CN112213579A - Method and device for identifying faults of turnout switch machine - Google Patents
Method and device for identifying faults of turnout switch machine Download PDFInfo
- Publication number
- CN112213579A CN112213579A CN202011017317.8A CN202011017317A CN112213579A CN 112213579 A CN112213579 A CN 112213579A CN 202011017317 A CN202011017317 A CN 202011017317A CN 112213579 A CN112213579 A CN 112213579A
- Authority
- CN
- China
- Prior art keywords
- current time
- curve
- distance
- identified
- time series
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/0092—Arrangements for measuring currents or voltages or for indicating presence or sign thereof measuring current only
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The embodiment of the invention provides a method and a device for identifying a fault of a switch point, wherein only when the LB _ Kim lower bound distance and the LB _ Keogh lower bound distance between a sample current time sequence curve and a current time sequence curve to be identified are determined to be smaller than the current minimum DTW distance, the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified is calculated, and fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances is used as the fault information corresponding to the current time sequence curve to be identified. Therefore, a large number of curves which are not similar to the time series curve of the current to be identified in the turnout switch machine fault current database can be eliminated, the retrieval time can be greatly shortened, and the identification efficiency of fault information is further improved.
Description
Technical Field
The invention relates to the technical field of rail transit, in particular to a method and a device for identifying a turnout point machine fault.
Background
The switch machine is a switch device of the switch, and can be used for switching the switch or locking the switch. The failure of the switch machine may result in derailment of the on-track train, resulting in significant economic loss and casualties. The current time sequence curve of the turnout switch machine can be used for judging the working state of the turnout switch machine, for example, judging whether the turnout switch machine is in a normal working state, and if the turnout switch machine is in an abnormal working state, further judging which abnormal working state the turnout switch machine is in.
The core of judging the working state of the switch machine according to the current time series curve of the switch machine is to search whether a curve similar to the current time series curve to be searched in shape exists in a given fault current database of the switch machine or not for the current time series curve to be searched of the given switch machine. Judging whether the curve shapes are similar, wherein the simplest method is to adopt Euclidean distance measurement: that is, the euclidean distance between the two curves is calculated point by point, and then the sum of the euclidean distances of all points is averaged.
The similarity of two curves evaluated by the Euclidean distance has the following problems: firstly, the Euclidean distance cannot correctly calculate the true similarity of a curve after the curve is distorted and deformed; second, euclidean distance is less well behaved when faced with complex curves. As shown in fig. 1, the similarity between the curves Q and C with the two bar shapes shifted cannot be correctly calculated by calculating the euclidean distance. Therefore, a Dynamic Time Warping (DTW) distance is often used for calculating the similarity between two curves, and the DTW distance can deal with twisting, translation and transformation between the curves and has strong robustness. However, because the DTW distance is high in complexity, if the DTW distance between each current time-series curve in the turnout switch machine fault current database and the current time-series curve to be retrieved is calculated, a large amount of time is consumed for the whole turnout switch machine fault current database, and the integral DTW distance cannot be used practically.
Disclosure of Invention
The embodiment of the invention provides a method and a device for identifying a fault of a switch machine, which are used for solving the defects in the prior art.
The embodiment of the invention provides a method for identifying a fault of a switch machine, which comprises the following steps:
acquiring a current time series curve to be identified of a turnout switch machine, and preprocessing the current time series curve to be identified;
for each sample current time sequence curve in a turnout switch machine fault current database, determining an LB _ Kim lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum Dynamic Time Warping (DTW) distance, determining an LB _ Keogh lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity;
if judging that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance, calculating the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified;
and taking the fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time sequence curve to be identified.
According to an embodiment of the invention, the method for identifying the fault of the turnout switch machine further comprises the following steps:
and if the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified is judged and obtained to be smaller than the current minimum DTW distance, updating the current minimum DTW distance to be the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified.
According to an embodiment of the invention, the method for identifying the fault of the turnout switch machine further comprises the following steps:
and if judging that the LB _ Kim lower bound distance is larger than or equal to the current minimum DTW distance, finishing the subsequent calculation of the sample current time sequence curve, and continuously calculating the LB _ Kim lower bound distance between other sample current time sequence curves and the current time sequence curve to be identified.
