An intelligent detection system and method applied to a power adapter

Abstract

The application discloses an intelligent detection system and method applied to a power adapter, relates to the technical field of intelligent detection of equipment, and comprises the following steps: analyzing the equipment abnormality degree of the power adapter in historical equipment detection records; acquiring abnormal historical equipment detection records of the power adapter, acquiring characteristic operation data of the power adapter, and evaluating the equipment operation parameters in the characteristic operation data and the equipment fitting degree between the power adapter; acquiring historical charging records and historical charging priority setting records of the power adapter, constructing a charging priority evaluation model of the power adapter based on the historical charging priority setting records; charging an electronic device by using the power adapter, performing priority evaluation on the electronic device by using the charging priority evaluation model, and performing equipment abnormality risk detection on the power adapter in combination with target equipment operation data, so that the charging strategy of the power adapter is adjusted.

Description

TECHNICAL FIELD

[0001] The application relates to an intelligent detection system and method applied to a power adapter. BACKGROUND

[0002] The power adapter is a device for converting alternating current into direct current, and is mainly used for converting power into a specific voltage and current suitable for electronic devices, so as to provide the required power supply for the electronic devices. Different electronic devices have great differences in the requirements for the current and voltage of the power supply. The voltage and current of the power adapter must be matched with the electronic device to be charged. Therefore, under normal circumstances, a power adapter of a certain model only has good charging effect on one or several specific models of electronic devices. The power adapter cannot intelligently detect the running power adapter according to the actual situation of the electronic device and the power adapter, so as to adjust the power supply of the power adapter to the connected electronic devices. This results in the fact that the user needs to purchase power adapters of different models, causing resource waste, or the user uses a power charger that does not match the electronic device to charge the electronic device, causing damage to the electronic device and the power charger. SUMMARY

[0003] The application aims to provide an intelligent detection system and method applied to a power adapter to solve the problems in the prior art.

[0004] To achieve the above-mentioned purpose, the application provides the following technical scheme: an intelligent detection method applied to a power adapter, the method comprising:

[0005] Step S100: obtaining the historical device detection record of the power adapter, analyzing the device abnormality degree of the power adapter in the historical device detection record, and obtaining the abnormal historical device detection record;

[0006] Step S200: obtaining the abnormal historical device detection record of the power adapter, obtaining the characteristic operation data of the power adapter, evaluating the device operation parameter in the characteristic operation data, and obtaining the target device operation data;

[0007] Step S300: obtaining the historical charging record and the historical charging priority setting record of the power adapter, and constructing a charging priority evaluation model of the power adapter based on the historical charging priority setting record;

[0008] Step S400: charging the electronic device by using the power adapter, performing priority evaluation on the electronic device by using the charging priority evaluation model, performing device abnormality risk detection on the power adapter by combining the target device operation data, and adjusting the charging strategy of the power adapter.

[0009] Further, the step S100 comprises:

[0010] Step S101: obtaining each historical equipment detection record of the power adapter, and obtaining data corresponding to the output voltage of the power adapter from the historical equipment detection record;

[0011] Step S102: analyzing the equipment abnormality degree of the power adapter in the historical equipment detection record, wherein the specific analysis process comprises:

[0012] obtaining the average value of the output voltage of the power adapter from the historical equipment detection record, and obtaining the mean value μ of the average value of the output voltage of the power adapter in each historical equipment detection record;

[0013] Step S103: calculating the characteristic dispersion value B of the power adapter:

[0014] ,

[0015] wherein j is the total number of each historical equipment detection record; B i is the average value of the output voltage of the power adapter in the i-th historical equipment detection record;

[0016] Step S104: obtaining the characteristic running range C of the output voltage of the power adapter, wherein C ∈ [μ-k×B, μ+k×B], wherein k is a preset characteristic range coefficient, when the average value of the output voltage of the power adapter in the historical equipment detection record is not within the characteristic running range C, it is determined that the equipment of the power adapter in the historical equipment detection record is abnormal, and the historical equipment detection record is recorded as an abnormal historical equipment detection record of the power adapter.

