Artificial Intelligence-Based Fault Prediction Method, Device and Medium for Traction Drive System

Abstract

The present application discloses a method, device and medium for predicting faults of a traction drive system based on artificial intelligence, the method comprising: obtaining first fault deviation data based on traction machine information, traction rope information and historical fault information; obtaining second fault deviation data based on traction rope information, guide wheel information, anti-ropes wheel information and the historical fault information; obtaining third fault deviation data based on guide wheel information, anti-ropes wheel information and the historical fault information; predicting faults of the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information. The present application aims to accurately predict possible faults of the traction drive system.

Description

Technical Field

[0001] The present application relates to the technical field of traction drive system management, and in particular to a method, device and medium for predicting traction drive system faults based on artificial intelligence. Background Art

[0002] The traction drive system is the core component of the elevator, mainly composed of a traction machine, traction rope, guide wheel, anti-ropes wheel, etc. Its working principle is based on friction. The traction machine drives the traction wheel to rotate, and relies on the friction between the traction rope and the traction wheel to drive the car and the counterweight to move relative to each other, so as to realize the vertical lifting of the car in the shaft, thereby completing the task of transporting people and goods. Various sensors, such as acceleration sensors, temperature sensors, current sensors, etc., are installed on the key components of the traction drive system to collect the operating status data of the components in real time. By analyzing these data, it is determined whether the components are abnormal. For example, by monitoring the current changes of the traction machine, its load condition can be understood; by monitoring the bearing temperature, it can be determined whether there is excessive friction. When the data collected by the sensor exceeds the normal range, the system will issue a warning signal to indicate that there may be a fault.

[0003] However, a single sensor can only provide information on one aspect, and may not be able to fully reflect the characteristics of some complex faults. At the same time, the data sensed by multiple sensors requires the knowledge and experience of domain experts to comprehensively judge the situation in order to make manual predictions of possible future faults. It is inevitable that the prediction accuracy is low, the consideration is not thorough enough, or there is human negligence. Summary of the invention

[0004] The main purpose of this application is to provide a method, device and medium for predicting faults of a traction drive system based on artificial intelligence, aiming to accurately predict possible faults of the traction drive system.

[0005] To achieve the above objectives, the present application provides a traction drive system fault prediction method based on artificial intelligence, the method comprising:

[0006] Acquire the traction machine information in the traction drive system, the traction rope information in the traction drive system, the guide wheel information in the traction drive system, the anti-rope wheel information in the traction drive system, and the historical fault information of the traction drive system;

[0007] Based on the traction machine information, the traction rope information and the historical fault information, first fault deviation data is obtained, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system;

[0008] Based on the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information, second fault deviation data is obtained, wherein the second fault deviation data is used to characterize the degree of correlation between the guide wheel, the traction rope, the anti-ropes wheel and the historical faults of the traction drive system;

[0009] Based on the guide wheel information, the anti-ropes wheel information and the historical fault information, third fault deviation data is obtained, wherein the third fault deviation data is used to characterize the degree of correlation between the guide wheel, the anti-ropes wheel and the historical faults of the traction drive system;

[0010] Based on the first fault deviation data, the second fault deviation data and the third fault deviation data, fault prediction is performed on the traction drive system to obtain target prediction information.

[0011] Specifically, obtaining the first fault deviation data based on the traction machine information, the traction rope information and the historical fault information includes:

[0012] By presetting a first model, the first fault deviation data is obtained according to the traction machine information, the traction rope information and the historical fault information.

[0013] Specifically, the preset first model includes: a first input layer, a first hidden layer, a second hidden layer and a first output layer, and the first output layer includes a neuron;

[0014] The first fault deviation data is obtained by presetting the first model according to the traction machine information, the traction rope information and the historical fault information, including:

[0015] Obtaining a first input vector through the first input layer according to the traction machine information, the traction rope information and the historical fault information;

[0016] Obtaining a first feature vector according to the first input vector through the first hidden layer;

[0017] Obtaining a second eigenvector according to the first eigenvector through the second hidden layer;

[0018] The first fault deviation data is obtained through the first output layer according to the second feature vector.

[0019] Specifically, obtaining the second fault deviation data based on the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information includes:

[0020] By presetting the second model, the second fault deviation data is obtained according to the traction rope information, the guide wheel information, the return rope wheel information and the historical fault information.

[0021] Specifically, the preset second model includes a second input layer, a self-attention layer, a feedforward neural network layer and a second output layer;

[0022] The second fault deviation data is obtained by presetting the second model according to the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information, including:

[0023] Obtaining a second input vector through the second input layer according to the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information;

[0024] Obtaining an attention output matrix according to the second input vector through the self-attention layer;

[0025] Obtaining a feedforward neural network output matrix according to the attention output matrix through the feedforward neural network layer;

[0026] The second fault deviation data is obtained through the second output layer according to the feedforward neural network output matrix.