According to an embodiment of the invention, the method for identifying the fault of the turnout switch machine further comprises the following steps:
and if judging that the LB _ Keogh lower bound distance is larger than or equal to the current minimum DTW distance, finishing the subsequent calculation of the sample current time sequence curve, and continuously calculating LB _ Kim lower bound distances between other sample current time sequence curves and the current time sequence curve to be identified.
According to the method for identifying the fault of the point switch machine, the preprocessing of the current time series curve to be identified specifically comprises the following steps:
sampling the current time series curve to be identified to a specified length;
and carrying out zero-mean value normalization processing on the current time series curve to be identified with the specified length.
According to the method for identifying the fault of the turnout switch machine, the determining of the LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be identified specifically comprises the following steps:
calculating an LB _ Keogh lower bound of the current time series curve to be recognized;
and calculating the distance between the LB _ Keogh lower bound of the current time series curve to be recognized and the predetermined LB _ Keogh lower bound of the sample current time series curve, and taking the distance as the LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be recognized.
According to the method for identifying the fault of the point switch machine, the step of determining the LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified specifically comprises the following steps:
respectively calculating the LB _ Kim lower bound of the current time series curve to be identified and the LB _ Kim lower bound of the sample current time series curve;
and calculating the distance between the LB _ Kim lower bound of the current time series curve to be identified and the LB _ Kim lower bound of the sample current time series curve, and taking the distance as the LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified.
The embodiment of the invention also provides a switch machine fault identification system, which comprises: the device comprises a curve acquisition module, a lower boundary distance determination module, a DTW distance determination module and a fault information determination module. Wherein the content of the first and second substances,
the curve acquisition module is used for acquiring a current time series curve to be identified of the turnout switch machine and preprocessing the current time series curve to be identified;
the lower bound distance determining module is used for determining LB _ Kim lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified for each sample current time sequence curve in the turnout switch machine fault current database, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum dynamic time warping DTW distance, determining LB _ Keogh lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity;
the DTW distance determining module is used for calculating the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified if judging that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance;
and the fault information determining module is used for taking the fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time sequence curve to be identified.
The embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above methods for identifying a fault of a point switch machine when executing the program.
Embodiments of the present invention further provide a non-transitory computer readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the method for identifying a fault of a switch machine as described in any one of the above.
Before calculating the DTW distance between each sample current time sequence curve and the current time sequence curve to be identified, the method and the device for identifying the fault of the turnout switch machine provided by the embodiment of the invention successively judge whether the LB _ Kim lower boundary distance and the LB _ Keogh lower boundary distance between the sample current time sequence curve and the current time sequence curve to be identified are smaller than the current minimum DTW distance, only if the LB _ Kim lower boundary distance and the LB _ Keogh lower boundary distance between the sample current time sequence curve and the current time sequence curve to be identified are both determined to be smaller than the current minimum DTW distance, the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified is calculated, and taking the fault information carried by the sample current time series curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time series curve to be identified. Therefore, a large number of curves which are not similar to the time series curve of the current to be identified in the turnout switch machine fault current database can be eliminated, the retrieval time can be greatly shortened, and the identification efficiency of fault information is further improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic representation of curves Q and C for a prior art translation of two bars;
fig. 2 is a schematic flow chart of a method for identifying a switch point fault according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a turnout switch machine fault identification device according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 2 is a schematic flow chart of a method for identifying a switch machine fault according to an embodiment of the present invention, as shown in fig. 2, the method includes:
s1, acquiring a current time series curve to be identified of the turnout switch machine, and preprocessing the current time series curve to be identified;
s2, for each sample current time series curve in the turnout switch machine fault current database, determining the LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum Dynamic Time Warping (DTW) distance, determining the LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity;
s3, if judging that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance, calculating the DTW distance between the sample current time series curve and the current time series curve to be identified;
and S4, taking the fault information carried by the sample current time series curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time series curve to be identified.
