Fault prediction device and method for mine car electric wheel speed reducer

Abstract

The invention provides a fault prediction device and method for a mine car electric wheel speed reducer, and the device comprises a data measurement module which is disposed on a mine car electric wheel, and a data collection module which is used for preprocessing a rotating speed signal, a vibration signal, and a temperature signal to obtain operation state data. The on-line monitoring module is used for carrying out signal processing on the operation state data and carrying out equipment state monitoring; and the fault diagnosis module is used for calculating parameters of a fault diagnosis model according to the operation state data and carrying out diagnosis on the operation of the equipment according to the fault diagnosis model. According to the fault prediction device and method for the mine car electric wheel speed reducer, the fault condition in equipment operation can be found and prevented in time through comprehensive analysis of vibration data of all detection points, and the residual life of equipment can be predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of electric wheel reducer state diagnosis, and in particular to a fault prediction device and method for an electric wheel reducer of a mine car. Background Art

[0002] As one of the most important assemblies of a mine car, the electric wheel has a high cost and a high cost for major repairs. The electric wheel reducer is currently monitored by offline regular testing, which has a large gap period. If the major repair is not timely, it is easy to cause serious mechanical accidents such as bearing damage and tooth breaking, resulting in large losses. For safety reasons, the major repair cycle is shortened, which increases costs.

[0003] Fault diagnosis is to collect the on-site equipment operation data from the inspection system and the online system, and establish an equipment diagnosis model based on the on-site equipment data, and use technical means such as mechanism-based diagnosis and regression algorithms to analyze the degree of degradation of the production equipment under condition monitoring. It automatically gives the cause of the unit's failure, the hazard of the failure and maintenance suggestions, and finally forms a diagnosis report.

[0004] Application number CN202010662489.4 discloses a method, device and computer program product for generating a travel reducer fault model. The method for generating a travel reducer fault model includes: obtaining vibration signals of the travel reducer under various fault states; decomposing the vibration signals of various fault states using a preset algorithm to obtain corresponding fault characteristic parameters; using the fault characteristic parameters of various fault states and a preset modeling algorithm to establish fault diagnosis models corresponding to various fault states, and generate a fault model library. Summary of the invention

[0005] The present invention aims to overcome the technical problem in the prior art that the electric wheel reducer is monitored by an offline periodic testing method, which makes it difficult to detect equipment failure in time. A fault prediction device and method for the electric wheel reducer of a mine car is provided. Through the comprehensive analysis of the vibration data of each detection point, faults in the operation of the equipment can be discovered and prevented in time, and the remaining life of the equipment can be predicted.

[0006] The present invention provides a fault prediction device for a mine car electric wheel reducer, comprising:

[0007] Data measurement module: arranged on the electric wheel of the mine car, including a speed sensor, a vibration sensor and a temperature sensor, used to measure the speed data, vibration data and ambient temperature data of the electric wheel of the mine car, used to convert the speed data into a speed signal and transmit it to the data acquisition module, used to convert the vibration data into a vibration signal and transmit it to the data acquisition module, used to convert the temperature data into a temperature signal and transmit it to the data acquisition module;

[0008] Data acquisition module: used to receive the speed signal, vibration signal and temperature signal transmitted by the data measurement module, used to pre-process the speed signal, vibration signal and temperature signal to obtain the operation status data, and used to transmit the speed signal, vibration signal, temperature signal and operation status data to the online monitoring module;

[0009] Online monitoring module: used to receive the speed signal, vibration signal, temperature signal and operation status data transmitted by the data acquisition module, to perform signal processing on the operation status data, to perform equipment status monitoring based on the processed operation status data, to display the equipment status monitoring results, and to transmit the signal processed operation status data to the fault diagnosis module;

[0010] Fault diagnosis module: used to receive the operating status data after signal processing transmitted by the online monitoring module, used to process the operating status data, used to calculate the parameters of the fault diagnosis model according to the operating status data, and used to diagnose the operation of the equipment according to the fault diagnosis model stored in the database; the fault diagnosis model is as follows:

[0011]

[0012] Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0013] When the status indication value Z is less than the fault threshold, it is judged to be normal;

[0014] When the status indication value Z is greater than the fault threshold, a warning is issued.

