Method and System for Detecting the Safety of Power Batteries of Driverless Vehicles Based on Big Data

Abstract

The present invention provides a method and system for safety detection of power batteries of unmanned vehicles based on big data, which relates to the technical field of power battery safety detection. The method includes: collecting battery appearance images of the power battery; determining an appearance safety detection score of the power battery according to the battery appearance images; obtaining historical operation data; determining an aging influence coefficient of the power battery according to the historical operation data; obtaining battery parameters of the power battery through a battery management system; obtaining the internal resistance of the power battery through a detection instrument; determining a battery operation safety detection score according to the battery internal resistance, the battery parameters and the aging influence coefficient of the power battery; and generating a power battery safety detection report according to the battery operation safety detection score and the battery appearance safety detection score. According to the present invention, the safety condition of the power battery can be comprehensively detected based on the appearance safety and operation safety of the power battery, and the comprehensiveness and accuracy of the power battery safety detection results can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power battery safety detection, and particularly to a method and system for power battery safety detection of unmanned vehicles based on big data. Background Art

[0002] In the related art, CN114137428A discloses a portable power battery safety detection system and device, including a main control module, an input module, a data acquisition module, an OBD data acquisition module, a network module, a cloud server module, and a connection module; by connecting the detection device in series between a charging pile and an electric vehicle, the health status detection function of the power battery during the charging process is realized, and a plurality of detections and analyses are performed on the collected data by using a health diagnosis algorithm and a fault diagnosis algorithm, and finally a comprehensive and accurate detection result report is obtained. This solution has the beneficial effects that the power battery safety detection process is simple and convenient, the evaluation result of the power battery health status is accurate, and the use safety of electric vehicles is effectively improved.

[0003] CN116298478A discloses a method, device and vehicle for functional safety detection of a power battery. Among them, the method includes: obtaining a first current signal, a second current signal, a voltage signal and a temperature signal corresponding to the power battery, where the first current signal is a digital signal and the second current signal is an analog signal; performing sensor fault diagnosis and overcurrent detection based on the first current signal and the second current signal to obtain a first detection result; performing rationality detection and overcurrent detection based on the voltage signal, the temperature signal and the opening and closing state of the contactor of the power battery to obtain a second detection result; determining a target overcurrent detection result corresponding to the power battery according to the first detection result and the second detection result. This solution solves the technical problems that the existing functional safety detection method of power batteries only uses current sensors for overcurrent detection, resulting in low accuracy and poor safety.

[0004] Based on the above related technologies, the accuracy of the power battery health status evaluation result can be improved. However, the related technologies do not consider the influence of the appearance condition of the power battery and the vehicle and usage conditions on the power battery safety status evaluation, that is, the related technologies cannot comprehensively evaluate the safety status of the power battery according to the appearance condition of the power battery and the historical usage conditions of the vehicle. Summary of the Invention

[0005] The present invention provides a method and system for power battery safety detection of unmanned vehicles based on big data, which can solve the technical problem that the related technologies cannot comprehensively evaluate the safety status of the power battery according to the appearance condition of the power battery and the historical usage conditions of the vehicle.

[0006] According to the first aspect of the present invention, there is provided a method for detecting the safety of a power battery of an unmanned vehicle based on big data, including:

[0007] Collecting an image of the appearance of the power battery through an image acquisition device;

[0008] Determining a safety detection score for the appearance of the power battery according to the battery appearance image;

[0009] Obtaining historical operation data of the unmanned vehicle in a historical operation cycle, where the historical operation data includes: historical operation environment data, historical operation mileage, total number of charging times, and battery discharge depth;

[0010] Determining an aging influence coefficient of the power battery according to the historical operation data;

[0011] At multiple moments in a detection cycle, obtaining battery parameters of the power battery through a battery management system, where the battery parameters include: battery current, battery voltage, and battery temperature;

[0012] At multiple moments in a detection cycle, obtaining the internal resistance of the power battery through a detection instrument;

[0013] Determining a safety detection score for battery operation according to the battery internal resistance, the battery parameters, and the aging influence coefficient of the power battery;

[0014] Generating a safety detection report for the power battery according to the safety detection score for battery operation and the safety detection score for the appearance of the battery;

[0015] According to the second aspect of the present invention, there is provided a system for detecting the safety of a power battery of an unmanned vehicle based on big data, including:

[0016] An image acquisition module for collecting an image of the appearance of the power battery through an image acquisition device;

[0017] An appearance safety module for determining a safety detection score for the appearance of the power battery according to the battery appearance image;

[0018] A historical data module for obtaining historical operation data of the unmanned vehicle in a historical operation cycle, where the historical operation data includes: historical operation environment data, historical operation mileage, total number of charging times, and battery discharge depth;

[0019] An aging influence module for determining an aging influence coefficient of the power battery according to the historical operation data;

[0020] A battery parameter module for obtaining battery parameters of the power battery through a battery management system at multiple moments in a detection cycle, where the battery parameters include: battery current, battery voltage, and battery temperature;

[0021] A battery internal resistance module, configured to obtain the internal resistance of a power battery through a detection instrument at multiple moments during a detection period;

[0022] An operation safety module, configured to determine an operation safety detection score of the battery according to the battery internal resistance, the battery parameters, and the power battery aging influence coefficient;

[0023] A detection report module, configured to generate a safety detection report of the power battery according to the operation safety detection score of the battery and the appearance safety detection score of the battery.

