CN112067632A - A kind of power equipment detection cloud platform and detection method - Google Patents

Abstract

The invention discloses a power equipment defect detection cloud platform and a detection method, wherein the detection cloud platform comprises the following steps: the device comprises an input module, a classification module, a detection module and an output module. Data of the power equipment is input to the cloud platform through the input module; the classification module identifies data input to the cloud platform and classifies the data according to a defect detection method corresponding to the data; the detection module simultaneously detects the data classified by the classification module according to a corresponding defect detection method and generates a corresponding detection result; and the output module receives the detection result and outputs the detection result. Through using above-mentioned cloud platform to detect power equipment's defect, can carry out multiple detection to power equipment simultaneously, detection cycle is short and detect the strong reliability.

Description

A kind of power equipment detection cloud platform and detection method

technical field

The invention relates to the field of equipment detection, in particular to a cloud platform for detection of electric power equipment and a detection method.

Background technique

In recent years, great progress has been made in the research on defect identification and detection technology of power equipment. A variety of defect detection methods for power equipment have been developed at home and abroad, such as X-ray detection method (using X-ray to scan the power equipment, and determine the Defect type and defect location), infrared detection method (using an infrared camera to scan the power equipment, and determining the defect type and defect position according to the scanning results) and voiceprint detection method (using a voiceprint acquisition device to scan the power equipment, according to the scanning results. Determine the defect type and defect location), etc.

Although there are many electrical equipment defect detection technologies, each detection method is independent, that is, if X-ray detection, infrared detection and voiceprint detection are required for the same electrical equipment, the operator needs to use three sets of equipment to detect the same electrical equipment. The equipment is tested separately, and the cycle of comprehensive testing is long. And with the increase of the number of power equipment, the data generated in the process of power equipment defect identification and detection increases geometrically, and it is difficult for the traditional power detection system to process the increasingly huge data.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a power equipment detection cloud platform and a detection method, the cloud platform can perform multiple detections on the same equipment at the same time, and the detection period is short and the operation reliability is high.

In order to achieve the above purpose, the present invention discloses a cloud platform for defect detection of electric power equipment, which includes: an input module, a classification module, a detection module and an output module. The data of the power equipment is input to the cloud platform through the input module; the classification module identifies the data input to the cloud platform and classifies the data according to the defect detection method corresponding to the data; The detection module simultaneously detects the data classified by the classification module according to the corresponding defect detection method and generates corresponding detection results; the output module receives the detection results of the detection module and outputs the detection results.

Preferably, the cloud platform further includes a platform support module, which ensures stable operation of the input module, the classification module, the detection module and the output module.

Preferably, the classification module includes a first classification unit for classifying data manually input by an operator through the input module and a second classification unit for classifying data input by other devices through the input module .

Preferably, the first classification unit divides the data input by the operator into the first X-ray data, the first infrared data and the first voiceprint data; the second classification unit divides the data input by the other devices It is divided into second X-ray data, second infrared data and second voiceprint data.

Preferably, the first sorting unit includes a first storage member that stores sorting rules for use by the first sorting unit; the second sorting unit includes a second storage member, the second sorting unit The storage means stores classification rules for use by the second classification unit.

Preferably, the detection module includes an X-ray detection unit that detects the first X-ray data and the second X-ray data; an infrared detection unit that detects the first X-ray data infrared data and the second infrared data; and a voiceprint detection unit that detects the first voiceprint data and the second voiceprint data.

Preferably, the X-ray detection unit includes a first detection member for performing X-ray fault detection and fault location on the first X-ray data and the second X-ray data; the infrared detection unit includes a A second detection component for infrared fault detection and fault location for the first infrared data and the second infrared data; the voiceprint detection unit includes a detection of the first voiceprint data and the second voiceprint data A third detection component for voiceprint fault detection and fault location.

Preferably, the detection module includes at least two of the X-ray detection units, at least two of the infrared detection units and at least two of the voiceprint detection units.