[0027] Specifically, obtaining the third fault deviation data based on the guide wheel information, the anti-ropes wheel information and the historical fault information includes:

[0028] By presetting the third model, the third fault deviation data is obtained according to the guide wheel information, the anti-ropes wheel information and the historical fault information.

[0029] Specifically, the preset third model includes a third input layer, a graph convolution layer, a global pooling layer and a third output layer;

[0030] The third fault deviation data is obtained by presetting the third model according to the guide wheel information, the anti-ropes wheel information and the historical fault information, including:

[0031] Obtaining a characteristic graph structure through the third input layer according to the guide wheel information, the anti-ropes wheel information and the historical fault information;

[0032] Based on the feature graph structure, obtaining a node feature matrix corresponding to the feature graph structure and an adjacency matrix corresponding to the feature graph structure;

[0033] Obtaining a convolved node feature matrix according to the node feature matrix and the adjacency matrix through the graph convolution layer;

[0034] Obtaining a global feature vector according to the convolved node feature matrix through the global pooling layer;

[0035] The third fault deviation data is obtained through the third output layer according to the global feature vector.

[0036] Specifically, the performing fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information includes:

[0037] Preprocessing the first fault deviation data, the second fault deviation data, and the third fault deviation data to obtain preprocessed fault deviation data;

[0038] Dividing the preprocessed fault deviation data into at least one cluster by a preset clustering algorithm;

[0039] Assigning a preset fault location label and a preset fault type label to each cluster to obtain a cluster label dictionary, wherein the cluster label dictionary includes the cluster number, the fault location corresponding to the cluster number, and the fault type corresponding to the cluster number;

[0040] Obtaining a target cluster according to the first fault deviation data, the second fault deviation data, the third fault deviation data and the cluster label dictionary by using a preset k-means clustering model;

[0041] Based on the target cluster, and according to the cluster label dictionary, determining a target fault location corresponding to the target cluster and a target fault type corresponding to the target cluster;

[0042] The target prediction information is determined based on the target fault location and the target fault type.

[0043] To achieve the above objectives, the present application also provides a traction drive system fault prediction device based on artificial intelligence, the device comprising:

[0044] The first unit is used to obtain the traction machine information in the traction drive system, the traction rope information in the traction drive system, the guide wheel information in the traction drive system, the anti-rope wheel information in the traction drive system and the historical fault information of the traction drive system;

[0045] A second unit is used to obtain first fault deviation data based on the traction machine information, the traction rope information and the historical fault information, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system;

[0046] A third unit is used to obtain second fault deviation data based on the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information, wherein the second fault deviation data is used to characterize the correlation degree between the guide wheel, the traction rope, the anti-rope wheel and the historical faults of the traction drive system;

[0047] A fourth unit is used to obtain third fault deviation data based on the guide wheel information, the anti-ropes wheel information and the historical fault information, wherein the third fault deviation data is used to characterize the correlation degree between the guide wheel, the anti-ropes wheel and the historical faults of the traction drive system;

[0048] The fifth unit is used to perform fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information.

[0049] To achieve the above objectives, the present application also provides a medium, wherein the medium stores a plurality of instructions, wherein the instructions are suitable for a processor to load to execute the steps in any one of the methods provided in the present application.

[0050] The present application provides a method, device and medium for predicting faults of a traction drive system based on artificial intelligence, which can first obtain the information of a traction machine in the traction drive system, the information of a traction rope in the traction drive system, the information of a guide wheel in the traction drive system, the information of a return pulley in the traction drive system and the historical fault information of the traction drive system; then, based on the information of the traction machine, the information of the traction rope and the historical fault information, obtain first fault deviation data, wherein the first fault deviation data is used to characterize the degree of correlation between the traction machine and the traction rope and the historical faults of the traction drive system; then, based on the information of the traction rope, the information of the guide wheel, the information of the return pulley and the historical fault information, obtain first fault deviation data. The method comprises the following steps: obtaining a first fault deviation data and a second fault deviation data based on the historical fault information of the guide wheel, the traction rope, the anti-rope pulley and the traction drive system; obtaining a third fault deviation data based on the guide wheel information, the anti-rope pulley information and the historical fault information; and obtaining a third fault deviation data based on the guide wheel information, the anti-rope pulley information and the historical fault information; wherein the third fault deviation data is used to characterize the correlation between the guide wheel, the anti-rope pulley and the traction drive system; and finally, performing fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information, so as to improve the fault prediction accuracy of the traction drive system. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of a process for a method provided in an embodiment of the present application;

[0052] Figure 2 A schematic diagram of the structure of the device provided in the embodiment of the present application;

[0053] Figure 3 A schematic diagram of the structure of a terminal provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0055] Since a single sensor can only provide information on one aspect, it may not be able to fully reflect the characteristics of some complex faults. At the same time, the data sensed by multiple sensors requires the knowledge and experience of domain experts to comprehensively judge the situation in order to manually predict possible future faults. It is inevitable that the prediction accuracy is low, the consideration is not thorough enough, or there is human negligence.