[0024] Technical effects: According to the present invention, the appearance safety condition of a power battery can be evaluated based on the appearance image of the power battery to determine an appearance safety detection score of the battery. The influence of the historical operation condition of an unmanned vehicle on the aging degree of the power battery can be accurately analyzed, and based on this influence and the internal resistance, current, voltage, and temperature of the power battery during a detection period, the operation safety condition of the power battery can be evaluated to determine an operation safety detection score of the battery. Further, a comprehensive evaluation of the safety of the power battery can be performed according to the operation safety detection score of the battery and the appearance safety detection score of the battery. When determining the appearance safety detection score of the battery, various elements in the appearance image of the battery can be detected through an artificial intelligence model, so as to evaluate the appearance safety condition of the power battery from multiple aspects, improving the accuracy and objectivity of the appearance safety detection score. When determining the power battery aging influence coefficient, the power battery aging influence coefficient can be determined according to the historical operation humidity function, historical operation temperature function, historical operation mileage, total charging times, and battery discharge depth. During the calculation process, the influence of the historical operation condition of the unmanned vehicle on the aging of the power battery can be evaluated respectively from four aspects of the operation environment, discharge depth, total charging times, and historical operation mileage of the unmanned vehicle, improving the comprehensiveness and accuracy of the power battery aging influence coefficient. When determining the operation safety detection score of the battery, the operation safety detection score of the battery can be determined according to the similarity change rate, battery internal resistance, first standard deviation, cell temperature, and power battery aging influence coefficient. During the calculation process, the operation safety condition of the power battery under high temperature and low temperature conditions can be accurately analyzed. Further, the safety performance of the power battery can be evaluated respectively from three aspects of the temperature distribution balance condition of each cell of the battery, the internal resistance safety condition of the battery, and the operation temperature condition of the battery, improving the comprehensiveness and objectivity of the operation safety detection score of the battery.

[0025] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present invention. Other features and aspects of the present invention will be clearer according to the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other embodiments can also be obtained based on these drawings;

[0027] Figure 1 Exemplarily shown is a schematic flowchart of a method for detecting the safety of a power battery of an unmanned vehicle based on big data according to an embodiment of the present invention;

[0028] Figure 2 Exemplarily shown is a schematic diagram of a system for detecting the safety of a power battery of an unmanned vehicle based on big data according to an embodiment of the present invention. Detailed Embodiments

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, rather than all, embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0030] The following will detail the technical solutions of the present invention with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0031] Figure 1 Exemplarily shown is a schematic flowchart of a method for detecting the safety of a power battery of an unmanned vehicle based on big data according to an embodiment of the present invention. The method includes:

[0032] Step S101: Collect an image of the appearance of the power battery through an image acquisition device;

[0033] Step S102: Determine a safety detection score for the appearance of the power battery based on the battery appearance image;

[0034] Step S103: Obtain historical operation data of the unmanned vehicle in a historical operation cycle, where the historical operation data includes: historical operation environment data, historical operation mileage, total number of charging times, and battery discharge depth;

[0035] Step S104: Determine an aging influence coefficient of the power battery based on the historical operation data;

[0036] Step S105, at multiple moments during the detection period, obtain the battery parameters of the power battery through the battery management system, where the battery parameters include: battery current, battery voltage, and battery temperature;

[0037] Step S106, at multiple moments during the detection period, obtain the internal resistance of the power battery through a detection instrument;

[0038] Step S107, determine the battery operation safety detection score according to the internal resistance of the battery, the battery parameters, and the power battery aging influence coefficient;

[0039] Step S108, generate a power battery safety detection report according to the battery operation safety detection score and the battery appearance safety detection score.

[0040] According to the method for detecting the safety of the power battery of an unmanned vehicle based on big data according to an embodiment of the present invention, the appearance safety condition of the power battery can be evaluated according to the appearance image of the power battery, the battery appearance safety detection score can be determined, the influence of the historical operation condition of the unmanned vehicle on the aging degree of the power battery can be accurately analyzed, and based on this influence and the internal resistance of the power battery, battery current, battery voltage, and battery temperature during the detection period, the operation safety condition of the power battery can be evaluated, the battery operation safety detection score can be determined. Further, according to the battery operation safety detection score and the battery appearance safety detection score, a comprehensive evaluation of the power battery safety can be performed.

[0041] According to an embodiment of the present invention, in step S101, collect the battery appearance image of the power battery through an image acquisition device.

[0042] For example, collect the appearance images of all angles of the power battery through a camera.

[0043] According to an embodiment of the present invention, in step S102, determine the battery appearance safety detection score according to the battery appearance image.

[0044] According to an embodiment of the present invention, step S102 includes:

[0045] Determining the battery appearance safety detection score according to the battery appearance image includes:

[0046] In the battery appearance image, identify whether there is a crack in the housing of the power battery through an image detection model, and determine the crack recognition result;

[0047] In the battery appearance image, identify whether there is a depression in the housing of the power battery through an image detection model, and determine the depression recognition result;

[0048] In the battery appearance image, use an image detection model to identify whether there is corrosion on the housing of the power battery and determine the corrosion recognition result;

[0049] In the battery appearance image, use an image detection model to identify whether there is damage to the cables of the power battery and determine the damage recognition result;

[0050] In the battery appearance image, use an image detection model to identify whether the battery connectors of the power battery are loose and determine the looseness recognition result;

[0051] According to the crack recognition result, the dent recognition result, the corrosion recognition result, the damage recognition result and the looseness recognition result, determine the appearance safety detection score of the power battery.

