Similarity-based turnout friction current intelligent analysis method and system

Abstract

The invention discloses a turnout friction current intelligent analysis method and system based on similarity, and relates to the technical field of turnout friction current. Comprising the steps that turnout friction current curves are obtained, and the friction current curves comprise a standard friction current curve and a disordered friction current curve; feature points of the standard friction current curve are extracted based on an ORB algorithm, and feature point data matching is carried out on the feature points and the disordered friction current curve; calculating the Hamming distance between the standard friction current curve and the disordered friction current curve according to the feature points; and setting a similarity threshold, determining an adjustment range of the disordered friction current curve according to a relationship between the Hamming distance and the similarity, and giving an adjustment result. According to the method, the ORB algorithm and the Hamming distance are used, whether the friction current is adjusted to meet the requirement of a standard curve or not can be automatically recognized, meanwhile, field maintainers are guided to debug the turnout friction current, and the effects of saving the maintenance time, reducing the maintenance cost and improving the maintenance efficiency are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of turnout friction current, and in particular to a similarity-based intelligent analysis method and system for turnout friction current. Background Art

[0002] At present, turnouts, as one of the key equipment for rail transit such as railways and subways, have become the focus of research on intelligent operation and maintenance of railway signals. The adjustment of turnout friction current requires a lot of maintenance time. At the same time, the parameters are set by maintenance personnel, which is highly subjective. Unreasonable friction current will affect the service life of the switch machine, and it also requires high professional ability and level of maintenance personnel.

[0003] The friction current of the turnout switch machine occurs when a splint is placed between the turnout point rail and the base rail when the turnout can be normally switched to the fixed and reverse positions. At this time, the turnout cannot be switched to the correct position and cannot be locked. The current generated at this time is called the turnout friction current.

[0004] Therefore, it is an urgent problem for those skilled in the art to propose an intelligent analysis method and system for turnout friction current based on similarity to solve the difficulties existing in the prior art. Summary of the invention

[0005] In view of this, the present invention provides a similarity-based intelligent analysis method and system for turnout friction current, which can quickly determine whether the friction current adjusted by the current maintenance personnel meets the standard value, saving a lot of judgment, analysis and adjustment time, and achieving the purpose of easily adjusting the turnout friction current to the standard value with simple training for maintenance personnel.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A similarity-based intelligent analysis method for turnout friction current includes the following steps:

[0008] S1. Obtaining curves: Obtaining the friction current curve of the turnout, the friction current curve includes a standard friction current curve and a disordered friction current curve;

[0009] S2. Feature extraction: Extract feature points of the standard friction current curve based on the ORB algorithm, and match the feature point data with the disordered friction current curve;

[0010] S3. Calculate the Hamming distance: calculate the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points;

[0011] S4. Determine the result: set a similarity threshold, determine the adjustment range of the disordered friction current curve according to the relationship between the Hamming distance and the similarity, and give the adjustment result.

[0012] In the above method, the specific contents of extracting characteristic points of the standard friction current curve based on the ORB algorithm in S2 and matching characteristic point data with the disordered friction current curve are as follows:

[0013] S21. Select feature points based on the FAST algorithm. The feature point discriminant is as follows:

[0014]

[0015] Among them, I p is the gray value of the center of the circle, I x The gray point value is obtained by taking p as the center and 3 gray values ​​as the radius. tex When the sum is greater than the threshold, p is judged to be a feature point;

[0016] S22. For each feature point p, calculate the intensity centroid O around it, and determine the direction of the feature point by the vector from the feature point to the intensity centroid. The expression is as follows:

[0017]

[0018] Where m = 0, 1, n = 0, 1, B is the coordinate point of the pixel image, and θ is the direction of the feature point;

[0019] S23. Create feature descriptors based on the BRIEF algorithm. With n feature points as the center, sample n pairs of grayscale areas of size 5*5, compare the size of the pixels and the size of the grayscale area, and retain the results to form a feature vector. The feature descriptor expression of n pairs of feature points is as follows:

[0020]

[0021] Among them, τ(p; x i ,y i ) is the BRIEF descriptor feature vector;

[0022] S24. Match the characteristic points of the disordered friction current curve with the characteristic points extracted from the standard friction current curve.

