A portable power supply fault detection system and detection method

Abstract

The present invention discloses a fault detection system and a detection method for a portable power supply, and relates to the technical field of power supply detection. A trained fault image recognition model is used to recognize a power supply image. If a power supply element has a fault, the portable power supply is divided into a plurality of battery cells and tested respectively. An abnormality degree is generated from test data. If the abnormality degree exceeds expectations, an abnormal component is screened out in the battery cell, and feature extraction is performed on the test data of the abnormal component to obtain corresponding maintenance features. A portable power supply maintenance knowledge graph is used according to the output maintenance features to match a corresponding maintenance plan for the abnormal component. A maintenance degree of the maintenance plan is constructed, and damaged components are screened out in the abnormal component according to the maintenance degree, and maintenance of the power supply is achieved through the damaged components. Before maintaining the abnormal component, a targeted inspection and maintenance plan is given to achieve targeted maintenance of the abnormal component, thereby improving maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply detection, and in particular to a portable power supply fault detection system and a detection method. Background Art

[0002] Portable power, also known as temporary power or mobile power, is a portable charging device that can provide power to handheld devices or other small appliances when there is no socket nearby. It is a power system that provides power distribution for projects, especially for scenes that are used in a short period of time or need to be moved. The working principle of portable power is based on electrochemical reactions and circuit technology. It usually uses lithium-ion batteries as energy storage elements. When the power supply is plugged into a power adapter or connected to an external charging source, the lithium-ion battery stores electricity through electrochemical reactions. When the device needs to be powered, the inverter converts the DC power in the battery into standard AC or DC power and outputs it to the device.

[0003] In the Chinese invention patent application with publication number CN118091480A, a power supply fault detection method is disclosed, which collects current or voltage signals of phase A, phase B, phase C and neutral line, and divides them according to the peak and valley positions of the signals, extracts rising interval signals and falling interval signals, respectively extracts rising feature vectors of the rising interval signals and falling feature vectors of the falling interval signals, calculates symmetry coefficients through rising feature vectors and falling feature vectors, and is used to analyze, identify and quantify asymmetry, distortion and other abnormal conditions in signal waveforms, and calculates neutral line signal strength coefficients through neutral line signals, and is used to evaluate the imbalance of three-phase power supply, thereby combining the symmetry coefficients of each phase signal and the neutral line signal strength coefficient to obtain a three-phase power supply fault value, which is used to evaluate the fault condition of the three-phase power supply, evaluate power supply faults, and improve detection accuracy.

[0004] Combined with the above application and the contents of the prior art:

[0005] When the portable power supply is in the power supply state, it is necessary to keep an eye on the remaining power of the power supply. When the remaining power of the power supply is insufficient, it is necessary to charge the power supply in time or stop supplying power to the outside. If the power supply drops too quickly, it means that the power supply may have failed. In particular, when the power supply is in a high temperature and high humidity usage scenario, the power consumption rate of the power supply will be further accelerated, resulting in a further increase in the risk of power supply failure.

[0006] When there is an abnormality in the usage status of the power supply, it is usually necessary to detect whether the power supply has a fault. The existing fault judgment method usually performs a comprehensive fault detection on the power supply and judges whether the power supply has a fault based on the detection data. However, the portable mobile power supply is usually composed of multiple battery cells. If all the battery cells are fully tested, although the fault detection effect is good, the actual efficiency is low. After the fault is determined, it takes a long time to restore the power supply to normal, which will greatly affect the normal use of the power supply.

[0007] To this end, the present invention provides a portable power supply fault detection system and detection method. Summary of the invention

[0008] 1. Technical issues to be resolved

[0009] In view of the deficiencies in the prior art, the present invention provides a fault detection system and method for a portable power supply. The power supply image is identified by using a trained fault image recognition model. If a power supply component is faulty, the portable power supply is divided into several battery cells and tested separately. The abnormality degree is generated from the test data. If the abnormality degree exceeds expectations, the abnormal components are screened out in the battery cells, and feature extraction is performed on the test data of the abnormal components to obtain corresponding maintenance features. Based on the output maintenance features, a portable power supply maintenance knowledge graph is used to match corresponding maintenance plans for the abnormal components. Before maintaining the abnormal components, targeted inspection and maintenance plans are given to achieve targeted maintenance of the abnormal components and improve maintenance efficiency, thereby solving the technical problems raised in the background technology.

[0010] (II) Technical solution

[0011] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0012] A portable power supply fault detection method includes: when there is an abnormality in the power supply use environment, detecting the power consumption state of the power supply, aggregating the detection data to form a power consumption data set, and generating a power consumption imbalance degree from the power consumption data set. If the imbalance If it exceeds expectations, a warning instruction will be issued to the outside world;

[0013] Test the state of the power supply, obtain test data and summarize to build a power supply test data set, and generate the state coefficient of the power supply from the power supply test data , if the state coefficient If the state is lower than the threshold, an identification instruction is issued to the outside;

[0014] After image acquisition, the trained fault image recognition model is used to identify the power supply image. If there is a fault in the power supply component, the fault location is marked and the abnormal density is constructed based on the location information. , if the abnormal density If it exceeds expectations, a fault detection instruction will be issued to the outside;

