A vehicle-mounted PHM diagnostic system and method for rail transit vehicle doors

Abstract

The present invention discloses an on-board PHM diagnostic system and method for rail transit vehicle doors, in which a ground server is connected to a PHM system in a rail vehicle; the PHM system is used to collect door operation data on the rail vehicle to which it belongs and then use an intelligent diagnostic algorithm to determine the sub-healthy state of each door; the ground server is used to update the intelligent diagnostic algorithm in the PHM system for determining the sub-healthy state of the door according to the working condition of each door. The PHM system is multi-level and has the function of local edge computing. During operation, only the sub-healthy determination results of the door are transmitted between the on-board PHM system and the ground server, reducing the load between the vehicle and the ground. In addition, the ground server can update the sub-healthy intelligent diagnostic algorithm corresponding to each door in combination with the actual operating conditions of each door, optimize the model and parameters of the intelligent diagnostic algorithm online, and improve the accuracy of diagnosis.

Description

Technical Field

[0001] The invention belongs to the field of urban rail transit, and in particular relates to an on-board PHM diagnostic system and method for rail transit vehicle doors. Background Art

[0002] The existing rail transit door PHM system puts the diagnostic algorithm on the ground server. The real-time data generated by the door opening and closing, various signals, etc., will be sent to the ground server through the vehicle-ground communication. The ground server deploys the database, diagnostic algorithm, and display interface. The diagnostic model is stored in the database in advance. When the door data accumulates to a certain amount, the diagnostic algorithm is triggered to determine whether the data is faulty or sub-healthy, and then the diagnostic results are pushed to the display interface or the ground server. However, the existing system cannot achieve instant diagnosis on the vehicle. Most importantly, the massive data is transmitted to the server through the vehicle-ground communication, which not only occupies the precious wireless bandwidth on the vehicle, but also fails to display the diagnostic results on the vehicle in time, which may delay the staff's judgment of the door failure at a critical moment. Summary of the invention

[0003] Purpose of the invention: The first purpose of the present invention is to provide an on-board PHM diagnostic system for rail transit vehicle doors. The second purpose of the present invention is to provide an on-board PHM diagnostic method for rail transit vehicle doors.

[0004] Technical solution: The on-board PHM diagnostic system for rail transit vehicle doors of the present invention comprises a PHM system arranged in a rail vehicle, wherein the PHM system is connected to a ground server;

[0005] The PHM system is used to collect the door operation data on the railway vehicle and then use the intelligent diagnosis algorithm to determine the sub-health status of each door;

[0006] The ground server is used to update the intelligent diagnosis algorithm for determining the sub-health status of each door in the PHM system according to the working condition of the door.

[0007] Preferably, the PHM system includes a vehicle-level PHM system, a car-level PHM subsystem and a door-level PHM subsystem; the door-level PHM subsystem is used to receive door controller data and send it to the car-level PHM subsystem, and the vehicle-level PHM system is used to receive door operation data sent by the car-level PHM subsystem and then use an intelligent diagnosis algorithm to determine the sub-health status of each door, and send the determination result to the ground server.

[0008] Preferably, the ground server sends the updated intelligent diagnosis algorithm data corresponding to each door to the vehicle-level PHM system, the car-level PHM subsystem and the door-level PHM subsystem step by step;

[0009] When the vehicle-level PHM system is shut down, the vehicle-level PHM subsystem uses the locally stored intelligent diagnosis algorithm to determine the sub-health status of the vehicle door and stores the determination result;

[0010] When the carriage-level PHM subsystem is shut down, the door-level PHM subsystem uses the locally stored intelligent diagnosis algorithm to determine the sub-health status of the door and stores the determination result.

[0011] Preferably, the ground server updates the intelligent diagnosis algorithm for determining the sub-healthy state of each door according to the working condition of the door as follows:

[0012] (1) All operating data feature values ​​and algorithm parameters of the intelligent diagnosis algorithm candidates are combined into a parameter library;

[0013] (2) Randomly select n characteristic values ​​and intelligent diagnosis algorithm parameters to form a combination item as the population of the genetic algorithm;

[0014] (3) Combined with the standard operating data of each door under the actual health status, the intelligent diagnosis algorithm in the PHM system is updated with each chromosome in sequence, and the chromosome data that meets the preset requirements is retained;

[0015] (4) Perform genetic manipulation on the retained chromosome data to generate a progeny population;

[0016] (5) Repeat steps (3) and (4) until the updated intelligent diagnosis algorithm meets the preset performance.

