Battery detection method, device, equipment and storage medium
Through the combined use of health detection model and life prediction model, battery failure is sensed in advance, and the problems of operation and maintenance lag and life prediction errors in the existing technology are solved, and the battery life and operation and maintenance costs are achieved at low failure rates. .
Patent Information
- Application Number
- CN202210970441.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-12
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-08-12
AI Technical Summary
The prior art is difficult to accurately and timely detect faults in battery operation and maintenance, resulting in large errors in operation and maintenance lag and life prediction, which cannot effectively extend the battery service life and reduce operation and maintenance costs.
The combined method of health detection model and life prediction model is adopted to obtain the battery feature vector to determine whether the performance reaches the qualitative change point, and health monitoring and life prediction are carried out before and after the qualitative change point, faults are sensed in advance, alarm information is generated, and operation and maintenance are guided.
On the premise of ensuring a low failure rate, extend the battery service life, reduce operation and maintenance costs, and improve the pertinence and accuracy of operation and maintenance.
Smart Images

Figure CN115291112B_ABST
Abstract
Description
Background Art
[0002] In battery operation and maintenance, the battery pre-maintenance status can be divided into three stages according to maturity: stage 1.0 - passive maintenance; stage 2.0 - preventive maintenance; stage 3.0 - predictive maintenance. The battery maintenance cost and the severity of the loss are inversely proportional to the time of hidden danger discovery. The earlier the hidden danger is discovered, the lower the maintenance cost and the smaller the loss caused by the mains failure to the data center. In the worst case, if the battery is not able to provide stable power when it is needed, it will cause the shutdown of the data center server and storage equipment, and even the loss of all data. The consequences include not only the cost of repairing and replacing the battery itself, but also the unpredictable business losses caused by the unavailability of data services to users or the loss of data.
[0003] Specifically, batteries are an important part of UPS (Uninterruptible Power System), which can provide power to data center equipment in a timely manner in the event of power outages or other emergencies. Since batteries are normally in a floating charge state, there is a risk of failure and loss of control. Current battery maintenance mostly uses passive maintenance or manual regular inspection methods, which is time-consuming and labor-intensive, and it is difficult to accurately and timely detect battery failures. Once a power outage occurs, it will cause huge losses.
[0004] At present, in the battery operation and maintenance solutions based on AI (artificial intelligence), health monitoring and life prediction functions are considered separately. Health monitoring alone cannot predict the occurrence of failures, and operation and maintenance are lagging. In different stages, single life prediction will lead to problems such as the lack of obvious trends in the service life prediction of newer batteries and the inability to obtain good prediction performance, while the life prediction of batteries in the performance degradation period does not take into account the characteristics of the battery performance change trend in the early stage.
[0005] Therefore, in view of the limitations brought about by the existing separate considerations of battery health monitoring and life prediction (operation and maintenance lags caused by single health monitoring, and large errors in results caused by single life prediction), how to extend the battery service life and reduce operation and maintenance costs while ensuring a low failure rate.
[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention
[0007] In view of the problems in the prior art, the purpose of the present invention is to provide a battery detection method, device, equipment and storage medium, which overcome the difficulties of the prior art and extend the service life of the battery and reduce the operation and maintenance costs while ensuring a low failure rate.
[0008] An embodiment of the present invention provides a battery detection method, comprising:
[0009] Obtaining a battery feature vector of the battery to be tested;
[0010] Inputting the battery feature vector into a health detection model;
[0011] Determining whether the performance of the battery to be tested has reached a qualitative change point according to the output of the health detection model;
[0012] In response to the performance of the battery to be tested reaching a qualitative change point, inputting the vector sequence of the battery feature vector into a life prediction model; and
[0013] According to the output of the life prediction model, it is determined whether to perform a detection alarm on the battery to be detected.
[0014] In some embodiments of the present application, it also includes:
[0015] In response to the performance of the battery to be detected reaching a qualitative change point, a battery feature vector of the battery to be detected is used as a training sample of the health detection model.
[0016] In some embodiments of the present application, the health detection model is a neural network model.
[0017] In some embodiments of the present application, it also includes:
[0018] In response to determining to perform a detection alarm on the battery to be detected, the life curve of the battery to be detected is stored as a training sample of the life prediction model.