[0015] The fault prediction device for the electric wheel reducer of a mine car described in the present invention is, as a preferred embodiment, the vibration signal includes vibration frequency, amplitude, phase, time domain, frequency domain, and energy spectrum; the time domain characteristics include peak value, variance, mean square value, effective value, skewness and kurtosis, peak index, margin index, waveform index, pulse index, etc., and the frequency domain characteristics include FFT spectrum, power spectrum, envelope spectrum, etc.

[0016] The fault prediction device for the electric wheel reducer of a mine car described in the present invention, as a preferred embodiment, the process of processing the vibration signal by the online monitoring platform module includes the following steps:

[0017] S1. Data classification: The vibration signal is divided into a state data set A0 and a time-sensitive data set B0 according to different data types; the state data set A0 includes vibration frequency data, amplitude data and phase data; the time-sensitive data set B0 includes time domain feature data, frequency domain feature data, time-frequency domain feature data and energy spectrum;

[0018] S2, data temperature effect optimization: assigning temperature effect to state data a0 in state data set A0 to obtain optimized state data set A1, assigning temperature effect to aging data b0 in aging data set B0 to obtain optimized aging data set B1;

[0019] S3, processed operating status data: the status data set A1 and the aging data set B1 constitute the processed operating status data set C.

[0020] The fault prediction device for the electric wheel reducer of a mine car described in the present invention, as a preferred embodiment, the data processing process of the operating status data set C by the fault diagnosis module includes the following steps:

[0021] Feature vector extraction: extract the state feature vector M based on the state data set A1, and extract the time feature vector N based on the time data set B1.

[0022] The fault prediction device for the electric wheel reducer of a mine car described in the present invention, as a preferred embodiment, the parameter calculation of the fault diagnosis model includes the following steps:

[0023] Select the state feature vectors M and time feature vectors N of several fault points, substitute them into the fault diagnosis model, and fit the parameters p, q, and f of the fault diagnosis model. The fault diagnosis model is as follows:

[0024]

[0025] Where: Z is the status indication value, 0≤Z, at the fault point, the status indication value Z is equal to the fault threshold; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t=i, and t is the data collection time.

[0026] The fault prediction device for the electric wheel reducer of a mine car described in the present invention, as a preferred embodiment, diagnoses the operation of the device including the following steps:

[0027] Substitute the state characteristic vector M and the time-effect characteristic vector N into the fault diagnosis model to obtain the state indication value Z:

[0028]

[0029] Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0030] A diagnosis result is generated according to the state indication value Z, and the diagnosis result includes an operating state and a remaining service life prediction value.

[0031] The fault prediction method based on a state predictive system for a mine car electric wheel reducer according to the present invention, as a preferred embodiment, comprises the following steps:

[0032] S1. Setting of fault threshold: comprehensively formulate the operation process fault threshold Y of each electric wheel reducer according to the design data of the equipment, the on-site operation status data, and the usage data of similar equipment in different usage scenarios;

[0033] S2. Establishment of fault diagnosis model: Extract the state feature vector M and time feature vector N of the equipment at the fault point according to the existing fault data, substitute them into the following model, and fit the parameters p, q, f of the fault diagnosis model:

[0034]

[0035] Where: Y is the fault threshold, 0≤Y; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0036] Get the fault diagnosis model:

[0037]

[0038] Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0039] S3, operation data collection: the data measurement module measures the speed data, vibration data and ambient temperature data of the electric wheel of the mining vehicle and converts the speed data, vibration data and temperature data into speed signals, vibration signals and temperature signals respectively;

[0040] S4, pre-processing of running status data: the data acquisition module removes noise from the vibration signal to obtain running status data;

[0041] S5, data classification: the vibration signal is divided into a state data set A0 and a time-effect data set B0 according to different data types;

[0042] S6, data temperature effect optimization: assigning the temperature effect to the state data a0 in the state data set A0 to obtain an optimized state data set A1, and assigning the temperature effect to the aging data b0 in the aging data set B0 to obtain an optimized aging data set B1;

[0043] S7, processed operation status data: the status data set A1 and the timeliness data set B1 constitute the processed operation status data set C;

[0044] S8, feature vector extraction: extract the state feature vector M according to the state data set A1, and extract the time feature vector N according to the time data set B1;

[0045] S9, diagnostic result generation: Substitute the state feature vector M and the time-effect feature vector N into the fault diagnosis model to obtain the state indication value Z:

[0046]

[0047] Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state characteristic vector at time t = i, Ni is the value of the time-effect characteristic vector at time t = i, and t is the data collection time;

[0048] Generate diagnostic results based on the status indication value Z:

[0049] When the status indication value Z is less than the fault threshold, it is judged to be normal;

[0050] When the status indication value Z is greater than the fault threshold, a warning is issued.