[0017] Preferably, in step (3), the sub-health data of various points pre-created under the vehicle and the sub-health rule list are updated into the PHM system, and the fitness value of each chromosome is calculated according to the fitness function in combination with the standard operating data under the actual health state of the vehicle door. When the calculated fitness value is the largest, the optimal parameter result is obtained. The fitness value is the confidence of the result calculated by the updated intelligent diagnosis algorithm and the sub-health rule list.

[0018] Preferably, the genetic operations performed on the retained chromosome data in step (4) include selection, crossover and mutation.

[0019] Preferably, the specific steps of the intelligent diagnosis algorithm for diagnosing the sub-health status of the vehicle door are as follows:

[0020] (S1) acquiring data: acquiring standard operation data, and triggering sub-health status diagnosis when the door accumulates m pieces of door opening and closing operation data;

[0021] (S2) feature extraction: obtaining operation data features for reflecting the movement state of the door;

[0022] (S3) Cluster diagnosis: using preset standard data features and real-time door operation data features, and utilizing classification algorithms, it is possible to determine whether the door is in a sub-healthy state.

[0023] Furthermore, the rotating motor of the rail transit door is used to collect signals using acceleration sensors and Hall devices. When the motor rotates during the door opening and closing process, the sensor can collect real-time data curves of the door's current, speed and other signals, and transmit them to the door-level PHM subsystem through the door controller. Each door controller corresponds to a door, and then all the door-level PHM subsystems in a car are connected through communication interfaces such as Ethernet and summarized into the car-level PHM subsystem. Each car is connected to the vehicle-level PHM system through an Ethernet interface. After the vehicle-level PHM system summarizes the real-time data, it can perform data reception, data filtering, data storage, data diagnosis, data outbound and other operations locally through edge computing, and the data is outbound to the ground server for cloud-edge collaborative processing.

[0024] Furthermore, the ground processor can simultaneously connect to the onboard PHM systems of multiple rail trains and receive the sub-health status diagnosis data of the doors sent by each train, thus realizing the algorithm parameter evaluation and correction and model optimization for all trains on the entire line. The onboard PHM system is used to evaluate the sub-health status of each door of the train, but due to the complexity of the on-site working conditions of the doors, the standard curves of the current, angle and other data of each door are different. However, in the onboard PHM system, the preset feature selection method, diagnostic algorithm parameters and diagnostic rules are consistent, resulting in slight deviations in the diagnostic results due to different doors.

[0025] Therefore, by entering the genetic algorithm into the ground server, a random search algorithm based on natural selection and genetic genetics is used. Genetic operations such as selection, crossover and mutation are used on the population composed of eigenvalues ​​and algorithm parameters to exchange information between string structures in an organized but random manner, so that the fittest survive and gradually approach the optimal solution.

[0026] In the genetic algorithm, the selection operation of the genetic algorithm adopts the tournament algorithm, the crossover operation of the genetic algorithm adopts the arithmetic crossover method, and the mutation operation of the genetic algorithm adopts the uniform mutation method.

[0027] Furthermore, the sub-health rule list is a special feature sequence under the sub-health item. The sub-health rule list is generated by rigorous experiments under the train. When the train PHM diagnostic system is installed, it has been pre-installed in each PHM system, and the sub-health rule list of each door is the same. The sub-health data of each point generated under the train is also the same.

[0028] Due to different working conditions of each door, the standard curve of each door in the actual health state on the train is different. When the clustering algorithm is used, the standard data and sub-health data are clustered to complete the judgment of the sub-health state. Since the standard curves of different doors in the actual health state on the train are different, different chromosomes are obtained by mutation algorithm and brought into the on-board PHM system. After the clustering algorithm with chromosomes calculates the results, it will perform confidence calculation with the rule list. This confidence calculation process is the fitness function. The greater the confidence, the better the fitness function, and the better the chromosome. After the loop iteration, the characteristic value and algorithm parameters suitable for the current working condition are obtained for each door. Under normal circumstances, the ground server does not receive the door operation data, and sends the population data generated by the genetic algorithm to the corresponding on-board PHM system. After the fitness evaluation is performed in the on-board PHM system, the results are returned to the ground server. In the above process, since the basis of fitness evaluation is to use the standard operation data of the door under the current working condition, the characteristics and data algorithm parameters finally screened are most suitable for the corresponding door, so that the sub-health prediction result is more accurate.

[0029] Furthermore, the PHM system uses open source tools such as Kettle to filter data and perform data filtering operations on defective parts of the collected data such as missing content, null values, format errors, high sampling duplication, etc.

[0030] Furthermore, when the PHM system compresses data, it annotates the data based on the semantic model, reduces the data items with the same meaning that can be converted, reduces duplicate data based on the sampling frequency, and reduces invalid data transmission; it also uses a data dictionary to efficiently encode the original data, improves information entropy, and reduces the amount of transmitted data. The cloud platform decodes based on the data dictionary; in addition, for real-time continuous data such as real-time current and speed, compressed sensing is used for processing and then uploaded, and the powerful computing power of the cloud platform is used to run iterative algorithms for recovery.

[0031] Furthermore, the vehicle-level PHM system, the car-level PHM subsystem and the door-level PHM subsystem are set up step by step. When the PHM system at the upper level fails or crashes, the subsystem at the lower level can be used to determine the sub-health status. Since data cannot be uploaded or sent when a PHM system at a certain level fails, the detection data and results can only be stored locally and sent to the upper level after the system is restored.

[0032] Furthermore, in step (S1), the intelligent diagnosis algorithm generally collects the initial s door opening and closing operation data as standard data. When the sub-health status diagnosis is triggered, the door accumulates m door opening and closing operation data each time, and under normal working conditions, m>2s.

[0033] In step (S2), the intelligent diagnosis algorithm screens the feature library through a feature dimensionality reduction algorithm or artificial experience to form target features of the initial state of the intelligent diagnosis algorithm.

[0034] In step (S3), during the cluster diagnosis process, all feature values ​​of each door opening and closing are defined as a high-dimensional data point. After accumulating to a set number, the pre-stored standard data and these points are classified by K-means. The evaluation basis is the distance principle to the center point. After iteration, if it can be divided into two categories, it means that there is sub-health data. If it cannot be classified into two categories, it is a normal door opening and closing. The k feature indexes with the greatest impact in the classification are compared with the sub-health rules pre-configured in the database. If the similarity reaches the specified confidence level, it is considered that the door is in this sub-health state.

[0035] The on-board PHM diagnostic method for rail transit vehicle doors of the present invention has the following specific steps:

[0036] Step 1, obtaining an intelligent diagnosis algorithm for determining a sub-healthy state of a vehicle door, and forming a parameter library with all operating data feature values ​​and algorithm parameters of the intelligent diagnosis algorithm candidates;

[0037] Step 2, randomly select n characteristic values ​​and intelligent diagnosis algorithm parameters to form a combination item as the population of the genetic algorithm;

[0038] Step 3, combining the standard operating data of each door under the actual health status, sequentially updating the intelligent diagnosis algorithm in the PHM system with each chromosome, and retaining the chromosome data that meets the preset requirements;

[0039] Step 4, genetically manipulate the retained chromosome data to generate a progeny population;

[0040] Step 5, repeat steps (3) and (4) until the updated intelligent diagnosis algorithm meets the preset performance, and use the updated intelligent diagnosis algorithm to determine the sub-health status of the corresponding door.

[0041] Preferably, in step 3, the sub-health data of various points pre-created under the vehicle and the sub-health rule list are updated into the PHM system, and the fitness value of each chromosome is calculated according to the fitness function in combination with the standard operating data under the actual health state of the door. When the calculated fitness value is the largest, the optimal parameter result is obtained. The fitness value is the confidence of the result calculated by the updated intelligent diagnosis algorithm and the sub-health rule list.

[0042] Preferably, steps 2 to 5 are executed in a set interval time period.

[0043] Furthermore, in step 5, after m generations of inheritance, when the confidence tends to be constant, it indicates that the characteristic values ​​of each door and the diagnostic algorithm parameters have reached a globally optimal state. After the train has been running for a period of time, the algorithm can still continue to repeat the above method to find the globally optimal state at this time.

[0044] Beneficial effects: The present invention adopts an on-board PHM system, which can realize algorithm diagnosis on the train and can efficiently feed back the diagnosis results to the train in real time. During the operation, only the sub-health judgment results of the door are transmitted between the on-board PHM system and the ground server, and real-time operation data are not transmitted. Therefore, the traffic load between the train and the ground can be greatly reduced; in addition, the ground server can update the sub-health intelligent diagnosis algorithm corresponding to each door in combination with the actual operating conditions of each door, and optimize the model and parameters of the intelligent diagnosis algorithm online to achieve the effect of the global optimal solution and improve the accuracy of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of the structure of the vehicle-mounted PHM diagnostic system of the present invention;

[0046] Figure 2 It is a schematic diagram of the structure of the PHM system in the present invention;

[0047] Figure 3 It is a flow chart of the genetic algorithm of the ground server in the present invention;

[0048] Figure 4 This is a schematic diagram of the trigger logic of the sub-health state diagnosis of the vehicle door in the present invention;

[0049] Figure 5 It is a flow chart for diagnosing the sub-healthy state of a vehicle door in the present invention;

[0050] Figure 6 It is the workflow diagram of the clustering algorithm in the present invention. DETAILED DESCRIPTION

[0051] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0052] Taking the urban rail transit as an example, the on-board PHM diagnostic system of the rail transit door is Figure 1 As shown, the ground server 105 is connected to the vehicle-level PHM systems 104 and 109 in the rail vehicle; the vehicle-level PHM system in the first rail vehicle is connected to the car-level PHM subsystems 101, 102, and 103, and the vehicle-level PHM system in the second rail vehicle is connected to the car-level PHM subsystems 106, 107, and 108. Each car-level PHM subsystem is connected to the door controller of each door through the door-level PHM subsystem.

[0053] In this embodiment, when the on-board PHM diagnostic system is working, the acceleration sensor and the Hall device are used to collect signals from the rotating motor of the rail transit door. When the motor rotates during the door opening and closing process, the sensor can collect the real-time data curve of the door's current, speed and other signals, and transmit them to the door-level PHM subsystem through the door controller. Each door controller corresponds to a door, and then all the door-level PHM subsystems in a car are connected through the Ethernet and other communication interfaces, and summarized into the car-level PHM subsystem. Figure 2 As shown in the figure, each carriage is connected to the vehicle-level PHM system through an Ethernet interface. After the vehicle-level PHM system aggregates the real-time data, it can perform data reception, data filtering, data storage, data diagnosis, data outbound operations, etc. through edge computing locally, and the data is outbound to the ground server for cloud-edge collaborative processing.

[0054] In this embodiment, the ground server deploys train-level PHM diagnostic software, which mainly realizes algorithm parameter evaluation and correction and model optimization for all trains on the entire line, and under special circumstances can also receive door operation data to diagnose the sub-health status of the door.

[0055] In this embodiment, since the preset intelligent diagnosis algorithm for each door is the same, the genetic algorithm is moved into the diagnostic software of the ground server, and the genetic operation is combined with the actual standard operation data of each door to optimize the intelligent diagnosis algorithm of each door, so as to obtain the optimal solution for the sub-health diagnosis of each door, such as Figure 3 As shown, the specific steps are as follows:

[0056] Step 1, obtaining an intelligent diagnosis algorithm for determining a sub-healthy state of a vehicle door, and forming a parameter library with all operating data feature values ​​and algorithm parameters of the intelligent diagnosis algorithm candidates;

[0057] Step 2, randomly select n characteristic values ​​and intelligent diagnosis algorithm parameters to form a combination item as the population of the genetic algorithm;

[0058] Step 3, combining the standard operating data of each door under the actual health status, sequentially updating the intelligent diagnosis algorithm in the PHM system with each chromosome, and retaining the chromosome data that meets the preset requirements;

[0059] Step 4, genetically manipulate the retained chromosome data to generate a progeny population;

[0060] Step 5, repeat steps (3) and (4) until the updated intelligent diagnosis algorithm meets the preset performance, and use the updated intelligent diagnosis algorithm to determine the sub-health status of the corresponding door.

[0061] In this embodiment, in step 3, the sub-health data of various points pre-created under the vehicle and the sub-health rule list are updated into the PHM system, and the fitness value of each chromosome is calculated according to the fitness function in combination with the standard operating data under the actual health state of the door. When the calculated fitness value is the largest, the optimal parameter result is obtained. The fitness value is the confidence of the result calculated by the updated intelligent diagnosis algorithm and the sub-health rule list.

[0062] The genetic operations in step 4 include selection, crossover and mutation, wherein the genetic algorithm selection operation adopts the tournament algorithm, the genetic algorithm crossover operation adopts the arithmetic crossover method, and the genetic algorithm mutation operation adopts the uniform mutation method.

[0063] In step 5, after a set generation of inheritance or when the confidence tends to be constant, it indicates that the characteristic values ​​and diagnostic algorithm parameters of each train have reached a globally optimal state. After the train has been running for a period of time, the algorithm can continue to repeat the above method to find the globally optimal state at this time.

[0064] In this embodiment, the PHM system is responsible for the core operations of data filtering, data compression, data storage, data diagnosis, and data outbound transmission of the real-time data of the vehicle's doors. When filtering data, the PHM system uses open source tools such as Kettle to perform data filtering operations on the defective parts of the collected data such as missing content, null values, format errors, and high sampling duplication. When compressing data, the PHM system performs data annotation based on the semantic model, reduces data items that can be converted with the same meaning, reduces duplicate data based on the sampling frequency, and reduces invalid data transmission; it also uses a data dictionary to efficiently encode the original data, improve information entropy, and reduce the amount of transmitted data. The cloud platform decodes based on the data dictionary; in addition, for real-time continuous data such as real-time current and speed, compressed sensing is used for processing and then uploaded, and the powerful computing power of the cloud platform is used to run the iterative algorithm for recovery.

[0065] In this embodiment, the vehicle-level PHM system, the car-level PHM subsystem and the door-level PHM subsystem are set up level by level and back up each other in terms of function. The only difference is the scale of data processing. When the PHM system at the upper level fails or crashes, the subsystem at the lower level can be used to determine the sub-health status. Since data cannot be uploaded or sent when a PHM system at a certain level fails, the detection data and results can only be stored locally and sent to the upper level after the system is restored.

[0066] In this embodiment, the process of the PHM system using the intelligent diagnosis algorithm to diagnose the sub-health status of the vehicle door is as follows: Figure 5 As shown, the specific steps are as follows:

[0067] (S1) Obtaining data: The first 100 pieces of door opening and closing operation data of each door are used as standard data. Whenever the door accumulates 200 pieces of operation data, a sub-health diagnosis task is triggered.

[0068] (S2) Feature extraction: Screening the feature library through feature dimension reduction algorithm or manual experience to obtain operation data features used to reflect the movement state of the door;

[0069] (S3) Cluster diagnosis: using preset standard data features and real-time door operation data features, and utilizing classification algorithms, it is possible to determine whether the door is in a sub-healthy state.

[0070] In this embodiment, the process of triggering a sub-health diagnosis task in step (S1) is as follows: Figure 4 As shown, each time a door switch data is received and processed, the number in the statistical table is increased by one. When the data volume of the same door reaches the set value, the current door number is stored in the corresponding table. The diagnostic thread continuously polls the table to trigger the sub-health diagnosis task of the door.

[0071] In this embodiment, the process of clustering diagnosis using the K-means algorithm in step (S3) is as follows: Figure 6 As shown in the figure, all the feature values ​​of each door opening and closing are defined as a high-dimensional data point. After accumulating to a set number, the pre-stored standard data and these points are classified by K-means. The evaluation basis is the distance principle to the center point. After iteration, if it can be divided into two categories, it means that there is sub-health data. If it cannot be classified into two categories, it means that the door is opened and closed normally. The k feature indexes with the greatest impact in the classification are compared with the sub-health rules pre-configured in the database. If the similarity reaches the specified confidence level, it is considered that the door is in this sub-health state.

[0072] In summary, the multi-level PHM system ensures the stability of data collection, distribution, and uplink, and can back up each other when determining the sub-health status of the door, greatly improving the stability of the system. This PHM diagnostic system can realize sub-health diagnosis on the train, and can feedback the diagnostic results to the train in real time and efficiently, greatly reducing the load on the train-to-ground communication system. In addition, the ground server can update the sub-health intelligent diagnosis algorithm corresponding to each door based on the actual operating conditions of each door, and optimize the model and parameters of the intelligent diagnosis algorithm online to achieve the effect of the global optimal solution and improve the accuracy of diagnosis.

Claims

1. An on-board PHM diagnostic system for rail transit doors. Features: It includes a PHM system disposed in a rail vehicle, wherein the PHM system is connected to a ground server; The PHM system includes a vehicle-level PHM system, a car-level PHM subsystem and a door-level PHM subsystem, and intelligent diagnostic algorithms are stored locally; the PHM system is used to collect door operation data on the rail vehicle and then use the intelligent diagnostic algorithm to determine the sub-health status of each door, the steps are as follows: (S1) acquiring data: acquiring standard operation data, and triggering sub-health status diagnosis when the door accumulates m pieces of door opening and closing operation data; (S2) feature extraction: obtaining operation data features for reflecting the movement state of the door; (S3) Cluster diagnosis: using the preset standard data features and the real-time door operation data features, and using the classification algorithm, to determine whether the door is in a sub-healthy state; The ground server is used to update the intelligent diagnosis algorithm for determining the sub-health status of each door in the PHM system according to the working condition of each door, and the steps are as follows: (1) All operating data feature values ​​and algorithm parameters of the intelligent diagnosis algorithm candidates are combined into a parameter library; (2) Randomly select n characteristic values ​​and intelligent diagnosis algorithm parameters to form a combination item as the population of the genetic algorithm; (3) Combined with the standard operating data of each door under the actual health status, the intelligent diagnosis algorithm in the PHM system is updated with each chromosome in sequence, and the chromosome data that meets the preset requirements is retained; (4) Perform genetic manipulation on the retained chromosome data to generate a progeny population; (5) Repeat steps (3) and (4) until the updated intelligent diagnosis algorithm meets the preset performance; In the step (3), the sub-health data of various points pre-created under the vehicle and the sub-health rule list are updated into the PHM system, and the fitness value of each chromosome is calculated according to the fitness function in combination with the standard operating data under the actual health state of the vehicle door. When the calculated fitness value is the largest, the optimal parameter result is obtained. The fitness value is the confidence of the result calculated by the updated intelligent diagnosis algorithm and the sub-health rule list.

2. The on-board PHM diagnostic system for rail transit vehicle doors according to claim 1, Features: The ground server sends the updated intelligent diagnosis algorithm data corresponding to each door to the vehicle-level PHM system, the car-level PHM subsystem and the door-level PHM subsystem step by step; When the vehicle-level PHM system is shut down, the vehicle-level PHM subsystem uses the locally stored intelligent diagnosis algorithm to determine the sub-health status of the vehicle door and stores the determination result; When the carriage-level PHM subsystem is shut down, the door-level PHM subsystem uses the locally stored intelligent diagnosis algorithm to determine the sub-health status of the door and stores the determination result.

3. The on-board PHM diagnostic system for rail transit vehicle doors according to claim 1, Features: The genetic operations performed on the retained chromosome data in step (4) include selection, crossover and mutation.

4. A vehicle-mounted PHM diagnostic method for rail transit vehicle doors, according to the vehicle-mounted PHM diagnostic system for rail transit vehicle doors according to claim 1, Features: The specific steps of the diagnostic method are as follows: Step 1, obtaining an intelligent diagnosis algorithm for determining a sub-healthy state of a vehicle door, and forming a parameter library with all operating data feature values ​​and algorithm parameters of the intelligent diagnosis algorithm candidates; Step 2, randomly select n characteristic values ​​and intelligent diagnosis algorithm parameters to form a combination item as the population of the genetic algorithm; Step 3, combining the standard operating data of each door under the actual health status, sequentially updating the intelligent diagnosis algorithm in the PHM system with each chromosome, and retaining the chromosome data that meets the preset requirements; Step 4, genetically manipulate the retained chromosome data to generate a progeny population; Step 5, repeat steps (3) and (4) until the updated intelligent diagnosis algorithm meets the preset performance, and use the updated intelligent diagnosis algorithm to determine the sub-health status of the corresponding door.

5. The on-board PHM diagnostic method for rail transit vehicle doors according to claim 4, Features: Run steps 2 to 5 according to the set interval time cycle.

Images