[0019] In some embodiments of the present application, the life prediction model is a neural network model.
[0020] In some embodiments of the present application, the health detection model and the life prediction model are jointly trained.
[0021] In some embodiments of the present application, the joint training objective of the health detection model and the life prediction model is to minimize the operation and maintenance cost, and the operation and maintenance cost is calculated according to the following formula:
[0022]
[0023] Among them, ω1 and ω2 are trade-off parameters, C iis the cost of the i-th battery replacement, n is the total number of battery replacements, G j is the loss of the jth battery failure, and k is the total number of battery failures.
[0024] According to another aspect of the present application, a battery detection device is also provided, comprising:
[0025] An acquisition module configured to acquire a battery feature vector of a battery to be detected;
[0026] A first input module, configured to input the battery feature vector into a health detection model;
[0027] A qualitative change point determination module, configured to determine whether the performance of the battery to be tested reaches a qualitative change point according to the output of the health detection model;
[0028] A second input module, configured to input the vector sequence of the battery characteristic vector into a life prediction model in response to the performance of the battery to be detected reaching a qualitative change point; and
[0029] The alarm determination module is configured to determine whether to perform a detection alarm on the battery to be detected according to the output of the life prediction model.
[0030] According to another aspect of the present invention, there is also provided a battery detection processing device, comprising:
[0031] processor;
[0032] a memory storing executable instructions of the processor;
[0033] The processor is configured to execute the steps of the battery detection method described above by executing the executable instructions.
[0034] An embodiment of the present invention further provides a computer-readable storage medium for storing a program, wherein the program implements the steps of the above-mentioned battery detection method when executed.
[0035] Compared with the prior art, the purpose of the present invention is:
[0036] Health monitoring and life prediction are considered together, and targeted treatment is made for different stages of the battery life cycle. The concept of qualitative change point is creatively introduced. The qualitative change point is found through health monitoring in the stage of stable performance before the qualitative change point. In the stage of sudden performance drop after the qualitative change point, the performance change characteristics of the monitoring process are taken into account for life prediction. Faults are perceived in advance, alarm information is generated, and more effective guidance is provided for operation and maintenance, thereby extending the battery service life and reducing operation and maintenance costs while ensuring a low failure rate. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Other features, objectives and advantages of the present invention will become more apparent from a reading of the detailed description of non-limiting embodiments made with reference to the following accompanying drawings.
[0038] Figure 1 It is a flow chart of an embodiment of the battery detection method of the present invention.
[0039] Figure 2 It is a flow chart of a specific embodiment of the battery detection method of the present invention.
[0040] Figure 3 Schematic diagram of battery performance curves according to an embodiment of the present invention.
[0041] Figure 4 It is a module schematic diagram of the long short-term memory unit of the life prediction model of the present invention.
[0042] Figure 5 It is a module diagram of an embodiment of a battery detection device of the present invention.
[0043] Figure 6 It is a module diagram of another embodiment of the battery detection device of the present invention.
[0044] Figure 7 It is a structural schematic diagram of the battery testing device of the present invention.
[0045] Figure 8 It is a schematic diagram of the structure of a computer-readable storage medium according to an embodiment of the present invention. DETAILED DESCRIPTION
[0046] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their repeated description will be omitted.
[0047] See also Figure 1 , Figure 1 1 is a flow chart of an embodiment of a battery detection method of the present invention. An embodiment of the present invention provides a battery detection method, comprising the following steps:
[0048] Step S110: Obtain a battery feature vector of the battery to be detected.
[0049] Specifically, step S110 may first obtain the voltage, current, resistance, usage time, operating temperature, etc. of the battery to be detected. After obtaining the voltage, current, resistance, and operating temperature of the battery to be detected, the obtained feature vectors may be quantized to obtain the battery feature vector.
[0050] Step S120: inputting the battery feature vector into a health detection model.
[0051] Step S130: Determine whether the performance of the battery to be tested has reached a qualitative change point based on the output of the health detection model.