[0051] 8. The fault prediction method based on a state predictive system for a mine car electric wheel reducer according to claim 7, characterized in that it also includes the following steps:

[0052] S10. Optimization of fault diagnosis model: When a device fails, the state feature vector M and the time feature vector N of the fault point are extracted and substituted into the fault diagnosis model to optimize the parameters p, q, and f.

[0053] The fault prediction method based on a state predictive system for a mine car electric wheel reducer according to the present invention, as a preferred embodiment, step S6 further comprises the following steps:

[0054] S61, Optimization of temperature effect of state data set:

[0055] A1=f(T,t)×A0

[0056] Where: function f(T, t) is a constant related to temperature T and time t, 1≤f(T, t)≤10;

[0057] The value of the function f(T, t):

[0058] When it is lower than the critical temperature T0, it decreases with the increase of temperature T;

[0059] When the temperature is greater than the critical temperature T0, it increases with the increase of temperature T;

[0060] Increases with time t

[0061] S62, Temperature effect optimization of time-dependent data set:

[0062] B1=g(T)×B0

[0063] Where: function g(T) is a constant with temperature T, 1≤g(T)≤10;

[0064] The value of the function g(T):

[0065] When the temperature is lower than the critical temperature T1, it decreases with the increase of temperature T;

[0066] When the temperature is greater than the critical temperature T1, it increases with the increase of temperature T.

[0067] The fault prediction method based on a state predictive system for a mine car electric wheel reducer described in the present invention is, as a preferred embodiment, the types of faults diagnosed by the fault diagnosis model include: rotor fault, shaft fault, bearing fault, gearbox fault, and auxiliary component fault.

[0068] (1) The present invention collects, processes and extracts feature vectors of the electric wheel reducer status data in real time, sets different fault thresholds according to different fault types, and predicts the equipment status in combination with the fault diagnosis model, so as to timely discover and prevent faults in the operation of the equipment.

[0069] (2) The present invention comprehensively analyzes the vibration data collected from different data temperatures, and can accurately analyze the type and time of the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a composition diagram of a fault prediction device for a mine car motor reducer;

[0071] Figure 2 This is a schematic diagram of the position of the sensor installed on the outer side of the electric wheel of a fault prediction device for a mine car motor reducer;

[0072] Figure 3 This is a schematic diagram of the position of the sensor installed on the inner side of the electric wheel of a fault prediction device for a mine car motor reducer;

[0073] Figure 4 A schematic diagram of the sensor location for the main generator of a fault prediction device for a mine car motor reducer;

[0074] Figure 5 A flow chart of a fault prediction method based on a fault prediction device for a mine car motor reducer;

[0075] Figure 6 This is a schematic diagram of the fault diagnosis results of Example 1;

[0076] Figure 7 This is a schematic diagram of the fault diagnosis diagram of Example 1;

[0077] Figure 8 This is a schematic diagram of the remaining life prediction of Example 1. DETAILED DESCRIPTION

[0078] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0079] Example 1

[0080] like Figure 1 As shown, a fault prediction device for a mine car electric wheel reducer comprises:

[0081] Data measurement module: It is set on the electric wheel of the mine car, including a speed sensor, a vibration sensor and a temperature sensor, which is used to measure the speed data, vibration data and ambient temperature data of the electric wheel of the mine car, and is used to convert the speed data into a speed signal and transmit it to the data acquisition module, to convert the vibration data into a vibration signal and transmit it to the data acquisition module, and to convert the temperature data into a temperature signal and transmit it to the data acquisition module; Figures 2 to 4 As shown, each mine car includes at least 3 collection points, which are respectively set on the outside of the motor, the inside of the motor (on the gearbox) and the main generator; the vibration signal includes vibration frequency, amplitude, phase, time domain, frequency domain, and energy spectrum; the time domain features include peak value, variance, mean square value, effective value, skewness and kurtosis, peak index, margin index, waveform index, pulse index, etc., and the frequency domain features include FFT spectrum, power spectrum, envelope spectrum, etc.