[0052] For example, the image detection model belongs to a type of deep learning model. The image detection model is trained with historical data so that it can identify whether there are various defects in the appearance and connection of the power battery. Use the image detection model to identify the housing of the power battery. If there is a crack in the housing, the crack recognition result is 0; if there is no crack in the housing, the crack recognition result is 1; if there is a dent in the housing, the dent recognition result is 0; if there is no dent in the housing, the dent recognition result is 1; if there is corrosion on the housing, the corrosion recognition result is 0; if there is no corrosion on the housing, the corrosion recognition result is 1; if there is damage to the cables of the power battery, the damage recognition result is 0; if there is no damage to the cables of the power battery, the damage recognition result is 1; if there is looseness in the battery connectors of the power battery, the looseness recognition result is 0; if there is no looseness in the battery connectors of the power battery, the looseness recognition result is 1; According to the crack recognition result, the dent recognition result, the corrosion recognition result, the damage recognition result and the looseness recognition result, evaluate the appearance safety condition of the power battery and determine the appearance safety detection score of the power battery.

[0053] According to an embodiment of the present invention, determining the appearance safety detection score of the power battery according to the crack recognition result, the dent recognition result, the corrosion recognition result, the damage recognition result and the looseness recognition result includes: determining the appearance safety detection score of the power battery according to formula (1) ,

[0054] (1)

[0055] Wherein, is the crack recognition result, is the dent recognition result, is the corrosion recognition result, is the damage recognition result, is the looseness recognition result, , , , , .

[0056] According to an embodiment of the present invention, is the crack recognition result. If there is a crack in the battery housing, it indicates a potential safety hazard. The value of is 0, otherwise is the dent recognition structure. If there is a dent in the battery housing, it indicates a potential safety hazard. The value of is 0, otherwise is the corrosion recognition result. If there is corrosion in the battery housing, it indicates a potential safety hazard. The value of is 0, otherwise is the damage recognition result. If there is damage to the power battery cable, it indicates a potential safety hazard. The value of is 0, otherwise is the looseness recognition result. If there is looseness in the battery connector of the power battery, it indicates a potential safety hazard. The value of is 0, otherwise ; Multiply the crack recognition result, dent recognition result, corrosion recognition result, damage recognition result and damage recognition result to obtain the appearance safety detection score of the power battery. A value of 1 for indicates that the appearance of the power battery is relatively intact, and the battery connector and cable are firm, without looseness and damage, and the appearance safety of the battery is relatively high.

[0057] In this way, multiple elements in the battery appearance image can be detected through an artificial intelligence model, so as to evaluate the appearance safety status of the power battery from multiple aspects, improving the accuracy and objectivity of the appearance safety detection score.

[0058] According to an embodiment of the present invention, in step S103, historical operation data of the unmanned vehicle in the historical operation cycle is obtained, where the historical operation data includes: historical operation environment data, historical operation mileage, total number of charging times, and battery discharge depth.

[0059] For example, during the operation of an autonomous vehicle, a large amount of historical operation data will be accumulated, including the total number of battery charges (the total number of charges of the vehicle in multiple historical operation cycles), the depth of discharge of the battery (the percentage of the discharged battery capacity relative to the total battery capacity), the historical operation mileage of the autonomous vehicle, and the road conditions and environmental data during the operation of the autonomous vehicle, that is, historical environmental data. Among them, the depth of discharge of the battery represents the depth of discharge of the autonomous vehicle in a historical operation cycle.

[0060] According to an embodiment of the present invention, in step S104, according to the historical operation data, a power battery aging influence coefficient is determined.

[0061] According to an embodiment of the present invention, step S104 includes:

[0062] According to the historical operation environment data, the historical operation temperature and the historical operation humidity are determined;

[0063] The historical operation temperature and the time in the historical operation cycle are fitted to obtain the historical operation temperature function in the historical operation cycle;

[0064] The historical operation humidity and the time in the historical operation cycle are fitted to obtain the historical operation humidity function in the historical operation cycle;

[0065] According to the historical operation humidity function, the historical operation temperature function, the historical operation mileage, the total number of charges, and the depth of discharge of the battery, a power battery aging influence coefficient is determined.

[0066] For example, the historical operation temperature and the time in the corresponding historical operation cycle are fitted to obtain the historical operation temperature function of the historical operation temperature in each historical operation cycle; the historical operation humidity and the time in the corresponding historical operation cycle are fitted to obtain the historical operation humidity function of the historical operation humidity in each historical operation cycle; according to the historical operation humidity function, the historical operation temperature function, the historical operation mileage of the autonomous vehicle, the number of charges of the power battery, and the depth of discharge of the power battery, the influence of the usage degree of the autonomous vehicle on the aging of the power battery is evaluated, and a power battery aging influence coefficient is determined.

[0067] According to an embodiment of the present invention, determining the power battery aging influence coefficient according to the historical operation humidity function, the historical operation temperature function, the historical operation mileage, the total number of charges, and the depth of discharge of the battery includes: determining the power battery aging influence coefficient according to formula (2) ,

[0068] (2)

[0069] Wherein, , , , and are preset weight values, is the start time of the historical operation period, is the nth moment of the historical operation period, is the preset temperature threshold, is the preset humidity threshold, is the historical operation temperature function of the kth historical operation period, is the historical operation humidity function of the kth historical operation period, is the total charging times of the unmanned vehicle, is the preset charging times threshold, is the depth of discharge of the battery in the kth historical operation period, is the preset depth of discharge threshold, is the historical operation mileage of the unmanned vehicle, is the preset operation mileage threshold, K is the number of historical operation periods, k ≤ K, and both k and K are positive integers.

[0070] According to an embodiment of the present invention, is the relative difference between the historical operation temperature at time t in the kth historical operation period and the preset temperature threshold. The larger this ratio is, the greater the difference between the historical operation temperature at time t in the kth historical operation period and the preset temperature threshold, and the greater the impact of the historical operation temperature at time t in the kth historical operation period on the aging of the power battery. When the historical operation temperature is too high, the negative electrode of the battery will reduce the electrolyte, resulting in the loss of active lithium ions and accelerating battery aging. When the historical operation temperature is too low, the activity of the active material of the battery decreases, resulting in a reduction in the available capacity of the battery and accelerating battery aging. is the integration of within the kth historical operation period, indicating the impact of the operation temperature in the kth historical operation period on the aging of the power battery. is the relative difference between the historical operation humidity at time t in the kth historical operation period and the preset humidity threshold. The larger this ratio is, the greater the historical operation humidity at time t in the kth historical operation period relative to the preset humidity threshold. Excessive humidity will accelerate the chemical reaction inside the battery, resulting in the aging of the electrode material and the decomposition of the electrolyte, accelerating battery aging, and the greater the impact of the historical operation humidity at time t in the kth historical operation period on the aging of the power battery. is the integration of within the kth historical operation period, indicating the impact of the operation humidity in the kth historical operation period on the aging of the power battery. Indicates the degree of influence of the operating environment of the unmanned vehicle in K historical cycles on the aging of the power battery. Is the relative difference between the depth of discharge of the battery in the k-th historical operating cycle and the preset depth-of-discharge threshold. The larger this ratio, the greater the depth of discharge in the k-th historical operating cycle relative to the preset depth-of-discharge threshold. During deep discharge, the chemical reactions inside the battery are more intense, causing greater damage to the battery structure and accelerating battery aging. The depth of discharge of the power battery in the k-th historical operating cycle has a greater impact on the aging of the power battery. Indicates the degree of influence of the depth of discharge of the unmanned vehicle in K historical cycles on the aging of the power battery. Is the relative difference between the total number of charge cycles of the unmanned vehicle and the preset number of charge cycles. The larger this ratio, the more the total number of charge cycles of the unmanned vehicle relative to the preset number of charge cycles. Frequent charging will accelerate the aging process of the battery and shorten its overall service life. The total number of charge cycles of the unmanned vehicle in K historical cycles has a greater impact on the aging of the power battery. Is the relative difference between the historical operating mileage of the unmanned vehicle and the preset operating mileage. The larger this ratio, the greater the historical operating mileage of the unmanned vehicle relative to the preset operating mileage. The historical operating mileage of the unmanned vehicle in K historical cycles has a greater impact on the aging of the power battery. Indicates that according to the operating environment, depth of discharge, total number of charge cycles, and historical operating mileage of the unmanned vehicle in K historical cycles, the power battery aging influence coefficient is determined.

[0071] In this way, the power battery aging influence coefficient can be determined according to the historical operating temperature function, historical operating mileage, total number of charge cycles, and battery depth of discharge. During the calculation process, the influence of the historical operating conditions of the unmanned vehicle on the aging of the power battery can be evaluated respectively from four aspects: the operating environment, depth of discharge, total number of charge cycles, and historical operating mileage of the unmanned vehicle, improving the comprehensiveness and accuracy of the power battery aging influence coefficient.

[0072] According to an embodiment of the present invention, in step S105, at multiple moments in the detection cycle, the battery management system obtains battery parameters of the power battery, where the battery parameters include: battery current, battery voltage, and battery temperature.

[0073] For example, in the detection cycle, the battery management system (BMS) obtains the battery current, battery voltage, and battery temperature of the battery, where the battery temperature represents the overall temperature of the power battery.

[0074] According to an embodiment of the present invention, in step S106, at multiple moments in the detection cycle, the battery internal resistance of the power battery is obtained through a detection instrument.

[0075] For example, through the battery dynamic test function of professional new energy equipment (such as 909CEV), precise battery internal resistance testing is performed on the power battery.

[0076] According to an embodiment of the present invention, in step S107, based on the battery internal resistance, the battery parameters, and the power battery aging influence coefficient, a battery operation safety detection score is determined.

[0077] According to an embodiment of the present invention, step S107 includes:

[0078] Based on the battery temperature, the cell temperatures of each cell of the power battery are determined;

[0079] Based on the cell temperatures of each cell, the first standard deviation of the cell temperatures is determined;

[0080] Based on the battery current, the battery voltage, and the battery temperature, a real-time battery state vector is determined;

[0081] The normal battery current, normal battery voltage, and normal battery temperature under normal battery operation conditions are obtained;

[0082] Based on the normal battery current, the normal battery voltage, and the normal battery temperature, a normal battery state vector is determined;

[0083] Based on the cell temperature, the battery internal resistance, the real-time battery state vector, the normal battery state vector, the first standard deviation, the battery parameters, and the power battery aging influence coefficient, a battery operation safety detection score is determined.

[0084] For example, the power battery is composed of multiple cells. Through the temperature acquisition system in the BMS, the cell temperatures of each cell of the power battery are determined; based on the cell temperatures of each cell at the j-th moment of the detection period, the first standard deviation of the cell temperatures at the j-th moment of the detection period is determined; based on the battery current, battery voltage, and battery temperature, a real-time battery state vector is combined to represent the battery state of the power battery during the detection period; based on the normal battery current, normal battery voltage, and normal battery temperature under normal battery operation conditions, a normal battery state vector is combined to represent the normal battery operation state; based on the battery internal resistance, real-time battery state vector, normal battery state, first standard deviation, cell temperature, and power battery aging influence coefficient, the safety condition of the battery during the detection period is evaluated to determine the battery operation safety detection score.

[0085] According to an embodiment of the present invention, determining a battery operation safety detection score based on the cell temperature, the battery internal resistance, the real-time battery state vector, the normal battery state vector, the first standard deviation, the battery parameters, and the power battery aging influence coefficient includes:

[0086] Determining a battery state similarity based on the real-time battery state vector and the normal battery state vector;

[0087] Fitting the battery state similarity and the time in the detection period to obtain a similarity function of the battery state similarity in the detection period;

[0088] Obtaining a similarity derivative function according to the similarity function;

[0089] Determining the similarity change rates at multiple times in the current detection period according to the similarity derivative function;

[0090] Determining a battery operation safety detection score based on the similarity change rate, the battery internal resistance, the first standard deviation, the cell temperature, and the power battery aging influence coefficient.

[0091] For example, performing a cosine similarity operation based on the real-time battery state vector and the normal battery state vector to determine the battery state similarity. The greater the battery state similarity, the closer the battery temperature, battery voltage, and battery current of the power battery in the detection period are to the normal battery current, normal battery voltage, and normal battery temperature under normal operating conditions, the more normal the operating condition of the power battery in the detection period, and the higher the safety of the power battery; fitting the battery state similarity and the time in the detection period to obtain a similarity function for describing the change law of the battery state similarity in the detection period; taking the derivative of the similarity function to obtain a similarity derivative function; substituting the time in the detection period into the similarity derivative function to determine the similarity change rates at multiple times in the current detection period; evaluating the operating safety condition of the power battery in the detection period according to the similarity change rate, the battery internal resistance, the first standard deviation, the cell temperature, and the power battery aging influence coefficient, and determining a battery operation safety detection score.

[0092] According to an embodiment of the present invention, determining a battery operation safety detection score based on the similarity change rate, the battery internal resistance, the first standard deviation, the cell temperature, and the power battery aging influence coefficient includes: determining the battery operation safety detection score according to formula (3) ,

[0093] (3)

[0094] Wherein, , and is a preset weight value, is a preset multiple, and if is a conditional function. is the j-th moment of the detection period, is the similarity change rate at the j-th moment of the detection period, is the first preset similarity change rate threshold, is the second preset similarity change rate threshold, is the internal resistance of the battery at the r-th moment of the detection period, is the internal resistance of the battery at the m-th moment of the detection period, is the preset battery internal resistance threshold, is the influence coefficient of power battery aging, is the temperature of the e-th battery cell at the r-th moment of the detection period, is the first standard deviation of the battery cell temperature at the r-th moment of the detection period, is the temperature of the e-th battery cell at the m-th moment of the detection period, is the first standard deviation of the battery cell temperature at the m-th moment of the detection period. E is the number of battery cells of the power battery, e ≤ E, m is the number of moments of the detection period, r is the first preset moment, s is the second preset moment, r < s < m, and e, E, r, s, and m are all positive integers.

[0095] According to an embodiment of the present invention, within the 1 to r moments of the detection period, the experimental environment temperature in the experimental space rises uniformly, and the experimental environment temperature at the r-th moment of the detection period is the maximum value of the experimental environment temperature of the detection period (for example, the experimental environment temperature at the r-th moment of the detection period is 50 degrees Celsius), which is used to simulate the operating safety status of the battery under high temperature. Within the r to m moments of the detection period, the experimental environment temperature in the experimental space drops uniformly, and the experimental environment temperatures at the s-th moment and the 1st moment of the detection period are normal room temperatures (for example, the experimental environment temperatures at the s-th moment and the 1st moment of the detection period are 20 degrees Celsius), and the experimental environment temperature at the m-th moment of the detection period is the minimum value of the experimental environment temperature of the detection period (for example, the experimental environment temperature at the m-th moment of the detection period is minus 20 degrees Celsius), r < s < m, which is used to simulate the operating safety status of the battery under low temperature.

[0096] According to an embodiment of the present invention, in formula (3), the conditional function has the following two cases for its value. When the condition of is satisfied, it means that the temperature of the e-th battery cell at the r-th moment of the detection period is within the interval centered on the average battery cell temperature and with a standard deviation of twice the preset multiple as the interval length, and the value of the conditional function is 1. When the condition of is not satisfied,When the condition is met, the value of the conditional function is 0. If the cell temperature of the e-th cell at the r-th moment of the detection period meets the above conditions, it means that the deviation between the cell temperature of the e-th cell at the r-th moment of the detection period and the average value is small, and the possibility of abnormal cell temperature of this cell is small. Otherwise, it can be considered that the possibility of abnormal cell temperature of this cell is large. It represents the ratio of the number of cells with normal cell temperature to the total number of cells. The larger this ratio is, the more balanced the temperature distribution of each battery of the power battery is under high-temperature conditions, and the better the safety condition of the power battery under high temperature is.

[0097] According to an embodiment of the present invention, in formula (3), the conditional function has the following two cases for its value. When meeting the condition, it means that the cell temperature of the e-th cell at the m-th moment of the detection period is within the interval centered on the average cell temperature and with a standard deviation of twice the preset multiple as the interval length, and the value of the conditional function is 1. When not meeting the condition, the value of the conditional function is 0. If the cell temperature of the e-th cell at the m-th moment of the detection period meets the above conditions, it means that the deviation between the cell temperature of the e-th cell at the m-th moment of the detection period and the average value is small, and the possibility of abnormal cell temperature of this cell is small. Otherwise, it can be considered that the possibility of abnormal cell temperature of this cell is large. It represents the ratio of the number of cells with normal cell temperature to the total number of cells. The larger this ratio is, the more balanced the temperature distribution of each battery of the power battery is under low-temperature conditions, and the better the safety condition of the power battery under low temperature is.

[0098] According to an embodiment of the present invention, It represents the safety condition of the temperature distribution of the cells of the power battery under extreme environmental conditions. The larger this value is, the better the safety condition of the temperature distribution of the cells of the power battery under extreme environmental conditions is.

[0099] According to an embodiment of the present invention, as the battery ages, its internal structure and chemical composition change, resulting in an increase in the battery internal resistance. is the product of the preset battery internal resistance threshold and the power battery aging influence coefficient, representing the battery internal resistance threshold considering the battery aging influence factor. is the relative difference between the battery internal resistance at the r-th moment of the detection period and the battery internal resistance threshold considering the battery aging influence factor. The larger this ratio is, the larger the battery internal resistance at the r-th moment of the detection period is relative to the battery internal resistance threshold considering the battery aging influence factor, indicating that the battery internal resistance of the power battery under high-temperature conditions is large, resulting in a large decline in battery performance and a large possibility of explosion risk, and the battery internal resistance safety condition of the power battery under high-temperature conditions is poor. For the relative difference between the battery internal resistance at the m-th moment of the detection period and the battery internal resistance threshold considering the influence of battery aging, the larger this ratio is, the greater the battery internal resistance at the m-th moment of the detection period is relative to the battery internal resistance threshold considering the influence of battery aging, indicating that the battery internal resistance of the power battery is relatively large under low-temperature conditions, resulting in a relatively large decline in battery performance and a relatively high risk of explosion, and the safety condition of the battery internal resistance of the power battery under low-temperature conditions is relatively poor. It represents the safety condition of the battery internal resistance under extreme temperature conditions. The smaller this value is, the better the safety condition of the battery internal resistance under extreme temperature conditions.

[0100] According to an embodiment of the present invention, It is the average value of the similarity change rates at the first r moments of the detection period, representing the change condition of the operating state of the power battery under high-temperature conditions. The smaller this average value is, the less the operating condition of the power battery is affected by high temperature, and the better the safety condition of the power battery is. It is the relative difference between the first preset similarity change rate threshold and the average value of the similarity change rates at the first r moments of the detection period. The larger this ratio is, the smaller the average value of the similarity change rates at the first r moments of the detection period is relative to the first preset similarity change rate threshold, the less the operating condition of the power battery is affected by high temperature, the higher the operating stability of the power battery under high-temperature conditions, and the better the safety condition of the power battery is. It is the relative difference between the second preset similarity change rate threshold and the average value of the similarity change rates from the s-th moment to the m-th moment of the detection period. The larger this ratio is, the smaller the average value of the similarity change rates from the s-th moment to the m-th moment of the detection period is relative to the second preset similarity change rate threshold, the less the operating condition of the power battery is affected by low temperature, the better the operating stability of the power battery under low-temperature conditions, and the better the safety condition of the power battery is. It represents the operating stability condition of the power battery under extreme environmental conditions. The larger this ratio is, the more stable the power battery operates under extreme environmental conditions, and the better the safety condition of the power battery is.

[0101] According to an embodiment of the present invention, It represents determining the battery operation safety detection score based on three aspects: the temperature distribution balance condition of each battery cell, the battery internal resistance safety condition, and the battery operating temperature condition.

[0102] In this way, the battery operation safety detection score can be determined according to the similarity change rate, battery internal resistance, first standard deviation, cell temperature, and the influence coefficient of power battery aging. During the calculation process, the operation safety status of the power battery under high and low temperature conditions can be accurately analyzed. Further, the safety performance of the power battery is evaluated from three aspects: the temperature distribution balance status of each cell of the battery, the battery internal resistance safety status, and the battery operation temperature status, improving the comprehensiveness and objectivity of the battery operation safety detection score.

[0103] According to an embodiment of the present invention, in step S108, a power battery safety detection report is generated based on the battery operation safety detection score and the battery appearance safety detection score.

[0104] For example, if the battery appearance safety detection score is less than the preset battery appearance safety detection score threshold, a battery appearance safety alarm message is generated. If the battery operation safety detection score is less than the preset battery operation safety detection score threshold, a battery operation safety alarm message is generated.

[0105] The method for safety detection of power batteries of driverless vehicles based on big data according to an embodiment of the present invention can evaluate the appearance safety condition of a power battery based on the appearance image of the power battery, determine the appearance safety detection score of the battery, accurately analyze the influence of the historical operation condition of the driverless vehicle on the aging degree of the power battery, and evaluate the operation safety condition of the power battery based on this influence and the internal resistance, battery current, battery voltage and battery temperature of the power battery in the detection cycle, determine the operation safety detection score of the battery. Further, based on the operation safety detection score and the appearance safety detection score of the battery, a comprehensive evaluation of the safety of the power battery is carried out. When determining the appearance safety detection score of the battery, various elements in the battery appearance image can be detected through an artificial intelligence model, so as to evaluate the appearance safety condition of the power battery from multiple aspects, improving the accuracy and objectivity of the appearance safety detection score. When determining the aging influence coefficient of the power battery, the aging influence coefficient of the power battery can be determined according to the historical operation humidity function, historical operation temperature function, historical operation mileage, total number of charge times and battery discharge depth. During the operation process, the influence of the historical operation condition of the driverless vehicle on the aging of the power battery can be evaluated respectively from four aspects: the operation environment, discharge depth, total number of charge times and historical operation mileage of the driverless vehicle, improving the comprehensiveness and accuracy of the aging influence coefficient of the power battery. When determining the operation safety detection score of the battery, the operation safety detection score of the battery can be determined according to the similarity change rate, internal resistance of the battery, first standard deviation, cell temperature and aging influence coefficient of the power battery. During the operation process, the operation safety condition of the power battery under high temperature and low temperature conditions can be accurately analyzed. Further, the safety performance of the power battery is evaluated respectively from three aspects: the temperature distribution balance condition of each cell of the battery, the internal resistance safety condition of the battery and the battery operation temperature condition, improving the comprehensiveness and objectivity of the operation safety detection score of the battery.

[0106] Figure 2 Exemplarily shown is a schematic diagram of a system for safety detection of power batteries of driverless vehicles based on big data according to an embodiment of the present invention. The system includes:

[0107] An image acquisition module, configured to acquire a battery appearance image of a power battery through an image acquisition device;

[0108] An appearance safety module, configured to determine an appearance safety detection score of the power battery according to the battery appearance image;

[0109] A historical data module, configured to obtain historical operation data of the driverless vehicle in a historical operation cycle, where the historical operation data includes: historical operation environment data, historical operation mileage, total number of charge times and battery discharge depth;

[0110] An aging impact module, configured to determine an aging impact coefficient of a power battery according to the historical operation data;

[0111] A battery parameter module, configured to obtain battery parameters of the power battery through a battery management system at multiple moments in a detection period, wherein the battery parameters include: battery current, battery voltage, and battery temperature;

[0112] A battery internal resistance module, configured to obtain the battery internal resistance of the power battery through a detection instrument at multiple moments in the detection period;

[0113] An operation safety module, configured to determine a battery operation safety detection score according to the battery internal resistance, the battery parameters, and the aging impact coefficient of the power battery;

[0114] A detection report module, configured to generate a power battery safety detection report according to the battery operation safety detection score and the battery appearance safety detection score.

[0115] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The object of the present invention has been fully and effectively achieved. The function and structural principle of the present invention have been shown and described in the embodiments. Without departing from the principle, the embodiments of the present invention can have any deformation or modification.

Claims

1. A method for detecting the safety of power batteries of driverless vehicles based on big data, characterized in that, Including: Collect the battery appearance image of the power battery through an image acquisition device; Determine the battery appearance safety detection score of the power battery according to the battery appearance image; Obtain the historical operation data of the unmanned vehicle in the historical operation cycle, where the historical operation data includes: historical operation environment data, historical operation mileage, total number of charging times, and battery discharge depth; Determine the power battery aging influence coefficient according to the historical operation data; At multiple moments in the detection cycle, obtain the battery parameters of the power battery through the battery management system, where the battery parameters include: battery current, battery voltage, and battery temperature; At multiple moments in the detection cycle, obtain the internal resistance of the power battery through a detection instrument; Determine the battery operation safety detection score according to the battery internal resistance, the battery parameters, and the power battery aging influence coefficient; Generate a power battery safety detection report according to the battery operation safety detection score and the battery appearance safety detection score; Determine the battery operation safety detection score according to the battery internal resistance, the battery parameters, and the power battery aging influence coefficient, including: According to the formula Determine the safety detection score of battery operation , where , and are preset weight values, is a preset multiple, if is a conditional function, is the j-th moment of the detection period, is the similarity change rate at the j-th moment of the detection period, is the first preset similarity change rate threshold, is the second preset similarity change rate threshold, is the internal resistance of the battery at the r-th moment of the detection period, is the internal resistance of the battery at the m-th moment of the detection period, is the preset battery internal resistance threshold, is the influence coefficient of power battery aging, is the cell temperature of the e-th cell at the r-th moment of the detection period, is the first standard deviation of the cell temperature at the r-th moment of the detection period, is the cell temperature of the e-th cell at the m-th moment of the detection period, is the first standard deviation of the cell temperature at the m-th moment of the detection period, E is the number of cells of the power battery, e ≤ E, m is the number of moments of the detection period, r is the first preset moment, s is the second preset moment, r < s < m, e, E, r, s, and m are all positive integers.

2. The method for detecting the safety of the power battery of an unmanned vehicle based on big data according to claim 1, wherein Determine the battery appearance safety detection score of the power battery according to the battery appearance image, including: In the battery appearance image, identify whether there are cracks in the outer shell of the power battery through an image detection model, and determine the crack identification result; In the battery appearance image, identify whether there are depressions in the outer shell of the power battery through an image detection model, and determine the depression identification result; In the battery appearance image, identify whether there is corrosion in the outer shell of the power battery through an image detection model, and determine the corrosion identification result; In the battery appearance image, identify whether there is damage to the cable of the power battery through an image detection model, and determine the damage identification result; In the battery appearance image, identify whether the battery connector of the power battery is loose through an image detection model, and determine the looseness identification result; Determine the battery appearance safety detection score of the power battery according to the crack identification result, the depression identification result, the corrosion identification result, the damage identification result, and the looseness identification result.

3. The method for detecting the safety of the power battery of an unmanned vehicle based on big data according to claim 2, wherein, Determine the battery appearance safety detection score according to the crack identification result, the depression identification result, the corrosion identification result, the damage identification result, and the looseness identification result, including: According to the formula Determine the battery appearance safety detection score of the power battery , where is the crack recognition result, is the dent recognition result, is the corrosion recognition result, is the damage recognition result, is the looseness recognition result, , , , , .

4. The method for detecting the safety of the power battery of an unmanned vehicle based on big data according to claim 1, wherein, Determine the power battery aging influence coefficient according to the historical operation data, including: Determine the historical operation temperature and historical operation humidity according to the historical operation environment data; Fit the historical operation temperature and the moments in the historical operation cycle to obtain the historical operation temperature function in the historical operation cycle; Fit the historical operation humidity and the moments in the historical operation cycle to obtain the historical operation humidity function in the historical operation cycle; Determine the power battery aging influence coefficient according to the historical operation humidity function, the historical operation temperature function, the historical operation mileage, the total number of charging times, and the battery discharge depth.

5. The method for detecting the safety of the power battery of an unmanned vehicle based on big data according to claim 4, characterized in that Determine the aging influence coefficient of the power battery according to the historical operating humidity function, the historical operating temperature function, the historical operating mileage, the total number of charge cycles, and the depth of discharge of the battery, including: According to the formula Determine the aging influence coefficient of the power battery , where , , , and are preset weights, is the start time of the historical operation cycle, is the nth moment of the historical operation cycle, is the preset temperature threshold, is the preset humidity threshold, is the historical operation temperature function of the kth historical operation cycle, is the historical operation humidity function of the kth historical operation cycle, is the total number of charge times of the unmanned vehicle, is the preset charge times threshold, is the depth of discharge of the battery in the kth historical operation cycle, is the preset depth of discharge threshold, is the historical operation mileage of the unmanned vehicle, is the preset operation mileage threshold, K is the number of historical operation cycles, k ≤ K, and both k and K are positive integers.

6. The method for detecting the safety of the power battery of an unmanned vehicle based on big data according to claim 1, wherein, Determine the battery operation safety detection score according to the battery internal resistance, the battery parameters, and the aging influence coefficient of the power battery, including: Determine the cell temperature of each cell of the power battery according to the battery temperature; Determine the first standard deviation of the cell temperature according to the cell temperatures of each cell; Determine the real-time battery state vector according to the battery current, the battery voltage, and the battery temperature; Obtain the normal battery current, the normal battery voltage, and the normal battery temperature under normal battery operating conditions; Determine the normal battery state vector according to the normal battery current, the normal battery voltage, and the normal battery temperature; Determine the battery operation safety detection score according to the cell temperature, the battery internal resistance, the real-time battery state vector, the normal battery state vector, the first standard deviation, the battery parameters, and the aging influence coefficient of the power battery.

7. The method for detecting the safety of the power battery of an unmanned vehicle based on big data according to claim 6, wherein Determine the battery operation safety detection score according to the cell temperature, the battery internal resistance, the real-time battery state vector, the normal battery state vector, the first standard deviation, the battery parameters, and the aging influence coefficient of the power battery, including: Determine the battery state similarity according to the real-time battery state vector and the normal battery state vector; Fit the battery state similarity and the time in the detection period to obtain the similarity function of the battery state similarity in the detection period; Obtain the similarity derivative function according to the similarity function; Determine the similarity change rate at multiple times in the current detection period according to the similarity derivative function; Determine the battery operation safety detection score according to the similarity change rate, the battery internal resistance, the first standard deviation, the cell temperature, and the aging influence coefficient of the power battery.

8. A big data-based power battery safety detection system for an unmanned vehicle for performing the method according to any one of claims 1-7, characterized in that, Including: An image acquisition module for acquiring the battery appearance image of the power battery through an image acquisition device; An appearance safety module for determining the battery appearance safety detection score of the power battery according to the battery appearance image; A historical data module for acquiring the historical operation data of the unmanned vehicle in the historical operation period, where the historical operation data includes: historical operation environment data, historical operation mileage, total number of charge cycles, and depth of discharge of the battery; An aging influence module for determining the aging influence coefficient of the power battery according to the historical operation data; A battery parameter module for acquiring the battery parameters of the power battery through the battery management system at multiple times in the detection period, where the battery parameters include: battery current, battery voltage, and battery temperature; A battery internal resistance module for acquiring the battery internal resistance of the power battery through a detection instrument at multiple times in the detection period; An operation safety module for determining the battery operation safety detection score according to the battery internal resistance, the battery parameters, and the aging influence coefficient of the power battery; A detection report module, configured to generate a power battery safety detection report based on the battery operation safety detection score and the battery appearance safety detection score.

Images