[0052] Specifically, the health detection model can be a neural network model. The health detection model is used to determine whether the battery performance of the battery to be detected at the current time has reached a qualitative change point. The qualitative change point is used to distinguish the changing trend of battery performance during the battery life cycle. Before reaching the qualitative change point, the battery performance changes relatively steadily over time. After reaching the qualitative change point, the battery performance drops sharply over time. Therefore, the health detection model can be used to identify different states of the battery to be detected.
[0053] Step S140: In response to the performance of the battery to be tested reaching a qualitative change point, the vector sequence of the battery feature vector is input into a life prediction model.
[0054] Specifically, when the performance of the battery to be tested has not reached the qualitative change point, since the battery performance changes relatively steadily over time, the health detection model can be continued to detect until the performance of the battery to be tested reaches the qualitative change point. When the performance of the battery to be tested reaches the qualitative change point, since the battery performance drops sharply over time, the battery life can be predicted by the life prediction model.
[0055] Specifically, when the performance of the battery to be tested reaches a qualitative change point, the battery feature vector of the battery to be tested can be used as a training sample of the health detection model to update the sample data and sample labels of the training samples of the health detection model, thereby updating the training health detection model.
[0056] Step S150: Determine whether to perform a detection alarm on the battery to be detected based on the output of the life prediction model.
[0057] Specifically, the life prediction model is a neural network model. The life prediction model includes but is not limited to a long short-term memory network, so as to be able to predict the life of the battery to be detected based on the vector sequence of the battery feature vector. Further, the life prediction model is used to determine whether the life of the battery to be detected has reached an alarm point. When the alarm point is reached, it is determined to detect and alarm the battery to be detected. When the alarm point is not reached, the life prediction model continues to perform life prediction.
[0058] Specifically, in response to determining to conduct a detection alarm on the battery to be detected, the life curve of the battery to be detected is stored as a training sample of the life prediction model to update the sample data and sample labels of the training sample of the life prediction model, so that the trained life prediction model can be updated.
[0059] Furthermore, in some embodiments, the health detection model and the life prediction model are jointly trained to ensure the normal operation of the two models during detection.
[0060] Specifically, the joint training objective of the health detection model and the life prediction model is to minimize the operation and maintenance cost, and the operation and maintenance cost is calculated according to the following formula:
[0061]
[0062] Among them, ω1 and ω2 are trade-off parameters, C i is the cost of the i-th battery replacement, n is the total number of battery replacements, G j is the loss of the jth battery failure, and k is the total number of battery failures.
[0063] Therefore, the present application obtains the performance qualitative change point through the health monitoring system, and then starts the life prediction algorithm and issues a life alarm.
[0064] Specifically, this application proposes the concept of qualitative change points in the battery life cycle. The performance change curve of the battery is generally divided into two stages. The early stage is a period of relatively stable performance changes, during which the battery can work normally for a long time; the later stage is a period of rapid performance decay. During this period, the battery will experience a rapid decline in performance due to failures and aging. It is necessary to focus on observation and predict the time when the battery performance is unqualified. The qualitative change point is proposed mainly to play a distinguishing role, so that the battery operation and maintenance strategy can be targeted according to each stage.
[0065] In model training, the health monitoring and life prediction models can be jointly trained based on the processed data collection results, and data will be continuously collected and added to the training set based on the actual situation of the battery. Specifically, it first labels the existing data and conducts initial model training, and then puts the trained models into the health monitoring module and life prediction module for use. Further collect the battery characteristic parameter data after use to update the data and labels, and collect feedback results to update the model parameters.
[0066] In health detection, the main focus is on the stable performance period of the battery life cycle. In the health detection stage, the AI model trained by the algorithm model training module is used to monitor the current performance-related parameters of the battery. The collected characteristic parameters are analyzed to determine whether the qualitative change point has been reached, and the current characteristic change curve is recorded for use by subsequent modules.
[0067] In life prediction, the main target of the battery life cycle is the performance drop period. The main function of this life prediction stage is to determine the time when the battery performance is unqualified, predict the location of the alarm point, and provide information for the alarm stage. By analyzing the changes in the early performance trend of the battery, the AI model provided by the algorithm model training module is used to predict the future performance trend curve of the battery, and finally a comprehensive result of the life prediction is given.