[0056] Therefore, the embodiments of the present application provide a traction drive system fault prediction method, device and medium based on artificial intelligence to solve practical technical problems.

[0057] In some embodiments, the device may be specifically integrated into an electronic device, which may be a terminal, a server or other device.

[0058] In some embodiments, the server may also be implemented in the form of a terminal.

[0059] Among them, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0060] The terminal may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal and the server may be directly or indirectly connected via wired or wireless communication, and this application does not limit this.

[0061] It should be noted that the serial numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0062] The present application embodiment provides a traction drive system fault prediction method based on artificial intelligence, such as Figure 1 , the specific process of the method can be as follows:

[0063] S110, obtaining the traction machine information, the traction rope information, the guide wheel information, the return rope wheel information and the historical fault information of the traction drive system.

[0064] In some embodiments, the traction machine information may include:

[0065] Voltage: includes the voltage value of the input traction machine motor. Stable voltage is the basis for the normal operation of the motor. Abnormally high or low voltage may affect the performance of the motor and even cause damage to the motor.

[0066] Current: The current of a motor when it is running reflects its load condition. Too much current may mean that the motor is overloaded, has mechanical jamming or other faults; too little current may mean that the motor is not working properly.

[0067] Power: reflects the working capacity of the traction machine. The change of power can reflect the operating efficiency and load status of the motor.

[0068] Speed: The speed of the traction machine directly affects the running speed of the elevator. If the speed is unstable or the deviation from the set value is too large, it may cause the elevator to run unsteadily, affecting the riding experience and even safety.

[0069] Vibration: The vibration amplitude of a normally operating traction machine is small and regular. Abnormal vibration may be caused by motor rotor imbalance, bearing wear, poor gear meshing, etc.

[0070] Temperature: including motor winding temperature, bearing temperature, etc. Excessive temperature may be caused by problems such as poor heat dissipation, overload, insufficient lubrication, etc. Long-term high-temperature operation will accelerate equipment aging and even cause serious accidents such as fire.

[0071] Braking torque: The braking torque must be large enough to ensure that the elevator can brake reliably when it stops. Insufficient braking torque may cause the elevator to slip, posing a safety hazard.

[0072] Braking clearance: Appropriate braking clearance is the key to ensure the normal operation of the brake. If the clearance is too large, the braking response time will be prolonged; if the clearance is too small, the brake may not be fully released, increasing the motor load.

[0073] In some embodiments, the traction rope information may include:

[0074] Tension: The tension of each traction rope should be kept uniform. Uneven tension will cause some traction ropes to be overstressed, accelerate wear, and may even cause the traction rope to break.

[0075] Wear degree: During long-term use, the traction rope will rub against the traction wheel, guide wheel and other parts, causing wear. When the wear is severe, the strength of the traction rope will decrease, affecting its safety.

[0076] Elongation: As the use time increases, the traction rope will elongate to a certain extent. Excessive elongation may cause inaccurate leveling of the elevator and affect normal use.

[0077] Broken wires: Broken wires on the surface of the traction rope are a common problem. Too many broken wires will significantly reduce the strength of the traction rope and need to be replaced in time.

[0078] Rust: If the traction rope is in a humid environment or there are corrosive substances, it may rust. Rust will weaken the strength of the traction rope and shorten its service life.

[0079] In some embodiments, the guide wheel information may include:

[0080] Rotation speed: The rotation speed of the guide wheel should match the running speed of the traction rope. Abnormal rotation speed may increase the friction between the traction rope and the guide wheel, accelerating wear.

[0081] Vibration: Excessive vibration of the guide wheel may be caused by loose installation, bearing damage, unbalanced wheel body, etc. Vibration will affect the smooth operation of the elevator and may also cause abnormal wear of the traction rope.

[0082] Lubrication: Good lubrication can reduce the friction between the guide wheel and the traction rope and reduce wear. Insufficient lubrication will increase friction, generate abnormal noise and heat.

[0083] Wear degree: The wheel groove of the guide wheel will wear out as the use time increases. Excessive wear of the wheel groove will affect the normal operation of the traction rope and may even cause the traction rope to fall out of the groove.

[0084] In some embodiments, the anti-ropes information may include:

[0085] Rotation speed: Similar to the guide wheel, the rotation speed of the anti-rope wheel should also be adapted to the running speed of the traction rope. Abnormal rotation speed may affect the normal operation of the traction system.

[0086] Vibration: The vibration of the return rope pulley reflects its operating status. Abnormal vibration may be caused by installation problems, bearing failure, etc.

[0087] Lubrication condition: Adequate lubrication can ensure smooth rotation of the anti-ropes pulley, reducing wear and noise.

[0088] Wear degree: The wheel groove and bearing of the anti-ropes will wear out. When the wear is serious, it is necessary to repair or replace it in time to ensure the safe operation of the traction drive system.

