Method and system for offline maintenance of control panel based on artificial intelligence

Abstract

The invention provides a method and system for offline maintenance of a control panel based on artificial intelligence, and relates to the technical field of artificial intelligence, and the method comprises the steps: an AI self-inspection module reads real-time state data of a sensor and an execution part, generates a fault diagnosis result based on an offline fault maintenance information base, generates a fault maintenance scheme, and sends the fault maintenance scheme to a server; and after determining that the user has the maintenance tool, generating a fault inspection instruction, guiding the user to complete inspection, comparing inspection data with a standard, and generating a maintenance guidance scheme. And guiding the user to complete device replacement and equipment assembly, rechecking the operation state of the equipment, and confirming that fault repair is completed. According to the invention, under the condition that the network is unstable or is completely offline, the control panel can also quickly respond to the fault by virtue of the built-in AI module, so that preliminary self-diagnosis and repair are realized. Once the fault information flow is detected, the AI module can immediately analyze the fault type and automatically deploy the corresponding maintenance strategy according to the data in the offline library, so that the downtime is reduced to the greatest extent.

Description

Technical Field

[0001] The present invention relates to artificial intelligence technology, and in particular to a method and system for maintaining a control panel offline based on artificial intelligence. Background Art

[0002] At present, the maintenance of various equipment is increasingly dependent on complex diagnostic tools and professional technicians. Traditional equipment maintenance methods usually require sending the equipment to a repair center or having professional maintenance personnel provide door-to-door service, which is not only time-consuming and costly, but also difficult to obtain timely maintenance support in some cases, such as remote areas or emergency situations. With the development of artificial intelligence technology, AI technology has begun to be applied to the field of equipment maintenance, which can realize equipment fault diagnosis and predictive maintenance.

[0003] However, existing AI-based equipment maintenance methods mainly rely on online connections and cloud server support. These methods require uploading the equipment's operating data to the cloud server for analysis and processing, and then feeding back the diagnosis results and maintenance plans to the user. This online maintenance method has the following defects and shortcomings: Network dependency: Online maintenance relies on a stable network connection and cannot work properly in scenarios with poor network signals or no network coverage. This limits its application in some special scenarios, such as remote areas, underground facilities, or offshore platforms.

[0004] Data security risks: Online maintenance requires uploading the device's operating data to a cloud server, which poses a risk of data leakage and tampering, especially for devices involving sensitive information, such as medical equipment, financial equipment, etc.

[0005] High maintenance costs: Online maintenance requires continuous network connection and cloud server maintenance, which increases maintenance costs, especially for some equipment that needs to run for a long time, such as industrial equipment and infrastructure. Summary of the invention

[0006] The embodiments of the present invention provide a method and system for offline maintenance of a control panel based on artificial intelligence, which can solve the problems in the prior art.

[0007] According to a first aspect of the embodiments of the present invention, Provide an artificial intelligence-based offline maintenance control panel method, including: The fault detection instructions input by the user's voice are collected through the built-in microphone of the device, and the voice recognition module converts the fault detection instructions into text instructions; the AI ​​self-test module is triggered to execute the self-test process according to the text instructions, and the AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board, and the real-time status data includes motor current, voltage, temperature and speed parameters; the AI ​​self-test module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library, and the fault diagnosis result includes fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device; Based on the fault diagnosis result, the AI ​​self-check module generates a fault repair plan, which includes a repair tool list and repair operation steps; the AI ​​self-check module asks the user by voice whether he has the repair tools in the repair tool list, and after the user confirms that he has the repair tools, the AI ​​self-check module determines the equipment disassembly path according to the fault location information, and generates disassembly guidance information according to the equipment disassembly path, and the disassembly guidance information includes power disconnection indication, protection requirements and disassembly operation sequence; the AI ​​self-check module broadcasts the disassembly guidance information to the user by voice to guide the user to complete the equipment disassembly; The AI ​​self-check module generates a fault inspection instruction based on the fault type, and the fault inspection instruction includes inspection items and inspection parameter standards; the AI ​​self-check module broadcasts the fault inspection instruction by voice to guide the user to complete the inspection items and obtain inspection data; the AI ​​self-check module compares the inspection data with the inspection parameter standards to generate a maintenance guidance plan, and the maintenance guidance plan includes information on components to be replaced and installation requirements; the AI ​​self-check module generates equipment assembly guidance information according to the equipment disassembly path, guides the user to complete component replacement and equipment assembly by voice, and reviews the equipment operation status based on the real-time status data to confirm that the fault repair is complete.

[0008] The fault detection instructions input by the user's voice are collected through the built-in microphone of the device, and the voice recognition module converts the fault detection instructions into text instructions; according to the text instructions, the AI ​​self-test module is triggered to execute the self-test process, and the AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board. The real-time status data includes motor current, voltage, temperature and speed parameters including: The fault detection instruction input by the user's voice is collected through the built-in microphone of the device, wherein the built-in microphone of the device adopts an electret condenser microphone, and the sensitivity of the electret condenser microphone is -38±2dB, and the frequency response range is 20Hz-20kHz; the fault detection instruction is pre-amplified and analog-to-digital converted to obtain a digital voice signal; Input the digital voice signal into a voice recognition module, the voice recognition module includes a temporal convolutional network structure, the temporal convolutional network structure includes eight convolutional layers and two fully connected layers, the convolution kernel size of each layer is 3, and the fault detection instruction is converted into a text instruction; The AI ​​self-test module is triggered to execute the self-test process according to the text instruction. The AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board through the microcontroller of the 32-bit ARMCortex-M4 core. The real-time status data includes motor current, voltage, temperature and speed parameters.

[0009] The AI ​​self-checking module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library. The fault diagnosis result includes the fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device, including: Performing time domain analysis on the real-time status data to obtain the root mean square value, peak factor, margin factor and waveform factor, performing frequency domain analysis to obtain the power spectrum density, inverse spectrum and envelope spectrum, and combining the results of the time domain analysis and the results of the frequency domain analysis to form a feature analysis vector; based on reading the standard feature fingerprints of various motor faults in an offline fault maintenance information library, the offline fault maintenance information library adopts a hierarchical storage structure, including a feature template layer, a fault description layer and a diagnostic rule layer; The AI ​​self-check module performs similarity matching between the feature vector and the standard feature fingerprint, calculates feature correlation using a fuzzy similarity algorithm based on adaptive weights, and generates a preliminary fault matching result; the preliminary fault matching result is input into a pre-trained deep neural network model, the deep neural network model includes a five-layer fully connected structure, uses Dropout to prevent overfitting, and outputs a probability distribution of fault types; The AI ​​self-check module extracts corresponding fault information from the offline fault maintenance information library according to the fault type probability distribution, and generates a fault diagnosis result including fault code, fault type and fault location information; inputs the fault diagnosis result into a speech synthesis model, and the speech synthesis model adaptively adjusts the voice tone and intonation parameters according to the severity of the fault to generate a diagnostic broadcast voice; and plays the diagnostic broadcast voice through the built-in speaker of the device.