[0015] The portable power source is divided into several battery cells and tested separately, and the abnormality degree is generated from the test data. , if the abnormality If the battery voltage exceeds the expected value, the abnormal components are screened out in the battery cell, and the test data of the abnormal components are extracted to obtain the corresponding maintenance features; wherein, the test voltage of the battery cell is Perform linear normalization and map the corresponding data values ​​to the interval The abnormality degree is constructed according to the following formula :

[0016] ;

[0017] in, is the test voltage on the i-th test node, is the qualified standard value of the test voltage, is the standard deviation of several test voltages; weight coefficient: , ,and ;

[0018] According to the output maintenance features, the portable power maintenance knowledge graph matches the corresponding maintenance plan for the abnormal component and builds the maintenance degree of the maintenance plan. , according to the degree of maintenance Screen out damaged components from abnormal components and use the damaged components to maintain the power supply; generate improvement coefficients based on the status data of the power supply after maintenance , if the improvement coefficient If the improvement threshold is not exceeded, an alarm command will be issued to the outside.

[0019] Furthermore, when the power supply is in the power supply state, the temperature and humidity data in the working area are collected, and after continuously acquiring the data of multiple monitoring nodes, the working environment data set is summarized and constructed; the condition coefficient of the environmental condition is generated from the working environment data set , if the conditional coefficient When the condition threshold is exceeded, a power detection instruction is sent to the outside; after receiving the power detection instruction, the power reduction speed of the portable power supply and the power consumption data of the power supply are recorded at each detection node, and the power consumption data set is constructed after being summarized.

[0020] Furthermore, the temperature and humidity Perform linear normalization and map the corresponding data values ​​to the interval The conditional coefficients are constructed as follows :

[0021] ;

[0022] Weight coefficient: , and ; For the The temperature on each monitoring node, is the qualified standard value of temperature; For the The humidity on each monitoring node, It is the qualified standard value of humidity.

[0023] Furthermore, the imbalance of power consumption is generated from the power consumption data set when the power source is used. , where the rate of charge decreases Under dimensionless conditions, the following is true:

[0024] ;

[0025] in, and is the weight coefficient: , and ; , is the number of detection nodes, For the The rate at which the power on each detection node decreases. is the corresponding qualified standard value, is the average rate of power reduction.

[0026] Further, after receiving the warning instruction, the power supply is connected to the test load and the load test is performed; the corresponding voltage and current data are recorded at each test node, and the acquired test data is aggregated to construct a power supply test data set;

[0027] Constructing the state coefficient of portable power supply when supplying power from power supply test data set , where the test voltage And test current Perform linear normalization and map the corresponding data values ​​to the interval According to the following formula:

[0028] ;

[0029] in, is the weight coefficient, and its value falls into , For the The test voltage on the test node, is the acceptable value of the test voltage, For the The test current on the test node is is an acceptable value for the test current.

[0030] Furthermore, after receiving the recognition instruction, the volume of the portable power supply is detected to determine whether the power supply volume has expanded. If so, the acquired fault image recognition model is trained; the components on the power management circuit board are imaged and power supply image data is acquired; the power supply image is recognized using the trained fault image recognition model to determine whether the power supply component has a fault, and the area with the fault is marked as a fault area.

[0031] Furthermore, a portable power supply three-dimensional model is constructed based on the structural data of the portable power supply, and each fault area is marked on the power supply three-dimensional model. After obtaining the location information of each fault area, the abnormal density is constructed based on the location information. ,in,

[0032] ;

[0033] in, is the number of fault areas, It is Fault area to The shortest distance between fault areas, is the average distance; weight coefficient: , ,and .

[0034] Furthermore, after receiving the fault detection instruction, the portable power supply is divided into several battery cells, and the key node voltages on the power management circuit board of each battery cell are measured respectively. The obtained key node voltage data and the operating temperature of the battery cell are summarized to construct a voltage data set of the battery cell; an abnormality degree is generated from the voltage data set of the battery cell. , if the abnormality If it exceeds the abnormal threshold, it will be treated as an abnormal component; otherwise, it will be treated as a qualified component.

[0035] Furthermore, the abnormal components are performance tested and the corresponding test data are obtained. After the qualified standard value is selected for each test indicator, feature recognition is performed on each test data to obtain the corresponding maintenance features, which are then aggregated to generate a maintenance feature set for the power supply components.

[0036] Furthermore, taking portable power supply detection and maintenance as the target word, a portable power supply maintenance knowledge graph is pre-built, and the maintenance features in the maintenance feature set of the power supply component are taken as input. According to the correspondence between the maintenance features and the maintenance plans of the power supply components, the portable power supply maintenance knowledge graph matches the corresponding maintenance plans for the abnormal components.

[0037] Furthermore, after obtaining the maintenance plan, the maintenance time and maintenance cost of the abnormal components are estimated and obtained, and the maintenance plan data set is generated after aggregation. The maintenance degree is constructed from the maintenance plan data set. ,in,

[0038] Maintenance time and maintenance costs Perform linear normalization and map the corresponding data values ​​to the interval According to the following formula:

[0039] ;

[0040] Weight coefficient, , ;

[0041] If the maintenance If the maintenance threshold is not exceeded, maintenance is performed on the abnormal component; if the maintenance threshold is exceeded, maintenance is performed on the abnormal component. If the maintenance threshold is exceeded, the abnormal component is treated as a damaged component and the damaged component is replaced.