[0017] Further, the step S200 comprises:

[0018] Step S201: obtaining each abnormal historical equipment detection record of the power adapter, and obtaining the average value of each equipment running parameter in the power adapter from the abnormal historical equipment detection record;

[0019] Step S202: obtaining the characteristic running data of the power adapter, wherein the characteristic running data comprises a preset threshold of each equipment running parameter in the power adapter;

[0020] Step S203: obtaining the characteristic threshold of each equipment running parameter in the power adapter, wherein the characteristic threshold of the f-th equipment running parameter in the power adapter is obtained by:

[0021] obtaining the preset threshold Q f, wherein the feature value of the fth device operating parameter in the a th abnormal historical device detection record is G f α =Q f / |D f α -Q f |, wherein D f α is an average value of the fth device operating parameter in the a th abnormal historical device detection record;

[0022] obtaining a maximum value of the feature value of the fth device operating parameter in each abnormal historical device detection record, and obtaining an average value of the fth device operating parameter in the abnormal historical device detection record corresponding to the maximum value, and denoted as a feature threshold value of the fth device operating parameter;

[0023] Step S204: evaluating the device fitting degree between each device operating parameter in the feature operating data and the power adapter, wherein the specific process of evaluating the device fitting degree between the h th device operating parameter in the feature operating data and the power adapter is as follows:

[0024] obtaining a feature threshold value U h of the h th device operating parameter in the power adapter; h calculating the device fitting degree V h =Q h / |Q h -U h |, wherein Q h is a preset threshold value of the h th device operating parameter in the feature operating data;

[0025] Step S205: when the device fitting degree V h is less than or equal to a preset device fitting degree threshold value, determining that the h th device operating parameter in the feature operating data is not fitted with the power adapter, and setting the feature threshold value U h as a target threshold value of the h th device operating parameter;

[0026] when the device fitting degree V h is greater than the device fitting degree threshold value, determining that the h th device operating parameter in the feature operating data is fitted with the power adapter, and setting Q

[0027] Step S206: obtaining the target threshold values of the device operating parameters of the power adapter, and collecting the target threshold values to obtain target device operating data of the power adapter.

[0028] Further, the step S300 comprises:

[0029] The step S301: obtaining each historical charging record of the power adapter, obtaining a historical charging priority setting record of the power adapter, extracting a priority score set by the user for the historical electronic device from the historical charging priority setting record, obtaining a historical time period to which the historical charging priority setting record belongs, and recording the historical time period as a characteristic historical time period of the power adapter;

[0030] The step S302: obtaining a historical charging record of the power adapter in the characteristic historical time period and recording the historical charging record as a characteristic historical charging record of the power adapter, and obtaining a historical electronic device on which the user performs the priority score from the characteristic historical charging record and recording the historical electronic device as a characteristic historical electronic device in the characteristic historical charging record;

[0031] The step S303: extracting charging data of the characteristic historical electronic device from the characteristic historical charging record, wherein the charging data comprises data corresponding to each characteristic charging index of the characteristic historical electronic device;

[0032] The step S304: obtaining a plurality of characteristic historical charging records of the power adapter, and constructing a charging priority evaluation model of the power adapter, wherein the specific construction process comprises:

[0033] Obtaining an average value of each characteristic charging index in the plurality of characteristic historical electronic devices in the characteristic historical charging record;

[0034] Taking the average value of each characteristic charging index of the characteristic historical electronic device in the characteristic historical charging record as input data of the charging priority evaluation model, and taking the priority score of the characteristic historical electronic device as target output data of the charging priority evaluation model;

[0035] The step S305: dividing the plurality of characteristic historical charging records based on a preset proportion to obtain a model training set and a model test set, training the charging priority evaluation model by using the model training set, and obtaining a feature activation function f(x) of the charging priority evaluation model:

[0036] ,

[0037] Obtaining a preset weight matrix W and a bias p of the charging priority evaluation model, and obtaining an output S (m) of an mth layer of the charging priority evaluation model:

[0038] ,

[0039] wherein, W (m) is the weight matrix of the mth layer of the charging priority evaluation model; and S (m-1)an output of an m-1th layer of the charging priority evaluation model; p (m) a bias of an mth layer of the charging priority evaluation model;

[0040] obtaining a loss function L of the charging priority evaluation model:

[0041]

[0042] wherein n is a total number of training samples in a model training set; y i a true value of a priority score of an i th training sample in the model training set; y i a predicted value of the priority score of the i th training sample in the training set;

[0043] obtaining a gradient δ （m） =∂L / ∂p (m) of the bias of the mth layer of the charging priority evaluation model, and calculating a gradient of a weight matrix of the mth layer of the charging priority evaluation model, and the specific formula is:

[0044]