[0023] In the above method, optionally, the specific content of calculating the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points in S3 is:

[0024] Assume that the input random friction current curve feature point set is I, the standard friction current curve feature point set is S, I = (i1, i2, ..., i m ), S=(s1,s2,......,s n ), the Hamming distance H between I and S is defined as:

[0025]

[0026] Among them, x i is the input characteristic point of the disordered friction current curve, y i is the characteristic point of the standard friction current curve, and h is the calculation function of the input disordered friction current curve characteristic point and the standard friction current curve characteristic point.

[0027] In the above method, optionally, a similarity threshold is set in S4, and the adjustment range of the disordered friction current curve is determined according to the relationship between the Hamming distance and the similarity, and the specific content of the adjustment result is given as follows:

[0028] According to the relationship between Hamming distance and similarity, the greater the distance, the lower the similarity, and the similarity is R, and the formula is as follows:

[0029]

[0030] The similarity threshold T is set to determine the adjustment range of the disordered friction current curve to be R ≥ T.

[0031] The above method optionally provides the adjustment result in S4 to guide the maintenance personnel to adjust the friction current.

[0032] A similarity-based intelligent analysis system for turnout friction current, applying any of the similarity-based intelligent analysis methods for turnout friction current, comprising a curve acquisition module, a feature extraction module, a Hamming distance calculation module, and a result determination module connected in sequence;

[0033] A curve acquisition module is used to acquire the friction current curve of the turnout, and the friction current curve includes a standard friction current curve and a disordered friction current curve;

[0034] The feature extraction module is used to extract the feature points of the standard friction current curve based on the ORB algorithm and match the feature point data with the disordered friction current curve;

[0035] A Hamming distance calculation module is used to calculate the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points;

[0036] The result determination module is used to set a similarity threshold, determine the adjustment range of the disordered friction current curve according to the relationship between the Hamming distance and the similarity, and give the adjustment result.

[0037] It can be seen from the above technical solution that compared with the prior art, the present invention provides a similarity-based intelligent analysis method and system for turnout friction current, which has the following beneficial effects: the present invention uses the ORB algorithm and Hamming distance, which can basically ignore the requirements for the professional ability of maintenance personnel, and can automatically identify whether the friction current is adjusted to meet the requirements of the standard curve. At the same time, it can guide on-site maintenance personnel to debug the turnout friction current, thereby saving maintenance time, reducing maintenance costs and increasing maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0039] Figure 1 A flow chart of a similarity-based intelligent analysis method for turnout friction current provided by the present invention;

[0040] Figure 2 A structural block diagram of a similarity-based intelligent analysis system for turnout friction current provided by the present invention;

[0041] Figure 3 A schematic diagram of extracting characteristic points of a standard friction current curve based on an ORB algorithm and matching characteristic point data with a disordered friction current curve provided by an embodiment of the present invention;

[0042] Figure 4 A diagram showing the disordered friction current adjustment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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.

[0044] Reference Figure 1 As shown, the present invention discloses a similarity-based intelligent analysis method for turnout friction current, comprising the following steps:

[0045] S1. Obtaining curves: Obtaining the friction current curve of the turnout, the friction current curve includes a standard friction current curve and a disordered friction current curve;

[0046] S2. Feature extraction: Extract feature points of the standard friction current curve based on the ORB algorithm, and match the feature point data with the disordered friction current curve;

[0047] S3. Calculate the Hamming distance: calculate the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points;

[0048] S4. Determine the result: set a similarity threshold, determine the adjustment range of the disordered friction current curve according to the relationship between the Hamming distance and the similarity, and give the adjustment result.