[0010] The AI ​​self-check module generates a fault check instruction based on the fault type, wherein the fault check instruction includes an inspection item and an inspection parameter standard; the AI ​​self-check module broadcasts the fault check instruction by voice, guides the user to complete the inspection item and obtains the inspection data, including: The AI ​​self-check module receives fault type information, and searches for corresponding fault diagnosis rules in a pre-established motor fault knowledge base according to the fault type information, wherein the motor fault knowledge base includes inspection items and inspection parameter standards for different fault types; the AI ​​self-check module generates a fault inspection instruction based on the fault diagnosis rule, wherein the fault inspection instruction includes inspection items to be executed and inspection parameter standards corresponding to each inspection item; The AI ​​self-check module inputs the fault check instruction into the speech synthesis model, and the speech synthesis model performs natural language processing on the fault check instruction to generate a speech broadcast text; the AI ​​self-check module generates a speech signal according to the speech broadcast text, and broadcasts the specific inspection item content and inspection parameter standards to the user through the built-in speaker of the device; The AI ​​self-check module receives voice information fed back by the user. When the voice information fed back by the user indicates that repeated broadcasting is required, the AI ​​self-check module re-broadcasts the current inspection item; when the voice information fed back by the user contains inspection data, the AI ​​self-check module compares the inspection data with the inspection parameter standard; the AI ​​self-check module determines whether additional inspection items are required based on the comparison results, and returns to generate new fault inspection instructions when additional inspection items are required, and stores the inspection result data when all inspection items are completed.

[0011] The AI ​​self-check module compares the inspection data with the inspection parameter standard to generate a maintenance guidance plan, which includes information about the component to be replaced and installation requirements including: The inspection data is compared with the inspection parameter standard to construct a feature vector including parameter deviation, change trend and associated influence; the AI ​​self-test module analyzes the feature vector using a fuzzy comprehensive evaluation method to generate a fault assessment result; the AI ​​self-test module determines the necessity of component replacement based on the fault assessment result; The AI ​​self-checking module extracts information about the device to be replaced from a preset device database according to the fault assessment result, wherein the information about the device to be replaced includes device model, specification parameters and performance indicators; the AI ​​self-checking module screens the information about the device to be replaced based on electrical characteristic matching, mechanical size adaptability and performance requirement satisfaction, and determines the final device to be replaced; The AI ​​self-check module extracts the installation requirements corresponding to the component to be replaced from a pre-established motor maintenance knowledge graph, and the installation requirements include installation tool requirements, installation step sequence, and tightening torque parameters; the AI ​​self-check module integrates the information of the component to be replaced and the installation requirements to generate a maintenance guidance plan.

[0012] The AI ​​self-check module generates device assembly guidance information according to the device disassembly path, guides the user to complete device replacement and device assembly through voice, and rechecks the device operation status based on the real-time status data to confirm that the fault repair is complete, including: The AI ​​self-check module receives the device disassembly path, establishes a spatial relationship model between components, and converts the device disassembly path into an assembly sequence matrix containing position constraints, assembly order, and interdependencies; the AI ​​self-check module uses a graph theory algorithm to analyze the assembly sequence matrix to generate device assembly guidance information; the AI ​​self-check module decomposes the device assembly guidance information into multiple assembly units, and extracts the mating surface processing method, fastener selection standard, and torque control parameter corresponding to each assembly unit from the assembly knowledge base; The AI ​​self-check module converts the device assembly guidance information into a structured voice command sequence, and guides the user to perform the device replacement operation through voice broadcast; the AI ​​self-check module receives the device information fed back by the user through voice recognition technology, and adjusts the voice command broadcast progress according to the device information fed back by the user; the AI ​​self-check module uses a sensor network to collect assembly status data in real time, and the assembly status data includes relative positions of components, contact stress distribution and dynamic response characteristics; The AI ​​self-check module collects real-time status data of the equipment, which includes electrical characteristics, mechanical properties and thermal characteristics parameters; the AI ​​self-check module performs multi-parameter joint analysis on the real-time status data of the equipment, and uses a status diagnosis algorithm to evaluate equipment performance indicators; the AI ​​self-check module collects performance data under different load conditions, and confirms that the fault repair is completed by comparing the performance data before and after the repair.

[0013] According to a second aspect of the embodiments of the present invention, Provides an AI-based offline maintenance control panel system, including: The first unit is used to collect the fault detection instructions input by the user's voice through the built-in microphone of the device, and the voice recognition module converts the fault detection instructions into text instructions; trigger the AI ​​self-checking module to execute the self-checking process according to the text instructions, and the AI ​​self-checking module reads the real-time status data of the sensors and actuators connected to the control board, and the real-time status data includes motor current, voltage, temperature and speed parameters; the AI ​​self-checking module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library, and the fault diagnosis result includes fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device; The second unit is used for the AI ​​self-checking module to generate a fault repair plan based on the fault diagnosis result, and the fault repair plan includes a repair tool list and repair operation steps; the AI ​​self-checking module asks the user through voice whether he has the repair tools in the repair tool list, and after the user confirms that he has the repair tools, the AI ​​self-checking module determines the equipment disassembly path according to the fault location information, and generates disassembly guidance information according to the equipment disassembly path, and the disassembly guidance information includes a power disconnection indication, protection requirements and disassembly operation sequence; the AI ​​self-checking module broadcasts the disassembly guidance information to the user through voice to guide the user to complete the equipment disassembly; The third unit is used for the AI ​​self-check module to generate a fault inspection instruction based on the fault type, and the fault inspection instruction includes inspection items and inspection parameter standards; the AI ​​self-check module broadcasts the fault inspection instruction by voice to guide the user to complete the inspection items and obtain inspection data; the AI ​​self-check module compares the inspection data with the inspection parameter standards to generate a maintenance guidance plan, and the maintenance guidance plan includes information on components to be replaced and installation requirements; the AI ​​self-check module generates equipment assembly guidance information according to the equipment disassembly path, guides the user to complete component replacement and equipment assembly by voice, and reviews the equipment operation status based on the real-time status data to confirm that the fault repair is completed.

[0014] According to a third aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0015] A fourth aspect of the embodiments of the present invention is: A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0016] The beneficial effects of this application are as follows: Simplify the fault diagnosis process: Through voice interaction and AI self-checking, the fault is automatically diagnosed and the results are broadcasted, without the need for professionals to be present, reducing the complexity and time cost of fault diagnosis. Lower the maintenance threshold: Provide detailed voice guidance, including disassembly, inspection, maintenance and assembly steps, even non-professionals can easily complete maintenance operations, reducing dependence on professionals. Improve maintenance efficiency: The AI ​​self-checking module quickly locates the fault location and generates a maintenance plan, guiding users to quickly complete the maintenance, shortening equipment downtime and improving equipment operation efficiency.