[0045] wherein S (m-1) is an output of an m-1th layer of the charging priority evaluation model;

[0046] based on the gradients of the weight matrix and the bias, updating the weight matrix and the bias using the model training set, and testing and verifying the charging priority evaluation model using a model test set, to obtain a verified charging priority evaluation model;

[0047] In the above steps, because in actual life, during the charging process of the power adapter to each electronic device, the device failure risk of the power adapter is likely to be caused by too many electronic devices being charged, as long as the charging strategy of the electronic device is adjusted, the device failure of the power adapter can be avoided, and through the establishment of the charging priority evaluation model, the power of the electronic device charging can be accurately evaluated, not only to ensure the normal operation of the power adapter, but also to minimize the impact of the charging adjustment of the electronic device on the user.

[0048] Further, the step S400 comprises:

[0049] Step S401: using the charging priority evaluation model, evaluating the priority of the power adapter to charge a plurality of electronic devices in the current period, to obtain a priority score of the plurality of electronic devices;

[0050] ​​Step S402: Monitor the device running state of the power adapter, and detect the device abnormal risk of the power adapter in combination with the target device running data, analyze the device abnormal degree of the power adapter, and the specific intelligent detection process is:

[0051] When any one of the device running parameters in the power adapter is greater than any one of the target threshold values of the device running parameters, it is determined that the power adapter has a device abnormal risk;

[0052] Step S403: According to the priority score of the power adapter of the power adapter in the current period, the power of the charging of the plurality of electronic devices is adjusted in order according to the score size, until there is no device running parameter greater than the target threshold value in the power adapter, and the charging strategy of the power adapter is adjusted.

[0053] In order to better realize the above method, an intelligent detection system applied to a power adapter is further provided, which comprises an abnormality analysis module, a device fitting evaluation module, an evaluation model construction module and an intelligent detection module.

[0054] The abnormality analysis module is used for analyzing the device abnormal degree of the power adapter in the historical device detection record, and obtaining the abnormal historical device detection record.

[0055] The device fitting evaluation module is used for evaluating the device fitting degree between the device running parameters in the feature running data and the power adapter, and obtaining the target device running data.

[0056] The evaluation model construction module is used for constructing the charging priority evaluation model of the power adapter.

[0057] The intelligent detection module is used for performing device intelligent detection on the power adapter, analyzing the device abnormal degree of the power adapter, and adjusting the charging strategy of the power adapter.

[0058] Further, the abnormality analysis module comprises a feature running range unit and an abnormality analysis unit.

[0059] The feature running range unit is used for acquiring the feature running range of the output voltage of the power adapter.

[0060] The abnormality analysis unit is used for analyzing the device abnormal degree of the power adapter in the historical device detection record according to the feature running range, and obtaining the abnormal historical device detection record.

[0061] Further, the device fitting evaluation module comprises a feature threshold unit and a device fitting evaluation unit.

[0062] A feature threshold unit is configured to acquire feature thresholds of various device operation parameters in the power adapter.

[0063] A device fitting evaluation unit is configured to evaluate the device fitting degree between the device operation parameters in the feature operation data and the power adapter according to the feature thresholds, to obtain target device operation data.

[0064] Further, the evaluation model construction module comprises a data acquisition unit and an evaluation model construction unit.

[0065] The data acquisition unit is configured to acquire historical charging records and historical charging priority setting records of the power adapter.

[0066] The evaluation model construction unit is configured to construct a charging priority evaluation model of the power adapter.

[0067] Further, the intelligent detection module comprises an intelligent detection unit.

[0068] The intelligent detection unit is configured to monitor the device operation state of the power adapter in the current period, and perform device intelligent detection on the power adapter in combination with the target device operation data, analyze the device abnormality degree of the power adapter, and adjust the charging strategy of the power adapter.