[0089] In some embodiments, the historical fault information of the traction drive system may include:

[0090] Fault occurrence time: Recording the specific time of each fault occurrence helps to analyze the pattern of fault occurrence, such as whether there are seasonal or periodic characteristics.

[0091] Fault type: Identify the specific type of fault, such as electrical fault (motor fault, control cabinet fault, etc.), mechanical fault (broken traction rope, bearing damage, etc.), so as to carry out targeted prevention and maintenance.

[0092] Fault location: Accurately record the specific location where the fault occurred, such as the traction machine, traction rope, guide wheel, etc., which helps to quickly locate the source of the fault and improve maintenance efficiency.

[0093] Fault handling method: record the handling measures and repair methods taken for each fault, as well as the effect evaluation after the repair. This is of great reference significance for the subsequent handling of similar faults.

[0094] S120. Obtain first fault deviation data based on the traction machine information, the traction rope information and the historical fault information, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system.

[0095] In some embodiments, the traction machine is the power source that drives the traction rope to move. Failure of the traction machine's motor will cause abnormal power output, making it impossible for the traction rope to drive the car and counterweight to operate normally. For example, if the motor winding is short-circuited and cannot operate, the car will stop. If the traction machine's brake fails and cannot brake normally or the braking force is insufficient, the car may slip, which will cause abnormal tension and impact on the traction rope, accelerate the wear of the traction rope, and may even cause the traction rope to break. In addition, if the rope groove of the traction sheave of the traction machine is worn, the friction between the traction rope and the traction sheave will decrease, making it easy for slipping to occur, affecting the normal operation of the elevator and the service life of the traction rope.

[0096] In some embodiments, obtaining the first fault deviation data based on the traction machine information, the traction rope information and the historical fault information includes:

[0097] By presetting a first model, the first fault deviation data is obtained according to the traction machine information, the traction rope information and the historical fault information.

[0098] Specifically, the preset first model includes: a first input layer, a first hidden layer, a second hidden layer and a first output layer, wherein the first output layer includes a neuron; the first fault deviation data is obtained according to the traction machine information, the traction rope information and the historical fault information by the preset first model, including the steps S121 to S124 as shown below:

[0099] S121. Obtain a first input vector through the first input layer according to the traction machine information, the traction rope information and the historical fault information.

[0100] S122. Obtain a first feature vector through the first hidden layer according to the first input vector.

[0101] S123. Obtain a second eigenvector through the second hidden layer according to the first eigenvector.

[0102] S124. Obtain the first fault deviation data according to the second feature vector through the first output layer.

[0103] Continuing with the above embodiment, the first input layer can concatenate the vectors corresponding to the traction machine information, the traction rope information and the historical fault information to obtain the first input vector.

[0104] Continuing with the above embodiment, each neuron in the first hidden layer receives all inputs of the first input layer, performs weighted summation, and then performs nonlinear transformation through an activation function.

[0105] Continuing with the above embodiment, each neuron in the second hidden layer receives the output of the first hidden layer as input.

[0106] Continuing with the above embodiment, the second feature vector is multiplied by the preset weight vector corresponding to the first output layer to obtain a product, and then the product is added with a preset bias to obtain a continuous value, namely the first fault deviation data.

[0107] S130. Based on the traction rope information, the guide wheel information, the return rope pulley information and the historical fault information, obtain second fault deviation data, wherein the second fault deviation data is used to characterize the degree of correlation between the guide wheel, the traction rope, the return rope pulley and the historical faults of the traction drive system.

[0108] In some embodiments, the guide wheel and the anti-rope pulley are used to guide the direction of the traction rope so that the traction rope can run along the designed path. If the bearings of the guide wheel or the anti-rope pulley are worn or poorly lubricated, the wheel body will not rotate flexibly, increasing the friction between the traction rope and the wheel body, causing the wear of the traction rope to increase. If the position of the guide wheel or the anti-rope pulley is offset, the traction rope will be unevenly stressed, further accelerating the wear of the traction rope, and may also cause the traction rope to jump or fall out of the groove during operation, which may cause safety accidents in severe cases. The wear and wire breakage of the traction rope will also have adverse effects on the guide wheel and the anti-rope pulley, such as aggravating the wear of the wheel groove and causing abnormal stress on the wheel body.

[0109] In some embodiments, obtaining the second fault deviation data based on the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information includes:

[0110] By presetting the second model, the second fault deviation data is obtained according to the traction rope information, the guide wheel information, the return rope wheel information and the historical fault information.

[0111] Specifically, the preset second model includes a second input layer, a self-attention layer, a feedforward neural network layer and a second output layer; the second fault deviation data is obtained by preset the second model according to the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information, including the steps S131 to S134 as shown below:

[0112] S131. Obtain a second input vector through the second input layer according to the traction rope information, the guide wheel information, the return rope wheel information and the historical fault information.

[0113] S132. Obtain an attention output matrix according to the second input vector through the self-attention layer.

[0114] S133. Obtain a feedforward neural network output matrix through the feedforward neural network layer according to the attention output matrix.

[0115] S134. Obtain the second fault deviation data through the second output layer according to the feedforward neural network output matrix.

[0116] In some embodiments, the second input layer concatenates the vectors corresponding to the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information to obtain the second input vector.

[0117] Continuing with the above embodiment, the self-attention layer calculates the correlation between elements at different positions in the input vector and assigns different weights to each position, thereby capturing long-distance dependencies in the data. Based on the second input vector, the corresponding query vector, key vector and value vector are calculated, and based on the query vector, key vector and value vector, an attention score matrix is ​​calculated, and a softmax function is applied to the attention score matrix to obtain an attention weight matrix. Based on the attention weight matrix and the value vector, the attention output matrix is ​​calculated.

[0118] Continuing with the above embodiment, the feedforward neural network layer is composed of two linear layers and a nonlinear activation function (such as ReLU) for performing a nonlinear transformation on the output of the self-attention layer. Through the feedforward neural network layer, the attention output matrix is ​​nonlinearly transformed to obtain a feedforward neural network output matrix of the same shape.

[0119] Continuing with the above embodiment, the second output layer may be a simple linear layer that maps the feedforward neural network output matrix to a single value, namely, the second fault deviation data.

[0120] S140. Obtain third fault deviation data based on the guide wheel information, the return rope pulley information and the historical fault information, wherein the third fault deviation data is used to characterize the degree of correlation between the guide wheel, the return rope pulley and the historical faults of the traction drive system.

[0121] In some embodiments, both the guide wheel and the anti-rope wheel play the role of guiding the traction rope in the elevator system, and their working environment and stress conditions are similar. When one of the wheels fails, such as bearing damage, wheel wear, etc., it may cause the tension distribution of the traction rope to change, thereby affecting the normal operation of the other wheel, causing the other wheel to also bear abnormal force, accelerating its wear and failure. For example, if the bearing of the guide wheel is damaged and the rotation is not flexible, the friction of the traction rope at the wheel will increase, resulting in changes in the tension of the traction rope, and then the force borne by the anti-rope wheel will also change, which may cause the failure of the anti-rope wheel.

[0122] In some embodiments, obtaining the third fault deviation data based on the guide wheel information, the anti-ropes wheel information and the historical fault information includes:

[0123] By presetting the third model, the third fault deviation data is obtained according to the guide wheel information, the anti-ropes wheel information and the historical fault information.

[0124] Specifically, the preset third model includes a third input layer, a graph convolution layer, a global pooling layer and a third output layer;

[0125] The third fault deviation data is obtained by presetting the third model according to the guide wheel information, the anti-ropes wheel information and the historical fault information, including the steps S141 to S145 as shown below:

[0126] S141. Obtain a characteristic graph structure through the third input layer according to the guide wheel information, the anti-ropes wheel information and the historical fault information.

[0127] S142. Based on the feature graph structure, obtain a node feature matrix corresponding to the feature graph structure and an adjacency matrix corresponding to the feature graph structure.

[0128] S143. Obtain a convolved node feature matrix according to the node feature matrix and the adjacency matrix through the graph convolution layer.

[0129] S144. Obtain a global feature vector through the global pooling layer according to the convolved node feature matrix.

[0130] S145. Obtain the third fault deviation data through the third output layer according to the global feature vector.

[0131] In some embodiments, the guide wheel information, the anti-ropes wheel information and the historical fault information are respectively represented by an information feature vector, and a feature graph structure is constructed through the third input layer according to the three information feature vectors, and the guide wheel, the anti-ropes wheel and the historical fault are respectively regarded as nodes in the graph, and the edges between the nodes can be defined according to the actual association relationship. For example, if the historical fault is related to certain state changes of the guide wheel, an edge is established between the guide wheel node and the historical fault node. Thus, the adjacency matrix of the feature graph structure represents the connection relationship between the three nodes, and at the same time, the three information feature vectors are combined into a node feature matrix.

[0132] Continuing with the above embodiment, the graph convolution layer updates the feature representation of the node by aggregating the neighbor information of the node. Through the convolution operation of the graph convolution layer, a convolved node feature matrix is ​​obtained.

[0133] Continuing with the above embodiment, the global pooling layer aggregates the features of all nodes in the graph to obtain a global feature representation. Average pooling is used here to average the node feature values ​​on each feature dimension. Specifically, an average operation is performed on each column of the convolved node feature matrix to obtain the global feature vector.

[0134] Continuing with the above embodiment, the third output layer may be a fully connected layer, which maps the global feature vector to a single value, namely, the third fault deviation data.

[0135] S150 . Perform fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data, and the third fault deviation data to obtain target prediction information.