[0017] In the case of unstable network or complete offline, the control panel can also rely on the built-in AI module to quickly respond to faults and achieve preliminary self-diagnosis and repair. Once the fault information flow is detected, the AI ​​module will immediately analyze the fault type and automatically deploy the corresponding maintenance strategy based on the data in the offline library, thereby minimizing downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of a method for offline maintenance of a control panel based on artificial intelligence according to an embodiment of the present invention; Figure 2 The present invention is a schematic diagram of the structure of a system for offline maintenance of a control panel based on artificial intelligence. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0020] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0021] Figure 1 FIG. 1 is a flow chart of a method for offline maintenance of a control panel based on artificial intelligence according to an embodiment of the present invention. Figure 1 As shown, the method includes: S11. The fault detection instruction input by the user's voice is collected through the built-in microphone of the device, and the voice recognition module converts the fault detection instruction into a text instruction; the AI ​​self-checking module is triggered to execute the self-checking process according to the text instruction, and the AI ​​self-checking module reads the real-time status data of the sensors and actuators connected to the control board, and the real-time status data includes motor current, voltage, temperature and speed parameters; the AI ​​self-checking module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library, and the fault diagnosis result includes the fault code, fault type and fault location information, and the fault diagnosis result is broadcast to the user through the built-in speaker of the device; S12. Based on the fault diagnosis result, the AI ​​self-check module generates a fault repair plan, which includes a repair tool list and repair operation steps; the AI ​​self-check module asks the user by voice whether he has the repair tools in the repair tool list, and after the user confirms that he has the repair tools, the AI ​​self-check module determines the equipment disassembly path according to the fault location information, and generates disassembly guidance information according to the equipment disassembly path, and the disassembly guidance information includes power disconnection indication, protection requirements and disassembly operation sequence; the AI ​​self-check module broadcasts the disassembly guidance information to the user by voice to guide the user to complete the equipment disassembly; S13. The AI ​​self-check module generates a fault inspection instruction based on the fault type, and the fault inspection instruction includes inspection items and inspection parameter standards; the AI ​​self-check module broadcasts the fault inspection instruction by voice to guide the user to complete the inspection items and obtain inspection data; the AI ​​self-check module compares the inspection data with the inspection parameter standards to generate a maintenance guidance plan, and the maintenance guidance plan includes information on components to be replaced and installation requirements; the AI ​​self-check module generates equipment assembly guidance information according to the equipment disassembly path, guides the user to complete component replacement and equipment assembly by voice, and reviews the equipment operation status based on the real-time status data to confirm that the fault repair is complete.

[0022] In an optional implementation, the fault detection instruction input by the user's voice is collected through the built-in microphone of the device, and the voice recognition module converts the fault detection instruction into a text instruction; according to the text instruction, the AI ​​self-test module is triggered to execute the self-test process, and the AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board, and the real-time status data includes the motor current, voltage, temperature and speed parameters including: The fault detection instruction input by the user's voice is collected through the built-in microphone of the device, wherein the built-in microphone of the device adopts an electret condenser microphone, and the sensitivity of the electret condenser microphone is -38±2dB, and the frequency response range is 20Hz-20kHz; the fault detection instruction is pre-amplified and analog-to-digital converted to obtain a digital voice signal; Input the digital voice signal into a voice recognition module, the voice recognition module includes a temporal convolutional network structure, the temporal convolutional network structure includes eight convolutional layers and two fully connected layers, the convolution kernel size of each layer is 3, and the fault detection instruction is converted into a text instruction; The AI ​​self-test module is triggered to execute the self-test process according to the text instruction. The AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board through the microcontroller of the 32-bit ARMCortex-M4 core. The real-time status data includes motor current, voltage, temperature and speed parameters.

[0023] The device collects the fault detection instructions input by the user's voice through the built-in microphone, and finally realizes the device self-check through a series of processing steps. The specific implementation method of this method is described in detail below: First, the user issues a fault detection command through the device's built-in microphone, such as "start self-test", "check motor", etc. The built-in microphone uses an electret condenser microphone with a sensitivity of -38±2dB and a frequency response range of 20Hz-20kHz, which can effectively capture voice signals within the human voice frequency range. The microphone converts sound wave signals into analog electrical signals. For example, when the user issues the command "start self-test", the microphone converts the sound wave vibration of the voice command into a corresponding analog electrical signal.

[0024] Next, the collected analog voice signal is preamplified and converted to digital. The preamplifier circuit amplifies the weak analog electrical signal to a suitable amplitude, such as 10 times. The analog-to-digital converter converts the amplified analog signal into a digital signal, for example, using a sampling rate of 16kHz and a quantization accuracy of 16 bits for conversion. Assuming that the amplitude of the analog signal output by the microphone is 1mV, it becomes 10mV after 10 times amplification, and then a series of digital values ​​are obtained after analog-to-digital conversion, such as [1024, 1025, 1026...], which represent the waveform of the voice signal.

[0025] Then, the digital voice signal is input into the voice recognition module for processing. The voice recognition module adopts a sequential convolutional network structure, which includes eight convolutional layers and two fully connected layers, and the convolution kernel size of each layer is 3. The sequential convolutional network performs feature extraction and pattern recognition on the input digital voice signal and converts it into corresponding text instructions. For example, the digital voice signal [1024, 1025, 1026...] is recognized as the text instruction "start self-test". The specific recognition process is that each convolutional layer performs convolution operation on the input data, extracts features at different levels, and then integrates the extracted features through the fully connected layer, and finally outputs the recognition result.

[0026] Finally, according to the text instructions output by the voice recognition module, the AI ​​self-test module is triggered to execute the self-test process. The AI ​​self-test module is implemented based on a microcontroller based on a 32-bit ARM Cortex-M4 core. The microcontroller reads the real-time status data of the sensors and actuators connected to the control board, including motor current, voltage, temperature and speed parameters. For example, after receiving the "start self-test" command, the AI ​​self-test module will read the data of the motor current sensor. Assume that the read current value is 1.5A, the voltage value is 24V, the temperature value is 30℃, and the speed value is 1000rpm. The AI ​​self-test module will compare these data with the preset thresholds to determine whether the device is in normal working condition. For example, if the motor current exceeds 2A, the motor is considered to be overloaded and the alarm mechanism is triggered.

[0027] The solution of this application can: Improve fault detection efficiency: Triggering self-test through voice commands simplifies the operation process and eliminates the need for manual operation, thereby improving the efficiency of fault detection. Users only need to speak the command to start the self-test process, without the need for cumbersome button operations or menu navigation. Enhance user experience: Voice interaction is more natural and convenient, which improves the user experience. Compared with traditional button operations, voice commands are more in line with human natural communication habits, reduce users' learning costs, and are easier to remember and operate. Realize automated fault diagnosis: The AI ​​self-test module can automatically read and analyze device status data to achieve automated fault diagnosis, reduce human intervention, and improve the accuracy and objectivity of diagnosis. There is no need for professionals to perform manual inspections, which reduces the requirements for personnel skills and avoids the possibility of human misjudgment.

[0028] In an optional implementation, the AI ​​self-check module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library, the fault diagnosis result includes the fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device, including: Performing time domain analysis on the real-time status data to obtain the root mean square value, peak factor, margin factor and waveform factor, performing frequency domain analysis to obtain the power spectrum density, inverse spectrum and envelope spectrum, and combining the results of the time domain analysis and the results of the frequency domain analysis to form a feature analysis vector; based on reading the standard feature fingerprints of various motor faults in an offline fault maintenance information library, the offline fault maintenance information library adopts a hierarchical storage structure, including a feature template layer, a fault description layer and a diagnostic rule layer; The AI ​​self-check module performs similarity matching between the feature vector and the standard feature fingerprint, calculates feature correlation using a fuzzy similarity algorithm based on adaptive weights, and generates a preliminary fault matching result; the preliminary fault matching result is input into a pre-trained deep neural network model, the deep neural network model includes a five-layer fully connected structure, uses Dropout to prevent overfitting, and outputs a probability distribution of fault types; The AI ​​self-check module extracts corresponding fault information from the offline fault maintenance information library according to the fault type probability distribution, and generates a fault diagnosis result including fault code, fault type and fault location information; inputs the fault diagnosis result into a speech synthesis model, and the speech synthesis model adaptively adjusts the voice tone and intonation parameters according to the severity of the fault to generate a diagnostic broadcast voice; and plays the diagnostic broadcast voice through the built-in speaker of the device.