[0042] Furthermore, the power supply is reassembled and functionally tested. After several stages of testing, the corresponding imbalance values ​​are constructed from the test data. , where each test phase before and after maintenance is sorted according to the timeline and corresponds one by one, and the imbalance degree is The change in construction improvement factor , as follows:

[0043] ;

[0044] in, For the The median improvement in the test phase, is its mean value, is its qualified standard value;

[0045] , is the number of test phases, and Before and after maintenance The imbalance of the test phase, and is the corresponding mean.

[0046] A portable power supply fault detection system includes a power detection unit, which detects the power consumption state of the power supply when there is an abnormality in the power supply use environment, aggregates the detection data to form a power consumption data set, and generates a power consumption imbalance degree from the power consumption data set. If the imbalance If it exceeds expectations, a warning instruction will be issued to the outside world;

[0047] The power supply status test unit tests the power supply status, obtains test data and summarizes and constructs a power supply test data set, and generates the power supply status coefficient from the power supply test data. , if the state coefficient If the state is lower than the threshold, an identification instruction is issued to the outside;

[0048] The fault detection unit uses the trained fault image recognition model to identify the power supply image after image acquisition. If there is a fault in the power supply component, the fault location is marked and the abnormal density is constructed based on the location information. , if the abnormal density If it exceeds expectations, a fault detection instruction will be issued to the outside;

[0049] The fault analysis unit divides the portable power supply into several battery cells and tests them separately, generating an abnormality degree from the test data. , if the abnormality If the number of abnormal components exceeds expectations, the abnormal components are screened out in the battery unit, and the test data of the abnormal components are subjected to feature extraction to obtain the corresponding maintenance features;

[0050] The maintenance unit matches the corresponding maintenance plan for the abnormal component based on the output maintenance features, and builds the maintenance degree of the maintenance plan. , according to the degree of maintenance Screen out damaged components from abnormal components and use the damaged components to maintain the power supply; generate improvement coefficients based on the status data of the power supply after maintenance , if the improvement coefficient If the improvement threshold is not exceeded, an alarm command will be issued to the outside.

[0051] (III) Beneficial effects

[0052] The present invention provides a portable power supply fault detection system and detection method, which have the following beneficial effects:

[0053] 1. Based on the degree of imbalance The power consumption status is evaluated. If the difference between the power consumption speed and the qualified speed is large, the power consumption status can be warned by detecting and recording the power supply power.

[0054] 2. If based on the obtained state coefficient If it is determined that the power supply is in a normal state, the detection process can be stopped. If it is abnormal, the power supply state can be adjusted in time to ensure the safe working state of the portable power supply.

[0055] 3. Use the trained image recognition model to identify faults in various components of the power supply to determine whether there is a fault in the power supply. If a fault is identified, it can be processed to ensure that the power supply can be used normally.

[0056] 4. Based on the density of faults Determine the current fault level of the power supply. If the fault density is If it exceeds expectations, it means that the current fault of the power supply is already quite serious and further fault detection is needed in time to determine the specific fault and its cause.

[0057] 5. Constructing anomaly degree from detection data , according to the degree of abnormality Screen out the faulty parts from several battery cells; by screening out the faulty parts from several battery cells, targeted processing can be achieved when necessary, reducing the time spent on battery inspection and maintenance.

[0058] 6. After obtaining the data of the abnormal component, perform data identification and obtain the corresponding maintenance features. After obtaining the maintenance features and determining the location of the fault, refer to the maintenance features to perform targeted maintenance on the abnormal component and improve maintenance efficiency.

[0059] 7. The portable power supply maintenance knowledge graph matches the corresponding maintenance plan for the abnormal component. After the abnormal component is detected, a targeted inspection and maintenance plan can be given before the abnormal component is maintained, thereby achieving targeted maintenance of the abnormal component and improving maintenance efficiency.

[0060] 8. Analyze and estimate the maintenance plan, and build the corresponding maintenance level based on the corresponding data. , roughly judge whether the maintenance plan is feasible. If the feasibility of the given maintenance plan is not high, you can re-formulate the corresponding maintenance plan, or give up maintenance on abnormal components to save maintenance time for the portable power supply and enable the power supply to be quickly restored to use.

[0061] 9. Continuously obtain the test data of several stages along the time axis, and Generate improvement factor , based on the improvement factor It is judged whether the current maintenance result meets the expectation. If it fails to meet the expectation, it can be maintained or suspended. After the power supply is completed, the maintenance effect is guaranteed so that the power supply can maintain normal operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic flow chart of a fault detection method of a portable power supply of the present invention;

[0063] Figure 2 It is a schematic diagram of the structure of the fault detection system of the portable power supply of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] See also Figure 1 The present invention provides a portable power supply fault detection method, comprising:

[0066] Step 1: When there is an abnormality in the power usage environment, detect the power consumption state of the power supply, aggregate the detection data to form a power consumption data set, and generate the power consumption imbalance degree from the power consumption data set. If the imbalance If it exceeds expectations, a warning instruction will be issued to the outside world;

[0067] The step 1 includes the following contents:

[0068] Step 101: When the power supply is in the power supply state, after determining the working area of ​​the power supply, collect temperature and humidity data in the working area, and after continuously acquiring the data of multiple monitoring nodes, summarize and construct a working environment data set;

[0069] Generate the condition coefficients of environmental conditions from the working environment data set , where the temperature and humidity Perform linear normalization and map the corresponding data values ​​to the interval In the following way:

[0070] ;

