Platform door equipment health degree monitoring method and system

Abstract

The invention relates to a platform door equipment health degree monitoring method and system. The method comprises the steps that abnormal values in historical monitoring data are extracted; calibrating abnormal values to form a data set; calibrating the platform door health score of the corresponding platform door based on the data set to form a training sample, and training to obtain a platform door health degree evaluation model; a platform door health degree evaluation model is utilized to complete monitoring of a target platform door, the health score of the target platform door and a corresponding abnormity troubleshooting scheme are used as samples to be input into a large model adopting a transformer architecture, and meanwhile, a LoRA fine tuning technology is adopted to train the samples to obtain a platform door maintenance scheme recommendation model; and using the platform door maintenance scheme recommendation model to complete management and control of a target platform door. According to the invention, by using a machine learning technology, the health degree of the target platform door can be automatically identified and a corresponding troubleshooting scheme can be given without human intervention, so that operation and maintenance personnel are helped to save time and resources, unnecessary inspection and maintenance are avoided, and the operation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of platform door early warning, and in particular to a method and system for monitoring the health of platform door equipment. Background Art

[0002] In urban rail transit systems, especially in subway operations, platform doors, as an important safety facility, play the key role of isolating the track from the passenger waiting area. Its main purpose is to improve the efficiency of overall operations while ensuring passenger safety. With the rapid development of urban rail transit, the use of platform doors is becoming more and more common, and the importance of their status monitoring is becoming increasingly prominent.

[0003] Currently, the status monitoring of platform doors mainly relies on the threshold judgment of a single indicator. This method usually involves manually setting the thresholds of multiple key parameters, such as current, voltage, and door opening and closing time. When the monitoring data exceeds the set threshold, the system will determine it as an abnormal state. However, the setting of such thresholds often lacks scientific basis, which can easily lead to unreasonable settings, resulting in frequent false alarms and missed alarms. Summary of the invention

[0004] In order to solve the above problems, an object of an embodiment of the present invention is to provide a method and system for monitoring the health of platform door equipment.

[0005] A method for monitoring the health of platform door equipment, comprising:

[0006] Step 1: Obtain historical monitoring data of platform doors;

[0007] Step 2: Determine whether there are abnormal values ​​in the monitoring data of the target platform door based on historical monitoring data;

[0008] Step 3: Calibrate the outliers to form a data set;

[0009] Step 4: Calibrate the platform door health score of the corresponding platform door based on the data set to form a training sample;

[0010] Step 5: Input the training samples into the deep learning model for training to obtain a platform door health assessment model;

[0011] Step 6: Use the platform door health evaluation model to complete the monitoring of the target platform door and obtain the health score of the target platform door;

[0012] Step 7: Input the health score of the target platform door and its corresponding abnormality troubleshooting plan as samples into the large model using the transformer architecture, and use the LoRA fine-tuning technology to train it to obtain the platform door maintenance plan recommendation model;

[0013] Step 8: Use the platform door maintenance plan recommendation model to complete the control of the target platform door.

[0014] Preferably, the step 2: determining whether there are abnormal values ​​in the historical monitoring data includes:

[0015] Step 2.1: Use the K-center clustering algorithm and elbow rule to remove abnormal data from historical monitoring data to obtain monitoring data after removing abnormal values;

[0016] Step 2.2: Obtain the central value of the monitoring data after removing outliers;

[0017] Step 2.3: Calculate the difference between the center value and the target platform door monitoring data; the difference includes the Euclidean distance and the Pearson correlation coefficient;

[0018] Step 2.4: The monitoring data corresponding to the difference value beyond the preset range is regarded as an abnormal value.

[0019] Preferably, in step 2.3, the formula is used:

[0020]

[0021] Calculate the difference between the center value and the target station door monitoring data; where the sample point X is represented by the vector (X1, X2, …, X n ), the center point Y is composed of the vector (Y1,Y2,…,Y n ), r is the Pearson coefficient, n represents the length of the vector, X i is the i-th observation value of sample point X, Table X's root mean square, σ X is the root mean square of sample point X, Y i The i-th observation value of the center point, represents the mean of Y, σ Y is the root mean square of Y.