Specifically, in the method for identifying a fault of a point switch machine provided in the embodiment of the present invention, for a current Time series curve to be identified of a given point switch machine, a closest sample current Time series curve in a point switch machine fault current database is found by determining a Dynamic Time Warping (DTW) distance between the current Time series curve to be identified and each sample current Time series curve in a point switch machine fault current database, and fault information carried by the sample current Time series curve is used as fault information corresponding to the current Time series curve to be identified.
Firstly, step S1 is executed to obtain the current time series curve to be identified of the point switch machine, and the current time series curve to be identified is preprocessed. The preprocessing specifically may be to sample the current time series curve to be identified, so that the length of the current time series curve becomes a specified length. The current time series curve to be identified can be normalized, so that the distance between the current time series curve and the sample current time series curve can be conveniently calculated.
And then executing a step S2, and determining the LB _ Kim lower bound distance between the sample current time series curve I and the current time series curve to be identified for each sample current time series curve I (I is more than or equal to 1 and less than or equal to I, and I is the total number of all the sample current time series curves in the turnout switch machine fault current database) in the turnout switch machine fault current database. The sample current time series curve i in the turnout switch machine fault current database is provided with a database main key id and used for identifying the sample current time series curve. The sample current time series curve needs to be preprocessed and carries fault information of the turnout switch machine. The LB _ Kim lower bound distance refers to the distance between the LB _ Kim lower bound of the sample current time series curve i and the LB _ Kim lower bound of the current time series curve to be identified. The LB _ Kim lower bound distance can obtain the lower bound of the DTW distance of the sample current time series curve i and the current time series curve to be identified with constant time complexity.
And judging the size relationship between the calculated LB _ Kim lower bound distance and the current minimum DTW distance, wherein the initial value of the current minimum DTW distance can be set to be infinite. If the LB _ Kim lower bound distance is less than the current minimum DTW distance, the sample current time series curve i is a curve which is probably the closest to the current time series curve to be identified. Therefore, the LB _ Keogh lower bound distance between the sample current time series curve i and the current time series curve to be identified can be further determined. The LB _ Keogh lower bound distance refers to the distance between the LB _ Keogh lower bound of the sample current-time series curve i and the LB _ Keogh lower bound of the current-time series curve to be recognized. The LB _ Keogh lower bound distance can obtain the lower bound of the DTW distance of the sample current time series curve i and the current time series curve to be identified with O (n) time complexity.
Then, step S3 is executed to determine the magnitude relationship between the calculated LB _ Keogh lower bound distance and the current minimum DTW distance, and if the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance, it indicates that the sample current time-series curve i may be the curve closest to the current time-series curve to be recognized. Thus, at this point the DTW distance between the sample current time series curve i and the current time series curve to be identified is calculated. On the basis, an early-stopping strategy can be adopted to prune the calculation.
And finally, executing step S4, determining all DTW distances, and then selecting a minimum DTW distance from all DTW distances, where the fault information carried by the sample current time series curve corresponding to the minimum DTW distance is used as the fault information corresponding to the current time series curve to be identified. The fault information is the fault information of the turnout switch machine.
Since the time complexity for calculating the LB _ Kim lower bound distance is constant, the time complexity for calculating the LB _ Keogh lower bound distance is O (n), and even the sum of the two is far less than the time complexity for directly calculating the DTW distance O (n)2)。
Before calculating the DTW distance between each sample current time sequence curve and the current time sequence curve to be identified, the method for identifying the fault of the turnout switch machine provided by the embodiment of the invention successively judges whether the LB _ Kim lower boundary distance and the LB _ Keogh lower boundary distance between the sample current time sequence curve and the current time sequence curve to be identified are smaller than the current minimum DTW distance, and only when the LB _ Kim lower boundary distance and the LB _ Keogh lower boundary distance between the sample current time sequence curve and the current time sequence curve to be identified are both smaller than the current minimum DTW distance, the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified is calculated, and taking the fault information carried by the sample current time series curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time series curve to be identified. Therefore, a large number of curves which are not similar to the time series curve of the current to be identified in the turnout switch machine fault current database can be eliminated, the retrieval time can be greatly shortened, and the identification efficiency of fault information is further improved.