[0049] Further, such as Figure 3 As shown, the specific contents of extracting the characteristic points of the standard friction current curve based on the ORB algorithm in S2 and matching the characteristic point data with the disordered friction current curve are as follows:

[0050] S21. Select feature points based on the FAST algorithm. The feature point discriminant is as follows:

[0051]

[0052] Among them, I p is the gray value of the center of the circle, I x The gray point value is obtained by taking p as the center and 3 gray values ​​as the radius. tex When the sum is greater than the threshold, p is judged to be a feature point;

[0053] S22. For each feature point p, calculate the intensity centroid O around it, and determine the direction of the feature point by the vector from the feature point to the intensity centroid. The expression is as follows:

[0054]

[0055] Where m = 0, 1, n = 0, 1, B is the coordinate point of the pixel image, and θ is the direction of the feature point;

[0056] S23. Create feature descriptors based on the BRIEF algorithm. With n feature points as the center, sample n pairs of grayscale areas of size 5*5, compare the size of the pixels and the size of the grayscale area, and retain the results to form a feature vector. The feature descriptor expression of n pairs of feature points is as follows:

[0057]

[0058] Among them, τ(p; x i ,y i ) is the BRIEF descriptor feature vector;

[0059] S24. Match the characteristic points of the disordered friction current curve with the characteristic points extracted from the standard friction current curve.

[0060] Specifically, in S21, when f tex When the sum is greater than the threshold (usually 12), p is judged to be a feature point.

[0061] Furthermore, the specific content of calculating the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points in S3 is:

[0062] Assume that the input random friction current curve feature point set is I, the standard friction current curve feature point set is S, I = (i1, i2, ..., i m ), S=(s1,s2,......,s n ), the Hamming distance H between I and S is defined as:

[0063]

[0064] Among them, x i is the input characteristic point of the disordered friction current curve, y i is the characteristic point of the standard friction current curve, and h is the calculation function of the input disordered friction current curve characteristic point and the standard friction current curve characteristic point.

[0065] Specifically, the Hamming distance is to calculate the matching degree between the characteristic points of the standard friction current curve of the input turnout and the characteristic points of the disordered friction current curve through the binary string distance, find the overall characteristic difference between the two, and identify whether the disordered friction current curve meets the target requirements of the standard curve;

[0066] In current railways and urban rail transit, the ORB algorithm is used to analyze the turnout current curves collected by the microcomputer monitoring system;

[0067] According to the Hamming distance, the matching degree between the I feature point and the S feature point can be obtained. The similarity is inversely proportional to the Hamming distance, that is, the lower the Hamming distance value, the higher the graphic matching degree and the higher the similarity.

[0068] Furthermore, a similarity threshold is set in S4, and the adjustment range of the disordered friction current curve is determined according to the relationship between the Hamming distance and the similarity, and the specific content of the adjustment result is given as follows:

[0069] According to the relationship between Hamming distance and similarity, the greater the distance, the lower the similarity, and the similarity is R, and the formula is as follows:

[0070]

[0071] The similarity threshold T is set to determine the adjustment range of the disordered friction current curve to be R ≥ T.

[0072] Furthermore, after the adjustment result is given in S4, the maintenance personnel are guided to adjust the friction current.

[0073] In a specific embodiment, the following are included

[0074] (1) obtaining a friction current curve of a turnout, wherein the friction current curve includes a standard friction current curve and a disordered friction current curve;

[0075] The data of a group of turnouts are selected to establish a database, and according to the opinions of turnout experts, the disordered friction current curves that do not comply with the "Technical Standards for Signal Maintenance Rules" and the disordered friction current curves that comply with the "Technical Standards for Signal Maintenance Rules" are selected as the control group.