[0068] In the alarm, based on the predicted battery health status and fault occurrence time, an alarm message is generated to remind the operation and maintenance personnel to replace the battery in time.
[0069] Therefore, health monitoring and life prediction are considered together, and targeted treatment is made for different stages of the battery life cycle. The concept of qualitative change point is creatively introduced. The qualitative change point is found through health monitoring in the stage of stable performance before the qualitative change point. In the stage of sudden performance drop after the qualitative change point, the performance change characteristics of the monitoring process are taken into account to predict the life, perceive faults in advance, generate alarm information, and provide more effective guidance for operation and maintenance, thereby extending the battery service life and reducing operation and maintenance costs while ensuring a low failure rate.
[0070] See below Figure 2 , Figure 2 It is a flow chart of a specific embodiment of the battery detection method of the present invention. Figure 2 The steps performed in the health detection phase, life prediction interval, alarm phase, and model training and optimization phase are shown.
[0071] During the health monitoring phase, perform:
[0072] Step S201: Acquire characteristic data of the battery to be tested;
[0073] Step S202: converting battery characteristic data into a characteristic vector;
[0074] Step S203: inputting the battery feature vector into the health detection model.
[0075] Step S204: Determine whether the battery performance has reached a qualitative change point based on the output of the health detection model.
[0076] Specifically, a feature vector can be constructed by collecting the characteristic parameters of the battery. The dimensions of each sample input data include but are not limited to input=(U t,I t ,R t ,T t ), respectively monitoring the battery voltage, current, resistance, and operating temperature, based on which the peak-valley voltage, current, resistance, peak-valley temperature, etc. can be further monitored. By inputting the monitoring data into the deep neural network, the output result output=(γ t ), indicating whether the qualitative change point has been reached
[0077] The computational structure of the i-th layer of the neural network: υ [i] =max(0,W [i] υ [i-1] +ξ [i] )
[0078] The output of the neural network: Y = softmax (υ [I-1] )
[0079] Among them, W [i] is the weight parameter of the i-th layer, υ [i-1] is the output of the previous layer, ξ [i] is the bias parameter of the i-th layer.
[0080] If the judgment result of step S204 is no, the process returns to the health monitoring stage and executes step S201 ; if the judgment result of step S204 is yes, the process enters the life prediction stage.
[0081] During the life prediction phase, perform:
[0082] Step S205: storing the current feature parameters as samples (model training and optimization phases may be performed synchronously or asynchronously after step S205);
[0083] Step S206: Acquire characteristic parameters and previous change curve (ie, the change curve of battery performance relative to time before the current time). Specifically, the battery performance can be calculated based on the characteristic parameters.
[0084] Step S207: Input the data obtained in step S206 into the life prediction model.
[0085] Specifically, the life prediction model finds the warning point based on the performance characteristic index before the qualitative change point and the current index change. The life prediction model is a long short-term memory model, in which the structure of each long short-term memory unit 410 can be seen in Figure 4 .
[0086] The input data dimensions of each sample include but are not limited to input=(U t ,I t ,R t ,T t,ρ), by analyzing the characteristic parameters and performance change curve ρ, the LSTM network is used to predict the life span, and the current alarm status indication is given according to the prediction results
[0087] Forget Gate:
[0088] Input Gate:
[0089] Based on this, we can get the current unit state c of the network k .
[0090] Output Gate:
[0091] Based on this, we can get the current hidden state h of the network k .
[0092] Among them, c k-1 is the unit state of the previous unit, h k-1 is the hidden state of the previous unit, x k is the input feature of the current unit, is the weight matrix of each computing gate, b f , b i , b o is the bias parameter of each computing gate.
[0093] Step S208: Display the life prediction result.
[0094] Step S209: Determine whether the life prediction result reaches the warning point.
[0095] If the judgment result of step S209 is no, the process returns to the life prediction stage and executes step S206; if the judgment result of step S209 is yes, the process enters the alarm stage.
[0096] During the warning phase, execute:
[0097] Step S210: store the completed change curve of the current characteristic parameter as a sample (ie, the complete battery performance change curve of the battery to be tested relative to time. After step S210, the model training and optimization phases can be performed synchronously or asynchronously).
[0098] Step S211: Output an alarm instruction.
[0099] During model training and optimization:
[0100] Step S212: Setting a label mark for the existing data;
[0101] Step S213: updating the label mark, which is performed after step S205. If step S205 is not performed, step S213 can be omitted;
[0102] Step S214: Update the loss function and weight bias matrix;
[0103] Step S215: Update the detection model according to the threshold of the loss function.
[0104] Step S216: training according to the existing performance-time curve;
[0105] Step S217: updating the performance-time curve, which is performed after step S210. If step S210 is not performed, step S217 can be omitted;
[0106] Step S218: Update the loss function and weight bias matrix;
[0107] Step S219: Update the detection model according to the threshold of the loss function.
[0108] Battery operation and maintenance requires long-term monitoring to ensure that the battery can be discovered and replaced as soon as possible when a performance failure occurs. The existing technology only focuses on a single stage of the battery life cycle, has a limited scope of application, and has poor accuracy. This application combines health monitoring with life prediction, makes targeted treatments for different stages of the battery life cycle, and creatively introduces the concept of qualitative change points. In the stage of stable performance before the qualitative change point, the qualitative change point is found through health monitoring. In the stage of sudden performance drop after the qualitative change point, the performance change characteristics of the monitoring process are taken into account for life prediction, which can sense faults in advance, generate alarm information, and provide more effective guidance for operation and maintenance.
[0109] See below Figure 3 , Figure 3 3 is a schematic diagram of a battery performance curve 300 of an embodiment of the present invention. The battery performance curve 300 shows the battery performance curves of positive electrode oxidation, electrolyte drying, negative electrode salinization and early capacity loss, as well as the qualitative change points and warning points of each curve.
[0110] Positive electrode oxidation: Positive electrode oxidation during charging causes grid corrosion, which will cause voltage drop, rapid capacity reduction, increased internal resistance and other qualitative changes. Specifically, the battery performance does not change significantly in the early stage, but the battery performance drops rapidly after this problem occurs.
[0111] Electrolyte drying up: Abnormal temperature during use causes water loss, which can lead to quality changes such as reduced battery capacity. Specifically, the battery performance shows a clear downward trend during the early health monitoring process, and as the problem becomes more serious, the performance decline gradually accelerates.
[0112] Negative electrode salting: The interaction during discharge hinders normal charging and discharging, causing the effective capacity of the battery to decrease, the internal resistance to increase, and excessive heating, etc.
[0113] Early capacity loss: Improper use such as long-term overcharging will cause qualitative changes such as reduced battery capacity. Specifically, there will be a period of performance degradation during the health monitoring stage (caused by improper operation), and the performance trend in the later stage will be similar to that of naturally aged batteries.
[0114] The above is only a schematic description of the specific implementation of the present invention, and the present invention is not limited thereto. The splitting, merging, changing of the execution order of steps, splitting, merging of modules, and changing of information transmission are all within the protection scope of the present invention.
[0115] Figure 5 1 is a block diagram of an embodiment of a battery detection device of the present invention. The battery detection device 500 of the present invention, such as Figure 5 As shown, it includes but is not limited to an acquisition module 510, a first input module 520, a qualitative change point determination module 530, a second input module 540 and an alarm determination module 550.
[0116] The acquisition module 510 is configured to acquire a battery feature vector of a battery to be detected;
[0117] The first input module 520 is configured to input the battery feature vector into the health detection model;
[0118] The quality change point determination module 530 is configured to determine whether the performance of the battery to be tested reaches the quality change point according to the output of the health detection model;
[0119] The second input module 540 is configured to input the vector sequence of the battery feature vector into the life prediction model in response to the performance of the battery to be detected reaching a qualitative change point; and
[0120] The alarm determination module 550 is configured to determine whether to issue a detection alarm for the battery to be detected according to the output of the life prediction model.
[0121] Figure 6 The battery detection device 600 of the present invention includes but is not limited to: an acquisition module 610, a first input module 620, a quality change point determination module 630, a first sample storage module 640, a second input module 650, and an alarm determination module 660.
[0122] The acquisition module 610 is configured to acquire a battery feature vector of a battery to be detected;
[0123] The first input module 620 is configured to input the battery feature vector into the health detection model;
[0124] The quality change point determination module 630 is configured to determine whether the performance of the battery to be tested reaches the quality change point according to the output of the health detection model;
[0125] The first sample storage module 640 is configured to use the battery feature vector of the battery to be detected as a training sample of the health detection model in response to the performance of the battery to be detected reaching a qualitative change point.
[0126] The second input module 650 is configured to input the vector sequence of the battery feature vector into the life prediction model in response to the performance of the battery to be tested reaching a qualitative change point; and
[0127] The alarm determination module 660 is configured to determine whether to issue a detection alarm for the battery to be detected according to the output of the life prediction model.
[0128] The second sample storage module 670 is configured to store the life curve of the battery to be detected as a training sample of the life prediction model in response to determining to perform a detection alarm on the battery to be detected.
[0129] The training module 680 is configured to jointly train the health detection model and the life prediction model.
[0130] The implementation principle of the above modules can be found in the relevant introduction of the battery detection method, which will not be repeated here.
[0131] The battery detection device of the present invention combines health monitoring with life prediction, makes targeted treatments for different stages of the battery's life cycle, and creatively introduces the concept of qualitative change point. In the stage before the qualitative change point when performance is stable, the qualitative change point is found through health monitoring. In the stage after the qualitative change point when performance drops sharply, the performance change characteristics of the monitoring process are taken into account to predict the life, perceive faults in advance, generate alarm information, and provide more effective guidance for operation and maintenance, thereby extending the battery service life and reducing operation and maintenance costs while ensuring a low failure rate.
[0132] Figure 5 and Figure 6 The battery detection devices 500 and 600 provided by the present invention are only schematically shown. Without violating the concept of the present invention, the splitting, merging, and adding of modules are all within the protection scope of the present invention. The battery detection devices 500 and 600 provided by the present invention can be implemented by software, hardware, firmware, plug-ins, and any combination thereof, and the present invention is not limited thereto.
[0133] The embodiment of the present invention further provides a battery detection processing device, comprising a processor and a memory, wherein executable instructions of the processor are stored, wherein the processor is configured to execute the steps of the battery detection method by executing the executable instructions.
[0134] As shown above, the battery detection and processing equipment of this embodiment of the present invention jointly considers health monitoring and life prediction, makes targeted treatments for different stages of the battery's life cycle, creatively introduces the concept of qualitative change point, finds the qualitative change point through health monitoring in the stage of stable performance before the qualitative change point, and takes into account the performance change characteristics of the monitoring process in the stage of sudden performance drop after the qualitative change point to predict the life, perceive faults in advance, generate alarm information, and provide more effective guidance for operation and maintenance, thereby extending the battery service life and reducing operation and maintenance costs while ensuring a low failure rate.
[0135] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be collectively referred to herein as "circuits", "modules" or "platforms".
[0136] Figure 7 This is a schematic diagram of the structure of the battery detection and processing equipment of the present invention. Figure 7 An electronic device 800 according to this embodiment of the present invention will be described. Figure 7 The electronic device 800 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0137] like Figure 7 As shown, the electronic device 800 is in the form of a general computing device. The components of the electronic device 800 may include, but are not limited to: at least one processing unit 810, at least one storage unit 820, a bus 830 connecting different platform components (including the storage unit 820 and the processing unit 810), a display unit 840, etc.
[0138] The storage unit stores program codes, which can be executed by the processing unit 810, so that the processing unit 810 performs the steps of various exemplary embodiments of the present invention described in the battery detection method section above. For example, the processing unit 810 can perform the following steps: Figure 1 Follow the steps shown in .
[0139] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 8201 and / or a cache memory unit 8202 , and may further include a read-only memory unit (ROM) 8203 .
[0140] The storage unit 820 may also include a program / utility 8204 having a set (at least one) of program modules 8205, such program modules 8205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0141] Bus 830 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.
[0142] The electronic device 800 may also communicate with one or more external devices 8001 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 850. Furthermore, the electronic device 800 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 860. The network adapter 860 may communicate with other modules of the electronic device 800 via a bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.
[0143] The embodiment of the present invention also provides a computer-readable storage medium for storing a program, and the steps of the battery detection method are implemented when the program is executed. In some possible implementations, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps of various exemplary embodiments of the present invention described in the battery detection method section above in this specification.
[0144] As shown above, the computer-readable storage medium used to perform battery detection in this embodiment takes health monitoring and life prediction into consideration together, makes targeted treatments for different stages of the battery life cycle, creatively introduces the concept of qualitative change point, finds the qualitative change point through health monitoring in the stable performance stage before the qualitative change point, and takes the performance change characteristics of the monitoring process into consideration to perform life prediction in the performance drop stage after the qualitative change point, perceives faults in advance, generates alarm information, and provides more effective guidance for operation and maintenance, thereby extending the battery service life and reducing operation and maintenance costs while ensuring a low failure rate.
[0145] Figure 8 Schematic diagram of the structure of the computer-readable storage medium of the present invention. Figure 8 As shown, a program product 900 for implementing the above method according to an embodiment of the present invention is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.
[0146] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0147] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.
[0148] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).
[0149] In summary, this application jointly considers health monitoring and life prediction, makes targeted treatments for different stages of the battery's life cycle, creatively introduces the concept of qualitative change point, finds the qualitative change point through health monitoring in the stage of stable performance before the qualitative change point, and takes into account the performance change characteristics of the monitoring process in the stage of sudden performance drop after the qualitative change point to predict the life, perceive faults in advance, generate alarm information, and provide more effective guidance for operation and maintenance, thereby extending the battery service life and reducing operation and maintenance costs while ensuring a low failure rate.
[0150] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.
Claims
1. A battery detection method, characterized in that: include: Obtaining a battery feature vector of the battery to be tested; Inputting the battery feature vector into a health detection model; Determining whether the performance of the battery to be tested reaches a qualitative change point according to the output of the health detection model; In response to the performance of the battery to be tested reaching a qualitative change point, inputting the vector sequence of the battery feature vector into a life prediction model; and According to the output of the life prediction model, it is determined whether to perform a detection alarm on the battery to be detected. The health detection model and the life prediction model are jointly trained, and the joint training goal of the health detection model and the life prediction model is to minimize the operation and maintenance cost.
2. The battery detection method according to claim 1, characterized in that: Also includes: In response to the performance of the battery to be detected reaching a qualitative change point, a battery feature vector of the battery to be detected is used as a training sample of the health detection model.
3. The battery detection method according to claim 1, characterized in that: The health detection model is a neural network model.
4. The battery detection method according to claim 1, characterized in that: Also includes: In response to determining to perform a detection alarm on the battery to be detected, the life curve of the battery to be detected is stored as a training sample of the life prediction model.
5. The battery detection method according to claim 1, characterized in that: The life prediction model is a neural network model.
6. The battery detection method according to claim 1, characterized in that: The operation and maintenance cost is calculated according to the following formula: Among them, ω1 and ω2 are trade-off parameters, C i is the cost of the i-th battery replacement, n is the total number of battery replacements, G j is the loss of the jth battery failure, and k is the total number of battery failures.
7. A battery detection device, characterized in that: include: An acquisition module configured to acquire a battery feature vector of a battery to be detected; A first input module, configured to input the battery feature vector into a health detection model; A qualitative change point determination module, configured to determine whether the performance of the battery to be tested reaches a qualitative change point according to the output of the health detection model; A second input module is configured to input the vector sequence of the battery characteristic vector into the life prediction model in response to the performance of the battery to be detected reaching a qualitative change point; as well as The alarm determination module is configured to determine whether to issue a detection alarm for the battery to be detected based on the output of the life prediction model. The health detection model and the life prediction model are jointly trained, and the joint training goal of the health detection model and the life prediction model is to minimize the operation and maintenance cost.
8. A battery detection and processing device, characterized in that: include: processor; a memory storing executable instructions of the processor; The processor is configured to perform, via executing the executable instructions: The battery testing method according to any one of claims 1 to 6.
9. A computer-readable storage medium for storing a program, characterized in that: When the program is executed, it achieves: The battery testing method according to any one of claims 1 to 6.
