Equipment operation condition monitoring method and device of rail transit system

Abstract

The invention relates to the technical field of rail transit monitoring, and provides an equipment operation condition monitoring method and device of a rail transit system, and the method comprises the steps: determining at least one data collection point based on an equipment list of the rail transit system and a class list of the rail transit system; the equipment list comprises attribute information corresponding to at least one piece of equipment of the rail transit system, and the class list comprises attribute information corresponding to at least one equipment type; performing feature extraction on the first monitoring data of each data acquisition point in the rail transit system to obtain second monitoring data of each data acquisition point; performing differential processing on the second monitoring data of each data acquisition point to obtain a first monitoring data sequence corresponding to each data acquisition point; and according to the first monitoring data sequence corresponding to each data acquisition point, performing prediction by using the first recurrent neural network to obtain first prediction data corresponding to each data acquisition point. According to the invention, the monitoring efficiency of equipment operation condition monitoring of the rail transit system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit monitoring, and in particular to a method and device for monitoring the equipment operating status of a rail transit system. Background Art

[0002] The safety management of urban rail transit is of vital importance. During the safety management of urban rail transit, the operating status of various electromechanical equipment needs to be uniformly monitored and visually managed.

[0003] At present, the existing rail transit operation status monitoring system collects the real-time data of the equipment one by one, and then displays the collected data in real time. However, as the number of equipment in the rail transit system increases, the monitoring efficiency of the method of collecting and displaying the data of the equipment one by one is low. Summary of the invention

[0004] In view of the defects existing in the prior art, the present invention provides a method and device for monitoring the equipment operating status of a rail transit system, which improves the monitoring efficiency of the equipment operating status monitoring of the rail transit system.

[0005] In a first aspect, the present invention provides a method for monitoring the equipment operating status of a rail transit system, the method comprising the following steps.

[0006] Determine at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system; the device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; Performing feature extraction on the first monitoring data of each of the data collection points in the rail transit system to obtain second monitoring data of each of the data collection points; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

[0007] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the equipment list and the class table both contain an equipment class code field, and the equipment class code field is used to characterize the equipment type; the method of determining at least one data collection point based on the equipment list of the rail transit system and the class table of the rail transit system includes: Perform a Cartesian product of the attribute information corresponding to the device class code field in the device list and the attribute information corresponding to the device class code field in the class table to obtain a Cartesian product result; Based on the set corresponding to the Cartesian product result, each of the data collection points is determined.

[0008] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the feature extraction of the first monitoring data of each data collection point in the rail transit system to obtain the second monitoring data corresponding to each data collection point includes: Determine a preset feature; the preset feature is a feature used to predict abnormal operation of the equipment; The first monitoring data of each data collection point is extracted according to the preset characteristics to obtain the second monitoring data corresponding to each data collection point.

[0009] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the first recurrent neural network is a long short-term memory network, including a forget gate, an input gate, and an output gate; the first monitoring data sequence corresponding to each data collection point is predicted by the first recurrent neural network to obtain the first prediction data corresponding to each data collection point, including: Performing sequence windowing on the first monitoring data sequence corresponding to each of the data collection points to obtain monitoring data corresponding to a preset number of time steps corresponding to each of the data collection points; For each of the data collection points, the forget gate is used to determine the old state of the unit to be forgotten in the monitoring data corresponding to the current time step; Determining, using the input gate, a candidate state of the unit in the monitoring data corresponding to the current time step; Determine the state of the unit in the current time step according to the old state of the unit to be forgotten in the monitoring data corresponding to the current time step, the candidate state of the unit in the monitoring data corresponding to the current time step, and the state of the unit in the previous time step; Determining the hidden state of the current time step according to the cell state in the current time step using the output gate; The first prediction data corresponding to each of the data collection points is determined according to the hidden state corresponding to the last time step of the preset number of time steps.

[0010] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the method further includes: For any of the data collection points, determining whether to trigger an alarm for the data collection point according to the first prediction data corresponding to the data collection point and a preset alarm threshold; After determining that an alarm of the data collection point is triggered, an alarm management component is used to generate an alarm.

[0011] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the method further includes: Creating a data panel in a data visualization component; the data panel includes at least two monitoring indicators; Determine the first monitoring data sequence corresponding to each of the data collection points as the data source in the data panel; According to each of the monitoring indicators and the data source in the data panel, an indicator prediction curve of the rail transit system is drawn in the data panel.

[0012] In a second aspect, the present invention further provides a device for monitoring the equipment operation status of a rail transit system, the device comprising the following modules: A determination module, configured to determine at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system; the device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; A monitoring module, used for extracting features from the first monitoring data of each data collection point in the rail transit system to obtain second monitoring data of each data collection point; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

[0013] In a third aspect, the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for monitoring the equipment operating status of a rail transit system as described in any one of the above-mentioned methods is implemented.

[0014] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for monitoring the equipment operating status of a rail transit system as described in any one of the above.

[0015] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for monitoring the equipment operating status of a rail transit system.