[0076] (ii) extracting characteristic points of the standard friction current curve based on the ORB algorithm and matching the characteristic point data with the disordered friction current curve;

[0077] Feature points are selected based on the FAST algorithm, and the feature point discriminant is as follows:

[0078]

[0079] Among them, I p is the gray value of the center of the circle, I x The gray point value is obtained by taking p as the center and 3 gray values ​​as the radius. tex When the sum is greater than the threshold (usually 12), p is judged as a feature point;

[0080] For each feature point p, calculate the intensity centroid O around it, and determine the direction of the feature point by the vector from the feature point to the intensity centroid. The expression is as follows:

[0081]

[0082] Where m = 0, 1, n = 0, 1, B is the coordinate point of the pixel image, and θ is the direction of the feature point;

[0083] Based on the BRIEF algorithm, a feature descriptor is created. With n feature points as the center, n pairs of grayscale areas of size 5*5 are sampled, the size of the pixels and the size of the grayscale area are compared, and the result is retained to form a feature vector. The feature descriptor expression of n pairs of feature points is as follows:

[0084]

[0085] Among them, τ(p; x i ,y i ) is the BRIEF descriptor feature vector;

[0086] Match the characteristic points of the disordered friction current curve with the characteristic points extracted from the standard friction current curve.

[0087] (iii) calculating the Hamming distance between the standard friction current curve and the disordered friction current curve based on the characteristic points;

[0088] Assume that the input random friction current curve feature point set is I, the standard friction current curve feature point set is S, I = (i1, i2, ..., i m ), S=(s1,s2,......,s n ), the Hamming distance H between I and S is defined as:

[0089]

[0090] Among them, x i is the input characteristic point of the disordered friction current curve, y i is the characteristic point of the standard friction current curve, and h is the calculation function of the input disordered friction current curve characteristic point and the standard friction current curve characteristic point.

[0091] The relationship between the Hamming distance and the similarity is that the larger the distance, the lower the similarity. The disordered friction current curve that does not meet the standard of the turnout friction current has a larger Hamming distance and a lower similarity with the standard friction current curve; the disordered friction current curve that meets the standard of the turnout friction current has a smaller Hamming distance and a higher similarity with the standard friction current curve.

[0092] (iv) setting a similarity threshold, determining the adjustment range of the disordered friction current curve according to the relationship between the Hamming distance and the similarity, and providing the adjustment result;

[0093] After a large amount of actual field data analysis and verification, the friction current curve with the Hamming distance threshold set below 6 and the similarity threshold set above 94.5% is in line with the friction current curve specified in the "Technical Standards for Signal Maintenance Rules" and has a recognition accuracy of 100%.

[0094] After obtaining the real-time friction current value, after processing by the ORB algorithm and the Hamming distance, combined with the comparison between the actual current value and the standard value, if the similarity does not match and the friction current is too large, the on-site debugging personnel will be prompted "The friction current is too large, please loosen the friction connector screws"; if the similarity does not match and the friction current is too small, the system will prompt the on-site debugging personnel "The friction current is too small, please tighten the friction connector screws". After multiple adjustments, until the actual current meets the standard friction current similarity set by the system, and the difference between the positioning and anti-position friction currents does not exceed the requirement of 0.3A. The maintenance suggestions are displayed in the interface maintenance support module, such as Figure 4 shown.

[0095] and Figure 1 Corresponding to the method described above, the embodiment of the present invention also provides a similarity-based intelligent analysis system for turnout friction current, which is used to Figure 1The specific implementation of the method is shown in the following diagram: Figure 2 As shown, it specifically includes a curve acquisition module, a feature extraction module, a Hamming distance calculation module, and a result determination module connected in sequence;

[0096] A curve acquisition module is used to acquire the friction current curve of the turnout, and the friction current curve includes a standard friction current curve and a disordered friction current curve;

[0097] The feature extraction module is used to extract the feature points of the standard friction current curve based on the ORB algorithm and match the feature point data with the disordered friction current curve;

[0098] A Hamming distance calculation module is used to calculate the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points;

[0099] The result determination module is used to set a similarity threshold, determine the adjustment range of the disordered friction current curve according to the relationship between the Hamming distance and the similarity, and give the adjustment result.