[0082] Data acquisition module: used to receive the speed signal, vibration signal and temperature signal transmitted by the data measurement module, used to pre-process the speed signal, vibration signal and temperature signal to obtain the operation status data, and used to transmit the speed signal, vibration signal, temperature signal and operation status data to the online monitoring module;

[0083] Online monitoring module: used to receive the speed signal, vibration signal, temperature signal and operation status data transmitted by the data acquisition module, used to perform signal processing on the operation status data, used to perform equipment status monitoring based on the processed operation status data, used to display the equipment status monitoring results, and used to transmit the signal-processed operation status data to the fault diagnosis module; the processing process of the vibration signal by the online monitoring platform module includes the following steps:

[0084] S1. Data classification: The vibration signal is divided into a state data set A0 and a time-sensitive data set B0 according to different data types; the state data set A0 includes vibration frequency data, amplitude data and phase data; the time-sensitive data set B0 includes time domain feature data, frequency domain feature data, time-frequency domain feature data and energy spectrum;

[0085] S2, data temperature effect optimization: assigning temperature effect to state data a0 in state data set A0 to obtain optimized state data set A1, assigning temperature effect to aging data b0 in aging data set B0 to obtain optimized aging data set B1;

[0086] S3, processed operation status data: the status data set A1 and the timeliness data set B1 constitute the processed operation status data set C;

[0087] Fault diagnosis module: used to receive the operating status data after signal processing transmitted by the online monitoring module, used to process the operating status data, used to calculate the parameters of the fault diagnosis model according to the operating status data, and used to diagnose the operation of the equipment according to the fault diagnosis model stored in the database; the data processing process of the operating status data set C by the fault diagnosis module includes the following steps:

[0088] Feature vector extraction: extract the state feature vector M based on the state data set A1, and extract the time feature vector N based on the time data set B1.

[0089] The fault diagnosis model is as follows:

[0090]

[0091] Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N iis the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0092] When the status indication value Z is less than the fault threshold, it is judged to be normal;

[0093] When the status indication value Z is greater than the fault threshold, a warning is issued.

[0094] like Figure 5 As shown, during use, this embodiment includes the following steps:

[0095] S1. Setting of fault threshold: comprehensively formulate the operation process fault threshold Y of each electric wheel reducer according to the design data of the equipment, the on-site operation status data, and the usage data of similar equipment in different usage scenarios;

[0096] S2. Establishment of fault diagnosis model: Extract the state feature vector M and time feature vector N of the equipment at the fault point according to the existing fault data, substitute them into the following model, and fit the parameters p, q, f of the fault diagnosis model:

[0097]

[0098] Where: Y is the fault threshold, 0≤Y; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0099] Get the fault diagnosis model:

[0100]

[0101] Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0102] S3, operation data collection: the data measurement module measures the speed data, vibration data and ambient temperature data of the electric wheel of the mining vehicle and converts the speed data, vibration data and temperature data into speed signals, vibration signals and temperature signals respectively;

[0103] S4, pre-processing of running status data: the data acquisition module removes noise from the vibration signal to obtain running status data;

[0104] S5, data classification: the vibration signal is divided into a state data set A0 and a time-effect data set B0 according to different data types;

[0105] S6, data temperature effect optimization: assigning the state data a0 in the state data set A0 with the temperature effect to obtain the optimized state data set A1, assigning the aging data b0 in the aging data set B0 with the temperature effect to obtain the optimized aging data set B1; step S6 further includes the following steps:

[0106] S61, Optimization of temperature effect of state data set:

[0107] A1=f(T,t)×A0

[0108] Where: function f(T, t) is a constant related to temperature T and time t, 1≤f(T, t)≤10;

[0109] The value of the function f(T,t):

[0110] When it is lower than the critical temperature T0, it decreases with the increase of temperature T;

[0111] When the temperature is greater than the critical temperature T0, it increases with the increase of temperature T;

[0112] Increases with the increase of time t;

[0113] The critical temperature of the motor is T0≈30.1℃, and the critical temperature of the main generator is T0≈35.2℃

[0114] S62, Temperature effect optimization of time-dependent data set:

[0115] B1=g(T)×B0

[0116] Where: function g(T) is a constant with temperature T, 1≤g(T)≤10;

[0117] The value of the function g(T):

[0118] When the temperature is lower than the critical temperature T1, it decreases with the increase of temperature T;

[0119] When it is greater than the critical temperature T1, it increases with the increase of temperature T;

[0120] The critical temperature of the motor is T1≈29.1℃, and the critical temperature of the main generator is T1≈33.7℃

[0121] S7, processed operation status data: the status data set A1 and the timeliness data set B1 constitute the processed operation status data set C;

[0122] S8, feature vector extraction: extract the state feature vector M according to the state data set A1, and extract the time feature vector N according to the time data set B1;

[0123] S9, diagnostic result generation: Substitute the state feature vector M and the time-effect feature vector N into the fault diagnosis model to obtain the state indication value Z:

[0124]

[0125] Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time;

[0126] Generate diagnostic results based on the status indication value Z, such as Figures 6 to 8 As shown:

[0127] When the status indication value Z is less than the fault threshold, it is judged to be normal;

[0128] When the status indication value Z is greater than the fault threshold, a warning is issued;

[0129] S10. Optimization of fault diagnosis model: When a device fails, the state feature vector M and the time feature vector N of the fault point are extracted and substituted into the fault diagnosis model to optimize the parameters p, q, and f.

[0130] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A fault prediction device for a mine car electric wheel reducer, characterized in that: include Data measurement module: arranged on the electric wheel of the mine car, including a speed sensor, a vibration sensor and a temperature sensor, used to measure the speed data, vibration data and ambient temperature data of the electric wheel of the mine car, used to convert the speed data into a speed signal and transmit it to the data acquisition module, used to convert the vibration data into a vibration signal and transmit it to the data acquisition module, used to convert the temperature data into a temperature signal and transmit it to the data acquisition module; Data acquisition module: used for receiving the speed signal, the vibration signal and the temperature signal transmitted by the data measurement module, for pre-processing the speed signal, the vibration signal and the temperature signal to obtain the operation status data, and for transmitting the speed signal, the vibration signal, the temperature signal and the operation status data to the online monitoring module; Online monitoring module: used for receiving the speed signal, the vibration signal, the temperature signal and the operating status data transmitted by the data acquisition module, for performing signal processing on the operating status data, for performing equipment status monitoring according to the processed operating status data, for displaying equipment status monitoring results, and for transmitting the operating status data after signal processing to the fault diagnosis module; Fault diagnosis module: used to receive the operating status data after signal processing transmitted by the online monitoring module, used to process the operating status data, used to calculate the parameters of the fault diagnosis model according to the operating status data, and used to diagnose the operation of the equipment according to the fault diagnosis model stored in the database; the fault diagnosis model is as follows: Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time; When the status indication value Z is less than the fault threshold, it is judged to be normal; When the status indication value Z is greater than the fault threshold, a warning is issued.

2. A fault prediction device for a mine car electric wheel reducer according to claim 1, characterized in that: The vibration signal includes vibration frequency, amplitude, phase, time domain, frequency domain, and energy spectrum.

3. A state predictive system for a mine car electric wheel reducer according to claim 2, characterized in that: The processing process of the vibration signal by the online monitoring platform module includes the following steps: S1. Data classification: the vibration signal is divided into a state data set A0 and a time-effect data set B0 according to different data types; S2, data temperature effect optimization: assigning the state data a0 in the state data set A0 with temperature effect to obtain an optimized state data set A1, and assigning the aging data b0 in the aging data set B0 with temperature effect to obtain an optimized aging data set B1; S3, processed operating status data: the status data set A1 and the aging data set B1 constitute the processed operating status data set C.

4. A state predictive system for a mine car electric wheel reducer according to claim 3, characterized in that: The data processing process of the operating status data set C by the fault diagnosis module includes the following steps: Feature vector extraction: extract the state feature vector M according to the state data set A1, and extract the time feature vector N according to the time data set B1.

5. A state predictive system for a mine car electric wheel reducer according to claim 4, characterized in that: The parameter calculation of the fault diagnosis model comprises the following steps: Select the state feature vectors M and time-effect feature vectors N of several fault points, substitute them into the fault diagnosis model, and fit to obtain the parameters p, q, and f of the fault diagnosis model. The fault diagnosis model is as follows: Wherein: Z is the state indication value, 0≤Z, and at the fault point, the state indication value Z is equal to the fault threshold; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t=i, and t is the data collection time.

6. A state predictive system for a mine car electric wheel reducer according to claim 4, characterized in that: The operation of the equipment for diagnosis includes the following steps: Substitute the state feature vector M and the time-effect feature vector N into the fault diagnosis model to obtain the state indication value Z: Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time; A diagnosis result is generated according to the state indication value Z, wherein the diagnosis result includes an operating state and a remaining service life prediction value.

7. The fault prediction method based on a state predictive system for a mine car electric wheel reducer according to any one of claims 5 to 6, characterized in that: The following steps are involved: S1. Setting of fault threshold: comprehensively formulate the fault threshold Y of the operation process of each electric wheel reducer according to the design data of the equipment, the on-site operation status data, and the usage data of similar equipment in different usage scenarios; S2. Establishment of fault diagnosis model: Extract the state feature vector M and time feature vector N of the equipment at the fault point according to the existing fault data, substitute them into the following model, and fit the parameters p, q, f of the fault diagnosis model: Where: Y is the fault threshold, 0≤Y; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time; The fault diagnosis model is obtained: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time; S3, operation data collection: the data measurement module measures the speed data, the vibration data and the temperature data of the surrounding environment of the electric wheel of the mining vehicle and converts the speed data, the vibration data and the temperature data into the speed signal, the vibration signal and the temperature signal respectively; S4, pre-processing of the running status data: the data acquisition module performs denoising on the vibration signal to obtain the running status data; S5, data classification: the vibration signal is divided into a state data set A0 and a time-effect data set B0 according to different data types; S6, data temperature effect optimization: assigning the state data a0 in the state data set A0 with temperature effect to obtain an optimized state data set A1, and assigning the aging data b0 in the aging data set B0 with temperature effect to obtain an optimized aging data set B1; S7, processed operation status data: the status data set A1 and the timeliness data set B1 constitute the processed operation status data set C; S8, feature vector extraction: extracting a state feature vector M according to the state data set A1, and extracting a time-effect feature vector N according to the time-effect data set B1; S9, generating diagnostic results: Substitute the state feature vector M and the time-effect feature vector N into the fault diagnosis model to obtain the state indication value Z: Where: Z is the state indication value, 0≤Z; p, q, f are constants, 0≤p≤1, 0≤q≤1, 0≤f≤1; M i is the value of the state eigenvector at time t = i, N i is the value of the time-dependent characteristic vector at time t = i, where t is the data collection time; Generate a diagnostic result according to the state indication value Z: When the status indication value Z is less than the fault threshold, it is judged to be normal; When the status indication value Z is greater than the fault threshold, a warning is issued.

8. The fault prediction method based on a state predictive system for a mine car electric wheel reducer according to claim 7, characterized in that: The following steps are also included: S10, optimization of the fault diagnosis model: when a device fails, the state feature vector M and the time-effect feature vector N of the fault point are extracted and substituted into the optimization parameters p, q, and f of the fault diagnosis model.

9. The fault prediction method based on a state predictive system for a mine car electric wheel reducer according to claim 7, characterized in that: Step S6 further comprises the following steps: S61, Optimization of temperature effect of state data set: A1=f(T,t)×A0 Where: function f(T, t) is a constant related to temperature T and time t, 1≤f(T, t)≤10; The value of the function f(T,t) is: When it is lower than the critical temperature T0, it decreases with the increase of temperature T; When the temperature is greater than the critical temperature T0, it increases with the increase of temperature T; Increases with time t S62, Optimization of temperature effect of time-dependent data set: B1=g(T)×B0 Where: function g(T) is a constant with temperature T, 1≤g(T)≤10; The value of the function g(T): When the temperature is lower than the critical temperature T1, it decreases with the increase of temperature T; When the temperature is greater than the critical temperature T1, it increases with the increase of temperature T.

10. The fault prediction method based on a state predictive system for a mine car electric wheel reducer according to claim 7, characterized in that: The fault types diagnosed by the fault diagnosis model include: rotor fault, shaft fault, bearing fault, gearbox fault, and auxiliary component fault.

Images