Active equalization energy efficiency monitoring system based on intelligent base station lead-acid storage battery

Abstract

The invention belongs to the technical field of lead-acid storage batteries, and particularly relates to an active equalization energy efficiency monitoring system based on an intelligent base station lead-acid storage battery. The system comprises a data acquisition module, a big data processing module, a battery equalization module and a pulse repair module, the data acquisition module is used for monitoring various state data of the battery in real time, and the state data comprises voltage, electric quantity and temperature; and when the high-frequency pulse is applied to the battery, the data acquisition module records response characteristics of the battery. A base station battery pack is managed by an intelligent system through the whole process of integrating data monitoring, analysis, decision making and execution, big data screening is adopted, and meanwhile pulse repairing parameters are recommended through a similarity matching algorithm; through intelligent monitoring, analysis and repair decision making, the rejection rate of the batteries can be reduced, the cost for maintaining and replacing the batteries is reduced, and the overall service life of the battery pack is prolonged.

Description

Technical Field

[0001] The present invention belongs to the technical field of lead-acid batteries, and in particular, relates to an active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries. Background Art

[0002] In current battery maintenance, active battery balancing technology can only solve the problem of imbalance among the batteries in the battery pack; battery repair technology mainly deals with the failure of a single battery and can only solve the failure problems of a single battery, such as sulfation, insufficient electrolyte, etc. Therefore, it is difficult to ensure the health state (SOH) of the battery system by relying solely on battery balancing technology or repair technology; especially in the application of lead-acid battery systems in medium and large communication base stations such as iron towers, the health of the battery is crucial.

[0003] High-energy pulse repair technology In the sulfation repair of lead-acid batteries, high-energy pulse technology has a cleaning and repair effect on mild and moderate sulfation, but before repair, the batteries must first be identified and mild and moderate sulfation batteries must be screened out.

[0004] Traditional battery monitoring and data analysis methods can usually only identify surface problems of the battery, such as a drop in voltage and capacity, but it is difficult to accurately assess the degree of sulfation inside the battery; there is a lack of accurate and effective screening methods.

[0005] In addition, most existing battery repair technologies use fixed pulse repair schemes, which lack flexibility and precision to address individual differences in batteries, resulting in lower battery repair efficiency.

[0006] Therefore, developing an intelligent battery sulfide screening method, selecting targeted high-energy pulse repair solutions and combining them with active balancing technology have become important research directions for improving the health of battery systems. Summary of the invention

[0007] In order to solve the technical problems existing in the above-mentioned background technology that it is difficult to accurately evaluate the degree of sulfation inside the battery; and the pulse repair scheme lacks specificity, the present invention provides an active balancing energy efficiency monitoring system based on an intelligent base station lead-acid battery.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] Active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries, including data acquisition module, big data processing module, battery balancing module and pulse repair module;

[0010] The data acquisition module is used to monitor various status data of the battery in real time, including voltage, power and temperature; and when a high-frequency pulse is applied to the battery, the data acquisition module is used to record the response characteristics of the battery;

[0011] Preferably, the response characteristics include at least three of response time, response amplitude, frequency response bandwidth, transient voltage drop and energy loss.

[0012] The big data processing module is used to pre-process, analyze and make decisions on the collected data;

[0013] The big data processing module includes a data preprocessing unit, a screening unit, and a technical decision-making unit;

[0014] The data preprocessing unit is used for data cleaning, denoising and standardization;

[0015] The screening unit identifies batteries with unbalanced energy efficiency in the battery pack according to various status data of the batteries; and screens out lead-acid batteries with a mild or moderate degree of sulfation or those in between according to the response characteristics of different lead-acid batteries;

[0016] The technical decision unit formulates a battery balancing method according to the identified batteries with energy efficiency imbalance; and selects the best pulse repair parameters according to the physical characteristics of the screened batteries;

[0017] The implementation steps of the screening unit include:

[0018] Data annotation: Use the response characteristic data of batteries with known sulfidation degree as the training set, including response time, response amplitude and transient voltage drop; and annotate the sulfidation degree of each sample.