Through tests, aiming at a given time sequence, the most similar curve in hundreds of thousands of time sequences is searched, the speed of the turnout switch machine fault identification method provided by the embodiment of the invention is improved by nearly 250 times compared with the original DTW-based search method, and the millisecond-level response speed is reached compared with the minute-level response time of single search.
On the basis of the above embodiment, the method for identifying a fault of a point switch machine provided in the embodiment of the present invention further includes:
and if the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified is judged and obtained to be smaller than the current minimum DTW distance, updating the current minimum DTW distance to be the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified.
Specifically, if the DTW distance between the sample current time series curve i and the current time series curve to be identified is smaller than the current minimum DTW distance, it is indicated that the current minimum DTW distance is not the minimum DTW distance, and therefore, the current minimum DTW distance needs to be updated to the DTW distance between the sample current time series curve i and the current time series curve to be identified.
In the embodiment of the invention, the current minimum DTW distance is updated according to the DTW distance between the sample current time sequence curve i and the current time sequence curve to be identified, so that the accuracy of the current minimum DTW distance can be ensured.
On the basis of the above embodiment, the method for identifying a fault of a point switch machine provided in the embodiment of the present invention further includes:
and if judging that the LB _ Kim lower bound distance is larger than or equal to the current minimum DTW distance, finishing the subsequent calculation of the sample current time sequence curve, and continuously calculating the LB _ Kim lower bound distance between other sample current time sequence curves and the current time sequence curve to be identified.
Specifically, if the LB _ Kim lower bound distance is greater than or equal to the current minimum DTW distance, it indicates that the DTW distance between the sample current time series curve i and the current time series curve to be identified is not necessarily minimum, so that the subsequent calculation of the sample current time series curve i can be finished without performing subsequent calculation on the sample current time series curve i, and the LB _ Kim lower bound distances between other sample current time series curves and the current time series curve to be identified are continuously calculated.
In the embodiment of the invention, when the LB _ Kim lower bound distance is more than or equal to the current minimum DTW distance, the subsequent calculation of the sample current time sequence curve i is finished, so that the time waste caused by the calculation of the DTW distance between the sample current time sequence curve i and the current time sequence curve to be identified can be avoided, the curve retrieval efficiency can be improved, and the fault identification efficiency of the point switch machine can be further improved.
On the basis of the above embodiment, the method for identifying a fault of a point switch machine provided in the embodiment of the present invention further includes:
and if judging that the LB _ Keogh lower bound distance is larger than or equal to the current minimum DTW distance, finishing the subsequent calculation of the sample current time sequence curve, and continuously calculating LB _ Kim lower bound distances between other sample current time sequence curves and the current time sequence curve to be identified.
Specifically, if the LB _ Kim lower bound distance is smaller than the current minimum DTW distance, but the LB _ Keogh lower bound distance is greater than or equal to the current minimum DTW distance, it can also be described that the DTW distance between the sample current time series curve i and the current time series curve to be recognized is certainly not the minimum, so that the subsequent calculation of the sample current time series curve i is finished without performing the subsequent calculation on the sample current time series curve i, and the LB _ Kim lower bound distances between other sample current time series curves and the current time series curve to be recognized are continuously calculated.
In the embodiment of the invention, when the LB _ Kim lower bound distance is less than the current minimum DTW distance and the LB _ Keogh lower bound distance is more than or equal to the current minimum DTW distance, the subsequent calculation of the sample current time sequence curve i is finished, so that the time waste caused by the calculation of the DTW distance between the sample current time sequence curve i and the current time sequence curve to be identified can be avoided, the efficiency of curve retrieval can be improved, and the efficiency of fault identification of the turnout switch machine can be further improved.
On the basis of the above embodiment, the method for identifying a fault of a point switch machine provided in the embodiment of the present invention preprocesses the current time series curve to be identified, and specifically includes:
sampling the current time series curve to be identified to a specified length;
and carrying out zero-mean value normalization processing on the current time series curve to be identified with the specified length.
Specifically, in the embodiment of the present invention, the preprocessing operation may specifically include: and sampling the current time series curve to be identified, and changing the current time series curve to be identified into a specified length. Sampling the current time series curve to be identified to a specified length w, performing up-sampling interpolation if the length of the current time series curve to be identified is lower than w, and performing down-sampling if the length of the current time series curve to be identified is higher than w.