[0029] First, collect the real-time status data of the motor during operation. For example, use sensors to collect the vibration signal, current signal, and temperature signal of the motor. In a specific example, the motor runs for 1 minute, and 1,000 vibration data points are collected per second, resulting in a vibration signal sequence consisting of 60,000 data points.

[0030] Then, the collected real-time status data is analyzed in the time domain and frequency domain. In the time domain analysis, the root mean square value, peak factor, margin factor and waveform factor of the signal are calculated. For example, the root mean square value of the above vibration signal sequence is calculated to be 0.8g, the peak factor is 3.5, the margin factor is 1.2, and the waveform factor is 1.1. In the frequency domain analysis, the signal is fast Fourier transformed to obtain the power spectrum density, inverse spectrum and envelope spectrum of the signal. For example, after fast Fourier transforming the above vibration signal, it is found that the frequency components of 50Hz and 100Hz are prominent, indicating that there may be rotor imbalance or stator winding fault. The time domain analysis results and the frequency domain analysis results are combined into a feature analysis vector. For example, the root mean square value, peak factor, margin factor, waveform factor and power spectrum density values ​​at 50Hz and 100Hz calculated above are combined into a six-dimensional feature vector [0.8, 3.5, 1.2, 1.1, 0.5, 0.2].

[0031] Next, the standard feature fingerprints of various motor faults are read from the offline fault maintenance information library. The information library adopts a hierarchical storage structure, including a feature template layer, a fault description layer, and a diagnosis rule layer. The feature template layer stores the standard feature fingerprints corresponding to various faults, the fault description layer stores the text description information of the fault, and the diagnosis rule layer stores the rules for fault diagnosis based on the feature fingerprints.

[0032] The feature analysis vector is matched with the standard feature fingerprint for similarity. The feature correlation is calculated using a fuzzy similarity algorithm based on adaptive weights to generate preliminary fault matching results.

[0033] The preliminary fault matching results are input into the pre-trained deep neural network model. The model contains a five-layer fully connected structure, uses Dropout to prevent overfitting, and outputs the probability distribution of fault types. For example, a similarity of 0.95 is input into the deep neural network model, and the probability of "rotor imbalance" is 0.8, the probability of "stator winding fault" is 0.15, and the probability of "bearing fault" is 0.05.

[0034] According to the probability distribution of fault types, the corresponding fault information is extracted from the offline fault maintenance information library to generate a fault diagnosis result containing fault code, fault type and fault location information. For example, according to the probability distribution, the probability of "rotor imbalance" is the highest, so the corresponding fault information is extracted to generate a fault diagnosis result: the fault code is "E001", the fault type is "rotor imbalance", and the fault location is "motor rotor".

[0035] The fault diagnosis results are input into the speech synthesis model. The model adaptively adjusts the voice tone and intonation parameters according to the severity of the fault and generates a diagnosis announcement voice. For example, according to the severity of the "rotor imbalance" fault, the speech synthesis model generates a more urgent announcement voice: "Warning: Rotor imbalance fault detected, fault code E001, fault location motor rotor".

[0036] Finally, the diagnostic announcement voice is played through the device's built-in speaker to inform the user of the diagnosis results.

[0037] The solution of this application can: Improve fault diagnosis efficiency: AI algorithms are used to automatically analyze data and generate diagnostic results without human intervention, greatly shortening fault diagnosis time. Improve fault diagnosis accuracy: Combine time domain and frequency domain analysis to extract more comprehensive feature information, and use deep learning models for intelligent diagnosis to improve the accuracy of fault identification. Improve user experience: Voice broadcast of fault information allows users to quickly understand the fault situation and take appropriate maintenance measures in a timely manner.

[0038] In an optional implementation, the AI ​​self-check module generates a fault check instruction based on the fault type, the fault check instruction includes an inspection item and an inspection parameter standard; the AI ​​self-check module broadcasts the fault check instruction by voice, guides the user to complete the inspection item and obtains the inspection data, including: The AI ​​self-check module receives fault type information, and searches for corresponding fault diagnosis rules in a pre-established motor fault knowledge base according to the fault type information, wherein the motor fault knowledge base includes inspection items and inspection parameter standards for different fault types; the AI ​​self-check module generates a fault inspection instruction based on the fault diagnosis rule, wherein the fault inspection instruction includes inspection items to be executed and inspection parameter standards corresponding to each inspection item; The AI ​​self-check module inputs the fault check instruction into the speech synthesis model, and the speech synthesis model performs natural language processing on the fault check instruction to generate a speech broadcast text; the AI ​​self-check module generates a speech signal according to the speech broadcast text, and broadcasts the specific inspection item content and inspection parameter standards to the user through the built-in speaker of the device; The AI ​​self-check module receives voice information fed back by the user. When the voice information fed back by the user indicates that repeated broadcasting is required, the AI ​​self-check module re-broadcasts the current inspection item; when the voice information fed back by the user contains inspection data, the AI ​​self-check module compares the inspection data with the inspection parameter standard; the AI ​​self-check module determines whether additional inspection items are required based on the comparison results, and returns to generate new fault inspection instructions when additional inspection items are required, and stores the inspection result data when all inspection items are completed.

[0039] The motor AI self-test method includes the following steps: First, build a motor fault knowledge base. This knowledge base stores the diagnostic rules corresponding to various motor fault types in a structured manner. Each diagnostic rule contains a series of inspection items and the inspection parameter standards corresponding to each inspection item. For example, for the "motor overheating" fault, its diagnostic rules may include inspection items such as "check the insulation resistance of the motor winding" and "measure the motor bearing temperature", and specify the parameter standards for each item, such as the insulation resistance should be greater than 10 megohms and the bearing temperature should be lower than 80 degrees Celsius. The data structure of the knowledge base can be in the form of key-value pairs, for example, using the fault type as the key and the corresponding diagnostic rule as the value.

[0040] Receive fault type information input by the user. For example, the user may input "motor overheat" by voice input, text input, or selecting a preset option.

[0041] According to the fault type information entered by the user, the corresponding diagnostic rules are retrieved in the motor fault knowledge base. For example, when the user enters "motor overheating", the system will search for the corresponding diagnostic rules of "motor overheating" in the knowledge base and extract the corresponding inspection items and inspection parameter standards.

[0042] Generate fault inspection instructions. The fault inspection instructions contain specific inspection items that the user needs to perform and the parameter standards for each item. For example, "Step 1, use an insulation resistance tester to measure the insulation resistance of the motor winding, the standard value is greater than 10 megohms; Step 2, use an infrared thermometer to measure the motor bearing temperature, the standard value is less than 80 degrees Celsius."