[0071] Weight coefficient: , and ; The weight coefficient can be obtained by referring to the hierarchical analysis method, For the The temperature on each monitoring node is is the qualified standard value of temperature; For the The humidity on each monitoring node, is the qualified standard value of humidity;

[0072] If the conditional coefficient When the preset condition threshold is exceeded, a power detection instruction is sent to the outside;

[0073] Step 102: after receiving the power detection instruction, a detection cycle including a plurality of detection nodes is set, and the power reduction speed and power consumption data of the portable power source are recorded at each detection node, and the power consumption data set is constructed after being summarized;

[0074] Generate a power consumption imbalance from a collection of power consumption data during power usage , in order to judge whether the imbalance degree of power supply change is high, among which, the power decrease speed Under dimensionless conditions, the following is true:

[0075] ;

[0076] in, and is the weight coefficient: , and ;

[0077] , To detect the number of nodes, the weight coefficient remains consistent with the previous value; For the The rate at which the power on each detection node decreases. is the corresponding qualified standard value, is the average value of the power reduction rate;

[0078] Based on historical data and management expectations for power consumption, the imbalance threshold is pre-set; if the imbalance degree If the imbalance threshold is exceeded, it means that the portable power supply may have some faults when in use. At this time, an early warning command is issued to the outside;

[0079] High temperatures can cause the battery temperature to rise, which may damage the battery or reduce the battery life. The heating of the electrodes of lithium-ion batteries at high temperatures will accelerate aging and affect battery performance. In places where the temperature is too low, the internal resistance and overvoltage of lithium-ion batteries will increase, resulting in performance degradation and shortening the battery life. Avoid direct storage in low temperatures below 20°C in cold areas in winter, which can easily cause problems such as failure to charge. Humidity and moisture are common problems in outdoor environments and may cause short circuits in power devices or damage to circuit boards.

[0080] When using, combine the contents in steps 101 and 102:

[0081] When the portable power source is in the power supply state, the power consumption in each stage is detected, and the corresponding imbalance degree is constructed based on the power consumption data obtained. , according to the imbalance Evaluate the power consumption status. If the difference between the power consumption rate and the qualified rate is large under the same usage status or in different scenarios, it means that the power supply may be faulty. At this time, by detecting and recording the power supply power, an early warning of the power usage status can be provided.

[0082] Combined with the above application and the contents of the prior art:

[0083] When the portable power supply is in the power supply state, it is necessary to keep an eye on the remaining power of the power supply. When the remaining power of the power supply is insufficient, it is necessary to charge the power supply in time or stop supplying power to the outside. If the power supply drops too quickly, it means that the power supply may have failed. In particular, when the power supply is in a high temperature and high humidity usage scenario, the power consumption rate of the power supply will be further accelerated, resulting in a further increase in the risk of power supply failure.

[0084] When there is an abnormality in the usage status of the power supply, it is usually necessary to detect whether the power supply has a fault. The existing fault judgment method usually performs a comprehensive fault detection on the power supply and judges whether the power supply has a fault based on the detection data. However, the portable mobile power supply is usually composed of multiple battery cells. If all the battery cells are fully tested, although the fault detection effect is good, the actual efficiency is low. After the fault is determined, it takes a long time to restore the power supply to normal, which will greatly affect the normal use of the power supply.

[0085] Step 2: Test the state of the power supply, obtain test data and summarize to build a power supply test data set, and generate the state coefficient of the power supply from the power supply test data , if the state coefficient If the state is lower than the threshold, an identification instruction is issued to the outside;

[0086] The step 2 includes the following contents:

[0087] Step 201: After receiving the warning instruction, connect the power supply to the pre-prepared test load and perform a load test; set a number of continuous test nodes, record the corresponding voltage and current data at each test node, and construct a power supply test data set after aggregating the acquired test data;

[0088] Step 202: After presetting acceptable values ​​for voltage and current, construct the state coefficient of the portable power supply when supplying power from the power supply test data set. , judge whether there is an abnormality in the power supply according to the state coefficient, among which, the test voltage And test current Perform linear normalization and map the corresponding data values ​​to the interval According to the following formula:

[0089] ;

[0090] in, is the weight coefficient, and its value falls into , which can be obtained through simulation software analysis or by referring to the hierarchical analysis method; For the The test voltage on the test node, is the acceptable value of the test voltage, For the The test current on the test node is is the acceptable value of the test current;

[0091] According to historical data and management expectations of power supply status, the status threshold is set in advance; if the status coefficient obtained If it is lower than the state threshold, it means that the power supply state of the portable power supply is poor, and the power supply may be faulty, and further fault identification and detection are required. At this time, an identification instruction is sent to the outside;

[0092] When using, combine the contents in steps 201 and 202:

[0093] After detecting the power supply, if the portable power supply may have a fault, continue to detect the power supply output status, including the voltage and current of the power supply output, and monitor the working status of the power supply to further determine whether the power supply is abnormal. If the status coefficient obtained is If it is determined that the power supply is in a normal state, the detection process can be stopped. If it is abnormal, the power supply state can be adjusted in time to ensure the safe working state of the portable power supply.