[0022] Preferably, in step 2.3, the formula is used:

[0023]

[0024] Calculate the difference between the center value and the target station door monitoring data; where the sample point x is represented by the vector (x1, x2, …, x n ), the center point y is composed of the vector (y1,y2,…,y n ), d(x,y) is the distance between the sample point and the center point, n represents the length of the x vector, x i is the i-th observation value of sample point x, y i The ith observation value of the center point.

[0025] Preferably, the step 4: calibrating the platform door health score of the corresponding platform door based on the data set to form a training sample includes:

[0026] Step 4.1: Calculate the health score of each platform door in the station to form a score set;

[0027] Step 4.2: Take the minimum value in the score set as the health score of the platform door in the corresponding station.

[0028] Preferably, in step 4.1, the health score calculation formula of the platform door is:

[0029]

[0030] Where s is the health score of the platform door, INT means rounding down to an integer, r means the number of times abnormal values ​​appear in the platform door monitoring data within a preset period, and R means the maximum tolerance threshold.

[0031] The present invention also provides a platform door equipment health monitoring system, comprising:

[0032] Historical data acquisition module, used to obtain historical monitoring data of platform doors;

[0033] An abnormal value judgment module is used to judge whether there are abnormal values ​​in the monitoring data of the target platform door based on the historical monitoring data;

[0034] An outlier marking module is used to mark outliers and form a data set;

[0035] A health score calibration module is used to calibrate the platform door health score of the corresponding platform door based on the data set to form a training sample;

[0036] A training module, used for inputting the training samples into a deep learning model for training to obtain a platform door health assessment model;

[0037] The monitoring module is used to monitor the target platform door using the platform door health assessment model.

[0038] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps in the above-mentioned method for monitoring the health of platform door equipment.

[0039] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned method for monitoring the health of platform door equipment are implemented.

[0040] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0041] The present invention relates to a method and system for monitoring the health of platform door equipment. Compared with the prior art, the present invention can automatically identify the health of a target platform door without human intervention by utilizing machine learning technology, thereby helping operation and maintenance personnel save time and resources, avoid unnecessary inspections and maintenance, and improve operational efficiency.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A flow chart of a method for monitoring the health of a platform door device in an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a platform door equipment health monitoring system in an embodiment provided by the present invention. DETAILED DESCRIPTION

[0046] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0047] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0048] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] See also Figure 1 , a platform door equipment health monitoring method, comprising:

[0050] Step 1: Obtain historical monitoring data of platform doors;

[0051] In the present invention, data such as the current, voltage, displacement, and door opening and closing time of the platform door are collected as profile indicators of the platform door equipment.

[0052] Step 2: Determine whether there are abnormal values ​​in the monitoring data of the target platform door based on historical monitoring data;

[0053] Furthermore, step 2 includes:

[0054] Step 2.1: Use the K-center clustering algorithm and elbow rule to remove abnormal data from historical monitoring data to obtain monitoring data after removing abnormal values;

[0055] Step 2.2: Obtain the central value of the monitoring data after removing outliers;

[0056] Step 2.3: Calculate the difference between the center value and the target platform door monitoring data; the difference includes the Euclidean distance and the Pearson correlation coefficient;

[0057] In step 2.3, the formula is used:

[0058]

[0059] Calculate the Pearson correlation coefficient between the center value and the target station door monitoring data; where the sample point X is represented by the vector (X1, X2, …, X n ), the center point Y is composed of the vector (Y1,Y2,…,Y n ), r is the Pearson coefficient, n represents the length of the vector, X i is the i-th observation value of sample point X, Table X's root mean square, σ X is the root mean square of sample point X, Y i The i-th observation value of the center point, represents the mean of Y, σ Y is the root mean square of Y.

[0060] Using the formula:

[0061]

[0062] Calculate the Euclidean distance between the center value and the target station door monitoring data; where the sample point x is represented by the vector (x1, x2, …, x n ), the center point y is composed of the vector (y1,y2,…,y n ), d(x,y) is the distance between the sample point and the center point, n represents the length of the x vector, x i is the i-th observation value of sample point x, y i The ith observation value of the center point.