[0043] The generated fault inspection instructions are input into the speech synthesis model. The speech synthesis model converts the text instructions into natural language speech, such as "Please use an insulation resistance tester to measure the insulation resistance of the motor winding. The standard value is greater than 10 megohms."

[0044] Voice instructions are broadcast through the device's built-in speaker to guide users in performing inspection operations.

[0045] Receive voice messages from users. Users can provide voice feedback of inspection results, such as "Insulation resistance is 15 megohms" or "Bearing temperature is 70 degrees Celsius". Users can also request a repeat of a command, such as "Please repeat."

[0046] Process user feedback. If the user requests a repeat report, the system will re-report the instructions for the current inspection item. If the user reports inspection data, the system will compare the data with the preset parameter standard. For example, if the user reports "insulation resistance is 15 megohms", the system will compare it with the standard value "greater than 10 megohms".

[0047] Determine whether additional inspection items are needed based on the comparison results. If the inspection results do not meet the standards, or there are subsequent inspection items in the diagnostic rules for the current fault type, a new fault inspection instruction is generated based on the preset rules, and the system returns to step five to continue execution. For example, if the bearing temperature reported by the user exceeds the standard value, the system may add an instruction to "check the operation of the motor cooling fan." If all inspection items have been completed, store the inspection result data, such as saving all inspection items and their corresponding measurement data in a database for subsequent analysis and maintenance.

[0048] The solution of this application can: Improve maintenance efficiency: Through voice guidance and automated comparison, the motor troubleshooting process is simplified, manual operation and judgment are reduced, and thus maintenance efficiency is improved. Lower maintenance threshold: Users do not need to have professional motor knowledge, they only need to follow voice instructions to complete troubleshooting, which lowers the maintenance threshold and facilitates non-professional operations. Improve maintenance accuracy: Systematic inspection process and standardized parameter comparison reduce human errors, improve the accuracy of fault diagnosis, and avoid unnecessary maintenance costs.

[0049] In an optional implementation, the AI ​​self-check module compares the inspection data with the inspection parameter standard to generate a maintenance guidance plan, wherein the maintenance guidance plan includes information about the component to be replaced and installation requirements including: The inspection data is compared with the inspection parameter standard to construct a feature vector including parameter deviation, change trend and associated influence; the AI ​​self-test module analyzes the feature vector using a fuzzy comprehensive evaluation method to generate a fault assessment result; the AI ​​self-test module determines the necessity of component replacement based on the fault assessment result; The AI ​​self-checking module extracts information about the device to be replaced from a preset device database according to the fault assessment result, wherein the information about the device to be replaced includes device model, specification parameters and performance indicators; the AI ​​self-checking module screens the information about the device to be replaced based on electrical characteristic matching, mechanical size adaptability and performance requirement satisfaction, and determines the final device to be replaced; The AI ​​self-check module extracts the installation requirements corresponding to the component to be replaced from a pre-established motor maintenance knowledge graph, and the installation requirements include installation tool requirements, installation step sequence, and tightening torque parameters; the AI ​​self-check module integrates the information of the component to be replaced and the installation requirements to generate a maintenance guidance plan.

[0050] The motor maintenance guidance scheme generation method of the AI ​​self-check module is specifically implemented as follows: First, data collection and preprocessing. The AI ​​self-test module collects motor operation data in real time from data sources such as sensor networks and monitoring systems, such as current, voltage, temperature, vibration, etc. The collected raw data may have problems such as noise and missing values, so preprocessing is required. The preprocessing steps include data cleaning, noise filtering, missing value filling, etc. For example, Kalman filtering is used to remove noise and the missing values ​​are filled with the average value of historical data. Taking current data as an example, suppose that a section of current data collected is [10.2, 10.1, 9.8, 10.3, -1, 10.0, 9.9], where "-1" is an outlier and needs to be processed. Using linear interpolation, replace "-1" with the average value of the two previous and next data (10.3+10.0) / 2=10.15.

[0051] Next, the feature vector is constructed. The preprocessed inspection data is compared with the preset inspection parameter standards, and the parameter deviation, change trend and associated influence are calculated to construct the feature vector. Parameter deviation refers to the difference between the actual value and the standard value or the difference percentage. The change trend is expressed by calculating the rate of change or slope of the parameter over a period of time. The associated influence analyzes the correlation between different parameters, such as the positive correlation between current and temperature. Assuming that the rated current of the motor is 10A and the actual current is 10.5A, the current deviation is (10.5-10) / 10=0.05. Assuming that the motor temperatures in the past 5 minutes were 40℃, 42℃, 44℃, 46℃, and 48℃, respectively, the temperature change trend is (48-40) / 5=1.6℃ / minute.

[0052] Then, fault assessment. The AI ​​self-check module uses a fuzzy comprehensive evaluation method to analyze the feature vector and generate a fault assessment result. Each parameter indicator in the feature vector is fuzzified, for example, the deviation, change trend and associated impact degree are divided into three levels: "mild", "medium" and "serious". Then, a fuzzy rule base is established based on expert experience or historical data. For example, "if the deviation is serious and the change trend is rapidly rising, the possibility of fault is high". Finally, the fuzzy reasoning algorithm is used to calculate the fault assessment result. For example, the confidence level of the output fault possibility being "high" is 0.8.

[0053] After that, the device replacement is judged and selected. The AI ​​self-test module determines the necessity of device replacement based on the fault assessment results. For example, if the confidence level of the fault possibility exceeds the preset threshold (for example, 0.7), it is determined that the device needs to be replaced. According to the fault assessment results, the information of the device to be replaced is extracted from the preset device database, including the device model, specification parameters and performance indicators. For example, it is determined that the motor bearing needs to be replaced based on the type of fault. Then, the information of the device to be replaced is screened based on the electrical characteristics matching, mechanical size adaptability and performance requirements satisfaction to determine the final device to be replaced. For example, select a bearing model with the same size and similar performance as the original motor bearing.

[0054] Finally, a maintenance guidance plan is generated. The AI ​​self-test module extracts the installation requirements corresponding to the components to be replaced from the pre-established motor maintenance knowledge graph, including installation tool requirements, installation step sequence, and tightening torque parameters. For example, replacing a bearing requires the use of tools such as a bearing remover and a torque wrench, following specific steps, and tightening the bolts to the specified torque value. The information on the components to be replaced and the installation requirements are integrated to generate a maintenance guidance plan, which is presented to maintenance personnel in the form of text, charts, or videos. For example, a maintenance guidance document is generated that includes the model of the bearing to be replaced, an illustration of the installation steps, and a tightening torque table.

[0055] The solution of this application can: Improve maintenance efficiency: The AI ​​self-check module can automatically generate maintenance guidance plans, reducing the time for manual data search and plan formulation, thereby improving maintenance efficiency. Reduce maintenance costs: Through accurate fault assessment and component selection, unnecessary component replacement is avoided, reducing maintenance costs. Improve maintenance quality: Standardized maintenance guidance plans and detailed installation requirements help improve maintenance quality and reduce human errors.