[0094] Step 3: After image acquisition, use the trained fault image recognition model to identify the power supply image. If there is a fault in the power supply component, mark the fault location and construct the abnormal density based on the location information. , if the abnormal density If it exceeds expectations, a fault detection instruction will be issued to the outside;

[0095] The step three includes the following contents:

[0096] Step 301: after receiving the recognition instruction, the volume of the portable power supply is detected, and the obtained detection data is compared with the standard data to determine whether the volume of the power supply has expanded. If so, the convolutional neural network is trained with the sample data to obtain a trained fault image recognition model;

[0097] Step 302: After the power supply housing is disassembled, images of components on the power management circuit board, including capacitors, resistors, integrated circuits, etc., are collected and power supply image data is obtained;

[0098] The trained fault image recognition model is used to identify the power supply image to determine whether the power supply component has faults, such as signs of burning or desoldering, and the area with faults is marked as the fault area.

[0099] During use, after monitoring the use status of the power supply, if it is pre-determined that the power supply may have a fault, after collecting the power supply image data, the trained image recognition model is used to identify the faults of various components of the power supply to determine whether the power supply has a fault. If a fault is identified, it can be processed to ensure that the power supply can be used normally.

[0100] Step 303: construct a three-dimensional model of the portable power supply based on the structural data of the portable power supply, mark each fault area on the three-dimensional model of the power supply, obtain the position information of each fault area, and construct an abnormal density from the position information. ,in,

[0101] ;

[0102] in, is the number of fault areas, It is Fault area to The shortest distance between fault areas, is the average distance; weight coefficient: , ,and ;

[0103] Based on historical data and tolerance to power failures, the density threshold is pre-set; if the abnormal density If the density threshold is exceeded, it means that there are many faults in the components of the power supply, which will affect the power supply state of the power supply and cause the power supply to enter a fault state when supplying power. At this time, a fault detection instruction is sent to the outside.

[0104] When using, combine the contents in steps 301 to 303:

[0105] After completing fault identification, the location information of each external fault is obtained, and the fault density is constructed. , according to the fault density Determine the current fault level of the power supply. If the fault density is If it exceeds expectations, it means that the current fault of the power supply is already quite serious and further fault detection is needed in time to determine the specific fault and its cause.

[0106] Step 4: Divide the portable power source into several battery cells and test them separately, and generate anomaly degree from the test data , if the abnormality If the number of abnormal components exceeds expectations, the abnormal components are screened out in the battery unit, and the test data of the abnormal components are subjected to feature extraction to obtain the corresponding maintenance features;

[0107] The step 4 includes the following contents:

[0108] Step 401: After receiving the fault detection instruction, the portable power source is divided into a plurality of battery cells, and the key node voltages on the power management circuit board of each battery cell are measured respectively, such as the charging input terminal, the battery output terminal, etc.; the obtained key node voltage data and the operating temperature of the battery cell are aggregated to construct a voltage data set of the battery cell;

[0109] Step 402: Generate an abnormality degree from the voltage data set of the battery cell , according to the abnormality, it is judged whether the voltage of the battery cell is abnormal, wherein the test voltage of the battery cell Perform linear normalization and map the corresponding data values ​​to the interval According to the following formula:

[0110] ;

[0111] in, is the test voltage on the i-th test node, is the qualified standard value of the test voltage, is the standard deviation of several test voltages; weight coefficient: , ,and , the weight coefficient remains the same as the previous value;

[0112] Based on historical data and the voltage management expectations for the power supply unit, the abnormal threshold is pre-set; if the abnormal degree If it exceeds the abnormal threshold, it means that the abnormality of the current power supply unit is high and it is regarded as an abnormal component; otherwise, it can be regarded as a qualified component;

[0113] When using, when each battery cell of the portable power source is tested for the first time, the abnormality degree is constructed from the test data. Can be based on abnormality Screen out the faulty parts from several battery cells; by screening out the faulty parts from several battery cells, targeted processing can be achieved when necessary, reducing the time spent on battery inspection and maintenance.

[0114] Step 403: Use a test tool, such as a resistance meter, a capacitance meter, an integrated circuit tester, etc., to perform a performance test on the abnormal component, and obtain corresponding test data, including discharge curve, self-discharge rate, charge and discharge efficiency, state of charge, and internal resistance, etc.

[0115] After selecting qualified standard values ​​for each test indicator, feature recognition is performed on each test data to obtain corresponding maintenance features, and the obtained maintenance features are aggregated to generate a maintenance feature set of the power supply component;

[0116] When using, combine the contents in steps 401 to 403:

[0117] After screening out abnormal components and obtaining various data of abnormal components, data recognition is performed and corresponding maintenance features are obtained. After obtaining the maintenance features and determining the location of the fault, targeted maintenance can be performed on the abnormal components with reference to the maintenance features, thereby improving maintenance efficiency.

[0118] Step 5: Based on the output maintenance features, the portable power maintenance knowledge graph matches the corresponding maintenance plan for the abnormal component and constructs the maintenance degree of the maintenance plan. , according to the degree of maintenance Screen out damaged components from abnormal components and use the damaged components to maintain the power supply; generate improvement coefficients based on the status data of the power supply after maintenance , if the improvement coefficient If the improvement threshold is not exceeded, an alarm instruction is issued to the outside;

[0119] The step five includes the following contents:

[0120] Step 501: Taking portable power supply detection and maintenance as the target word, after deep retrieval and entity relationship building, pre-build a portable power supply maintenance knowledge graph, take the maintenance features in the maintenance feature set of the power supply component as input, and match the corresponding maintenance plan for the abnormal component according to the correspondence between the maintenance features and the maintenance plan of the power supply component by the portable power supply maintenance knowledge graph;

[0121] When in use, based on the maintenance features obtained, the portable power supply maintenance knowledge graph matches the corresponding maintenance plan for the abnormal component. After detecting the abnormal component, a targeted inspection and maintenance plan can be given before maintaining the abnormal component, which can achieve targeted maintenance of the abnormal component and improve maintenance efficiency.