[0063] Euclidean distance is the true distance between two points in m-dimensional space, and is usually used to calculate the straight-line distance between two points in two-dimensional or three-dimensional space. Euclidean distance is also called Euclidean metric or Euclidean distance, and is a common distance definition in mathematics. In statistics, the Pearson correlation coefficient, also known as the Pearson product-moment correlation coefficient (PPMCC or PCCs for short), is used to measure the correlation (linear correlation) between two variables X and Y, and its value is between -1 and 1. The present invention can reflect the difference between the center value and the target platform door monitoring data through the above two parameters, thereby determining the abnormal value in the target platform door monitoring data.

[0064] Step 2.4: The monitoring data corresponding to the difference value beyond the preset range is regarded as an abnormal value.

[0065] Step 3: Calibrate the outliers to form a data set;

[0066] Step 4: Calibrate the platform door health score of the corresponding platform door based on the data set to form a training sample;

[0067] Further, the step 4 comprises:

[0068] Step 4.1: Calculate the health score of each platform door in the station to form a score set;

[0069] Step 4.2: Take the minimum value in the score set as the health score of the platform door in the corresponding station.

[0070] For example, a station has 24 doors (represented by {B1, B2, ..., B23, B24}). Each door consists of 5 devices (represented by {D1, D2, ..., D4, D5}).

[0071] The process of calculating health points is as follows:

[0072] 1) First, use the "Build a Station Abnormal Identification Solution" to identify and count the abnormalities of each device

[0073] 2) Calculate the health score of each station

[0074] a) For device D1, the default value of R = 100, r is the number of statistical abnormalities = 30, and the health score s(D1) = 88 is calculated by the following formula:

[0075]

[0076] b) Calculate the scores of devices {D1, D2, …, D4, D5} respectively as {s1 = 88, s2 = 82, s3 = 86, s4 = 89, s5 = 90}.

[0077] c) Take the minimum value of {s1=88,s2=82,s3=86,s4=89,s5=90} as the corresponding platform gate score

[0078] 3) Calculate the health of the platform

[0079] a) Calculate the scores of the 24 gates (represented by {B1, B2, ..., B23, B24}) according to "Calculate the health score of each station"

[0080] b) Take the minimum score of {B1, B2, …, B23, B24} as the platform door health score.

[0081] Step 5: Input the training samples into the deep learning model for training to obtain a platform door health assessment model;

[0082] Step 6: Use the platform door health evaluation model to complete the monitoring of the target platform door and obtain the health score of the target platform door;

[0083] Step 7: Input the health score of the target platform door and its corresponding abnormality troubleshooting plan as samples into the large model using the transformer architecture, and use the LoRA fine-tuning technology to train it to obtain the platform door maintenance plan recommendation model;

[0084] Step 8: Use the platform door maintenance plan recommendation model to complete the control of the target platform door.

[0085] For example, if the health score of a platform door at a certain station is 59, the score will be automatically pushed to the big model, and the big model will generate an abnormality troubleshooting plan based on the score, including fault troubleshooting points, troubleshooting methods, and tools to be carried, for front-line employees to troubleshoot.

[0086] By utilizing machine learning technology, the present invention can automatically identify the health of the target platform door and its corresponding troubleshooting plan without human intervention, helping operation and maintenance personnel save time and resources, avoid unnecessary inspections and maintenance, and improve operational efficiency.

[0087] The present invention also provides a platform door equipment health monitoring system, comprising:

[0088] Historical data acquisition module, used to obtain historical monitoring data of platform doors;

[0089] An abnormal value judgment module is used to judge whether there are abnormal values ​​in the monitoring data of the target platform door based on the historical monitoring data;

[0090] An outlier marking module is used to mark outliers and form a data set;

[0091] A health score calibration module is used to calibrate the platform door health score of the corresponding platform door based on the data set to form a training sample;

[0092] A training module, used for inputting the training samples into a deep learning model for training to obtain a platform door health assessment model;

[0093] The monitoring module is used to monitor the target platform door using the platform door health assessment model.

[0094] Compared with the prior art, the beneficial effects of a platform door equipment health monitoring system provided by the present invention are the same as the beneficial effects of a platform door equipment health monitoring method described in the above technical solution, which will not be repeated here.

[0095] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory and the processor are connected via the bus, and is characterized in that when the computer program is executed by the processor, the steps in the above-mentioned method for monitoring the health of a platform door device are implemented. Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the method for monitoring the health of a platform door device described in the above-mentioned technical solution, which will not be elaborated here.

[0096] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned method for monitoring the health of platform door equipment are implemented. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the method for monitoring the health of platform door equipment described in the above-mentioned technical solution, which will not be repeated here.

[0097] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technical solution that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for monitoring the health of platform door equipment, characterized in that: include: Step 1: Obtain historical monitoring data of platform doors; Step 2: Determine whether there are abnormal values ​​in the monitoring data of the target platform door based on historical monitoring data; Step 3: Calibrate the outliers to form a data set; Step 4: Calibrate the platform door health score of the corresponding platform door based on the data set to form a training sample; Step 5: Input the training samples into the machine learning model for training to obtain a platform door health assessment model; Step 6: Use the platform door health evaluation model to complete the monitoring of the target platform door and obtain the health score of the target platform door; Step 7: Input the health score of the target platform door and its corresponding abnormality troubleshooting plan as samples into the large model using the transformer architecture, and use the LoRA fine-tuning technology to train it to obtain the platform door maintenance plan recommendation model; Step 8: Use the platform door maintenance plan recommendation model to complete the control of the target platform door.

2. A platform door equipment health monitoring method according to claim 1, characterized in that: The step 2: determining whether there are abnormal values ​​in the historical monitoring data includes: Step 2.1: Use the K-center clustering algorithm and elbow rule to remove abnormal data from historical monitoring data to obtain monitoring data after removing abnormal values; Step 2.2: Obtain the central value of the monitoring data after removing outliers; Step 2.3: Calculate the difference between the center value and the target platform door monitoring data; the difference includes the Euclidean distance and the Pearson correlation coefficient; Step 2.4: The monitoring data corresponding to the difference value beyond the preset range is regarded as an abnormal value.

3. A platform door equipment health monitoring method according to claim 2, characterized in that: In step 2.3, the formula is used: Calculate the difference between the center value and the target station door monitoring data; where the sample point X is represented by the vector (X1, X2, …, X n ), the center point Y is composed of the vector (Y1,Y2,…,Y n ), r is the Pearson coefficient, n represents the length of the vector, X i is the i-th observation value of sample point X, Table X's root mean square, σ X is the root mean square of sample point X, Y i The i-th observation value of the center point, represents the mean of Y, σ Y is the root mean square of Y.

4. A platform door equipment health monitoring method according to claim 2, characterized in that: In step 2.3, the formula is used: Calculate the difference between the center value and the target station door monitoring data; where the sample point x is represented by the vector (x1, x2, …, x n ), the center point y is composed of the vector (y1,y2,…,y n ), d(x,y) is the distance between the sample point and the center point, n represents the length of the x vector, x i is the i-th observation value of sample point x, y i The ith observation value of the center point.

5. A platform door equipment health monitoring method according to claim 2, characterized in that: The step 4: calibrating the platform door health score of the corresponding platform door based on the data set to form a training sample, includes: Step 4.1: Calculate the health score of each platform door in the station to form a score set; Step 4.2: Take the minimum value in the score set as the health score of the platform door in the corresponding station.

6. A platform door equipment health monitoring method according to claim 5, characterized in that: In step 4.1, the calculation formula for the health score of the platform door is: Where s is the health score of the platform door, INT means rounding down to an integer, r means the number of times abnormal values ​​appear in the platform door monitoring data within a preset period, and R means the maximum tolerance threshold.

7. A platform door equipment health monitoring system, characterized in that: include: Historical data acquisition module, used to obtain historical monitoring data of platform doors; An abnormal value judgment module is used to judge whether there are abnormal values ​​in the monitoring data of the target platform door based on the historical monitoring data; An outlier marking module is used to mark outliers and form a data set; A health score calibration module is used to calibrate the platform door health score of the corresponding platform door based on the data set to form a training sample; A training module, used for inputting the training samples into a deep learning model for training to obtain a platform door health assessment model; The monitoring module is used to monitor the target platform door using the platform door health assessment model.

8. An electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps in the method for monitoring the health of platform door equipment as described in any one of claims 1-6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in a method for monitoring the health of platform door equipment as described in any one of claims 1-6 are implemented.

Images