[0056] In an optional implementation, the AI ​​self-check module generates device assembly guidance information according to the device disassembly path, guides the user to complete device replacement and device assembly through voice, and rechecks the device operation status based on the real-time status data to confirm that the fault repair is complete, including: The AI ​​self-check module receives the device disassembly path, establishes a spatial relationship model between components, and converts the device disassembly path into an assembly sequence matrix containing position constraints, assembly order, and interdependencies; the AI ​​self-check module uses a graph theory algorithm to analyze the assembly sequence matrix to generate device assembly guidance information; the AI ​​self-check module decomposes the device assembly guidance information into multiple assembly units, and extracts the mating surface processing method, fastener selection standard, and torque control parameter corresponding to each assembly unit from the assembly knowledge base; The AI ​​self-check module converts the device assembly guidance information into a structured voice command sequence, and guides the user to perform the device replacement operation through voice broadcast; the AI ​​self-check module receives the device information fed back by the user through voice recognition technology, and adjusts the voice command broadcast progress according to the device information fed back by the user; the AI ​​self-check module uses a sensor network to collect assembly status data in real time, and the assembly status data includes relative positions of components, contact stress distribution and dynamic response characteristics; The AI ​​self-check module collects real-time status data of the equipment, which includes electrical characteristics, mechanical properties and thermal characteristics parameters; the AI ​​self-check module performs multi-parameter joint analysis on the real-time status data of the equipment, and uses a status diagnosis algorithm to evaluate equipment performance indicators; the AI ​​self-check module collects performance data under different load conditions, and confirms that the fault repair is completed by comparing the performance data before and after the repair.

[0057] Specific implementation methods of AI self-checking module to assist equipment assembly and troubleshooting; First, the AI ​​self-test module obtains the disassembly path of the device to be repaired. For example, the disassembly path of a certain smartphone is: remove the SIM card tray, remove the back cover, unscrew the fixing screws, disconnect the battery connector, remove the battery, and disassemble the motherboard. This path is usually provided by the device manufacturer or obtained by disassembling and analyzing the device.

[0058] Then, the AI ​​self-check module establishes a spatial relationship model between components based on the disassembly path. Taking a smartphone as an example, it will record the position, connection method and assembly order between the back cover and the body, the battery and the motherboard, and the screws and the corresponding components. This information can be represented by a tree structure or a graph structure, with nodes representing components and edges representing the connection relationship between components. For example, the "back cover" node is connected to the "body" node, and the "battery" node is connected to the "motherboard" node.

[0059] Next, the AI ​​self-check module converts the disassembly path into an assembly order matrix. This matrix reflects the order and dependencies of component assembly. For example, the motherboard must be installed before the battery can be connected; all screws must be fastened before the back cover can be installed. Taking a smartphone as an example, the assembly order matrix can be represented as a table, with rows and columns representing components, and cell values ​​representing the order and dependencies of assembly. For example, the cell value of the "Battery" column in the "Motherboard" row is "Before", indicating that the motherboard must be installed before the battery is connected.

[0060] Subsequently, the AI ​​self-check module uses graph theory algorithms to analyze the assembly sequence matrix and generate equipment assembly guidance information. It breaks down the assembly process into multiple assembly units, such as "installing the motherboard", "connecting the battery", and "fixing the screws". For each assembly unit, the AI ​​self-check module extracts the corresponding mating surface treatment method, fastener selection criteria, and torque control parameters from the assembly knowledge base. For example, the mating surface treatment method of the "installing the motherboard" unit is "applying thermal grease", and the torque control parameter of the "fixing screws" unit is "0.5 Nm".

[0061] After that, the AI ​​self-test module converts the device assembly guidance information into a structured voice command sequence, and guides the user to perform device replacement operations through voice broadcast. For example, it will broadcast "Please install the motherboard to the body frame first, and make sure all the clips are aligned", and then broadcast "Please apply an appropriate amount of thermal conductive silicone grease to the motherboard processor position." At the same time, the AI ​​self-test module receives device information from the user through voice recognition technology. For example, if the user says "the battery is connected", the AI ​​self-test module will adjust the voice command broadcast progress based on this information and enter the next assembly step.

[0062] During the assembly process, the AI ​​self-inspection module uses a sensor network to collect assembly status data in real time, such as the relative position of components, contact stress distribution, and dynamic response characteristics. For example, the tightness of screws is detected by pressure sensors, and the position and alignment of components are identified by image sensors.

[0063] After the device is assembled, the AI ​​self-test module collects real-time status data of the device, including electrical characteristics, mechanical properties and thermal characteristics. For example, it measures battery voltage, screen brightness, CPU temperature, etc. Then, the AI ​​self-test module performs multi-parameter joint analysis on these data and uses a status diagnosis algorithm to evaluate device performance indicators. For example, it determines whether the device is working properly based on changes in CPU temperature and battery voltage.

[0064] Finally, the AI ​​self-check module collects performance data under different load conditions, such as running games and playing videos, and confirms that the fault has been repaired by comparing the performance data before and after the repair. For example, if the frame rate and power consumption of the device when running games after the repair are basically the same as before the repair, the fault is considered to have been repaired.

[0065] The solution of this application can: Improve maintenance efficiency: Voice guidance and real-time status monitoring simplify the assembly process, reduce human errors, and thus shorten maintenance time. Reduce maintenance costs: Standardized assembly processes and precise torque control reduce the risk of component damage and maintenance costs. Improve maintenance quality: Multi-parameter joint analysis and status diagnosis algorithms can more accurately evaluate equipment performance and ensure maintenance quality.

[0066] User voice input and command parsing stage; The user first clearly issues a fault detection command, such as "detect motor fault", through the device's built-in microphone or other voice input device. This voice signal is then captured and processed by the voice recognition module. This module uses advanced voice recognition technology to accurately parse the user's voice input and convert it into text instructions that can be understood by the system. Once the command is successfully parsed, the system will automatically trigger the self-test process and send the corresponding self-test command to the AI ​​self-test module.

[0067] AI self-check module operation stage; After receiving the self-test command, the AI ​​self-test module immediately starts to read the real-time status data of all sensors and actuators connected to the control board. These data include but are not limited to key parameters such as motor current, voltage, temperature, and speed. By comprehensively analyzing these data, the AI ​​self-test module can preliminarily determine the operating status of the motor and try to locate possible fault points.

[0068] The AI ​​self-check module will then further analyze and determine the current fault code based on the information flow fed back by the sensor. During this process, the system will refer to a large offline fault maintenance information library, which contains various possible fault codes and their corresponding fault causes and solutions. Once a matching fault code is found, the system will broadcast the possible cause of the fault to the user through the device's built-in speaker or other audio output device, such as "Your actuator motor may have a power problem or connection failure, please pay attention to check."

[0069] User guidance and troubleshooting phase; While announcing the cause of the fault, the system will further guide the user to conduct detailed troubleshooting. This includes reminding the user to check whether the power cord is tightly connected, whether it is damaged or aged, and suggesting that the user try to restart the device to eliminate temporary faults. Before guiding the user to perform these operations, the system will first ask the user through voice whether they have the corresponding skills and tools, and emphasize the importance of safe operation to avoid accidents during the user's operation.

[0070] Disassembly tool preparation and disassembly steps guidance stage; If the user decides to conduct more in-depth troubleshooting or repair work, the system will guide the user step by step through voice to prepare the necessary disassembly tools, such as screwdrivers, pliers, wrenches, etc. At the same time, the system will remind the user to ensure the safety of the disassembly environment (such as turning off the power, wearing protective gloves, etc.) and cleanliness (such as cleaning the work surface, preparing a waste bin, etc.).