[0069] Compared with the prior art, the present application has the beneficial effects that: the present application realizes intelligent detection of the power adapter, acquires abnormal historical device detection records with device abnormalities by analyzing the device abnormality degree of the power adapter in the historical device detection records, acquires target device operation data of the power adapter, and establishes a charging priority evaluation model, which not only enables accurate and timely detection of device faults of the power adapter, but also enables adjustment of the charging strategy of the power adapter by acquiring the priority of the electronic device, so that the normal use of the power adapter can be ensured, and the adjustment of the power adapter strategy can minimize the impact on the user. BRIEF DESCRIPTION OF DRAWINGS

[0070] Fig. 1 is a method flowchart of an intelligent detection method applied to a power adapter;

[0071] Fig. 2 is a module schematic diagram of an intelligent detection system applied to a power adapter. DETAILED DESCRIPTION

[0072] With reference to the drawings and embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0073] Embodiment: As shown in the figure, the present application provides a technical solution, an intelligent detection method applied to a power adapter, the method comprising: Figs. 1-2

[0074] Step S100: obtaining historical equipment detection records of the power adapter, analyzing the equipment abnormality degree of the power adapter in the historical equipment detection records, and obtaining abnormal historical equipment detection records;

[0075] Step S100 comprises:

[0076] Step S101: obtaining each historical equipment detection record of the power adapter, and obtaining data corresponding to the output voltage of the power adapter from the historical equipment detection records;

[0077] Step S102: analyzing the equipment abnormality degree of the power adapter in the historical equipment detection records, wherein the specific analysis process comprises:

[0078] Step S102 comprises:

[0079] Step S103: calculating the characteristic dispersion value B of the power adapter:

[0080] ,

[0081] Wherein, j is the total number of each historical equipment detection record; B i is the average value of the output voltage of the power adapter in the i-th historical equipment detection record;

[0082] For example, j is 5; B1 is 10; B2 is 11; B3 is 13; B4 is 12; B5 is 15; the characteristic dispersion value B of the power adapter is calculated as:

[0083] ,

[0084] ​Step S104: Obtain the characteristic operating range C of the output voltage of the power adapter, C ∈ [μ-k×B, μ+k×B], where k is a preset characteristic range coefficient. When the average value of the output voltage of the power adapter in the historical device detection record is not within the characteristic operating range C, it is determined that the power adapter in the historical device detection record is abnormal, and the historical device detection record is recorded as an abnormal historical device detection record of the power adapter.

[0085] Step S200: Obtain the abnormal historical device detection record of the power adapter, obtain the characteristic operating data of the power adapter, evaluate the device operating parameters in the characteristic operating data and the device fitting degree between the power adapter, and obtain target device operating data.

[0086] Step S200 includes:

[0087] Step S201: Obtain each abnormal historical device detection record of the power adapter, and obtain the average value of each device operating parameter in the power adapter from the abnormal historical device detection record.

[0088] For example, the device operating parameters include device temperature, device voltage, etc.

[0089] Step S202: Obtain the characteristic operating data of the power adapter, wherein the characteristic operating data includes a preset threshold of each device operating parameter in the power adapter.

[0090] Step S203: Obtain the characteristic threshold of each device operating parameter in the power adapter, wherein the characteristic threshold of the fth device operating parameter in the power adapter is obtained by:

[0091] Obtain the preset threshold Q of the fth device operating parameter of the power adapter from the characteristic operating data f Calculate the characteristic value of the fth device operating parameter in each abnormal historical device detection record, wherein the characteristic value G of the fth device operating parameter in the αth abnormal historical device detection record is f α =Q f / |D f α -Q f |, wherein D f α is the average value of the fth device operating parameter in the αth abnormal historical device detection record.

[0092] Obtain the maximum value of the characteristic value of the fth device operating parameter in each abnormal historical device detection record, obtain the average value of the fth device operating parameter in the abnormal historical device detection record corresponding to the maximum value, and record it as the characteristic threshold of the fth device operating parameter.

[0093] Step S204: Evaluate the compatibility between each device operating parameter in the characteristic operating data and the power adapter. Specifically, the process for evaluating the compatibility between the h-th device operating parameter in the characteristic operating data and the power adapter is as follows:

[0094] Obtain the characteristic threshold U of the h-th device operating parameter in the power adapter. h Calculate the h-th device operating parameter in the characteristic operating data and its compatibility V with the power adapter. h =Q h / |Q h -U h |, where Q h The preset threshold for the h-th item of the device operating parameters in the feature operation data;

[0095] Step S205: When the equipment compatibility level V h If the device operating parameter in the h-th item of the characteristic operating data is less than or equal to the preset device compatibility threshold, it is determined that the device is not compatible with the power adapter, and the characteristic threshold U is set to... h , denoted as the target threshold of the h-th equipment operating parameter;

[0096] When the equipment compatibility level V h If the value exceeds the device compatibility threshold, determine the h-th device operating parameter in the characteristic operating data; if it matches the power adapter, then Q... h , denoted as the target threshold of the h-th equipment operating parameter;