Preferably, the platform support module distributes the X-ray data input into the detection module to at least one X-ray detection unit according to the configuration and load rate of each X-ray detection unit; the platform support module according to each infrared The configuration and load rate of the detection unit distribute the infrared data input into the detection module to at least one infrared detection unit; the platform support module will input the infrared data into the detection module according to the configuration and load rate of each voiceprint detection unit The voiceprint data is distributed to at least one voiceprint detection unit.

An embodiment of the present invention also provides a method for detecting electrical equipment defects using the above-mentioned cloud platform to detect electrical equipment, which includes the following steps: S1. Input the data of the electrical equipment into the cloud platform through the input module S2, the classification module identifies the data entered into the cloud platform and classifies the data according to the corresponding defect detection method; S3, the detection module simultaneously classifies the data through the classification module according to the corresponding The defect detection method is used to detect and generate corresponding detection results; S4, and the output module receives the detection results and outputs the detection results.

式中，C(S)为所述检测模块中被选中的检测单元的总连接数、W(S)为所述检测模块中被选中的检测单元被分配的权重、C(Sn)为所述检测模块中未被选中的检测单元的总连接数、W(Sn)为所述检测模块中未被选中的检测单元被分配的权重。Preferably, between the steps S2 and S3, the following steps are further included: assign a higher weight to the detection unit with high configuration and low load in the detection module, so that it can process more data; The detection unit with low configuration and high load is assigned a lower weight to allow it to process a small amount of data; and in order to ensure load balance, the formula needs to be satisfied

In the formula, C(S) is the total number of connections of the selected detection units in the detection module, W(S) is the weight assigned to the selected detection units in the detection module, and C(Sn) is the The total number of connections of unselected detection units in the detection module, W(Sn), is the weight assigned to the unselected detection units in the detection module.

Preferably, the step S2 is specifically as follows: if the detection data of the power equipment is manually input by the operator into the input module, use the classification rules stored in the first storage member of the classification module to detect the power equipment classifying the data; if the detection data of the electric equipment is input into the input module by the detection equipment, classify the detection data of the electric equipment according to the classification rules stored in the second storage component of the classification module.

Preferably, the step S3 is specifically: the X-ray detection unit in the detection module detects the first X-ray data manually input by the operator, the second X-ray data input by the detection device, and the infrared detection in the detection module. The unit detects the first infrared data manually input by the operator and the second infrared data input by the detection device and/or the voiceprint detection unit in the detection module detects the first voiceprint data manually input by the operator and the first voiceprint data input by the detection device. Two voiceprint data;

The X-ray detection unit, the infrared detection unit and the voiceprint detection unit respectively analyze the first X-ray data and the second X-ray data, the first infrared data and the first X-ray data through a convolutional neural network. The two infrared data, the first voiceprint data and the second voiceprint data are analyzed and processed to obtain the fault type and fault location, thereby obtaining the detection result.

By adopting the above-mentioned technical scheme, the present invention mainly has the following technical effects:

1. The detection unit performs X-ray detection, infrared detection and voiceprint detection at the same time, which can significantly shorten the cycle of comprehensive detection of power equipment;

2. The cloud platform can classify the data manually input to the power equipment in the cloud platform according to the manual operation of the operator or automatically classify the data input by other devices to the power equipment in the cloud platform. The cloud platform can satisfy automatic operation or manual operation. The working mode of the cloud platform is high, and the reliability of the cloud platform is high;

3. According to the load situation in the detection unit, the data of the power equipment is reasonably allocated, so as to improve the reliability of the cloud platform operation, and avoid the cloud platform from crashing due to too much data to be processed by a certain detection unit.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings used in the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

1 is a schematic structural diagram of a cloud platform for power equipment detection according to an embodiment of the present invention;

2 is a schematic structural diagram of a classification module and a detection module according to an embodiment of the present invention;

FIG. 3 is a flowchart of a detection method according to an embodiment of the present invention.

Detailed ways

In order to make those skilled in the art better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Reference herein to an "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor a separate or alternative embodiment that is mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

Referring to FIG. 1 , this embodiment discloses a power equipment detection cloud platform, which includes an input module, a classification module, a detection module, and an output module. The data of the power equipment is input to the cloud platform through the input module; the classification module identifies the data input to the cloud platform and classifies the data according to the corresponding defect detection method; the detection module also corresponds to the data classified by the classification module The defect detection method detects and generates corresponding detection results; the output module receives the detection results and outputs the detection results.