And then, carrying out zero-mean normalization processing on the current time series curve to be identified with the specified length. The zero-mean normalization processing can also be called Z normalization processing, and can be specifically realized by the following formula:
wherein TS' is a result of Z normalization processing of each sampling point in the current time series curve to be recognized with a specified length, TS is a value of each sampling point in the current time series curve to be recognized with a specified length, mean (TS) is a mean value of values of all sampling points in the current time series curve to be recognized with a specified length, std (TS) is a variance of values of all sampling points in the current time series curve to be recognized with a specified length.
On the basis of the foregoing embodiment, the method for identifying a fault of a point switch machine provided in the embodiment of the present invention, where the determining of the LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be identified specifically includes:
calculating an LB _ Keogh lower bound of the current time series curve to be recognized;
and calculating the distance between the LB _ Keogh lower bound of the current time series curve to be recognized and the predetermined LB _ Keogh lower bound of the sample current time series curve, and taking the distance as the LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be recognized.
Specifically, in the embodiment of the invention, when calculating the LB _ Keogh lower bound distance, the LB _ Keogh lower bound of the current time series curve to be identified may be calculated in real time, and for the LB _ Keogh lower bound of the sample current time series curve, the LB _ Keogh lower bound may be predetermined when constructing the switch machine fault current database, because the LB _ Keogh lower bound has higher time complexity and longer time consumption for calculation, and thus, the time waste generated by calculating the LB _ Keogh lower bound in real time may be reduced in a predetermined manner.
In addition, the LB _ Kim lower bound of the sample current time series curve can be predetermined when a fault current database of the turnout switch machine is constructed, so that the time waste caused by calculating the LB _ Kim lower bound in real time is saved.
On the basis of the above embodiment, the method for identifying a fault of a point switch machine provided in the embodiment of the present invention, where the determining of the LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified specifically includes:
respectively calculating the LB _ Kim lower bound of the current time series curve to be identified and the LB _ Kim lower bound of the sample current time series curve;
and calculating the distance between the LB _ Kim lower bound of the current time series curve to be identified and the LB _ Kim of the sample current time series curve, and taking the distance as the LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified.
Specifically, in the embodiment of the present invention, when determining the LB _ Kim lower bound distance, the LB _ Kim lower bound of the current-time series curve to be recognized and the LB _ Kim lower bound of the sample current-time series curve may be calculated in real time, because the LB _ Kim lower bound has a low time complexity and a short calculation time, and a time waste caused by calculating the LB _ Keogh lower bound in real time is negligible.
Fig. 3 is a schematic structural diagram of a switch point fault identification system according to an embodiment of the present invention, as shown in fig. 3, the system includes: a curve acquisition module 31, a lower boundary distance determination module 32, a DTW distance determination module 33, and a fault information determination module 34. Wherein the content of the first and second substances,
the curve obtaining module 31 is configured to obtain a current time series curve to be identified of a turnout switch machine, and preprocess the current time series curve to be identified;
the lower bound distance determining module 32 is configured to determine, for each sample current time series curve in the turnout switch machine fault current database, an LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified, and if it is determined that the LB _ Kim lower bound distance is smaller than the current minimum dynamic time warping DTW distance, determine an LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity;
the DTW distance determining module 33 is configured to calculate a DTW distance between the sample current time series curve and the current time series curve to be identified if it is determined that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance;
the fault information determining module 34 is configured to use fault information carried by a sample current time series curve corresponding to a minimum DTW distance among all the DTW distances as fault information corresponding to the current time series curve to be identified.
Specifically, the functions of the modules in the turnout switch machine fault identification system provided in the embodiment of the present invention correspond to the operation flows of the steps in the foregoing method embodiments one to one, and the implementation effects are also consistent.