[0097] Step S206: Obtain the target thresholds of various device operating parameters of the power adapter, and aggregate them to obtain the target device operating data of the power adapter;

[0098] Step S300: Obtain the historical charging records and historical charging priority setting records of the power adapter, and construct a charging priority evaluation model for the power adapter based on the historical charging priority setting records;

[0099] Step S300 includes:

[0100] Step S301: Obtain each historical charging record of the power adapter, obtain the historical charging priority setting record of the power adapter, extract the user's priority rating for the historical electronic device settings from the historical charging priority setting record, obtain the historical time period to which the historical charging priority setting record belongs, and record the historical time period as the characteristic historical time period of the power adapter.

[0101] Step S302: Obtain the historical charging records of the power adapter in a feature history period, and record them as the feature historical charging records of the power adapter. From the feature historical charging records, obtain the historical electronic devices that are scored by the user in priority, and record them as the feature historical electronic devices in the feature historical charging records.

[0102] Step S303: Extract the charging data of the feature historical electronic devices from the feature historical charging records. The charging data includes the data corresponding to each feature charging indicator of the feature historical electronic devices.

[0103] For example, the feature charging indicators include the remaining power of the feature historical electronic devices, the total battery power, etc.

[0104] Step S304: Obtain a plurality of feature historical charging records of the power adapter, and construct a charging priority evaluation model of the power adapter. The specific construction process includes:

[0105] Obtain the average values of each feature charging indicator in the plurality of feature historical electronic devices in the feature historical charging records.

[0106] Take the average values of each feature charging indicator of the feature historical electronic devices in the feature historical charging records as the input data of the charging priority evaluation model, and take the priority scores of the feature historical electronic devices as the target output data of the charging priority evaluation model.

[0107] Step S305: Divide the plurality of feature historical charging records based on a preset proportion to obtain a model training set and a model test set. Use the model training set to train the charging priority evaluation model, and obtain the feature activation function f(x) of the charging priority evaluation model:

[0108] ,

[0109] Obtain the preset weight matrix W and the bias p of the charging priority evaluation model, and obtain the output S (m) of the mth layer of the charging priority evaluation model.

[0110] ,

[0111] Wherein, W (m) is the weight matrix of the mth layer of the charging priority evaluation model; S (m-1) is the output of the (m-1)th layer of the charging priority evaluation model; p (m) is the bias of the mth layer of the charging priority evaluation model.

[0112] Obtain the loss function L of the charging priority evaluation model:

[0113] ,

[0114] wherein n is the total number of training samples in the model training set; y i is the true value of the priority score of the i-th training sample in the model training set; y' i is the predicted value of the priority score of the i-th training sample in the training set;

[0115] obtain the gradient δ （m） =∂L / ∂p (m) of the weight matrix of the m-th layer of the charging priority evaluation model, and the specific formula is:

[0116] ,

[0117] wherein S (m-1) is the output of the m-1-th layer of the charging priority evaluation model;

[0118] Based on the gradients of the weight matrix and the bias, the weight matrix and the bias are updated using the model training set, and the charging priority evaluation model is tested and verified using the model test set, to obtain the verified charging priority evaluation model.

[0119] Step S400: using the power adapter to charge the electronic device, and using the charging priority evaluation model to evaluate the priority of the electronic device, combining the target device running data to detect the device abnormal risk of the power adapter, and adjusting the charging strategy of the power adapter.

[0120] wherein step S400 comprises:

[0121] Step S401: using the charging priority evaluation model to evaluate the priority of the power adapter to charge a plurality of electronic devices in the current period, to obtain the priority score of the plurality of electronic devices;

[0122] Step S402: monitoring the device running state of the power adapter, and combining the target device running data to detect the device abnormal risk of the power adapter, and analyzing the device abnormal degree of the power adapter, and the specific intelligent detection process is:

[0123] When any one of the device running parameters in the power adapter is greater than the target threshold of any one of the device running parameters, it is determined that the power adapter has a device abnormal risk;

[0124] Step S403: according to the priority score of the plurality of electronic devices charged by the power adapter in the current period, adjusting the charging power of the plurality of electronic devices in turn according to the score size, until there is no device running parameter greater than the target threshold in the power adapter, and the charging strategy of the power adapter is adjusted.

[0125] For better implementation of the above method is also proposed, a smart detection system applied to the power adapter, the system includes abnormal analysis module, equipment fit evaluation module, evaluation model construction module, intelligent detection module;

[0126] Abnormal analysis module, for analyzing the degree of abnormality of the power adapter in the historical equipment detection record, obtaining the abnormal historical equipment detection record;

[0127] Equipment fit evaluation module, for evaluating the equipment fit degree between the equipment running parameters in the feature running data and the power adapter, obtaining the target equipment running data;

[0128] Evaluation model construction module, for constructing the charging priority evaluation model of the power adapter;

[0129] Intelligent detection module, for performing equipment intelligent detection on the power adapter, analyzing the degree of abnormality of the power adapter, and adjusting the charging strategy of the power adapter;

[0130] Among them, the abnormal analysis module includes a feature running range unit, an abnormal analysis unit;

[0131] Feature running range unit, for obtaining the feature running range of the output voltage of the power adapter;

[0132] Abnormal analysis unit, for analyzing the degree of abnormality of the power adapter in the historical equipment detection record according to the feature running range, obtaining the abnormal historical equipment detection record;

[0133] Among them, the equipment fit evaluation module includes a feature threshold unit, an equipment fit evaluation unit;

[0134] Feature threshold unit, for obtaining the feature threshold of each equipment running parameter in the power adapter;

[0135] Equipment fit evaluation unit, for evaluating the equipment fit degree between the equipment running parameters in the feature running data and the power adapter according to the feature threshold, obtaining the target equipment running data;

[0136] Among them, the evaluation model construction module includes a data acquisition unit, an evaluation model construction unit;

[0137] Data acquisition unit, for obtaining the historical charging record and the historical charging priority setting record of the power adapter;

[0138] Evaluation model construction unit, for constructing the charging priority evaluation model of the power adapter;

[0139] The intelligent detection module comprises an intelligent detection unit;

[0140] The intelligent detection unit is configured to monitor the device running state of the power adapter in the current period, and perform device intelligent detection on the power adapter in combination with the target device running data, analyze the device abnormality degree of the power adapter, and adjust the charging strategy of the power adapter.

[0141] It will be apparent to those skilled in the art that the application is not limited to the details of the above-exemplified embodiments, and that the application can be implemented in other particular forms without departing from the spirit or essential characteristics of the application. The embodiments should therefore be considered in all respects as illustrative and not restrictive, the scope of the application being defined by the appended claims rather than by the above description, and it is therefore intended that all changes and modifications that fall within the meaning and range of equivalency of the elements of the claims are to be embraced by the application. No reference signs in the claims should be considered as limiting the scope of the claims to the features to which the reference signs are attached.

Claims

1. An intelligent detection method for power adapters, characterized in that, The method includes: Step S100: Obtain the historical device detection records of the power adapter, analyze the degree of device abnormality of the power adapter in the historical device detection records, and obtain the abnormal historical device detection records; Step S200: Obtain the abnormal historical device detection record of the power adapter, obtain the characteristic operating data of the power adapter, evaluate the device operating parameters in the characteristic operating data and the degree of device compatibility between them and the power adapter, and obtain the target device operating data; Step S300: Obtain the historical charging records and historical charging priority setting records of the power adapter, and construct a charging priority evaluation model for the power adapter based on the historical charging priority setting records; Step S300 includes: Step S301: Obtain each historical charging record of the power adapter, obtain the historical charging priority setting record of the power adapter, extract the user's priority rating for the historical electronic device settings from the historical charging priority setting record, obtain the historical time period to which the historical charging priority setting record belongs, and record the historical time period as the characteristic historical time period of the power adapter. Step S302: Obtain the historical charging records of the power adapter within the characteristic historical period and record them as the characteristic historical charging records of the power adapter. From the characteristic historical charging records, obtain the historical electronic devices for which the user gave priority ratings and record them as the characteristic historical electronic devices in the characteristic historical charging records. Step S303: Extract the charging data of the historical electronic device from the historical charging record. The charging data includes data corresponding to various characteristic charging indicators of the historical electronic device. Step S304: Obtain several characteristic historical charging records of the power adapter, and construct a charging priority evaluation model for the power adapter. The specific construction process includes: Obtain the average value of various characteristic charging indicators in several characteristic historical electronic devices in the characteristic historical charging records; The average value of each characteristic charging index of the characteristic historical electronic device in the characteristic historical charging record is used as the input data of the charging priority evaluation model, and the priority score of the characteristic historical electronic device is used as the target output data of the charging priority evaluation model. Step S305: Based on a preset ratio, divide the several feature historical charging records to obtain a model training set and a model test set. Use the model training set to train the charging priority evaluation model and obtain the feature activation function f(x) of the charging priority evaluation model. Obtain the preset weight matrix W and bias p of the charging priority evaluation model, and obtain the output S of the m-th layer of the charging priority evaluation model. (m) ; Obtain the loss function L of the charging priority evaluation model; Obtain the gradient δ of the bias of the m-th layer of the charging priority evaluation model. （m） Calculate the gradient of the weight matrix of the m-th layer of the charging priority evaluation model; Based on the gradient of the weight matrix and the bias, the weight matrix and the bias are updated using the model training set, and the charging priority evaluation model is tested and verified using the model test set to obtain the verified charging priority evaluation model. Step S400: Use the power adapter to charge the electronic device, and use the charging priority evaluation model to evaluate the priority of the electronic device. Combine the target device operation data to detect the device abnormality risk of the power adapter and adjust the charging strategy of the power adapter.

2. The intelligent detection method for power adapters according to claim 1, characterized in that, Step S100 includes: Step S101: Obtain the historical device detection records of the power adapter, and obtain the data corresponding to the output voltage of the power adapter from the historical device detection records; Step S102: Analyze the degree of device abnormality of the power adapter in the historical device detection records, wherein the specific analysis process includes: From the historical device detection records, obtain the average value of the output voltage of the power adapter, and obtain the mean value μ of the average value of the output voltage of the power adapter in each historical device detection record; Step S103: Calculate the characteristic dispersion value B of the power adapter: ， Where j is the total number of detection records for each historical device; B i The average output voltage of the power adapter in the i-th historical device detection record; Step S104: Obtain the characteristic operating range C∈[μ-k×B,μ+k×B] of the output voltage of the power adapter, where k is a preset characteristic range coefficient. When the average value of the output voltage of the power adapter in the historical device detection record is not within the characteristic operating range C, determine that the power adapter in the historical device detection record is abnormal, and record the historical device detection record as the abnormal historical device detection record of the power adapter.

3. The intelligent detection method for power adapters according to claim 2, characterized in that, Step S200 includes: Step S201: Obtain the historical device detection records of each abnormality of the power adapter, and obtain the average value of each device operating parameter in the power adapter from the historical device detection records of abnormalities; Step S202: Obtain the characteristic operating data of the power adapter, wherein the characteristic operating data includes preset thresholds for various device operating parameters in the power adapter; Step S203: Obtain the feature thresholds of various device operating parameters in the power adapter, wherein the process of obtaining the feature threshold of the f-th device operating parameter in the power adapter is as follows: From the characteristic operation data, obtain the preset threshold Q of the f-th device operation parameter of the power adapter. f Calculate the feature value of the f-th equipment operating parameter in each of the abnormal historical equipment detection records, wherein the feature value G of the f-th equipment operating parameter in the α-th abnormal historical equipment detection record is... f α =Q f / |D f α -Q f |, where D f α The average value of the f-th equipment operating parameter in the α-th abnormal historical equipment detection record; Obtain the maximum value of the feature value of the f-th device operating parameter in each abnormal historical device detection record, obtain the average value of the f-th device operating parameter in the abnormal historical device detection record corresponding to the maximum value, and record it as the feature threshold of the f-th device operating parameter; Step S204: Evaluate the compatibility between each device operating parameter in the characteristic operating data and the power adapter. Specifically, the process for evaluating the compatibility between the h-th device operating parameter in the characteristic operating data and the power adapter is as follows: Obtain the feature threshold U of the h-th device operating parameter in the power adapter. h Calculate the device operating parameter h in the characteristic operating data and its compatibility V with the power adapter. h =Q h / |Q h -U h |, where Q h The preset threshold for the h-th device operating parameter in the feature operation data; Step S205: When the device compatibility level V h If the value of the h-th device operating parameter in the characteristic operating data is less than or equal to a preset device compatibility threshold, it is determined that the power adapter is incompatible, and the characteristic threshold U is set to... h , denoted as the target threshold of the h-th equipment operating parameter; When the device compatibility level V h If the value is greater than the device compatibility threshold, the h-th device operating parameter in the characteristic operating data is determined to be compatible with the power adapter, and the Q value is set to... h , denoted as the target threshold of the h-th equipment operating parameter; Step S206: Obtain the target thresholds of various device operating parameters of the power adapter, and aggregate them to obtain the target device operating data of the power adapter.