[0019] Preferably, the degree of sulfidation is classified as unsulfated, slightly sulfated, moderately sulfated or heavily sulfated.

[0020] Model selection: Choose a suitable machine learning algorithm, including support vector machine (SVM), random forest (RandomForest), or convolutional neural network (CNN).

[0021] Model training and optimization: Use the training data set to train the machine learning model. After training, use the test set to evaluate the model performance. Use the K-fold cross-validation method to verify the generalization ability of the model. And adjust the model's hyperparameters to improve model performance.

[0022] Sulfation screening: Input new data into the trained machine learning model; identify the sulfation degree of the battery and screen out lead-acid batteries with mild or moderate sulfation and those in between;

[0023] Preferably, the machine learning model selects a random forest model, wherein the hyperparameter settings of the random forest model include the number of trees n, the maximum depth d, and the minimum number of sample splits s; the evaluation indicators of the random forest model include accuracy, precision, recall, and F1-score.

[0024] Preferably, the implementation steps of the technical decision unit include:

[0025] Collect a large number of physical characteristics of historical repair batteries and corresponding pulse repair parameters;

[0026] The physical properties of the battery include internal resistance, ambient temperature, service life and degree of sulfation; the pulse repair parameters include pulse frequency, pulse intensity and duration;

[0027] The physical characteristics of historical repaired batteries are quantized, and the similarity between the current battery and the repaired battery is calculated (e.g., Euclidean distance and cosine similarity of features such as internal resistance, ambient temperature, age, and degree of sulfation).

[0028] According to the historical repaired battery with the highest similarity score to the current battery, its corresponding pulse repair parameters are selected as the repair solution for the current battery.

[0029] Preferably, the similarity calculation method comprises the following steps:

[0030] Firstly, a historical repair database is constructed, which includes the cosine similarity, Euclidean distance and pulse repair parameters of the battery in terms of physical characteristics.

[0031] Among them, cosine similarity is used to measure the directional similarity of the physical characteristics of the battery, and its calculation formula is:

[0032]

[0033] Where: A and B are the physical characteristic vectors of the two batteries, i.e., the vectors of the internal resistance, ambient temperature, service life or sulfation degree of the batteries; the value of cosine similarity is between [-1, 1], and the closer to 1, the more similar the two batteries are in the direction of physical characteristics;

[0034] ||A|| and ||B|| are the moduli of vectors A and B respectively, and their formulas are expressed as follows:

[0035]

[0036] A·B represents the dot product of vectors A and B, and its formula is:

[0037]

[0038] Where: A i and B i are the values ​​of the ith physical characteristic of battery A and battery B respectively; n is the number of dimensions of the physical characteristics of the battery;

[0039] The Euclidean distance is used to measure the dimensional differences in the physical properties of the battery; the calculation formula is:

[0040]

[0041] In the formula, the smaller the value of the Euclidean distance is, the closer the values ​​of the two batteries are;

[0042] Combining cosine similarity and Euclidean distance, the differences in physical properties of batteries are considered in a weighted manner to obtain a similarity score; the calculation formula is:

[0043]

[0044] Where: w1 and w2 are the weights of cosine similarity and Euclidean distance respectively, and satisfy w1+w2=1; is the normalized Euclidean distance, d max is the maximum possible Euclidean distance of the physical characteristics of the battery in the historical repair data, which is used to scale the Euclidean distance to the [0,1] interval; the final similarity score is between [-1,1], and the larger the value, the more similar the two batteries are.

[0045] The battery balancing module is used to control the charging and discharging of the battery through active balancing technology to balance the voltage and power of each battery in the battery pack.

[0046] The pulse repair module controls the frequency, intensity and duration of the pulse according to the pulse repair parameters selected by the technical decision unit to perform a high-energy pulse repair process on the screened batteries.