Fig. 4 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 4: a processor (processor)410, a communication interface (communication interface)420, a memory (memory)430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 are communicated with each other via the communication bus 440. The processor 410 may invoke logic instructions in the memory 430 to perform a switch machine fault identification method comprising: acquiring a current time series curve to be identified of a turnout switch machine, and preprocessing the current time series curve to be identified; for each sample current time sequence curve in a turnout switch machine fault current database, determining an LB _ Kim lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum Dynamic Time Warping (DTW) distance, determining an LB _ Keogh lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity; if judging that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance, calculating the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified; and taking the fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time sequence curve to be identified.
In addition, the logic instructions in the memory 430 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, an embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by a computer, the computer is capable of executing the method for identifying a fault of a turnout switch provided by the above-mentioned embodiments of the method, where the method includes: acquiring a current time series curve to be identified of a turnout switch machine, and preprocessing the current time series curve to be identified; for each sample current time sequence curve in a turnout switch machine fault current database, determining an LB _ Kim lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum Dynamic Time Warping (DTW) distance, determining an LB _ Keogh lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity; if judging that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance, calculating the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified; and taking the fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time sequence curve to be identified.
In still another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented by a processor to execute the method for identifying a fault of a switch machine provided in the foregoing embodiments, where the method includes: acquiring a current time series curve to be identified of a turnout switch machine, and preprocessing the current time series curve to be identified; for each sample current time sequence curve in a turnout switch machine fault current database, determining an LB _ Kim lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum Dynamic Time Warping (DTW) distance, determining an LB _ Keogh lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity; if judging that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance, calculating the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified; and taking the fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time sequence curve to be identified.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for identifying a switch machine failure, comprising:
acquiring a current time series curve to be identified of a turnout switch machine, and preprocessing the current time series curve to be identified;
for each sample current time sequence curve in a turnout switch machine fault current database, determining an LB _ Kim lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum Dynamic Time Warping (DTW) distance, determining an LB _ Keogh lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity;
if judging that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance, calculating the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified;
and taking the fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time sequence curve to be identified.
2. The method of identifying a malfunction of a switch machine as claimed in claim 1, further comprising:
and if the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified is judged and obtained to be smaller than the current minimum DTW distance, updating the current minimum DTW distance to be the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified.
3. The method of identifying a malfunction of a switch machine as claimed in claim 1, further comprising:
and if judging that the LB _ Kim lower bound distance is larger than or equal to the current minimum DTW distance, finishing the subsequent calculation of the sample current time sequence curve, and continuously calculating the LB _ Kim lower bound distance between other sample current time sequence curves and the current time sequence curve to be identified.
4. The method of identifying a malfunction of a switch machine as claimed in claim 1, further comprising:
and if judging that the LB _ Keogh lower bound distance is larger than or equal to the current minimum DTW distance, finishing the subsequent calculation of the sample current time sequence curve, and continuously calculating LB _ Kim lower bound distances between other sample current time sequence curves and the current time sequence curve to be identified.
5. The method for identifying a malfunction of a point switch machine as claimed in any one of claims 1-4, wherein the preprocessing of the current time series curve to be identified comprises:
sampling the current time series curve to be identified to a specified length;
and carrying out zero-mean value normalization processing on the current time series curve to be identified with the specified length.
6. The method for identifying a fault of a switch machine as claimed in any one of claims 1 to 4, wherein the determining of the LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be identified specifically comprises:
calculating an LB _ Keogh lower bound of the current time series curve to be recognized;
and calculating the distance between the LB _ Keogh lower bound of the current time series curve to be recognized and the predetermined LB _ Keogh lower bound of the sample current time series curve, and taking the distance as the LB _ Keogh lower bound distance between the sample current time series curve and the current time series curve to be recognized.
7. The method for identifying a malfunction of a switch machine as claimed in any one of claims 1-4, wherein the determining the LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified comprises:
respectively calculating the LB _ Kim lower bound of the current time series curve to be identified and the LB _ Kim lower bound of the sample current time series curve;
and calculating the distance between the LB _ Kim lower bound of the current time series curve to be identified and the LB _ Kim lower bound of the sample current time series curve, and taking the distance as the LB _ Kim lower bound distance between the sample current time series curve and the current time series curve to be identified.