4. The intelligent detection method for power adapters according to claim 1, characterized in that, In step S305, the feature activation function f(x) and the output S (m) The specific formulas for the loss function L, the weight matrix of the m-th layer of the charging-first evaluation model, and the gradient of the bias include: The specific formula for the feature activation function f(x) is as follows: ， The output S (m) The specific formula is as follows: ， Among them, W (m) S is the weight matrix of the m-th layer of the charging priority evaluation model; (m-1) The output of the (m-1)th layer of the charging priority evaluation model; p (m) The bias of the m-th layer of the charging priority evaluation model; The loss function L has the following formula: ， Where n is the total number of training samples in the model training set; y i y' is the true value of the priority score for the i-th training sample in the model training set; i The predicted value of the priority score for the i-th training sample in the training set; The gradient of the bias at the m-th layer of the charging priority evaluation model is specifically formulated as: δ （m） =∂L / ∂p (m) ; The gradient of the weight matrix of the m-th layer of the charging priority evaluation model is specifically formulated as follows: ， Among them, S (m-1) The output of the (m-1)th layer of the charging priority evaluation model.

5. The intelligent detection method for power adapters according to claim 3, characterized in that, Step S400 includes: Step S401: Using the charging priority evaluation model, evaluate the priority of the power adapter for several electronic devices charging in the current cycle, and obtain the priority score of the several electronic devices. Step S402: Monitor the operating status of the power adapter and, in conjunction with the target device's operating data, perform device anomaly risk detection on the power adapter, analyzing the degree of device anomaly. The specific intelligent detection process is as follows: If any device operating parameter in the power adapter is greater than the target threshold of any device operating parameter, the power adapter is determined to have a device malfunction risk. Step S403: Based on the priority scores of several electronic devices being charged by the power adapter in the current cycle, adjust the charging power of several electronic devices in turn according to the score size until there are no devices in the power adapter with operating parameters greater than the target threshold, and adjust the charging strategy of the power adapter.

6. An intelligent detection system for power adapters, used to execute the intelligent detection method for power adapters as described in any one of claims 1-5, characterized in that, The system includes an anomaly analysis module, an equipment compatibility assessment module, an assessment model construction module, and an intelligent detection module. The anomaly analysis module is used to analyze the degree of anomaly of the power adapter in the historical device detection records to obtain the anomaly historical device detection records. The device compatibility assessment module is used to assess the degree of device compatibility between the device operating parameters in the characteristic operating data and the power adapter, and to obtain the target device operating data. The evaluation model construction module is used to construct a charging priority evaluation model for the power adapter. The intelligent detection module is used to perform intelligent device detection on the power adapter, analyze the degree of device abnormality of the power adapter, and adjust the charging strategy of the power adapter.

7. The intelligent detection system for power adapters according to claim 6, characterized in that, The anomaly analysis module includes a feature operating range unit and an anomaly analysis unit; The characteristic operating range unit is used to acquire the characteristic operating range of the output voltage of the power adapter; The anomaly analysis unit is used to analyze the degree of device anomaly of the power adapter in the historical device detection records based on the characteristic operating range, and to obtain the anomaly historical device detection records.

8. The intelligent detection system for power adapters according to claim 6, characterized in that, The device compatibility assessment module includes a feature threshold unit and a device compatibility assessment unit. The feature threshold unit is used to acquire the feature thresholds of various device operating parameters in the power adapter; The device compatibility assessment unit is used to assess the degree of device compatibility between the device operating parameters in the feature operating data and the power adapter based on the feature threshold, thereby obtaining the target device operating data.

9. The intelligent detection system for power adapters according to claim 6, characterized in that, The evaluation model construction module includes a data acquisition unit and an evaluation model construction unit; The data acquisition unit is used to acquire the historical charging records and historical charging priority setting records of the power adapter; The evaluation model construction unit is used to construct a charging priority evaluation model for the power adapter.

10. The intelligent detection system for power adapters according to claim 6, characterized in that, The intelligent detection module includes an intelligent detection unit; The intelligent detection unit is used to monitor the device operating status of the power adapter in the current cycle, and combine the target device operating data to perform intelligent device detection on the power adapter, analyze the degree of device abnormality of the power adapter, and adjust the charging strategy of the power adapter.

Images