[0122] Step 502: After obtaining the maintenance plan, estimate the maintenance time and maintenance cost of the abnormal component, summarize and generate a maintenance plan data set, and construct a maintenance degree from the maintenance plan data set. ,in,

[0123] Maintenance time and maintenance costs Perform linear normalization and map the corresponding data values ​​to the interval According to the following formula:

[0124] ;

[0125] Weight coefficient, , ;The weight coefficient can be obtained by referring to the hierarchical analysis method;

[0126] Based on historical data and expectations of maintenance difficulty, maintenance thresholds are set in advance; if the maintenance difficulty is too high, If the maintenance threshold is not exceeded, it means that the maintenance difficulty is low and acceptable, then the abnormal component will be maintained; if the maintenance difficulty is too high, the abnormal component will be maintained. If the maintenance threshold is exceeded, it means that the maintenance of the abnormal component is difficult, and the abnormal component is regarded as a damaged component and needs to be replaced;

[0127] When in use, after obtaining the maintenance plan corresponding to the abnormal component, the maintenance plan is analyzed and estimated, and the corresponding maintenance degree is constructed based on the corresponding data. , roughly judge whether the maintenance plan is feasible. If the feasibility of the given maintenance plan is not high, you can re-formulate the corresponding maintenance plan, or give up maintenance on abnormal components to save maintenance time for the portable power supply and enable the power supply to be quickly restored to use.

[0128] Step 503: Replace the damaged components and reassemble the power supply and perform functional testing. After several stages of testing, the corresponding imbalance degree is constructed from the test data. , where each test phase before and after maintenance is sorted according to the timeline and corresponds one by one, and the imbalance degree is The change in construction improvement factor , as follows:

[0129] ;

[0130] in, For the The median improvement in the test phase, is its mean value, is its qualified standard value;

[0131] , is the number of test phases, and Before and after maintenance The imbalance of the test phase, and is the corresponding mean.

[0132] Pre-set improvement thresholds based on historical data and expectations for portable power maintenance management;

[0133] If the improvement factor If the improvement threshold is exceeded, it means that the current replacement and maintenance have played the expected role, and the portable power supply can be restored to normal use after power detection and fault detection;

[0134] If the improvement factor If it does not exceed the improvement threshold, it means that the improvement after the current replacement maintenance has not played the expected role. The portable power supply may have other faults, which leads to abnormal power consumption during power detection. Further fault detection of the portable power supply is required. At this time, an alarm command is issued to the outside to facilitate further processing.

[0135] When using, combine the contents in steps 501 to 503:

[0136] After the power supply is maintained and put back into use, the detection data of several stages are continuously obtained along the time axis, and the imbalance degree is calculated. Generate improvement factor , at this time, according to the improvement coefficient It is judged whether the current maintenance result meets the expectation. If it fails to meet the expectation, it can be maintained or suspended. After the power supply is completed, the maintenance effect is guaranteed and the power supply can maintain normal operation.

[0137] The Analytic Hierarchy Process (AHP) is a decision-making method that decomposes decision-related elements into levels such as goals, criteria, and plans, and conducts qualitative and quantitative analysis on this basis. It is particularly suitable for dealing with target systems with hierarchical and staggered evaluation indicators, and when the target value is difficult to describe quantitatively, the AHP is an effective decision-making tool.

[0138] The core of the hierarchical analysis method is to decompose the decision problem into multiple levels to form a hierarchical structure, which usually includes the target level, the criterion level, the sub-criterion level and the solution level. By solving the eigenvector of the judgment matrix, the priority weight of each element of each level to an element of the previous level is obtained, and the final weight of each alternative solution to the total goal is recursively merged by the weighted sum method, so as to find the optimal solution.

[0139] The construction flow of the portable power maintenance knowledge graph construction method is as follows:

[0140] Data Collection and Preprocessing,Data Source:,Data on portable power supply maintenance are collected from,multiple channels, including user manuals, repair guides, product,instructions, online forum discussions, technical blogs, etc.

[0141] Preprocessing: Clean, convert and integrate the collected data, remove noise, stop words, punctuation marks, etc., and perform word segmentation and part-of-speech tagging to facilitate subsequent processing.

[0142] Entity recognition and linking,Entity recognition: Use natural language processing technology (such as CNN+LSTM+CRF model) to identify specific entities from the text, such as portable power supply model, component name, fault type, etc.

[0143] Entity linking: Connect nodes representing the same entity in different data sources to ensure the uniqueness and consistency of entities in the knowledge graph.

[0144] Attribute and relationship extraction,Attribute extraction: Extract attribute information related to entities from text, such as the weight, battery capacity, and charging time of a portable power source.

[0145] Relationship extraction: Extract relationship information between entities from text, such as the correspondence between fault types and solutions, the connection relationship between components, etc.

[0146] Knowledge fusion and entity alignment: Solve the problem of inconsistent entity names in different data sources and ensure the uniqueness of the same entity in different data sources.

[0147] Attribute value filling: When different values ​​appear for the same attribute, a decision is made based on the number and reliability of data sources to provide a more accurate attribute value.

[0148] Knowledge processing and model layer construction, concept abstraction: conceptual abstraction of the constructed data layer to form a model layer. This includes defining the top-level ontology and refining the concepts and relationships to form a well-structured concept hierarchy tree.

[0149] Quality assessment: Perform quality assessment on the constructed knowledge graph, including indicators such as entity coverage, attribute coverage, relationship coverage, and accuracy and consistency of entities, attributes, and relationships.

[0150] Knowledge graph update and maintenance, regular update: The knowledge graph is updated regularly according to time intervals to reflect the latest technology and product information.

[0151] Event-driven updates: Update the knowledge graph based on the occurrence of specific events (such as new product launches, technology updates, etc.).

[0152] User feedback updates: Optimize and improve the knowledge graph based on user feedback and suggestions.

[0153] See also Figure 2 The present invention provides a portable power supply fault detection system, comprising:

[0154] The power detection unit detects the power consumption status of the power supply when there is an abnormality in the power use environment, aggregates the detection data to form a power consumption data set, and generates the power consumption imbalance degree from the power consumption data set. If the imbalance If it exceeds expectations, a warning instruction will be issued to the outside world;

[0155] The power supply status test unit tests the power supply status, obtains test data and summarizes and constructs a power supply test data set, and generates the power supply status coefficient from the power supply test data. , if the state coefficient If the state is lower than the threshold, an identification instruction is issued to the outside;

[0156] The fault detection unit uses the trained fault image recognition model to identify the power supply image after image acquisition. If there is a fault in the power supply component, the fault location is marked and the abnormal density is constructed based on the location information. , if the abnormal density If it exceeds expectations, a fault detection instruction will be issued to the outside;

[0157] The fault analysis unit divides the portable power supply into several battery cells and tests them separately, generating an abnormality degree from the test data. , if the abnormality If the number of abnormal components exceeds expectations, the abnormal components are screened out in the battery unit, and the test data of the abnormal components are subjected to feature extraction to obtain the corresponding maintenance features;

[0158] The maintenance unit matches the corresponding maintenance plan for the abnormal component based on the output maintenance features, and builds the maintenance degree of the maintenance plan. , according to the degree of maintenance Screen out damaged components from abnormal components and use the damaged components to maintain the power supply; generate improvement coefficients based on the status data of the power supply after maintenance , if the improvement coefficient If the improvement threshold is not exceeded, an alarm command will be issued to the outside.

[0159] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0160] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0161] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0162] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0163] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0164] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0165] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0166] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A portable power supply fault detection method, characterized in that: include, When there is an abnormality in the power usage environment, the power consumption status of the power supply is detected, and the detection data is aggregated to form a power consumption data set, and the power consumption imbalance is generated from the power consumption data set. If the imbalance If it exceeds expectations, a warning instruction will be issued to the outside world; Test the state of the power supply, obtain test data and summarize to build a power supply test data set, and generate the state coefficient of the power supply from the power supply test data , if the state coefficient If the state is lower than the threshold, an identification instruction is issued to the outside; After image acquisition, the trained fault image recognition model is used to identify the power supply image. If there is a fault in the power supply component, the fault location is marked and the abnormal density is constructed based on the location information. , if the abnormal density If it exceeds expectations, a fault detection instruction will be issued to the outside; The portable power source is divided into several battery cells and tested separately, and the abnormality degree is generated from the test data. , if the abnormality If the battery voltage exceeds the expected value, the abnormal components are screened out in the battery cell, and the test data of the abnormal components are extracted to obtain the corresponding maintenance features; wherein, the test voltage of the battery cell is Perform linear normalization and map the corresponding data values ​​to the interval The abnormality degree is constructed according to the following formula : ； in, is the test voltage on the i-th test node, is the qualified standard value of the test voltage, is the standard deviation of several test voltages; weight coefficient: , ,and ; According to the output maintenance features, the portable power maintenance knowledge graph matches the corresponding maintenance plan for the abnormal component and builds the maintenance degree of the maintenance plan. , according to the degree of maintenance Screen out damaged components from abnormal components and use the damaged components to maintain the power supply; generate improvement coefficients based on the status data of the power supply after maintenance , if the improvement coefficient If the improvement threshold is not exceeded, an alarm command will be issued to the outside.

2. A portable power supply fault detection method according to claim 1, characterized in that: When the power supply is in the power supply state, the temperature and humidity data in the working area are collected, and after continuously acquiring the data of multiple monitoring nodes, the working environment data set is summarized and constructed; Generate the condition coefficients of environmental conditions from the working environment data set , if the conditional coefficient When the condition threshold is exceeded, a power detection instruction is sent to the outside; after receiving the power detection instruction, the power reduction speed and power consumption data of the portable power supply are recorded at each detection node, and the power consumption data set is constructed after aggregation.

3. A portable power supply fault detection method according to claim 2, characterized in that: Temperature and humidity Perform linear normalization and map the corresponding data values ​​to the interval The conditional coefficients are constructed as follows : ； Weight coefficient: , and ; For the The temperature on each monitoring node, is the qualified standard value of temperature; For the The humidity on each monitoring node, It is the qualified standard value of humidity.

4. A portable power supply fault detection method according to claim 2, characterized in that: Generate a power consumption imbalance from a collection of power consumption data during power usage , where the rate of charge decreases Under dimensionless conditions, the following is true: ； in, and is the weight coefficient: , and ; , is the number of detection nodes, For the The rate at which the power on each detection node decreases. is the corresponding qualified standard value, is the average rate of power reduction.

5. The portable power supply fault detection method according to claim 1, characterized in that: After receiving the warning instruction, the power supply is connected to the test load and the load test is performed; the corresponding voltage and current data are recorded at each test node, and the acquired test data is aggregated to construct a power supply test data set; Constructing the state coefficient of portable power supply when supplying power from power supply test data set , where the test voltage And test current Perform linear normalization and map the corresponding data values ​​to the interval According to the following formula: ； in, is the weight coefficient, For the The test voltage on the test node, is the acceptable value of the test voltage, For the The test current on the test node is is an acceptable value for the test current.

6. A portable power supply fault detection method according to claim 1, characterized in that: After receiving the recognition instruction, the portable power supply is tested for volume to determine whether the power supply volume has expanded. If so, the acquired fault image recognition model is trained; the components on the power management circuit board are imaged and the power supply image data is acquired; The trained fault image recognition model is used to identify the power supply image to determine whether the power supply component is faulty, and the area with the fault is marked as the fault area.

7. A portable power supply fault detection method according to claim 6, characterized in that: A portable power supply 3D model is constructed based on the structural data of the portable power supply. The fault area is marked on the 3D model of the power supply. After obtaining the location information of each fault area, the abnormal density is constructed based on the location information. ,in, ； in, is the number of fault areas, It is Fault area to The shortest distance between fault areas, is the average distance; weight coefficient: , ,and .

8. The portable power supply fault detection method according to claim 1, characterized in that: After receiving the fault detection instruction, the portable power supply is divided into several battery cells, and the key node voltages on the power management circuit board of each battery cell are measured respectively. The obtained key node voltage data and the operating temperature of the battery cell are summarized to construct a voltage data set of the battery cell; the abnormality degree is generated from the voltage data set of the battery cell. , if the abnormality If it exceeds the abnormal threshold, it will be treated as an abnormal component; otherwise, it will be treated as a qualified component.

9. A portable power supply fault detection method according to claim 8, characterized in that: Perform performance tests on abnormal components and obtain corresponding test data. After selecting qualified standard values ​​for each test indicator, perform feature recognition on each test data to obtain corresponding maintenance features. The acquired maintenance features are summarized to generate a maintenance feature set for the power supply component.

10. A portable power supply fault detection method according to claim 9, characterized in that: Taking portable power supply detection and maintenance as the target word, a portable power supply maintenance knowledge graph is pre-built, and the maintenance features in the maintenance feature set of the power supply component are taken as input. According to the correspondence between the maintenance features and the maintenance plans of the power supply components, the portable power supply maintenance knowledge graph matches the corresponding maintenance plans for the abnormal components.

11. A portable power supply fault detection method according to claim 10, characterized in that: After obtaining the maintenance plan, the maintenance time and maintenance cost of the abnormal components are estimated and summarized to generate a maintenance plan data set, and the maintenance degree is constructed from the maintenance plan data set. ,in, Maintenance time and maintenance costs Perform linear normalization and map the corresponding data values ​​to the interval According to the following formula: ； Weight coefficient, , ; If the maintenance If the maintenance threshold is not exceeded, maintenance is performed on the abnormal component; if the maintenance threshold is exceeded, maintenance is performed on the abnormal component. If the maintenance threshold is exceeded, the abnormal component is treated as a damaged component and the damaged component is replaced.

12. A portable power supply fault detection system, characterized in that: include, The power detection unit detects the power consumption status of the power supply when there is an abnormality in the power use environment, aggregates the detection data to construct a power consumption data set, and generates the power consumption imbalance degree from the power consumption data set If the imbalance If it exceeds expectations, a warning instruction will be issued to the outside world; The power supply status test unit tests the power supply status, obtains test data and summarizes and constructs a power supply test data set, and generates the power supply status coefficient from the power supply test data. , if the state coefficient If the state is lower than the threshold, an identification instruction is issued to the outside; The fault detection unit uses the trained fault image recognition model to identify the power supply image after image acquisition. If there is a fault in the power supply component, the fault location is marked and the abnormal density is constructed based on the location information. , if the abnormal density If it exceeds expectations, a fault detection instruction will be issued to the outside; The fault analysis unit divides the portable power supply into several battery cells and tests them separately, generating an abnormality degree from the test data. , if the abnormality If the battery voltage exceeds the expected value, the abnormal components are screened out in the battery cell, and the test data of the abnormal components are extracted to obtain the corresponding maintenance features; wherein, the test voltage of the battery cell is Perform linear normalization and map the corresponding data values ​​to the interval The abnormality degree is constructed according to the following formula : ； in, is the test voltage on the i-th test node, is the qualified standard value of the test voltage, is the standard deviation of several test voltages; weight coefficient: , ,and ; The maintenance unit matches the corresponding maintenance plan for the abnormal component based on the output maintenance features, and builds the maintenance degree of the maintenance plan. , according to the degree of maintenance Screen out damaged components from abnormal components and use the damaged components to maintain the power supply; generate improvement coefficients based on the status data of the power supply after maintenance , if the improvement coefficient If the improvement threshold is not exceeded, an alarm command will be issued to the outside.

Images