[0136] In some embodiments, the performing of fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information includes the following steps S151 to S156:

[0137] S151 . Preprocess the first fault deviation data, the second fault deviation data, and the third fault deviation data to obtain preprocessed fault deviation data.

[0138] In some embodiments, standardization processing may be used for preprocessing, so that the preprocessed fault deviation data is more suitable for subsequent cluster analysis.

[0139] S152: Divide the preprocessed fault deviation data into at least one cluster using a preset clustering algorithm.

[0140] Continuing with the above embodiment, the K-Means algorithm may be used as a preset clustering algorithm, assuming that the number of clusters is determined to be 3 by the elbow rule.

[0141] S153 . Allocate a preset fault location label and a preset fault type label to each cluster to obtain a cluster label dictionary, wherein the cluster label dictionary includes the cluster number, the fault location corresponding to the cluster number, and the fault type corresponding to the cluster number.

[0142] Continuing with the above embodiment, the following preset fault location label and preset fault type label may be allocated to each cluster according to historical fault information, an existing fault dictionary library and expert experience.

[0143] S154 , obtaining a target cluster according to the first fault deviation data, the second fault deviation data, the third fault deviation data and the cluster label dictionary by using a preset k-means clustering model.

[0144] Continuing with the above embodiment, the first fault deviation data, the second fault deviation data, the third fault deviation data and the cluster label dictionary are input into a preset k-means clustering model that has been trained to predict the target cluster.

[0145] S155 . Based on the target cluster and according to the cluster label dictionary, determine a target fault location corresponding to the target cluster and a target fault type corresponding to the target cluster.

[0146] S156. Determine the target prediction information based on the target fault location and the target fault type.

[0147] Continuing with the above embodiment, the fault location and fault type corresponding to the target cluster, namely the target fault location and the target fault type, can be found according to the cluster label dictionary, and the target prediction information can be obtained.

[0148] In summary, the present application provides a traction drive system fault prediction method based on artificial intelligence to accurately predict possible faults in the traction drive system.

[0149] In order to better implement the above method, the embodiment of the present application also provides a traction drive system fault prediction device based on artificial intelligence, which can be integrated in an electronic device, which can be a terminal, a server, etc. Among them, the terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, a personal computer, etc.; the server can be a single server or a server cluster composed of multiple servers.

[0150] For example, in this embodiment, the method of the embodiment of the present application will be described in detail by taking the specific integration of an artificial intelligence-based traction drive system fault prediction device in a terminal as an example.

[0151] For example, Figure 2 As shown, the artificial intelligence-based traction drive system fault prediction device 200 may include a first unit 201, a second unit 202, a third unit 203, a fourth unit 204 and a fifth unit 205, and the device includes:

[0152] The first unit is used to obtain the traction machine information in the traction drive system, the traction rope information in the traction drive system, the guide wheel information in the traction drive system, the anti-rope wheel information in the traction drive system and the historical fault information of the traction drive system;

[0153] A second unit is used to obtain first fault deviation data based on the traction machine information, the traction rope information and the historical fault information, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system;

[0154] A third unit is used to obtain second fault deviation data based on the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information, wherein the second fault deviation data is used to characterize the correlation degree between the guide wheel, the traction rope, the anti-rope wheel and the historical faults of the traction drive system;

[0155] A fourth unit is used to obtain third fault deviation data based on the guide wheel information, the anti-ropes wheel information and the historical fault information, wherein the third fault deviation data is used to characterize the correlation degree between the guide wheel, the anti-ropes wheel and the historical faults of the traction drive system;

[0156] The fifth unit is used to perform fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information.

[0157] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0158] It can be seen from the above that the embodiments of the present application can improve the fault prediction accuracy of the traction drive system.

[0159] The embodiment of the present application also provides an electronic device, which can be a terminal, a server, etc. The terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, a personal computer, etc. The server can be a single server or a server cluster composed of multiple servers, etc.

[0160] In some embodiments, the product processing device can also be integrated into multiple electronic devices. For example, the product processing device can be integrated into multiple servers, and the artificial intelligence-based traction drive system fault prediction method of the present application is implemented by multiple servers.

[0161] In this embodiment, the electronic device of this embodiment is a terminal as an example for detailed description, for example, Figure 3 As shown, it shows a schematic diagram of the structure of the terminal 300 involved in the embodiment of the present application, specifically:

[0162] The terminal 300 may include one or more processors 301 of processing cores, one or more storage media 302, a power supply 303, an input module 304, and a communication module 305. Those skilled in the art will appreciate that Figure 3 The structure of the terminal 300 shown in the figure does not constitute a limitation on the terminal 300, and the terminal 300 may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0163] The processor 301 is the control center of the terminal 300. It uses various interfaces and lines to connect various parts of the entire terminal 300. It executes various functions of the terminal 300 and processes data by running or executing software programs and / or modules stored in the memory 302, and calling data stored in the memory 302, so as to monitor the terminal 300 as a whole. In some embodiments, the processor 301 may include one or more processing cores; in some embodiments, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 301.