[0071] During the disassembly step guidance stage, the system will explain the disassembly method and possible problems in detail through voice according to the disassembly sequence and precautions of the lower computer. For example, "Please remove the screws at the bottom of the device first, and be careful not to damage the components or connecting wires on the circuit board; during the disassembly process, please be sure to handle it with care to avoid secondary damage to the device." Inspection and repair phase; After disassembly, the system will guide the user to perform fault checks using sensors and actuators, including observing whether the circuit board is burnt or short-circuited, measuring the resistance of the motor coil, etc. Based on the inspection results, the system will provide specific repair suggestions or guide the user to perform further repair operations, such as replacing damaged capacitors, resistors, or motors.

[0072] Reassembly and testing phase; After completing the inspection and repair, the system will guide the user through voice to reassemble the device in the reverse order of disassembly. During the assembly process, the system will remind the user to pay attention to the installation position and fixing method of the components to ensure that each component can be installed correctly and firmly in place. After the assembly is completed, the system will guide the user to test the equipment, including checking whether the motor can start and run normally, observing whether there are abnormal sounds or vibrations, etc. Through this series of tests, the user can confirm whether the fault has been successfully resolved.

[0073] In the case of unstable network or complete offline, the control panel can also rely on the built-in AI module to quickly respond to faults and achieve preliminary self-diagnosis and repair. Once the fault information flow is detected, the AI ​​module will immediately analyze the fault type and automatically deploy the corresponding maintenance strategy based on the data in the offline library, thereby minimizing downtime.

[0074] Figure 2 FIG. 1 is a schematic diagram of the structure of a system for offline maintenance control panel based on artificial intelligence according to an embodiment of the present invention. Figure 2 As shown, the system comprises: The first unit is used to collect the fault detection instructions input by the user's voice through the built-in microphone of the device, and the voice recognition module converts the fault detection instructions into text instructions; trigger the AI ​​self-checking module to execute the self-checking process according to the text instructions, and the AI ​​self-checking module reads the real-time status data of the sensors and actuators connected to the control board, and the real-time status data includes motor current, voltage, temperature and speed parameters; the AI ​​self-checking module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library, and the fault diagnosis result includes fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device; The second unit is used for the AI ​​self-checking module to generate a fault repair plan based on the fault diagnosis result, and the fault repair plan includes a repair tool list and repair operation steps; the AI ​​self-checking module asks the user through voice whether he has the repair tools in the repair tool list, and after the user confirms that he has the repair tools, the AI ​​self-checking module determines the equipment disassembly path according to the fault location information, and generates disassembly guidance information according to the equipment disassembly path, and the disassembly guidance information includes a power disconnection indication, protection requirements and disassembly operation sequence; the AI ​​self-checking module broadcasts the disassembly guidance information to the user through voice to guide the user to complete the equipment disassembly; The third unit is used for the AI ​​self-check module to generate a fault inspection instruction based on the fault type, and the fault inspection instruction includes inspection items and inspection parameter standards; the AI ​​self-check module broadcasts the fault inspection instruction by voice to guide the user to complete the inspection items and obtain inspection data; the AI ​​self-check module compares the inspection data with the inspection parameter standards to generate a maintenance guidance plan, and the maintenance guidance plan includes information on components to be replaced and installation requirements; the AI ​​self-check module generates equipment assembly guidance information according to the equipment disassembly path, guides the user to complete component replacement and equipment assembly by voice, and reviews the equipment operation status based on the real-time status data to confirm that the fault repair is completed.

[0075] According to a third aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0076] A fourth aspect of the embodiments of the present invention is: A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0077] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for offline maintenance of control panels based on artificial intelligence, characterized in that: include: The fault detection instructions input by the user's voice are collected through the built-in microphone of the device, and the voice recognition module converts the fault detection instructions into text instructions; the AI ​​self-test module is triggered to execute the self-test process according to the text instructions, and the AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board, and the real-time status data includes motor current, voltage, temperature and speed parameters; the AI ​​self-test module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library, and the fault diagnosis result includes fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device; Based on the fault diagnosis result, the AI ​​self-check module generates a fault repair plan, which includes a repair tool list and repair operation steps; the AI ​​self-check module asks the user by voice whether he has the repair tools in the repair tool list, and after the user confirms that he has the repair tools, the AI ​​self-check module determines the equipment disassembly path according to the fault location information, and generates disassembly guidance information according to the equipment disassembly path, and the disassembly guidance information includes power disconnection indication, protection requirements and disassembly operation sequence; the AI ​​self-check module broadcasts the disassembly guidance information to the user by voice to guide the user to complete the equipment disassembly; The AI ​​self-check module generates a fault inspection instruction based on the fault type, and the fault inspection instruction includes inspection items and inspection parameter standards; the AI ​​self-check module broadcasts the fault inspection instruction by voice to guide the user to complete the inspection items and obtain inspection data; the AI ​​self-check module compares the inspection data with the inspection parameter standards to generate a maintenance guidance plan, and the maintenance guidance plan includes information on components to be replaced and installation requirements; the AI ​​self-check module generates equipment assembly guidance information according to the equipment disassembly path, guides the user to complete component replacement and equipment assembly by voice, and reviews the equipment operation status based on the real-time status data to confirm that the fault repair is complete.

2. The method according to claim 1, characterized in that The fault detection instructions input by the user's voice are collected through the built-in microphone of the device, and the voice recognition module converts the fault detection instructions into text instructions; according to the text instructions, the AI ​​self-test module is triggered to execute the self-test process, and the AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board. The real-time status data includes motor current, voltage, temperature and speed parameters including: The fault detection instruction input by the user's voice is collected through the built-in microphone of the device, wherein the built-in microphone of the device adopts an electret condenser microphone, and the sensitivity of the electret condenser microphone is -38±2dB, and the frequency response range is 20Hz-20kHz; the fault detection instruction is pre-amplified and analog-to-digital converted to obtain a digital voice signal; Input the digital voice signal into a voice recognition module, the voice recognition module includes a temporal convolutional network structure, the temporal convolutional network structure includes eight convolutional layers and two fully connected layers, the convolution kernel size of each layer is 3, and the fault detection instruction is converted into a text instruction; The AI ​​self-test module is triggered to execute the self-test process according to the text instruction. The AI ​​self-test module reads the real-time status data of the sensors and actuators connected to the control board through the microcontroller of the 32-bit ARMCortex-M4 core. The real-time status data includes motor current, voltage, temperature and speed parameters.