In order to ensure the stable operation of the cloud platform, the cloud platform in this embodiment further includes a platform support module, and the platform support module ensures stable operation of the input module, the classification module, the detection module and the output module. The platform support module in this embodiment has functions such as service fuse and service monitoring. Among them, since the modules in the cloud platform in this embodiment are deployed in a distributed cluster manner, service fuse refers to when one or more of the input module, the classification module, the detection module and the output module occur. In the event of a failure, the platform support module shuts down the failed module to ensure the normal operation of other modules, that is, to prevent the "avalanche effect"; service monitoring refers to the platform support module monitoring the working status of the input module, classification module, detection module and output module to ensure input The normal operation of modules, classification modules, detection modules, and output modules.

1 and 2, the data input in the cloud platform includes the data manually input by the operator in the cloud platform and the data input by other equipment in the cloud platform (such as infrared images input by inspection robots, video images captured by external cameras, related data returned by the voiceprint sensor, etc.). In order to facilitate the classification module to classify data from different sources, the classification module includes a first classification unit and a second classification unit, wherein the first classification unit classifies the data manually input by the operator through the input module, and the second classification unit classifies other data. Devices are classified by the data entered by the input module. The first classification unit includes a first storage member, the first storage member stores classification rules for the first classification unit, and the first classification unit classifies the data manually input by the operator through the input module according to the stored classification rules, It is divided into the first X-ray data, the first infrared data and the first voiceprint data; similarly, the second classification unit includes a second storage member, and the second storage member stores the identification rules of the second classification unit, and the second classification unit includes the second storage member. The second classification unit classifies the data input by other devices through the input module according to the classification rules stored therein, and divides it into second X-ray data, second infrared data and second voiceprint data. The classification rules stored in the first storage component and the classification rules stored in the second storage component are not limited herein. Preferably, the classification rules stored in the first storage member in this embodiment are selected by the operator, that is, when the operator inputs the data of the power equipment, the data of the power equipment is manually classified; the data stored in the second storage member is The classification rule is to classify according to the device serial number. For example, set the number of the inspection robot, camera and other equipment that collects infrared images to 1, and set the number of the inspection robot, camera and other equipment to collect X-ray images to 2. The number of voiceprint data acquisition devices such as sensors is set to 3, and the classification module divides the data input by the device numbered 1 into the second infrared data according to the data source, and the data input by the device numbered 2 is divided into the second X-ray data and dividing the data input by the device numbered 3 into second voiceprint data.

In order to detect the classified data at the same time, the detection module of the cloud platform includes an X-ray detection unit, an infrared detection unit, and a voiceprint detection unit. The X-ray detection unit detects the first X-ray data and the second X-ray data, the infrared detection unit detects the first infrared data and the second infrared data, and the voiceprint detection unit detects the first voiceprint data and the second voiceprint data. The X-ray detection unit includes a first detection member, and the first detection member performs X-ray fault detection and fault location on the first X-ray data and the second X-ray data; the infrared detection unit includes a second detection member. An infrared data and a second infrared data are used for infrared fault detection and fault location; the voiceprint detection unit includes a third detection member, and the third detection member performs voiceprint fault detection and fault location on the first voiceprint data and the second voiceprint data . The working modes of the first detection component, the second detection component, and the third detection component are not limited here. In this embodiment, the first detection component, the second detection component, and the third detection component all use a convolutional neural network to analyze the data Perform analysis and processing to obtain the fault location and fault type of the equipment.

In order to improve the detection efficiency of the cloud platform and ensure that the cloud platform can process the data of multiple power devices at the same time, the detection module in this embodiment includes at least two X-ray detection units, at least two infrared detection units, and at least two voiceprint detection units. unit. Through the above settings, the detection module can process the data of multiple power devices at the same time, ensuring that when the amount of data input into the cloud platform is large, the cloud platform can quickly obtain the detection result, thereby improving the detection efficiency of the cloud platform.