[0047] Preferably, it also includes a temperature management module, which receives temperature data from the data acquisition module, automatically adjusts and ensures that the battery is in an appropriate temperature range by controlling the corresponding actuator; the temperature management module includes a microcontroller, a heater and a radiator, and the microcontroller is electrically connected to the heater and the radiator respectively; the microcontroller is used to receive temperature data and decide whether to activate the heater or the radiator according to a preset temperature range; when the temperature data is lower than the operating temperature threshold, the microcontroller will instruct the heater to start to increase the temperature; when the temperature is higher than the operating temperature threshold, the microcontroller will activate the radiator to lower the temperature.

[0048] Preferably, it also includes an alarm module, which receives various status data from the data acquisition module and is used to detect various abnormal data and issue an alarm in time;

[0049] The alarm module includes a microcontroller, an alarm, a communication unit and a protection circuit; the communication unit is connected to the microcontroller via a UART or SPI serial interface; and an alarm signal is sent to a system server and staff using a GSM, LoRa or Wi-Fi network;

[0050] The protection circuit is used for physical circuit breaking protection of the system, and cuts off the system circuit when abnormal data occurs to avoid damage to the battery and the system circuit.

[0051] Preferably, the abnormal data includes current surge, voltage drastic change or temperature surge, overheating.

[0052] Beneficial effects of the present invention:

[0053] 1. The present invention is based on base station lead-acid batteries. It manages base station battery packs through a full-process, intelligent system that integrates data monitoring, analysis, decision-making and execution, and uses big data to screen out batteries with sulfation problems in the battery pack. At the same time, it uses a similarity matching algorithm to recommend pulse repair parameters in historical repair data. Through intelligent monitoring, analysis and repair decisions, it can reduce the battery scrap rate, reduce the cost of maintenance and replacement of batteries, and extend the overall service life of the battery pack.

[0054] 2. Based on the high-energy pulse response characteristics of different batteries as training data, a machine learning model is introduced to analyze the degree of battery sulfation, which can accurately identify and screen out lightly / moderately sulfated lead-acid batteries, thereby improving recognition accuracy and efficiency.

[0055] 3. Through big data analysis and similarity calculation, and adding the identified battery sulfide degree to the battery physical properties, the system can calculate the similarity based on the physical characteristics of each dimension of the battery, select the most matching pulse repair parameters in the historical repair data, avoid blind repair, and improve the targetedness and efficiency of the repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0057] Figure 1 The present invention is a structural schematic diagram of an active balancing energy efficiency monitoring system based on an intelligent base station lead-acid battery.

[0058] Figure 2 The present invention is a flow chart of the screening unit of the active balancing energy efficiency monitoring system based on the intelligent base station lead-acid battery.

[0059] Figure 3 The present invention is a flow chart of a similarity algorithm in an active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries. DETAILED DESCRIPTION

[0060] 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.

[0061] See also Figure 1-Figure 3 As shown, the active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries includes a data acquisition module, a big data processing module, a battery balancing module and a pulse repair module;

[0062] The data acquisition module is used to monitor various status data of the battery in real time, including voltage, power and temperature; and when a high-frequency pulse is applied to the battery, the data acquisition module is used to record the response characteristics of the battery;

[0063] Furthermore, the response characteristics include at least three of response time, response amplitude, frequency response bandwidth, transient voltage drop and energy loss.

[0064] It should be noted that in this embodiment, high-frequency pulse response characteristics are used as training data for the machine learning model. After repeated verification, it is found that the high-frequency pulse response characteristics and the degree of battery sulfation have a strong correlation. Using these data as training samples, the machine learning model can help build an accurate classification and screening model by learning the response characteristics of these historical samples, thereby enhancing the reliability of the model.

[0065] Response time: refers to the time required for the battery to reach a stable state after receiving external high-frequency pulse stimulation. Among them, the electrochemical reaction of a healthy battery is active, and the charge can be transferred quickly, so the response time is short. On the contrary, the formation of lead sulfate crystals in sulfated batteries hinders the transfer of charge, and the internal electrochemical reaction is slow, resulting in a longer response time.

