Artificial intelligence electrical circuit aging degree detection system and detection method thereof

Abstract

The invention provides an artificial intelligence electrical circuit aging degree detection system, which comprises the following modules: a data acquisition module which can collect detected parameters in a target electrical circuit through a physical sensor and convert a received analog signal into a digital signal, the data processing module is arranged in a specific receiving area of the data acquisition module, the data processing module can receive data exported by the data acquisition module, the data processing module can preprocess the data, and an electrical circuit node graph structure is constructed by using a graph neural network algorithm, so that the electrical circuit node graph structure is constructed; and the fault line nodes can be quickly positioned by analyzing the node connection relationship and the parameter change, so that compared with the traditional manual troubleshooting, the fault positioning time is greatly shortened, a clear direction is provided for maintenance, and the operation loss caused by circuit faults is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of electric power detection, and in particular to an artificial intelligence electric circuit aging degree detection system and a detection method thereof. Background Art

[0002] Detecting the aging of electrical circuits is an important means of ensuring the safe and stable operation of electrical systems. As electrical equipment ages and the operating environment changes, circuit aging becomes increasingly prominent. Currently, there's a problem with standardizing testing standards for different circuit specifications when testing the aging of electrical circuits. Due to the diverse specifications of electrical circuits, traditional testing methods struggle to directly detect and compare aging values for circuits of different specifications, making subsequent holistic data analysis difficult. Aging testing also suffers from poor versatility, making it difficult to efficiently test multiple electrical circuits of different specifications simultaneously. This limits both testing efficiency and the scope of testing.

[0003] Publication No. CN111159650B, this artificial intelligence electrical circuit aging degree detection method and system obtains multi-dimensional historical data of each electrical circuit, and then obtains the load temperature rise characteristic curve and performs linear fitting. Then, the safe current range of each electrical circuit is determined based on the cross-sectional area ratio of each electrical circuit, and electrical circuits of different line specifications are converted to the same line specification. At the same time, the aging degree detection is converted into a temperature rise value detection corresponding to the middle value of the safe current range, and then converted into an aging value based on the temperature rise value. Through these means, the aging value detection and comparison between different electrical circuits are realized, which to a certain extent solves the problem of difficult detection of lines of different specifications.

[0004] However, in older residential communities, aging wiring cannot be quickly identified and located. The detection process doesn't fully consider the use of advanced algorithms to quickly locate the electrical wiring nodes where the problem occurs. When aging wiring is detected, the fault point cannot be quickly located, and manual troubleshooting may still be required, which is time-consuming, affects maintenance efficiency, and fails to provide a clear direction for subsequent repairs. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides an artificial intelligence electrical circuit aging degree detection system and a detection method thereof, which are used to solve the problems raised in the background technology.

[0006] An artificial intelligence electrical circuit aging degree detection system includes the following modules:

[0007] Data acquisition module: The data acquisition module can collect the detected parameters in the target electrical circuit through physical sensors, and the data acquisition module converts the received analog signals into digital signals;

[0008] Data processing module: The data processing module is arranged in a specific receiving area of the data acquisition module. The data processing module can receive the data exported by the data acquisition module and pre-process the data;

[0009] Artificial Intelligence Analysis Module: The artificial intelligence analysis module is used to extract and analyze the characteristic parameters of the data in the data processing module. The artificial intelligence analysis module also includes a submodule for rapid location of problem lines, a submodule for rapid solution generation, and a submodule for generating an optimal plan;

[0010] Communication module: The communication module is arranged between the data acquisition module, the data processing module, the artificial intelligence analysis module and the external device, and the communication module is connected to the terminal device;

[0011] Power module: The power module can provide stable power.

[0012] An artificial intelligence electrical circuit aging degree detection method includes the following steps:

[0013] Data acquisition S1: The data acquisition module collects multiple parameters such as current, voltage, temperature, humidity, partial discharge signal, insulation resistance, etc. through physical sensors, and converts the collected analog signals into digital signals and transmits them to the data processing module;

[0014] Data preprocessing S2: The data processing module performs filtering, denoising, and normalization operations on the received data, and removes interference and noise in the data. The data processing module uniformly maps the data to the [0, 1] interval, thereby improving the quality and usability of the data;

[0015] Feature extraction S3: The artificial intelligence analysis module receives the data processed by the data processing module, performs feature extraction, and maps the high-dimensional data imported by the data processing module into a low-dimensional space;

[0016] Aging degree assessment S4: The artificial intelligence analysis module is provided with an aging degree assessment model based on deep learning, and the model can extract local features of the data through a convolutional neural network;

[0017] Problem line rapid location S5: The problem line rapid location submodule in the artificial intelligence analysis module can use the graph neural network algorithm to locate the electrical line node where the problem occurs;

[0018] Rapid solution generation S6: The rapid solution generation submodule in the artificial intelligence analysis module can extract corresponding solutions from the knowledge graph through natural language processing technology and rule engine;

[0019] Optimal plan generation S7: The optimal plan generation submodule in the artificial intelligence analysis module can generate an optimal plan based on the cost database, the personnel skill database and the time estimation model, and the plan can be presented in a structure combining charts and text;

[0020] Result feedback and viewing S8: The communication module can import various text parameters expressed in the artificial intelligence analysis module into the central control terminal and the user mobile terminal. The central control terminal corresponds to professionals, and the user mobile terminal corresponds to general users.

[0021] Preferably, the physical sensing parameters connected to the data acquisition module include current, voltage, temperature, humidity, partial discharge signal, and insulation resistance of the electrical circuit.

[0022] Preferably, the artificial intelligence analysis module may adopt a principal component analysis method, which may perform feature extraction operations on the data preprocessed by the data processing module.

[0023] Preferably, the characteristic parameters extracted by the principal component analysis may correspond to the degree of aging of the electrical circuit.

[0024] Preferably, the deep learning-based aging assessment model in the artificial intelligence analysis module can utilize a long short-term memory network to process time series information of the data.

[0025] Preferably, the structure of the aging degree assessment model based on deep learning is a structure combining a convolutional neural network and a long short-term memory network. After receiving the characteristic parameters extracted by the data processing module, the aging degree assessment model based on deep learning can assess the aging degree of the electrical circuit and generate an electronic aging degree report;

[0026] Preferably, the algorithm used by the optimal plan generation submodule in the artificial intelligence analysis module is a multi-objective optimization algorithm;

[0027] Preferably, the plan derived by the optimal plan generation submodule includes professional technical guidance and popular explanations, and is understandable to both professionals and general users;

[0028] Preferably, the general user can install special software through a mobile phone, and the special software can be connected to the communication module in the detection system, wherein the communication module can import the aging status of the circuit of the general user's house, the solution and the optimal plan into the user's mobile phone.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1. In use, the present invention utilizes a graph neural network algorithm to construct an electrical circuit node graph structure. By analyzing node connection relationships and parameter changes, it can quickly locate problematic circuit nodes. Compared with traditional manual troubleshooting, this significantly shortens fault location time, provides a clear direction for maintenance, reduces operational losses caused by circuit faults, avoids blind troubleshooting that interferes with normal operations, improves fault location efficiency, and ensures stable operation of the electrical system.

[0031] 2. During use, this invention leverages natural language processing technology and a rule engine to extract information from the knowledge graph. Based on line aging characteristics, such as reduced insulation resistance and abnormal partial discharge, it quickly generates solutions. Compared to traditional methods that rely on manual experience, this approach reduces generation time and improves accuracy, providing scientific maintenance solutions, avoiding misjudgments and inappropriate solutions, reducing maintenance costs and safety hazards, and improving the relevance and effectiveness of maintenance work.

[0032] 3. During use, the present invention uses a multi-objective optimization algorithm based on the cost database, personnel skill database and time estimation model, comprehensively considers factors such as maintenance cost, personnel arrangement, and maintenance time, and generates a scientific and reasonable plan, which is presented in a combination of charts and text, and contains professional technical guidance and popular explanations. Compared with traditional manual planning, it reduces maintenance costs, shortens maintenance cycles, improves personnel utilization, realizes optimal resource allocation, and ensures that maintenance work is carried out efficiently and orderly.

[0033] 4. During use, the communication module transmits the text parameters of the AI analysis module to the central control terminal and the user's mobile terminal. This provides professional data reports, analysis reports, and concise, easy-to-understand charts and text for both professionals and general users. This shortens information access time, improves user satisfaction, and ensures that test results are readily available, facilitating decision-making for professionals and helping general users understand circuit conditions, thereby facilitating electrical line maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a schematic diagram of the module layout of the artificial intelligence electrical circuit aging degree detection system of the present invention;

[0035] Figure 2 This is a flow chart of the artificial intelligence electrical circuit aging degree detection method of the present invention;

[0036] Figure 3 This is the overall architecture and feedback diagram of the artificial intelligence electrical circuit aging degree detection system of the present invention. DETAILED DESCRIPTION

[0037] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0038] An artificial intelligence electrical circuit aging degree detection system includes the following modules:

[0039] Data acquisition module: The data acquisition module can collect the detected parameters in the target electrical circuit through physical sensors, and the data acquisition module converts the received analog signals into digital signals;

[0040] Among them, the physical sensor can be set as a current sensor, a voltage sensor, a temperature sensor and a partial discharge sensor, and the current sensor can convert the current signal into a voltage signal based on the Hall effect or electromagnetic induction principle;

[0041] The voltage sensor can use a voltage divider circuit or a voltage transformer to measure high voltage;

[0042] Temperature sensors use thermistors or thermocouples to convert temperature changes into electrical signals;

[0043] Partial discharge sensors can detect electromagnetic waves or ultrasonic signals generated by discharge;

[0044] In addition, the conversion process formula of analog signal to digital signal is:

[0045] And V analog is the input analog voltage value, V max With V min is the input voltage range of the ADC, n is the number of bits of the ADC, V digital is the converted numeric value.

[0046] Data processing module: The data processing module is arranged in a specific receiving area of the data acquisition module. The data processing module can receive the data exported by the data acquisition module and pre-process the data;

[0047] Among them, the data processing module can adopt a filtering algorithm, and the sliding average filtering formula is Where x(n) is the input signal, y(n) is the filtered signal, and N is the window size.

[0048] Artificial Intelligence Analysis Module: The artificial intelligence analysis module is used to extract and analyze the characteristic parameters of the data in the data processing module. The artificial intelligence analysis module also includes a submodule for rapid location of problem lines, a submodule for rapid solution generation, and a submodule for generating an optimal plan;

[0049] Communication module: The communication module is arranged between the data acquisition module, the data processing module, the artificial intelligence analysis module and the external device, and the communication module is connected to the terminal device;

[0050] Power module: The power module can provide stable power.

[0051] An artificial intelligence electrical circuit aging degree detection method includes the following steps:

[0052] Data acquisition S1: The data acquisition module collects multiple parameters such as current, voltage, temperature, humidity, partial discharge signal, insulation resistance, etc. through physical sensors, and converts the collected analog signals into digital signals and transmits them to the data processing module;

[0053] Data preprocessing S2: The data processing module performs filtering, denoising, and normalization operations on the received data, and removes interference and noise in the data. The data processing module uniformly maps the data to the [0, 1] interval, thereby improving the quality and usability of the data;

[0054] Among them, the normalization formula is:

[0055] Feature extraction S3: The artificial intelligence analysis module receives the data processed by the data processing module, performs feature extraction, and maps the high-dimensional data imported by the data processing module into a low-dimensional space;

[0056] Aging degree assessment S4: The artificial intelligence analysis module is provided with an aging degree assessment model based on deep learning, and the model can extract local features of the data through a convolutional neural network;

[0057] Problem line rapid location S5: The problem line rapid location submodule in the artificial intelligence analysis module can use the graph neural network algorithm to locate the electrical line node where the problem occurs;

[0058] Rapid solution generation S6: The rapid solution generation submodule in the artificial intelligence analysis module can extract corresponding solutions from the knowledge graph through natural language processing technology and rule engine;

[0059] Optimal plan generation S7: The optimal plan generation submodule in the artificial intelligence analysis module can generate an optimal plan based on the cost database, the personnel skill database and the time estimation model, and the plan can be presented in a structure combining charts and text;

[0060] Result feedback and viewing S8: The communication module can import various text parameters expressed in the artificial intelligence analysis module into the central control terminal and the user mobile terminal. The central control terminal corresponds to professionals, and the user mobile terminal corresponds to general users.

[0061] The artificial intelligence analysis module may adopt a principal component analysis method, which may perform a feature extraction operation on the data preprocessed by the data processing module.

[0062] Among them, the calculated covariance matrix in the principal component analysis (PCA) feature extraction is:

[0063]

[0064] Perform eigendecomposition on the covariance matrix: C = WΛW T ;

[0065] Select the eigenvectors corresponding to the first k largest eigenvalues and project the data into a low-dimensional space:

[0066] The structure of the deep learning-based aging assessment model is a combination of a convolutional neural network and a long short-term memory network. After receiving the characteristic parameters extracted by the data processing module, the deep learning-based aging assessment model can assess the aging degree of the electrical circuit and generate an electronic aging degree report.

[0067] Among them, the convolution operation formula in the convolutional neural network (CNN) aging assessment model is: And x is the input feature map, w is the convolution kernel, and b is the bias;

[0068] The node embedding update formula in the graph neural network (GNN) localization algorithm is:

[0069]

[0070] The deep learning-based aging assessment model in the artificial intelligence analysis module can use long short-term memory networks to process the time series information of the data.

[0071] Among them, the forget gate formula in the long short-term memory network is f t =σ(W f ·[h t-1 ,x t ]+b f );

[0072] Input gate formula: i t =σ(W i ·[h t-1 ,x t ]+b i );

[0073] Candidate cell states:

[0074] Cell status update:

[0075] Output gate: o t =σ(W o ·[h t-1 ,xt ]+b o );

[0076] Hidden state: h t =o t tanh(C t ).

[0077] The algorithm used by the optimal plan generation submodule in the artificial intelligence analysis module is a multi-objective optimization algorithm;

[0078] The plan exported by the optimal plan generation submodule includes professional technical guidance and popular explanations, and is understandable to both professionals and general users.

[0079] Among them, the multi-objective optimization algorithm of the optimal plan generation submodule is minF(x)=[f1(x),f2(x),...,f k (x)] T ;

[0080] The constraints are: g i (x)≤0,i=1,2,...,m and h j (x) = 0, j = 1, 2, ..., p, and f1(x) is the maintenance cost function, f2(x) is the maintenance time function, and f3(x) is the personnel arrangement rationality function;

[0081] The Pareto optimal solution set is

[0082] In addition, the data transmission protocol in the communication module is the MQTT protocol, and the MQTT data packet = fixed header + variable header + message body;

[0083] Communication quality assessment:

[0084]

[0085] Example 1: In a large commercial complex, during the operation of the artificial intelligence electrical circuit aging detection system, the data acquisition module collects parameters such as current, voltage, and temperature of the electrical circuits in each area in real time. One day, the system detected abnormal voltage fluctuations in some areas on the third floor of the mall. The data processing module pre-processed the collected data by filtering and denoising, and then transmitted the data to the artificial intelligence analysis module. The problem line rapid positioning submodule in the artificial intelligence analysis module is started, and the graph neural network algorithm is used to analyze the electrical line network of the entire mall. Each node in the electrical line is constructed into a graph structure. By analyzing the connection relationship and parameter changes between the nodes, it is quickly located that a line in a distribution room on the third floor of the mall has an aging problem, and the faulty line node is accurately locked, providing a clear direction for subsequent maintenance;

[0086] Among them, the data acquisition module ensures the acquisition of comprehensive and real-time electrical line parameters, providing basic data for subsequent analysis; the data processing module purifies the original data, improves data quality, and ensures analysis accuracy; the graph neural network algorithm of the problem line rapid positioning sub-module can fully explore the network structure characteristics and parameter associations of the electrical line to locate the fault node.

[0087] Compared with traditional manual troubleshooting, this method shortens the fault location time from several hours to a few minutes, greatly improving the efficiency of fault location, reducing operational losses caused by circuit failures in commercial complexes, and avoiding interference with normal operations caused by blind troubleshooting.

[0088] Please refer to Table 1 for relevant parameters.

[0089] Table 1: Comparison of efficiency of rapid location of problem lines

[0090]

[0091]

[0092] Example 2: During the inspection of electrical lines in an old residential area, the system detected that multiple lines in a residential building were aging. After the steps of data collection, preprocessing and feature extraction, the solution rapid generation submodule in the artificial intelligence analysis module began to work. This submodule extracts relevant information from a pre-built knowledge graph based on natural language processing technology and rule engine. Since the knowledge graph stores a large number of cases, causes and solutions to electrical line aging problems, the submodule quickly extracts corresponding solutions by analyzing the specific characteristics of the aging of the residential building lines, such as reduced insulation resistance and abnormal partial discharge signals, including measures such as replacing aging lines, adjusting line layout, and strengthening line insulation, providing a specific operation plan for subsequent maintenance work;

[0093] The data collection and processing module provides reliable data for solution generation. The natural language processing technology in the rapid solution generation submodule enables semantic interaction between data and the knowledge graph. The rule engine efficiently searches for and matches solutions. The knowledge graph, as a knowledge base, stores solutions to various aging problems and supports decision-making.

[0094] In the past, relying on manual experience to develop solutions was time-consuming and difficult to guarantee accuracy. This system can generate accurate solutions within seconds, avoiding misjudgments or inappropriate solutions due to lack of experience, providing scientific and efficient guidance for line maintenance in old communities, and reducing maintenance costs and safety hazards.

[0095] Please refer to Table 2 for relevant parameters.

[0096] Table 2: Solution Rapid Generation Efficiency Improvement Table

[0097]

[0098] Example 3: In a certain industrial park, the artificial intelligence electrical circuit aging degree detection system detected that the electrical circuits in multiple workshops had aging problems to varying degrees. After completing data collection, processing and analysis, the optimal plan generation submodule in the artificial intelligence analysis module uses a multi-objective optimization algorithm based on the cost database, personnel skill database and time estimation model, taking into account factors such as maintenance costs, personnel arrangements and maintenance time, to generate an optimal plan. The plan is presented in a structure that combines charts and text, and lists in detail the material costs required for line maintenance in each workshop, the expected number of maintenance personnel and skill requirements, maintenance time nodes for each stage and other information. At the same time, the plan also includes professional and technical guidance to guide maintenance personnel in their operations, as well as popular explanations to facilitate park managers to understand the overall situation of the maintenance plan;

[0099] Among them, the cost and personnel skills database provides a basis for decision-making, and the time estimation model ensures the accuracy of time prediction; the multi-objective optimization algorithm of the optimal plan generation sub-module balances multi-dimensional factors, generates a scientific and reasonable plan, and realizes optimal resource allocation. Traditional manual maintenance plans often lose sight of one thing while focusing on another. The optimal plan generated by this system reduces maintenance costs by 20%-30%, shortens maintenance cycles by 15%-25%, and improves personnel utilization, ensuring that line maintenance in industrial parks is carried out efficiently and orderly, reducing the impact on production.

[0100] Please refer to Table 3 for relevant parameters.

[0101] Table 3: Optimal plan generation benefit optimization table

[0102]

[0103] Example 4: In a residential circuit inspection project in a city, after the inspection system completes the aging inspection of each household's electrical circuits, the communication module imports the various text parameters generated by the artificial intelligence analysis module, such as the aging assessment of each household's circuits, specific solutions, and optimal plans, into the central control terminal and the user's mobile terminal. For professionals, the central control terminal presents information in the form of professional data tables and detailed analysis reports, facilitating in-depth analysis and decision-making. For ordinary users, through dedicated software installed on their mobile phones, the communication module displays the aging status of their own house's circuits in concise and easy-to-understand charts and text, and provides popular explanations of solutions and optimal plans, allowing ordinary users to clearly understand the status of their own circuits and know how to deal with them.

[0104] Among them, the communication module is responsible for data security, efficient transmission and distribution; the central control terminal and user mobile terminal provide targeted data display and interactive interface according to different user needs, meet users' needs for viewing and using test results, and achieve accurate access to test results. Professionals can formulate maintenance strategies based on detailed data, and general users can intuitively understand the status of their own circuits, improve users' awareness and attention to circuit safety, enhance users' trust in the detection system, and promote the smooth development of electrical line maintenance work.

[0105] Please refer to Table 4 for relevant parameters.

[0106] Table 4: Result feedback, viewing time and satisfaction improvement

[0107]

[0108] Based on the above four embodiments, we can draw innovative conclusions about the four key steps of rapid problem line location, rapid solution generation, optimal plan generation, and result feedback and review: The problem line rapid location submodule of this system uses a graph neural network algorithm to construct each node of the electrical line into a graph structure. By analyzing the node connection relationship and parameter changes, it can quickly and accurately locate the faulty line node. At the same time, traditional manual troubleshooting relies on manpower to check each section, which is not only time-consuming but also easily affected by subjective factors, making it difficult to quickly locate the fault location. This system fully exploits the network structure characteristics of the electrical line and the relationship between parameters, avoiding the interference of blind troubleshooting on normal operations, greatly improving the efficiency of fault location, and reducing operational losses caused by circuit failures.

[0109] The rapid solution generation submodule, based on natural language processing technology and a rule engine, extracts relevant information from the knowledge graph and can quickly generate precise solutions based on the specific characteristics of line aging. Furthermore, in electrical line inspections in older residential areas, the system generates solutions within seconds, including replacing aging lines and adjusting line layouts, for situations such as reduced insulation resistance and abnormal partial discharge signals. This reduces the time required to generate solutions. Furthermore, the system stores a large number of cases and solutions through the knowledge graph, enabling semantic interaction between data and the knowledge graph, efficient retrieval and matching, and providing scientific and efficient guidance for maintenance work, reducing maintenance costs and safety hazards.

[0110] The optimal plan generation submodule utilizes a multi-objective optimization algorithm, comprehensively considering the cost database, personnel skill database, and time estimation model to generate a scientific and reasonable plan. Traditional manual maintenance planning often fails to balance multiple factors and is difficult to balance. The optimal plan generated by this system is presented in a combination of charts and text, including professional technical guidance and popular explanations. This not only facilitates maintenance personnel's operation but also makes it easier for management to understand the overall situation, ensuring that maintenance work is carried out efficiently and orderly, and reducing the impact on production.

[0111] The communication module accurately transmits the test results to the central control terminal and the user's mobile terminal, and provides targeted data display according to different user needs. Professionals can obtain detailed data reports and analysis reports through the central control terminal to make decisions. General users can understand their own circuit conditions and treatment plans through the mobile terminal with simple and easy-to-understand charts and texts. In addition, this system realizes the rapid access to test results, improves users' awareness and attention to circuit safety, enhances users' trust in the detection system, and promotes the smooth implementation of electrical line maintenance work.

[0112] All aspects of the present invention are within the scope of protection of this patent.

[0113] The embodiments of the present invention are provided for the purpose of illustration and description. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present invention. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. An artificial intelligence electrical circuit aging degree detection system, characterized in that: Includes the following modules: Data acquisition module: The data acquisition module can collect the detected parameters in the target electrical circuit through physical sensors, and the data acquisition module converts the received analog signals into digital signals; Data processing module: The data processing module is arranged in a specific receiving area of the data acquisition module. The data processing module can receive the data exported by the data acquisition module and pre-process the data; Artificial Intelligence Analysis Module: The artificial intelligence analysis module is used to extract and analyze the characteristic parameters of the data in the data processing module. The artificial intelligence analysis module also includes a submodule for rapid location of problem lines, a submodule for rapid solution generation, and a submodule for generating an optimal plan; Communication module: The communication module is arranged between the data acquisition module, the data processing module, the artificial intelligence analysis module and the external device, and the communication module is connected to the terminal device; Power module: The power module can provide stable power.

2. An artificial intelligence electrical circuit aging degree detection method, characterized in that: The following steps are involved: Data acquisition S1: The data acquisition module collects multiple parameters such as current, voltage, temperature, humidity, partial discharge signal, insulation resistance, etc. through physical sensors, and converts the collected analog signals into digital signals and transmits them to the data processing module; Data preprocessing S2: The data processing module performs filtering, denoising, and normalization operations on the received data, and removes interference and noise in the data. The data processing module uniformly maps the data to the [0, 1] interval, thereby improving the quality and usability of the data; Feature extraction S3: The artificial intelligence analysis module receives the data processed by the data processing module, performs feature extraction, and maps the high-dimensional data imported by the data processing module into a low-dimensional space; Aging degree assessment S4: The artificial intelligence analysis module is provided with an aging degree assessment model based on deep learning, and the model can extract local features of the data through a convolutional neural network; Problem line rapid location S5: The problem line rapid location submodule in the artificial intelligence analysis module can use the graph neural network algorithm to locate the electrical line node where the problem occurs; Rapid solution generation S6: The rapid solution generation submodule in the artificial intelligence analysis module can extract corresponding solutions from the knowledge graph through natural language processing technology and rule engine; Optimal plan generation S7: The optimal plan generation submodule in the artificial intelligence analysis module can generate an optimal plan based on the cost database, the personnel skill database and the time estimation model, and the plan can be presented in a structure combining charts and text; Result feedback and viewing S8: The communication module can import various text parameters expressed in the artificial intelligence analysis module into the central control terminal and the user mobile terminal. The central control terminal corresponds to professionals, and the user mobile terminal corresponds to general users.

3. The artificial intelligence electrical circuit aging degree detection method according to claim 2, characterized in that: The physical sensing parameters connected to the data acquisition module include current, voltage, temperature, humidity, partial discharge signal, and insulation resistance of the electrical circuit.

4. The artificial intelligence electrical circuit aging degree detection method according to claim 2, characterized in that: The artificial intelligence analysis module may adopt a principal component analysis method, which may perform a feature extraction operation on the data preprocessed by the data processing module.

5. The artificial intelligence electrical circuit aging degree detection method according to claim 4 is characterized in that: The characteristic parameters extracted by the principal component analysis may correspond to the degree of aging of the electrical circuit.

6. The artificial intelligence electrical circuit aging degree detection method according to claim 2, characterized in that: The deep learning-based aging assessment model in the artificial intelligence analysis module can use the long short-term memory network to process the time series information of the data.

7. The artificial intelligence electrical circuit aging degree detection method according to claim 2, characterized in that: The structure of the deep learning-based aging assessment model is a combination of a convolutional neural network and a long short-term memory network. After receiving the characteristic parameters extracted by the data processing module, the deep learning-based aging assessment model can assess the aging degree of the electrical circuit and generate an electronic aging degree report.

8. The artificial intelligence electrical circuit aging degree detection method according to claim 2, characterized in that: The algorithm used by the optimal plan generation submodule in the artificial intelligence analysis module is a multi-objective optimization algorithm.

9. The artificial intelligence electrical circuit aging degree detection method according to claim 2, characterized in that: The plan derived by the optimal plan generation submodule includes professional technical guidance and popular explanations, and is understandable to both professionals and general users.

10. The artificial intelligence electrical circuit aging degree detection method according to claim 2, characterized in that: The general user can install special software through a mobile phone, and the special software can be connected to the communication module in the detection system, wherein the communication module can import the aging status of the circuit of the general user's house, the solution and the optimal plan into the user's mobile phone.

Images