[0164] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and data processing by running the software programs and modules stored in the memory 302. The memory 302 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the terminal 300, etc. In addition, the memory 302 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 302 can also include a memory controller to provide the processor 301 with access to the memory 302.

[0165] The terminal 300 also includes a power supply 303 for supplying power to various components. In some embodiments, the power supply 303 can be logically connected to the processor 301 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 303 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0166] The terminal 300 may further include an input module 304, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.

[0167] The terminal 300 may also include a communication module 305. In some embodiments, the communication module 305 may include a wireless module. The terminal 300 may perform short-range wireless transmission through the wireless module of the communication module 305, thereby providing the user with wireless broadband Internet access. For example, the communication module 305 may be used to help the user send and receive emails, browse web pages, and access streaming media.

[0168] Although not shown, the terminal 300 may also include a display unit, etc., which will not be described in detail here. Specifically in this embodiment, the processor 301 in the terminal 300 will load the executable files corresponding to the processes of one or more applications into the memory 302 according to the following instructions, and the processor 301 will run the applications stored in the memory 302, thereby realizing various functions, as follows:

[0169] Acquire the traction machine information in the traction drive system, the traction rope information in the traction drive system, the guide wheel information in the traction drive system, the anti-rope wheel information in the traction drive system, and the historical fault information of the traction drive system;

[0170] Based on the traction machine information, the traction rope information and the historical fault information, first fault deviation data is obtained, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system;

[0171] Based on the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information, second fault deviation data is obtained, wherein the second fault deviation data is used to characterize the degree of correlation between the guide wheel, the traction rope, the anti-ropes wheel and the historical faults of the traction drive system;

[0172] Based on the guide wheel information, the anti-ropes wheel information and the historical fault information, third fault deviation data is obtained, wherein the third fault deviation data is used to characterize the degree of correlation between the guide wheel, the anti-ropes wheel and the historical faults of the traction drive system;

[0173] Based on the first fault deviation data, the second fault deviation data and the third fault deviation data, fault prediction is performed on the traction drive system to obtain target prediction information.

[0174] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0175] It can be seen from the above that the embodiments of the present application can accurately predict possible faults that may occur in the traction drive system.

[0176] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a medium and loaded and executed by a processor.

[0177] To this end, an embodiment of the present application provides a medium in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any one of the artificial intelligence-based traction drive system fault prediction methods provided in the embodiments of the present application. For example, the instructions can execute the following steps:

[0178] Acquire the traction machine information in the traction drive system, the traction rope information in the traction drive system, the guide wheel information in the traction drive system, the anti-rope wheel information in the traction drive system, and the historical fault information of the traction drive system;

[0179] Based on the traction machine information, the traction rope information and the historical fault information, first fault deviation data is obtained, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system;

[0180] Based on the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information, second fault deviation data is obtained, wherein the second fault deviation data is used to characterize the degree of correlation between the guide wheel, the traction rope, the anti-ropes wheel and the historical faults of the traction drive system;

[0181] Based on the guide wheel information, the anti-ropes wheel information and the historical fault information, third fault deviation data is obtained, wherein the third fault deviation data is used to characterize the degree of correlation between the guide wheel, the anti-ropes wheel and the historical faults of the traction drive system;

[0182] Based on the first fault deviation data, the second fault deviation data and the third fault deviation data, fault prediction is performed on the traction drive system to obtain target prediction information.

[0183] The medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0184] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes a computer instruction, the computer instruction is stored in a medium. A processor of a computer device reads the computer instruction from the medium, and the processor executes the computer instruction, so that the computer device executes the method provided in various optional implementations provided in the above embodiments.

[0185] Since the instructions stored in the medium can execute the steps in any one of the artificial intelligence-based traction drive system fault prediction methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the artificial intelligence-based traction drive system fault prediction methods provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0186] The above is a detailed introduction to the artificial intelligence-based traction drive system fault prediction method, device and medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for predicting faults in a traction drive system based on artificial intelligence, characterized in that: The method comprises: Acquire the traction machine information in the traction drive system, the traction rope information in the traction drive system, the guide wheel information in the traction drive system, the anti-rope wheel information in the traction drive system, and the historical fault information of the traction drive system; Based on the traction machine information, the traction rope information and the historical fault information, first fault deviation data is obtained, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system; Based on the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information, second fault deviation data is obtained, wherein the second fault deviation data is used to characterize the degree of correlation between the guide wheel, the traction rope, the anti-ropes wheel and the historical faults of the traction drive system; Based on the guide wheel information, the anti-ropes wheel information and the historical fault information, third fault deviation data is obtained, wherein the third fault deviation data is used to characterize the degree of correlation between the guide wheel, the anti-ropes wheel and the historical faults of the traction drive system; Based on the first fault deviation data, the second fault deviation data and the third fault deviation data, fault prediction is performed on the traction drive system to obtain target prediction information.