8. A switch machine fault identification system, comprising:
the curve acquisition module is used for acquiring a current time series curve to be identified of the turnout switch machine and preprocessing the current time series curve to be identified;
the lower bound distance determining module is used for determining LB _ Kim lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified for each sample current time sequence curve in the turnout switch machine fault current database, and if judging that the LB _ Kim lower bound distance is smaller than the current minimum dynamic time warping DTW distance, determining LB _ Keogh lower bound distance between the sample current time sequence curve and the current time sequence curve to be identified; the sample current time series curve is preprocessed and carries fault information of a turnout switch machine; the initial value of the current minimum DTW distance is infinity;
the DTW distance determining module is used for calculating the DTW distance between the sample current time sequence curve and the current time sequence curve to be identified if the judgment result shows that the LB _ Keogh lower bound distance is smaller than the current minimum DTW distance;
and the fault information determining module is used for taking the fault information carried by the sample current time sequence curve corresponding to the minimum DTW distance in all the DTW distances as the fault information corresponding to the current time sequence curve to be identified.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of the method for identifying a malfunction of a switch machine as claimed in any one of claims 1 to 7.
10. A non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program when executed by a processor implements the steps of the switch machine fault identification method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011017317.8A CN112213579A (en) | 2020-09-24 | 2020-09-24 | Method and device for identifying faults of turnout switch machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011017317.8A CN112213579A (en) | 2020-09-24 | 2020-09-24 | Method and device for identifying faults of turnout switch machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112213579A true CN112213579A (en) | 2021-01-12 |
Family
ID=74051838
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011017317.8A Pending CN112213579A (en) | 2020-09-24 | 2020-09-24 | Method and device for identifying faults of turnout switch machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112213579A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112950023A (en) * | 2021-03-02 | 2021-06-11 | 西门子电力自动化有限公司 | Method and device for on-line monitoring switch equipment |
| CN114325188A (en) * | 2021-12-28 | 2022-04-12 | 国网河北省电力有限公司经济技术研究院 | Fault detection method and device and server |
-
2020
- 2020-09-24 CN CN202011017317.8A patent/CN112213579A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112950023A (en) * | 2021-03-02 | 2021-06-11 | 西门子电力自动化有限公司 | Method and device for on-line monitoring switch equipment |
| CN114325188A (en) * | 2021-12-28 | 2022-04-12 | 国网河北省电力有限公司经济技术研究院 | Fault detection method and device and server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112213579A (en) | Method and device for identifying faults of turnout switch machine | |
| CN111985930B (en) | Client number generation method and device | |
| CN111160329A (en) | Root cause analysis method and device | |
| CN109685805B (en) | Image segmentation method and device | |
| CN110609908A (en) | Case serial-parallel method and device | |
| CN107944866B (en) | Transaction record duplication elimination method and computer-readable storage medium | |
| CN114022748B (en) | Target identification method, device, equipment and storage medium | |
| CN110019762B (en) | Problem positioning method, storage medium and server | |
| CN111178037A (en) | Repeated defect report identification method and device and electronic equipment | |
| CN112287776A (en) | Bearing performance index analysis method and system, readable storage medium and electronic equipment | |
| CN116206319A (en) | Data processing system for clinical trials | |
| CN113592529B (en) | Potential customer recommendation method and device for bond products | |
| CN107423140B (en) | Return code identification method and device | |
| CN115424253A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN111488463B (en) | Test corpus generation method and device and electronic equipment | |
| CN113961725A (en) | Automatic label labeling method, system, equipment and storage medium | |
| CN114765574A (en) | Network anomaly delimitation positioning method and device | |
| CN111538669A (en) | Test case extraction method and device based on historical problem backtracking analysis | |
| CN112784903B (en) | Method, device and equipment for training target recognition model | |
| CN116681074B (en) | Method, device, equipment and storage medium for detecting misoperation of numerical control system | |
| CN111159998B (en) | Method and system for analyzing hierarchical structure of font outline | |
| CN115118793B (en) | BLF file analysis fault tolerance method and device and computer equipment | |
| CN107689846B (en) | Method and system for detecting data errors | |
| CN117453663A (en) | Intelligent management system for big data of bank | |
| CN112989021A (en) | Method, device and equipment for judging violation of counselor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |