Train traction motor fault online detection method

Abstract

The invention relates to a train traction motor fault on-line detection method. The method comprises the following steps that traction motor current data are partitioned according to specified unit time and preprocessed; and performing multi-classification fault prediction by using an inter-class difference network structure.

Description

Technical Field

[0001] The invention belongs to the field of intelligent operation and maintenance of rail transit, and relates to an online detection method for train traction motor faults. Background Art

[0002] The train traction motor is the driving power source of high-speed trains, subway trains and electric locomotives. It directly determines the running speed and driving stability of the train or train, and is the core key equipment of high-speed trains, subway trains and electric locomotives. The train traction motor is located in the train bogie, and is subjected to strong impact vibration between the wheel and rail. The working environment is harsh and it is a high-incidence point of the train traction system. Therefore, the online fault diagnosis of the train traction motor is of great significance to improving the operating reliability of high-speed trains, subway trains and electric locomotives.

[0003] At present, most of the fault diagnosis methods for train traction motors do not meet the rapidity requirements of online monitoring, and the false alarm and missed alarm rates of monitoring are far from meeting the requirements of operating units. Summary of the invention

[0004] Based on this, and in view of the deficiencies in the prior art, the present invention proposes a method for diagnosing a train traction motor fault.

[0005] Furthermore, the method comprises the following steps:

[0006] S1, traction motor current data is divided into blocks according to the specified unit time and pre-processed;

[0007] S2. Use the inter-class difference network structure for reasoning and make multi-classification fault prediction on the reasoning results.

[0008] Furthermore, the step S1 includes the following steps:

[0009] S11, assuming that the data acquisition frequency is f, obtain the data from the current time t0 to the previous cycle time t -1 Data set x at a fixed acquisition frequency f i (i=0,1,...,f-1);

[0010] S12. Use the following preprocessing function to obtain the feature vector The formula is as follows:

[0011]

[0012] Among them, min and max are x i The minimum and maximum values ​​in (i=0,1,...,f-1); a is the scale factor.

[0013] Furthermore, the step S2 comprises the following steps:

[0014] S21. Above is an eigenvector of length f×1, and is abbreviated as X1; construct a 1×4 zero kernel C0=(0,0,0,0) and a 1×4 difference kernel C1=(-3,-1,1,3), and calculate the eigenvector X of length f / 4 21 , the formula is as follows:

[0015]

[0016] Among them, C 211 It is a diagonal matrix with diagonal C1 and matrix size is f / 4×f.

[0017] S22, with a step size of 4 and a length of 4, calculate the maximum and minimum sliding differences of X1 to obtain a feature vector X with a length of f / 4 22 , the formula is as follows:

[0018]

[0019] S23, with a step size of 4 and a length of 4, calculate the sliding average of X1 to obtain a feature vector X with a length of f / 4 23 , the formula is as follows:

[0020]

[0021] Among them, mean represents the average value.

[0022] S24, construct a 1×3 convolution kernel C2, and combine the feature vectors of size f / 4×3 X3=(X 21 X 22 X 23 ) performs convolution calculation and uses ReLU function for activation to obtain a feature vector X4 with a length of f / 4. The formula is as follows:

[0023]

[0024] Among them, C 43 is a matrix of size f / 4×1.

[0025] S25. Perform multi-classification prediction on the feature vector X4, as shown in the following formula:

[0026] Y i =WX4+B

[0027] Where W is a matrix of size b×f / 4, B is a matrix of size b×1, and Y i is a vector of size b×1.

[0028] S26. Calculate the fault category y as shown in the following formula:

[0029]

[0030] Wherein, y represents the fault category, ranging from 0 to b-1, and function p represents the normalized exponential function. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a detection flow chart of the present invention. DETAILED DESCRIPTION

[0032] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] In order to facilitate the understanding of those skilled in the art, an embodiment is now used to illustrate the division and preprocessing of the traction motor current data according to a specified unit time: the data acquisition frequency is 256 Hz, and the data set x from the current moment to the previous cycle moment, that is, from the 1st second to the 2nd second, is obtained. i (i=0,1,...,255), obtain the feature vector after preprocessing function The scale factor 255 is selected so that the vector value ranges from 0 to 255.

[0034] In order to facilitate the understanding of those skilled in the art, an embodiment is now used to use the inter-class difference network structure for reasoning and to make multi-classification fault prediction for the reasoning result: X1 is a feature vector with a length of 256×1, and a 1×4 zero core C0=(0,0,0,0) and a 1×4 difference core C1=(-3,-1,1,3) are constructed; according to step S21, a feature vector X with a length of 64 is calculated. 21 , where C 211 The size is 64×256; according to step S22, the feature vector X with a length of 64 is calculated 22 , where C 43 The size is 64×1; according to step S23, the feature vector X with a length of 64 is calculated 23 ; According to step S24, a feature vector X4 with a length of 64 is calculated; according to step S25, a vector Y with a length of 5 is calculated, where the size of W is 5×64 and the size of B is 5×1; according to step S26, the fault category y=1 is calculated, where there are 5 fault categories: broken bar fault, inter-turn short circuit, air gap eccentricity, end ring break, and bearing fault.

Claims

1. An online detection method for train traction motor fault, characterized in that: The following steps are involved: S1, traction motor current data is divided into blocks according to the specified unit time and pre-processed; S2. Use the inter-class difference network structure for multi-classification fault prediction.

2. The detection method according to claim 1, characterized in that: The step S1 comprises the following steps: S11, assuming that the data acquisition frequency is f, obtain the data from the current time t0 to the previous cycle time t -1 Data set x at a fixed acquisition frequency f i (i=0,1,...,f-1); S12. Use the following preprocessing function to obtain the feature vector The formula is as follows: Among them, min and max are x i The minimum and maximum values ​​in (i=0,1,...,f-1); a is the scale factor.

3. The detection method according to claim 1, characterized in that: The step S2 comprises the following steps: S21. Above is an eigenvector of length f×1, and is abbreviated as X1; construct a 1×4 zero kernel C0=(0,0,0,0) and a 1×4 difference kernel C1=(-3,-1,1,3), and calculate the eigenvector X of length f / 4 21 , the formula is as follows: Among them, C 211 It is a diagonal matrix with diagonal C1 and matrix size is f / 4×f. S22, with a step size of 4 and a length of 4, calculate the maximum and minimum sliding differences of X1 to obtain a feature vector X with a length of f / 4 22 , the formula is as follows: S23, with a step size of 4 and a length of 4, calculate the sliding average of X1 to obtain a feature vector X with a length of f / 4 23 , the formula is as follows: Among them, mean represents the average value. S24, construct a 1×3 convolution kernel C2, and combine the feature vectors of size f / 4×3 X3=(X 21 X 22 X 23 ) performs convolution calculation and uses ReLU function for activation to obtain a feature vector X4 with a length of f / 4. The formula is as follows: Among them, C 43 is a matrix of size f / 4×1. S25. Perform multi-classification prediction on the feature vector X4. The formula is as follows: Y i =WX4+B Where W is a matrix of size b×f / 4, B is a matrix of size b×1, and Y i is a vector of size b×1. S26. Calculate the fault category y as shown in the following formula: Wherein, y represents the fault category, ranging from 0 to b-1, and function p represents the normalized exponential function.

Images