[0066] Response amplitude: Response amplitude refers to the amplitude of voltage or current change after the battery receives a high-frequency pulse, that is, the difference between the initial state and the maximum response value. It reflects the response strength of the battery under pulse stimulation. Among them, healthy batteries usually show a larger response amplitude under high-frequency pulses because they can efficiently convert external input electrical energy and produce larger voltage or current changes. Sulfated batteries have a reduced charge transfer efficiency due to the barrier effect of lead sulfate crystals, resulting in a reduced response amplitude.

[0067] Frequency response bandwidth: refers to the range of the battery's response to pulses of different frequencies, which usually reflects its dynamic performance and electrochemical activity. Among them, healthy batteries usually have a better response to pulses with a wider frequency range, while sulfated batteries have a narrow frequency response bandwidth, reflecting their reduced dynamic performance.

[0068] Transient voltage drop: refers to the magnitude of the instantaneous drop in battery voltage after the high-frequency pulse ends. The magnitude of the transient voltage drop can reflect the electrochemical response capability of the battery. Among them, sulfate batteries usually have a larger transient voltage drop because their internal impedance is high and the electrochemical reaction is unstable. The measurement method is to record the instantaneous voltage drop value at the moment the pulse ends, and calculate the difference between it and the stable voltage.

[0069] Energy loss: refers to the energy loss caused by internal resistance and other electrochemical losses during the pulse process. Batteries with large energy loss tend to show higher internal impedance and poor health. Among them, the energy loss of sulfated batteries is usually larger. The measurement method is to calculate the energy loss by analyzing the difference between the input pulse energy and the output energy.

[0070] The big data processing module is used to pre-process, analyze and make decisions on the collected data; the big data processing module includes a data pre-processing unit, a screening unit and a technical decision unit;

[0071] The data preprocessing unit is used for data cleaning, denoising and standardization;

[0072] The screening unit identifies batteries with unbalanced energy efficiency in the battery pack according to various status data of the batteries; and screens out lead-acid batteries with a mild or moderate degree of sulfation or those in between according to the response characteristics of different lead-acid batteries;

[0073] Specifically, the sensors collect various status data of the battery in real time, such as voltage and power, to provide data basis for the battery balancing module. The sensors collect the response characteristics of the battery, including response time, response amplitude, frequency response bandwidth, transient voltage drop, energy loss, etc., to provide data basis for the pulse repair module.

[0074] The technical decision unit formulates a battery balancing method according to the identified batteries with energy efficiency imbalance; and selects the best pulse repair parameters according to the physical characteristics of the screened batteries;

[0075] Furthermore, the implementation steps of the screening unit include:

[0076] Data annotation: Use the response characteristic data of batteries with known sulfidation degree as the training set, including response time, response amplitude and transient voltage drop; and annotate the sulfidation degree of each sample.

[0077] The degree of sulfidation is classified as unsulfated, slightly sulfated, moderately sulfated, or heavily sulfated.

[0078] Model selection: Choose a suitable machine learning algorithm, including support vector machine (SVM), random forest (RandomForest), or convolutional neural network (CNN).

[0079] Model training and optimization: Use the training data set to train the machine learning model. After training, use the test set to evaluate the model performance. Use the K-fold cross-validation method to verify the generalization ability of the model. And adjust the model's hyperparameters to improve model performance.

[0080] Sulfation screening: Input new data into the trained machine learning model; identify the sulfation degree of the battery and screen out lead-acid batteries with mild or moderate sulfation and those in between;

[0081] Furthermore, the machine learning model selects a random forest model, wherein the hyperparameter settings of the random forest model include the number of trees n, the maximum depth d, and the minimum number of sample splits s; the evaluation indicators of the random forest model include accuracy, precision, recall, and F1-score.

[0082] It should be noted that the preferred embodiment of the present invention is to select a random forest model for training for the following reasons:

[0083] Strong adaptability: Random forests can handle various types of input features and perform well in dealing with nonlinear relationships, making them suitable for evaluating the nonlinear behavior of battery response characteristics.

[0084] Anti-overfitting: Random forest reduces the risk of overfitting that may exist in a single model by integrating multiple decision trees.

[0085] Strong correlation of features: Random forests can provide strong correlation analysis of features, helping us understand the criticality of various response characteristics (such as response time, response amplitude, or transient voltage drop) to sulfation judgment.

[0086] In addition, the data set is divided into a training set and a test set, usually 70% is used as a training set and 30% is used as a test set. When dividing, the randomness of the data set should be ensured to avoid deviations in the distribution of the training set and the test set.

[0087] The meaning of the hyperparameters is as follows:

[0088] Number of trees (n): The number of decision trees in the random forest, which can be adjusted from 50 to below in the present invention.

[0089] Maximum depth (d): The maximum depth of each tree, which is used to control the complexity of the model and avoid overfitting.

[0090] Minimum number of sample splits(s): The minimum number of samples required at each node to prevent the tree from becoming too thin.

[0091] The screening unit combines big data analysis with machine learning models to accurately determine the health status and sulfation level of the battery pack, which not only improves the effectiveness of battery repair, but also significantly reduces unnecessary repair attempts.

[0092] Furthermore, the implementation steps of the technical decision unit include:

[0093] Collect a large number of physical characteristics of historical repair batteries and corresponding pulse repair parameters;

[0094] The physical properties of the battery include internal resistance, ambient temperature, service life and degree of sulfation; the pulse repair parameters include pulse frequency, pulse intensity and duration;

[0095] The physical characteristics of historical repaired batteries are quantized, and the similarity between the current battery and the repaired battery is calculated (e.g., Euclidean distance and cosine similarity of features such as internal resistance, ambient temperature, age, and degree of sulfation).

[0096] According to the historical repaired battery with the highest similarity score to the current battery, its corresponding pulse repair parameters are selected as the repair solution for the current battery.

[0097] Specifically, during the high-frequency pulse repair process of batteries, individual differences in batteries will significantly affect the repair effect; that is, the physical properties of different batteries, such as internal resistance, ambient temperature, age and degree of sulfation, will affect the selection and effect of pulse repair parameters. The analysis is as follows:

[0098] Internal resistance: The internal resistance of a battery reflects the conductivity of the internal materials of the battery and the efficiency of the electrochemical reaction, which directly affects the charge and discharge performance of the battery. Batteries with lower internal resistance are relatively easy to repair, while batteries with higher internal resistance require higher pulse energy to break the conductive bottleneck.

[0099] Ambient temperature: The electrochemical reaction rate of the battery at different temperatures will be significantly different, and the temperature will also affect the effect of high-frequency pulse repair. During repair, the pulse intensity and frequency need to be adjusted according to the temperature of the battery to avoid overheating or insufficient repair.

[0100] Service life: The service life of the battery will affect the degree of battery aging and changes in the internal structure (such as gradual degradation of active materials, deposition of by-products and decomposition of electrolytes), which in turn affects the effect of high-frequency pulse repair.

[0101] Sulfation degree: The sulfation degree of the battery is a core indicator that affects the repair effect. The degree of sulfation reflects the degree of crystallization of lead sulfate in the battery. Crystallized lead sulfate hinders the electrode reaction and causes the battery performance to deteriorate. The higher the degree of sulfation, the greater the intensity and duration of the pulse. To avoid excessive repair and damage to the battery, the repair parameters need to be more precise.

[0102] The technical decision-making unit uses the similarity matching between the physical characteristics of the battery and the historical repair data, and uses cosine similarity and Euclidean distance to calculate the similarity, and can recommend pulse repair parameters based on the most similar battery historical data. This process reduces the blindness of pulse repair parameter selection and improves the pertinence of pulse repair.

