation function; S744: Use historical data sets to train the neural network, including battery data with known health status as training samples. The training process optimizes weights and biases through the back propagation algorithm to minimize the difference between the predicted value and the actual value. The loss function used is the mean square error. The specific formula is: Among them, L is the loss function, y i is the actual value of the sample, is the predicted value, n is the number of samples; S745: Use the trained neural network to predict the health status of the newly collected battery data. The prediction result is used to estimate the remaining service life of the battery. By combining the predicted health status with the typical life curve of the battery, the expected life reduction is calculated. The specific algorithm is: RUL = L typical ×(1-SOH predicted ), where RUL is the remaining useful life, L typical is the typical life of the battery type, SOH predicted is the health status ratio predicted by the neural network.

10. An online detection device for an active equalization method for retired batteries, used to implement an active equalization method for retired batteries based on a DC converter and time-sharing multiplexing as claimed in any one of claims 1 to 9, characterized in that: Includes the following modules: Data acquisition module: used to collect key parameters of the connected battery pack in real time, including voltage, current, and temperature data; Signal generation and injection module: generates and injects sine waves of different frequencies into each battery pack for subsequent health status assessment; Signal analysis module: receives the battery response signal and analyzes the signal to evaluate the health of the battery; Data analysis and diagnosis module: Analyzes the data received from the signal analysis module, uses a preset health assessment algorithm to determine the battery status, and predicts the remaining service life; Communication control module: adjusts the battery balancing strategy and the operating parameters of the DC converter based on the diagnosis results; User Interface Module: Used to provide a user interface for operators to monitor battery status and adjust operating parameters, while displaying real-time diagnostic and prediction results.

Images