[0100] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0101] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A similarity-based intelligent analysis method for turnout friction current, characterized in that: The following steps are involved: S1. Obtaining curves: Obtaining the friction current curve of the turnout, the friction current curve includes a standard friction current curve and a disordered friction current curve; S2. Feature extraction: Extract feature points of the standard friction current curve based on the ORB algorithm, and match the feature point data with the disordered friction current curve; S3. Calculate the Hamming distance: calculate the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points; S4. Determine the result: set a similarity threshold, determine the adjustment range of the disordered friction current curve according to the relationship between the Hamming distance and the similarity, and give the adjustment result.

2. The intelligent analysis method of turnout friction current based on similarity according to claim 1 is characterized in that: In S2, the feature points of the standard friction current curve are extracted based on the ORB algorithm, and the feature point data are matched with the disordered friction current curve as follows: S21. Select feature points based on the FAST algorithm. The feature point discriminant is as follows: Among them, I p is the gray value of the center of the circle, I x The gray point value is obtained by taking p as the center and 3 gray values ​​as the radius. tex When the sum is greater than the threshold, p is judged to be a feature point; S22. For each feature point p, calculate the intensity centroid O around it, and determine the direction of the feature point by the vector from the feature point to the intensity centroid. The expression is as follows: Where m = 0, 1, n = 0, 1, B is the coordinate point of the pixel image, and θ is the direction of the feature point; S23. Create feature descriptors based on the BRIEF algorithm. With n feature points as the center, sample n pairs of grayscale areas of size 5*5, compare the size of the pixels and the size of the grayscale area, and retain the results to form a feature vector. The feature descriptor expression of n pairs of feature points is as follows: Among them, τ(p; x i ,y i ) is the BRIEF descriptor feature vector; S24. Match the characteristic points of the disordered friction current curve with the characteristic points extracted from the standard friction current curve.

3. The intelligent analysis method of turnout friction current based on similarity according to claim 1 or 2, characterized in that: The specific content of calculating the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points in S3 is: Assume that the input random friction current curve feature point set is I, the standard friction current curve feature point set is S, I = (i1, i2, ..., i m ), S=(s1,s2,......,s n ), the Hamming distance H between I and S is defined as: Among them, x i is the input characteristic point of the disordered friction current curve, y i is the characteristic point of the standard friction current curve, and h is the calculation function of the input disordered friction current curve characteristic point and the standard friction current curve characteristic point.

4. The intelligent analysis method of turnout friction current based on similarity according to claim 1 is characterized in that: In S4, a similarity threshold is set, and the adjustment range of the disordered friction current curve is determined according to the relationship between the Hamming distance and the similarity, and the specific content of the adjustment result is given as follows: According to the relationship between Hamming distance and similarity, the greater the distance, the lower the similarity, and the similarity is R, and the formula is as follows: The similarity threshold T is set to determine the adjustment range of the disordered friction current curve to be R ≥ T.

5. The intelligent analysis method of turnout friction current based on similarity according to claim 1 is characterized in that: After the adjustment result is given in S4, the maintenance personnel are guided to adjust the friction current.

6. A similarity-based intelligent analysis system for turnout friction current, characterized in that: A similarity-based intelligent analysis method for turnout friction current according to any one of claims 1 to 5 is applied, comprising a curve acquisition module, a feature extraction module, a Hamming distance calculation module, and a result determination module connected in sequence; A curve acquisition module is used to acquire the friction current curve of the turnout, and the friction current curve includes a standard friction current curve and a disordered friction current curve; The feature extraction module is used to extract the feature points of the standard friction current curve based on the ORB algorithm and match the feature point data with the disordered friction current curve; A Hamming distance calculation module is used to calculate the Hamming distance between the standard friction current curve and the disordered friction current curve according to the characteristic points; The result determination module is used to set a similarity threshold, determine the adjustment range of the disordered friction current curve according to the relationship between the Hamming distance and the similarity, and give the adjustment result.

Images