3. The method according to claim 1, characterized in that The AI ​​self-checking module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library. The fault diagnosis result includes the fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device, including: Performing time domain analysis on the real-time status data to obtain the root mean square value, peak factor, margin factor and waveform factor, performing frequency domain analysis to obtain the power spectrum density, inverse spectrum and envelope spectrum, and combining the results of the time domain analysis and the results of the frequency domain analysis to form a feature analysis vector; based on reading the standard feature fingerprints of various motor faults in an offline fault maintenance information library, the offline fault maintenance information library adopts a hierarchical storage structure, including a feature template layer, a fault description layer and a diagnostic rule layer; The AI ​​self-check module performs similarity matching between the feature vector and the standard feature fingerprint, calculates feature correlation using a fuzzy similarity algorithm based on adaptive weights, and generates a preliminary fault matching result; the preliminary fault matching result is input into a pre-trained deep neural network model, the deep neural network model includes a five-layer fully connected structure, uses Dropout to prevent overfitting, and outputs a probability distribution of fault types; The AI ​​self-check module extracts corresponding fault information from the offline fault maintenance information library according to the fault type probability distribution, and generates a fault diagnosis result including fault code, fault type and fault location information; inputs the fault diagnosis result into a speech synthesis model, and the speech synthesis model adaptively adjusts the voice tone and intonation parameters according to the severity of the fault to generate a diagnostic broadcast voice; and plays the diagnostic broadcast voice through the built-in speaker of the device.

4. The method according to claim 1, characterized in that: The AI ​​self-check module generates a fault check instruction based on the fault type, wherein the fault check instruction includes an inspection item and an inspection parameter standard; the AI ​​self-check module broadcasts the fault check instruction by voice, guides the user to complete the inspection item and obtains the inspection data, including: The AI ​​self-check module receives fault type information, and searches for corresponding fault diagnosis rules in a pre-established motor fault knowledge base according to the fault type information, wherein the motor fault knowledge base includes inspection items and inspection parameter standards for different fault types; the AI ​​self-check module generates a fault inspection instruction based on the fault diagnosis rule, wherein the fault inspection instruction includes inspection items to be executed and inspection parameter standards corresponding to each inspection item; The AI ​​self-check module inputs the fault check instruction into the speech synthesis model, and the speech synthesis model performs natural language processing on the fault check instruction to generate a speech broadcast text; the AI ​​self-check module generates a speech signal according to the speech broadcast text, and broadcasts the specific inspection item content and inspection parameter standards to the user through the built-in speaker of the device; The AI ​​self-check module receives voice information fed back by the user. When the voice information fed back by the user indicates that repeated broadcasting is required, the AI ​​self-check module re-broadcasts the current inspection item; when the voice information fed back by the user contains inspection data, the AI ​​self-check module compares the inspection data with the inspection parameter standard; the AI ​​self-check module determines whether additional inspection items are required based on the comparison results, and returns to generate new fault inspection instructions when additional inspection items are required, and stores the inspection result data when all inspection items are completed.

5. The method according to claim 1, characterized in that The AI ​​self-check module compares the inspection data with the inspection parameter standard to generate a maintenance guidance plan, which includes information about the component to be replaced and installation requirements including: The inspection data is compared with the inspection parameter standard to construct a feature vector including parameter deviation, change trend and associated influence; the AI ​​self-test module analyzes the feature vector using a fuzzy comprehensive evaluation method to generate a fault assessment result; the AI ​​self-test module determines the necessity of component replacement based on the fault assessment result; The AI ​​self-checking module extracts information about the device to be replaced from a preset device database according to the fault assessment result, wherein the information about the device to be replaced includes device model, specification parameters and performance indicators; the AI ​​self-checking module screens the information about the device to be replaced based on electrical characteristic matching, mechanical size adaptability and performance requirement satisfaction, and determines the final device to be replaced; The AI ​​self-check module extracts the installation requirements corresponding to the component to be replaced from a pre-established motor maintenance knowledge graph, and the installation requirements include installation tool requirements, installation step sequence, and tightening torque parameters; the AI ​​self-check module integrates the information of the component to be replaced and the installation requirements to generate a maintenance guidance plan.

6. The method according to claim 1, characterized in that The AI ​​self-check module generates device assembly guidance information according to the device disassembly path, guides the user to complete device replacement and device assembly through voice, and rechecks the device operation status based on the real-time status data to confirm that the fault repair is complete, including: The AI ​​self-check module receives the device disassembly path, establishes a spatial relationship model between components, and converts the device disassembly path into an assembly sequence matrix containing position constraints, assembly order, and interdependencies; the AI ​​self-check module uses a graph theory algorithm to analyze the assembly sequence matrix to generate device assembly guidance information; the AI ​​self-check module decomposes the device assembly guidance information into multiple assembly units, and extracts the mating surface processing method, fastener selection standard, and torque control parameter corresponding to each assembly unit from the assembly knowledge base; The AI ​​self-check module converts the device assembly guidance information into a structured voice command sequence, and guides the user to perform the device replacement operation through voice broadcast; the AI ​​self-check module receives the device information fed back by the user through voice recognition technology, and adjusts the voice command broadcast progress according to the device information fed back by the user; the AI ​​self-check module uses a sensor network to collect assembly status data in real time, and the assembly status data includes relative positions of components, contact stress distribution and dynamic response characteristics; The AI ​​self-check module collects real-time status data of the equipment, which includes electrical characteristics, mechanical properties and thermal characteristics parameters; the AI ​​self-check module performs multi-parameter joint analysis on the real-time status data of the equipment, and uses a status diagnosis algorithm to evaluate equipment performance indicators; the AI ​​self-check module collects performance data under different load conditions, and confirms that the fault repair is completed by comparing the performance data before and after the repair.

7. A system for offline maintenance control board based on artificial intelligence, used to implement the method described in any one of claims 1 to 6, characterized in that: include: The first unit is used to collect the fault detection instructions input by the user's voice through the built-in microphone of the device, and the voice recognition module converts the fault detection instructions into text instructions; trigger the AI ​​self-checking module to execute the self-checking process according to the text instructions, and the AI ​​self-checking module reads the real-time status data of the sensors and actuators connected to the control board, and the real-time status data includes motor current, voltage, temperature and speed parameters; the AI ​​self-checking module generates a fault diagnosis result based on the real-time status data and compares it with the offline fault maintenance information library, and the fault diagnosis result includes fault code, fault type and fault location information, and broadcasts the fault diagnosis result to the user through the built-in speaker of the device; The second unit is used for the AI ​​self-checking module to generate a fault repair plan based on the fault diagnosis result, and the fault repair plan includes a repair tool list and repair operation steps; the AI ​​self-checking module asks the user through voice whether he has the repair tools in the repair tool list, and after the user confirms that he has the repair tools, the AI ​​self-checking module determines the equipment disassembly path according to the fault location information, and generates disassembly guidance information according to the equipment disassembly path, and the disassembly guidance information includes a power disconnection indication, protection requirements and disassembly operation sequence; the AI ​​self-checking module broadcasts the disassembly guidance information to the user through voice to guide the user to complete the equipment disassembly; The third unit is used for the AI ​​self-check module to generate a fault inspection instruction based on the fault type, and the fault inspection instruction includes inspection items and inspection parameter standards; the AI ​​self-check module broadcasts the fault inspection instruction by voice to guide the user to complete the inspection items and obtain inspection data; the AI ​​self-check module compares the inspection data with the inspection parameter standards to generate a maintenance guidance plan, and the maintenance guidance plan includes information on components to be replaced and installation requirements; the AI ​​self-check module generates equipment assembly guidance information according to the equipment disassembly path, guides the user to complete component replacement and equipment assembly by voice, and reviews the equipment operation status based on the real-time status data to confirm that the fault repair is completed.

8. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images