In order to prevent the cloud platform from malfunctioning or even crashing due to excessive load in an X-ray detection unit, an infrared detection unit and/or a voiceprint detection unit in the cloud platform, the platform support module needs to reasonably classify the data into in each detection unit. The platform support module in this embodiment allocates the X-ray data in the detection module to at least one X-ray detection unit according to the configuration and load rate of each X-ray detection unit, that is, preferentially inputs the X-ray data to the high-configuration and In the X-ray detection unit with a low load rate; the platform support module allocates the infrared data in the detection module to at least one infrared detection unit according to the configuration and load rate of each infrared detection unit, that is, the infrared data is preferentially input to the configuration high. In the infrared detection unit with low load rate; the platform support module allocates the voiceprint data in the detection module to at least one voiceprint detection unit according to the configuration and load rate of each voiceprint detection unit, that is, prioritizes the voiceprint data Input to the voiceprint detection unit with high configuration and low load rate. Through the above settings, the flexibility and data processing capability of the cloud platform are improved.

In this embodiment, the output module outputs the detection result in the form of a report, that is, the output module generates a detection report including the image after frame selection and identification, defect type, defect quantity statistics, etc., and the operator repairs the power equipment according to the inspection report.

In order to better understand the working principle of the power equipment detection cloud platform in this embodiment, the cloud platform in this embodiment can be abstracted into a six-tuple Cloud, and Cloud={Ms, Mb, Input, Fault, Cnn, Kb} . Wherein Ms is the set of classification module, detection module and output module, Ms={Msg, Mai, Mse}, Msg is the entry of the classification module, Mai includes the first classification unit, the second classification unit, the first detection component and the second The detection component, Mse is the output module, and the detection report is generated and output according to the detection result. The triple of Ms indicates that the data enters the classification module for classification and is detected by the detection module to obtain the detection result and form the detection report and output through the output module. Mb is a platform support module, which ensures the normal operation of each module of the cloud platform. The platform support module includes a service registration center, and all detection units in the detection module are registered in the service registration center to form a service list. Specifically, In other words, the service registration center generates registration information including service name, service request address, port number, etc. according to the relevant information of each detection unit, and stores it in the form of a service list. For example {{"ser_name": "Infrared_ser", "ser_address": "192.168.1.1", "ser_prot": "5002"}, ...} where ser_name refers to the service name of the registered detection unit, and ser_address refers to the registered detection unit. The service request address of the detection unit, ser_prot refers to the port number of the registered detection unit. The Msg compares the obtained detection request information with the service list, obtains the address and port of the corresponding detection unit, and inputs the data of the power equipment into the detection unit for detection by means of gateway routing. Input is an input module. The data of the power equipment is input into Ms through the input module. The data in the Input includes the data manually input by the operator into the cloud platform and the data input by the detection device into the cloud platform. Fault is the detection result obtained by Ms, and Fault includes the fault location and defect type of the power equipment. Cnn is a set of convolutional neural networks {Cnnj}, and Cnnj represents the convolutional neural network that analyzes the jth fault detection. Kb is an event rule storage library, which includes a first storage component and a second storage component. The cloud platform classifies the input detection data of electric equipment according to the classification rules stored in Kb. For example, the inspection robot that collects infrared images, The number of the camera and other equipment is set to 1, the number of the inspection robot, camera and other equipment that collects X-ray images is set to 2, and the number of the voiceprint detection device is set to 3, and the request of the corresponding detection unit is obtained through the device number. method, as shown in the table below.

Table 1. How Kb works

Ser_name in the above table is the classification rule, and Req_data is the classification result.

Referring to FIG. 3 , this embodiment also includes a method for detecting power equipment using a cloud platform, which includes the following steps: S1, inputting data of the power equipment into the cloud platform through an input module; S2, classifying the input module for input Identify the data of the cloud platform and classify the data according to the corresponding defect detection method; S3. The detection module simultaneously detects the data classified by the classification module according to the corresponding defect detection method and generates corresponding detection results; S4. Output module Receive the test results and output the test results.

In step S1, inputting the data of the power equipment into the cloud platform through the input module is specifically: the operator manually inputs the data of the power equipment into the Input and/or the detection equipment (inspection robot, external camera, monitoring sensor, etc. is used to detect the power equipment) input the detection data of the power equipment into the Input, and the Input transmits the data to the Ms through a wireless bridge.

In step S2, the classification module identifies the data input to the cloud platform and classifies the data according to the corresponding defect detection method. Specifically, the cloud platform classifies the data input in the Input according to the classification rules stored in Kb, that is, if the power The detection data of the equipment is manually input by the operator into the input module, then the classification rules stored in the first storage member are used to classify the detection data of the electric equipment and/or if the detection data of the electric equipment is input to the input module by the detection equipment In the case, the detection data of the power equipment is classified according to the classification rules stored in the second storage component.

In step S3, the detection module simultaneously detects the data classified by the classification module according to the corresponding defect detection method and generates a corresponding detection result. Specifically, the X-ray detection unit is used to detect the first X-ray data and the second X-ray data; The infrared detection unit detects the first infrared data and the second infrared data and/or uses the voiceprint detection unit to detect the first voiceprint data and the second voiceprint data.

式中，C(S)为所述检测模块中被选中的检测单元的总连接数、W(S)为所述检测模块中被选中的检测单元被分配的权重、C(Sn)为所述检测模块中未被选中的检测单元的总连接数、W(Sn)为所述检测模块中未被选中的检测单元被分配的权重。Since the configuration and/or load rate of each X-ray detection unit, each infrared detection unit, and each voiceprint detection unit in the cloud platform are different, it is necessary to reasonably transmit the detection data of the power equipment to each X-ray In the detection unit, each infrared detection unit, and each voiceprint detection unit, in this embodiment, a load balancing strategy is used to reasonably distribute data, specifically: assign a higher weight to the detection unit with high configuration and low load, and let it process More data; detection units with low configuration and high load are assigned lower weights, allowing them to process small amounts of data. In addition, in order to ensure load balancing, the formula needs to be satisfied

In the formula, C(S) is the total number of connections of the selected detection units in the detection module, W(S) is the weight assigned to the selected detection units in the detection module, and C(Sn) is the The total number of connections of unselected detection units in the detection module, W(Sn), is the weight assigned to the unselected detection units in the detection module.

In step S3, the detection module simultaneously performs corresponding method detection on the data classified by the classification module and generates a corresponding detection result, and further includes: the first detection component, the second detection component and the third detection component detect X-rays through a convolutional neural network. Data, infrared data and voiceprint data are analyzed and processed. More specifically, after Mai receives the classified data, it calls Cnn to analyze the data, Cnn generates device image data, and Cnn deeply analyzes the device image data to obtain the fault type and fault location, thereby obtaining the detection result Fault and the detection result Fault. Input into Mse to generate inspection report. For example, in X-ray data defect recognition, the feature map of equipment image data is extracted through the convolution layer of the convolutional neural network, and the target feature map is output through the pooling layer of the defect recognition model, and then the target feature map is sent to the deep convolutional neural network to obtain The equipment defect type, and finally the frame regression process is used to obtain the precise position of the detection frame, and the judgment result of the power equipment defect identification is obtained.

In step S4, the output module receives the detection result and outputs the detection result to the cloud platform. Specifically: Mse receives the detection result Fault and automatically generates a detection report including the image after frame selection and identification, defect type, defect quantity statistics, etc. The output of the inspection report is for the operator to refer to when repairing or replacing the power equipment.

Finally, it should be noted that the embodiments of the present invention disclose only preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments It should be understood by those of ordinary skill in the art; it is still possible to modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements to some of the technical features; and these modifications or replacements do not make the corresponding technical solutions. The essence of the invention deviates from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims (10)