[0103] Furthermore, the similarity calculation method comprises the following steps:

[0104] Firstly, a historical repair database is constructed, which includes the cosine similarity, Euclidean distance and pulse repair parameters of the battery in terms of physical characteristics.

[0105] Among them, cosine similarity is used to measure the directional similarity of the physical characteristics of the battery, and its calculation formula is:

[0106]

[0107] Where: A and B are the physical characteristic vectors of the two batteries, i.e., the vectors of the internal resistance, ambient temperature, service life or sulfation degree of the batteries; the value of cosine similarity is between [-1, 1], and the closer to 1, the more similar the two batteries are in the direction of physical characteristics;

[0108] ||A|| and ||B|| are the moduli of vectors A and B respectively, and their formulas are expressed as follows:

[0109]

[0110] A·B represents the dot product of vectors A and B, and its formula is:

[0111]

[0112] Where: Ai and B i are the values ​​of the ith physical characteristic of battery A and battery B respectively; n is the number of dimensions of the physical characteristics of the battery;

[0113] The Euclidean distance is used to measure the dimensional differences in the physical properties of the battery; the calculation formula is:

[0114]

[0115] In the formula, A i and B i are the values ​​of battery A and battery B on the i-th physical characteristic respectively; n is the number of dimensions of the physical characteristic of the battery; the smaller the value of the Euclidean distance, the closer the values ​​of the two batteries are;

[0116] Combining cosine similarity and Euclidean distance, the differences in physical properties of batteries are considered in a weighted manner to obtain a similarity score; the calculation formula is:

[0117]

[0118] Where: w1 and w2 are the weights of cosine similarity and Euclidean distance respectively, and satisfy w1+w2=1; is the normalized Euclidean distance, d max is the maximum possible Euclidean distance of the physical characteristics of the battery in the historical repair data, which is used to scale the Euclidean distance to the [0,1] interval; the final similarity score is between [-1,1], and the larger the value, the more similar the two batteries are.

[0119] The battery balancing module is used to control the charging and discharging of the battery through active balancing technology to balance the voltage and power of each battery in the battery pack.

[0120] The pulse repair module controls the frequency, intensity and duration of the pulse according to the pulse repair parameters selected by the technical decision unit to perform a high-energy pulse repair process on the screened batteries.

[0121] Specifically, the similarity matching algorithm is combined with high-energy pulse technology to recommend the best pulse repair parameters based on historical repair data, so as to break up the sulfide crystals, reduce internal resistance, and improve the battery's charge and discharge capabilities.

[0122] Furthermore, it also includes a temperature management module, which receives temperature data from the data acquisition module, automatically adjusts and ensures that the battery is in an appropriate temperature range by controlling the corresponding actuator; the temperature management module includes a microcontroller, a heater, and a radiator, and the microcontroller is electrically connected to the heater and the radiator respectively; the microcontroller is used to receive temperature data and decide whether to activate the heater or the radiator according to a preset temperature range; when the temperature data is lower than the operating temperature threshold, the microcontroller will instruct the heater to start to increase the temperature; when the temperature is higher than the operating temperature threshold, the microcontroller will activate the radiator to lower the temperature.

[0123] Specifically, the temperature management module regulates the heating and cooling process of the battery according to the real-time situation of the temperature data to ensure that the battery is repaired within the optimal operating temperature threshold. When the temperature is low, preheating is performed first to prevent the inefficient operation of the equalization process or pulse repair under low temperature conditions; when the temperature is high, the radiator is started to prevent overheating and avoid battery damage.

[0124] Furthermore, it also includes an alarm module, which receives various status data from the data acquisition module and is used to detect various abnormal data and issue an alarm in time;

[0125] The alarm module includes a microcontroller, an alarm, a communication unit and a protection circuit; the communication unit is connected to the microcontroller via a UART or SPI serial interface; and an alarm signal is sent to a system server and staff using a GSM, LoRa or Wi-Fi network;

[0126] The protection circuit is used for physical circuit breaking protection of the system, and cuts off the system circuit when abnormal data occurs to avoid damage to the battery and the system circuit.

[0127] The abnormal data include current surge, voltage sharp change or temperature sudden rise, overheating.

[0128] Specifically, in the battery balancing module, current surges, voltage changes, or temperature rises and overheating can be used as abnormal data to determine the system, and the protection circuit automatically cuts off the relevant system circuits. In the pulse repair module, the pulse generator applies high-frequency, high-energy short pulse current or voltage to the battery, and current surges and voltage changes are normal repair processes; however, temperature rises and overheating can be used as abnormal data to determine the system, and the protection circuit automatically cuts off the working circuit of the pulse repair module to avoid accidents and protect battery safety.

[0129] In summary, the present invention is based on base station lead-acid batteries, and manages base station battery packs through a full-process, intelligent system that integrates data monitoring, analysis, decision-making and execution, and uses big data to screen out batteries with sulfation problems in the battery pack, while using a similarity matching algorithm to recommend pulse repair parameters in historical repair data; through intelligent monitoring, analysis and repair decisions, the battery scrap rate can be reduced, the cost of maintenance and battery replacement can be reduced, and the overall service life of the battery pack can be extended.

[0130] The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries of the present invention relies on the analysis and decision-making algorithm steps of the big data processing module. Through intelligent monitoring, analysis, decision-making, protection and early warning, the cost of maintenance and battery replacement is reduced, and the overall service life of the base station battery pack is extended.

[0131] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0132] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0133] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

[0134] In the description of the specification, the description with reference to the terms "further", "it should be noted that", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0135] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. Active balancing energy efficiency monitoring system based on intelligent base station lead-acid battery, characterized by: It includes data acquisition module, big data processing module, battery balancing module and pulse repair module; The data acquisition module is used to monitor various status data of the battery in real time, and the status data includes voltage, power and temperature; The data acquisition module is also used to record the response characteristics of the battery when high-frequency pulses are applied to the battery; The big data processing module is used to pre-process, analyze and make decisions on the collected data; The big data processing module includes a data preprocessing unit, a screening unit, and a technical decision-making unit; The data preprocessing unit is used for data cleaning, denoising and standardization; The screening unit identifies batteries with unbalanced energy efficiency in the battery pack according to various status data of the batteries; and screening out lead-acid batteries with a slight or moderate degree of sulfation according to the response characteristics of different lead-acid batteries; The technical decision unit is used to formulate a battery balancing method according to the identified battery with energy efficiency imbalance; and selecting optimal pulse repair parameters based on the physical properties of the screened batteries; The battery balancing module is used to control the charging and discharging of the battery through active balancing technology to balance the voltage and power of each battery in the battery pack; The pulse repair module is used to control the frequency, intensity and duration of the pulse according to the pulse repair parameters selected by the technical decision unit to perform a high-energy pulse repair process on the screened batteries.

2. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to claim 1 is characterized in that: The response characteristics include at least three of response time, response amplitude, frequency response bandwidth, transient voltage drop and energy loss.

3. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to claim 1 is characterized in that: The implementation steps of the screening unit include: Data annotation: Use the response characteristic data of batteries with known sulfidation degree as the training set, including response time, response amplitude and transient voltage drop; and annotate the sulfidation degree of each sample; Model selection: Select a suitable machine learning model, which can be support vector machine, random forest or convolutional neural network; Model training and optimization: Use the training data set to train the machine learning model. After training, use the test set to evaluate the model performance. Use the K-fold cross-validation method to verify the generalization ability of the model. And adjust the model's hyperparameters. Sulfation degree screening: Input new data into the trained machine learning model; identify the sulfation degree of the battery and screen out lead-acid batteries with mild or moderate sulfation.

4. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to claim 3 is characterized in that: The degree of sulfation is classified as unsulfated, slightly sulfated, moderately sulfated or heavily sulfated.

5. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to claim 3 is characterized in that: The machine learning model is a random forest model, wherein the hyperparameter settings of the random forest model include the number of trees n, the maximum depth d, and the minimum number of sample splits s; the evaluation indicators of the random forest model include accuracy, precision, recall, and F1-score.

6. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to claim 1 is characterized in that: The implementation steps of the technical decision-making unit include: Collect a large number of physical characteristics of historical repair batteries and corresponding pulse repair parameters; The physical properties of the battery include internal resistance, ambient temperature, service life and degree of sulfation; the pulse repair parameters include pulse frequency, pulse intensity and duration; The physical characteristics of the historical repaired battery are vectorized and the similarity between the current battery and the historical repaired battery is calculated; According to the historical repaired battery with the highest similarity score to the current battery, its corresponding pulse repair parameters are selected as the repair solution for the current battery.

7. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to claim 6 is characterized in that: The similarity calculation method comprises the following steps: Firstly, a historical repair database is constructed, which includes the cosine similarity, Euclidean distance and pulse repair parameters of the battery in terms of physical characteristics. Among them, cosine similarity is used to measure the directional similarity of the physical characteristics of the battery, and its calculation formula is: Where: A and B are the physical characteristic vectors of the two batteries, i.e., the vectors of the internal resistance, ambient temperature, service life or sulfation degree of the batteries; the value of cosine similarity is between [-1, 1], and the closer to 1, the more similar the two batteries are in the direction of physical characteristics; ||A|| and ||B|| are the moduli of vectors A and B respectively, and their formulas are expressed as follows: A·B represents the dot product of vectors A and B, and its formula is: Where: A i and B i are the values ​​of the ith physical characteristic of battery A and battery B respectively; n is the number of dimensions of the physical characteristics of the battery; The Euclidean distance is used to measure the dimensional differences in the physical properties of the battery; the calculation formula is: In the formula, the smaller the value of the Euclidean distance is, the closer the values ​​of the two batteries are; Combining cosine similarity and Euclidean distance, the differences in physical properties of batteries are considered in a weighted manner to obtain a similarity score; the calculation formula is: Where: w1 and w2 are the weights of cosine similarity and Euclidean distance respectively, and satisfy w1+w2=1; is the normalized Euclidean distance, d max is the maximum possible Euclidean distance of the physical characteristics of the battery in the historical repair data, which is used to scale the Euclidean distance to the [0,1] interval; the final similarity score is between [-1,1], and the larger the value, the more similar the two batteries are.

8. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to any one of claims 1 to 7, characterized in that: It also includes a temperature management module, which receives temperature data from the data acquisition module, automatically adjusts and ensures that the battery is in an appropriate temperature range by controlling the corresponding actuator; the temperature management module includes a microcontroller, a heater and a radiator, and the microcontroller is electrically connected to the heater and the radiator respectively; the microcontroller is used to receive temperature data, and decide whether to activate the heater or the radiator according to a preset temperature range; when the temperature data is lower than the operating temperature threshold, the microcontroller will instruct the heater to start to increase the temperature; when the temperature is higher than the operating temperature threshold, the microcontroller will activate the radiator to lower the temperature.

9. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to any one of claims 1 to 7, characterized in that: It also includes an alarm module, which receives various status data from the data acquisition module and is used to detect various abnormal data and issue an alarm in time; The alarm module includes a microcontroller, an alarm, a communication unit and a protection circuit; the communication unit is connected to the microcontroller via a UART or SPI serial interface; and an alarm signal is sent to a system server and staff using a GSM, LoRa or Wi-Fi network; The protection circuit is used for physical circuit breaking protection of the system and cuts off the system circuit when abnormal data occurs.

10. The active balancing energy efficiency monitoring system based on intelligent base station lead-acid batteries according to claim 9 is characterized in that: The abnormal data include current surge, voltage sharp change or temperature sudden rise, overheating.

Images