2. The method according to claim 1, characterized in that The obtaining of first fault deviation data based on the traction machine information, the traction rope information and the historical fault information includes: By presetting a first model, the first fault deviation data is obtained according to the traction machine information, the traction rope information and the historical fault information.

3. The method according to claim 2, characterized in that The preset first model includes: a first input layer, a first hidden layer, a second hidden layer and a first output layer, wherein the first output layer includes a neuron; The first fault deviation data is obtained by presetting the first model according to the traction machine information, the traction rope information and the historical fault information, including: Obtaining a first input vector through the first input layer according to the traction machine information, the traction rope information and the historical fault information; Obtaining a first feature vector according to the first input vector through the first hidden layer; Obtaining a second eigenvector according to the first eigenvector through the second hidden layer; The first fault deviation data is obtained through the first output layer according to the second feature vector.

4. The method according to claim 1, characterized in that The obtaining of second fault deviation data based on the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information includes: By presetting the second model, the second fault deviation data is obtained according to the traction rope information, the guide wheel information, the return rope wheel information and the historical fault information.

5. The method according to claim 4, characterized in that The preset second model includes a second input layer, a self-attention layer, a feedforward neural network layer and a second output layer; The second fault deviation data is obtained by presetting the second model according to the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information, including: Obtaining a second input vector through the second input layer according to the traction rope information, the guide wheel information, the anti-ropes wheel information and the historical fault information; Obtaining an attention output matrix according to the second input vector through the self-attention layer; Obtaining a feedforward neural network output matrix according to the attention output matrix through the feedforward neural network layer; The second fault deviation data is obtained through the second output layer according to the feedforward neural network output matrix.

6. The method according to claim 1, characterized in that The obtaining of third fault deviation data based on the guide wheel information, the anti-ropes wheel information and the historical fault information includes: By presetting the third model, the third fault deviation data is obtained according to the guide wheel information, the anti-ropes wheel information and the historical fault information.

7. The method according to claim 6, characterized in that The preset third model includes a third input layer, a graph convolution layer, a global pooling layer and a third output layer; The third fault deviation data is obtained by presetting the third model according to the guide wheel information, the anti-ropes wheel information and the historical fault information, including: Obtaining a characteristic graph structure through the third input layer according to the guide wheel information, the anti-ropes wheel information and the historical fault information; Based on the feature graph structure, obtaining a node feature matrix corresponding to the feature graph structure and an adjacency matrix corresponding to the feature graph structure; Obtaining a convolved node feature matrix according to the node feature matrix and the adjacency matrix through the graph convolution layer; Obtaining a global feature vector according to the convolved node feature matrix through the global pooling layer; The third fault deviation data is obtained through the third output layer according to the global feature vector.

8. The method according to claim 1, characterized in that The method of performing fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information includes: Preprocessing the first fault deviation data, the second fault deviation data, and the third fault deviation data to obtain preprocessed fault deviation data; Dividing the preprocessed fault deviation data into at least one cluster by a preset clustering algorithm; Assigning a preset fault location label and a preset fault type label to each cluster to obtain a cluster label dictionary, wherein the cluster label dictionary includes the cluster number, the fault location corresponding to the cluster number, and the fault type corresponding to the cluster number; Obtaining a target cluster according to the first fault deviation data, the second fault deviation data, the third fault deviation data and the cluster label dictionary by using a preset k-means clustering model; Based on the target cluster, and according to the cluster label dictionary, determining a target fault location corresponding to the target cluster and a target fault type corresponding to the target cluster; The target prediction information is determined based on the target fault location and the target fault type.

9. A traction drive system fault prediction device based on artificial intelligence, characterized in that: The device comprises: The first unit is used to obtain the traction machine information in the traction drive system, the traction rope information in the traction drive system, the guide wheel information in the traction drive system, the anti-rope wheel information in the traction drive system and the historical fault information of the traction drive system; A second unit is used to obtain first fault deviation data based on the traction machine information, the traction rope information and the historical fault information, wherein the first fault deviation data is used to characterize the degree of correlation between the historical faults of the traction machine and the traction rope and the traction drive system; A third unit is used to obtain second fault deviation data based on the traction rope information, the guide wheel information, the anti-rope wheel information and the historical fault information, wherein the second fault deviation data is used to characterize the correlation degree between the guide wheel, the traction rope, the anti-rope wheel and the historical faults of the traction drive system; A fourth unit is used to obtain third fault deviation data based on the guide wheel information, the anti-ropes wheel information and the historical fault information, wherein the third fault deviation data is used to characterize the correlation degree between the guide wheel, the anti-ropes wheel and the historical faults of the traction drive system; The fifth unit is used to perform fault prediction on the traction drive system based on the first fault deviation data, the second fault deviation data and the third fault deviation data to obtain target prediction information.

10. A medium, characterized in that The medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the method according to any one of claims 1 to 8.

Images