1. A power equipment defect detection cloud platform is characterized in that, comprising: an input module, through which the data of the power equipment is input to the cloud platform; a classification module, which identifies the data input to the cloud platform and classifies the data according to the defect detection method corresponding to the data; a detection module, which simultaneously detects several pieces of the data classified by the classification module according to a corresponding defect detection method; and An output module, the output module receives the detection result of the detection module and outputs the detection result. 2. cloud platform according to claim 1, is characterized in that: The cloud platform further includes a platform support module, which ensures stable operation of the input module, the classification module, the detection module and the output module. 3. cloud platform according to claim 1 and 2, is characterized in that: The classification module includes a first classification unit for classifying data manually input by an operator through the input module and a second classification unit for classifying data input by the detection device through the input module; The first classification unit includes a first storage component that stores classification rules for use by the first classification unit that divides the data entered by the operator into a first X optical data, first infrared data and first voiceprint data; The second classification unit includes a second storage component that stores classification rules for use by the second classification unit, and the second classification unit divides the data input by the electronic detection device into a second classification unit. X-ray data, second infrared data, and second voiceprint data. 4. cloud platform according to claim 3, is characterized in that: The detection module includes: An X-ray detection unit, the X-ray detection unit detects the first X-ray data and the second X-ray data; an infrared detection unit that detects the first infrared data and the second infrared data; and a voiceprint detection unit, the voiceprint detection unit detects the first voiceprint data and the second voiceprint data; There are at least two X-ray detection units, the infrared detection units and the voiceprint detection units. 5. cloud platform according to claim 4, is characterized in that: The X-ray detection unit includes a first detection component, and the first detection component performs X-ray fault detection and fault location on the first X-ray data and the second X-ray data; The infrared detection unit includes a second detection component, and the second detection component performs infrared fault detection and fault location on the first infrared data and the second infrared data; The voiceprint detection unit includes a third detection component, and the third detection component performs voiceprint fault detection and fault location on the first voiceprint data and the second voiceprint data. 6. cloud platform according to claim 4 or 5, is characterized in that: The platform support module distributes the X-ray data input in the detection module to at least one of the X-ray detection units according to the configuration and load rate of each of the X-ray detection units; The platform support module distributes the infrared data input into the detection module to at least one of the infrared detection units according to the configuration and load rate of each of the infrared detection units; The platform support module distributes the voiceprint data input into the detection module to at least one of the voiceprint detection units according to the configuration and load rate of each of the voiceprint detection units. 7. A method for detecting defects of power equipment based on a cloud platform, characterized in that, comprising the following steps: S1. Input the detection data of the power equipment into the cloud platform through the input module; S2, the classification module identifies the data input to the cloud platform and classifies the data according to the defect detection method corresponding to the data; S3, the detection module simultaneously detects the data classified by the classification module according to the corresponding defect detection method and generates corresponding detection results; and S4. The output module receives the detection result and outputs the detection result. 8 . The method according to claim 7 , further comprising the following steps between the steps S2 and S3 : assigning a higher weight to the detection unit with a high configuration and a low load in the detection module, allowing It processes more data; assigns a lower weight to the detection unit with low configuration and high load in the detection module, so that it can process a small amount of data; and in order to ensure load balance, the formula needs to be satisfied

In the formula, C(S) is the total number of connections of the selected detection units in the detection module, W(S) is the weight assigned to the selected detection units in the detection module, and C(Sn) is the The total number of connections of unselected detection units in the detection module, W(Sn), is the weight assigned to the unselected detection units in the detection module. 9. The method according to claim 7 or 8, wherein the step S2 is specifically: if the detection data of the power equipment is manually input by the operator into the input module, then adopt the classification module's detection data. The classification rules stored in the first storage member classify the detection data of the power equipment; if the detection data of the power equipment is input into the input module by the detection equipment, then according to the classification rules stored in the second storage member of the classification module The classification rule classifies the detection data of the power equipment. 10. The method according to claim 7 or 8, wherein the step S3 is specifically: the X-ray detection unit in the detection module detects the first X-ray data manually input by the operator and the first X-ray data input by the detection device. Second X-ray data, the infrared detection unit in the detection module detects the first infrared data manually input by the operator, and the second infrared data input from the detection device and/or the voiceprint detection unit in the detection module detects the operator The first voiceprint data inputted manually and the second voiceprint data inputted by the detection device; The X-ray detection unit, the infrared detection unit and the voiceprint detection unit respectively analyze the first X-ray data and the second X-ray data, the first infrared data and the first X-ray data through a convolutional neural network. The two infrared data, the first voiceprint data and the second voiceprint data are analyzed and processed to obtain the fault type and fault location, thereby obtaining the detection result.

Images