[0016] The method and device for monitoring the equipment operation status of a rail transit system provided by the present invention determine at least one data collection point based on an equipment list of the rail transit system and a class table of the rail transit system, wherein the equipment list contains attribute information corresponding to at least one equipment of the rail transit system, and the class table contains attribute information corresponding to at least one equipment type of the rail transit system; then, feature extraction is performed on the first monitoring data of each data collection point in the rail transit system to obtain second monitoring data of each data collection point; further, differential processing is performed on the second monitoring data of each data collection point to obtain a first monitoring data sequence corresponding to each data collection point, and according to the first monitoring data sequence corresponding to each data collection point, a first recurrent neural network is used to predict and obtain first predicted data corresponding to each data collection point.

[0017] The present invention first determines at least one data collection point based on the equipment list of the rail transit system and the class table of the rail transit system, that is, determines the data collection points that the rail transit system needs to collect, and performs differential processing on the second monitoring data of each data collection point to obtain a first monitoring data sequence corresponding to each data collection point, and then uses the first recurrent neural network to predict the first predicted data corresponding to each data collection point, thereby improving the monitoring efficiency of the equipment operation status monitoring of the rail transit system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are 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.

[0019] Figure 1 This is one of the flow charts of the method for monitoring the equipment operation status of the rail transit system provided by the present invention.

[0020] Figure 2 This is the second flow chart of the method for monitoring the equipment operating status of the rail transit system provided by the present invention.

[0021] Figure 3 It is a structural schematic diagram of the equipment operation status monitoring device of the rail transit system provided by the present invention.

[0022] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first" and "second" are generally a class, and the number of objects is not limited. For example, the first node can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the front and back associated objects are in a "or" relationship.

[0025] Combine the following Figure 1-Figure 4 The present invention describes a method and a device for monitoring the equipment operating status of a rail transit system.

[0026] Figure 1 FIG. 1 is one of the flow charts of the method for monitoring the equipment operation status of the rail transit system provided by the present invention, such as Figure 1 As shown, the method includes the following: Step 101: Determine at least one data collection point based on a device list and a class table of the rail transit system; the device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; Specifically, the executor of this embodiment is an electronic device, and this embodiment is used to improve the monitoring efficiency of equipment operation status monitoring of the rail transit system.

[0027] First, the data collection point can be determined. For example, based on the equipment list of the rail transit system and the class table of the rail transit system, at least one data collection point is determined. The equipment list contains attribute information corresponding to at least one equipment of the rail transit system, and the class table contains attribute information corresponding to at least one equipment type of the rail transit system.

[0028] Before performing data monitoring, components can be deployed in the system environment. For example, the automated deployment tool Ansible can be used to deploy the monitoring component Prometheus, the time series database IoTDB, the alarm management component Alertmanager, and the data visualization component Grafana in the system environment to ensure the normal operation of each component, thereby providing support for the monitoring and analysis of the operating status of rail transit system equipment.

[0029] Step 102: extracting features from the first monitoring data of each data collection point in the rail transit system to obtain second monitoring data of each data collection point; Specifically, after determining each data collection point, the original monitoring data corresponding to each data collection point, that is, the first monitoring data of each data collection point, can be further obtained. For example, data monitoring is performed through the pre-deployed monitoring component Prometheus. Among them, Prometheus is an open source system monitoring and alert toolkit, which collects and stores indicators (usually time series data) and provides a powerful data query language (PromQL) for users to query and analyze this data. Prometheus is usually used to record the status and performance indicators of machines, software applications, and network devices, and alarms can be configured to notify users when certain indicators exceed preset thresholds.

[0030] After obtaining the original monitoring data corresponding to each data collection point, feature data can be further extracted from the original monitoring data corresponding to each data collection point. For example, feature data corresponding to preset features for predicting abnormal operating conditions of equipment can be extracted from the first monitoring data of each data collection point and used as the second monitoring data of each data collection point.

[0031] Step 103: performing differential processing on the second monitoring data of each data collection point to obtain a first monitoring data sequence corresponding to each data collection point; Specifically, after obtaining the second monitoring data of each data collection point (feature data corresponding to the preset features of each data collection point), the second monitoring data of each data collection point can be further differentially processed to obtain a first monitoring data sequence corresponding to each data collection point.

[0032] Among them, differential processing is used to extract information from time series data with the following purposes: Eliminate trends: Through differential processing, the trend component in the data can be reduced or eliminated; Eliminate seasonality: Seasonal differences help to identify and eliminate seasonal patterns in the data; Data stabilization: Make time series data closer to a stationary series, which is convenient for applying models for analysis and prediction.

[0033] In time series analysis, differential processing is usually used to eliminate non-stationarity in data, that is, trend and seasonal factors. The basic idea of ​​differential processing is to calculate the difference between adjacent observations to generate a new series. The purpose of this new series is to make the data more stable and easier to analyze. The types of differential processing include first-order differential and high-order differential. This embodiment does not specifically limit the type of differential method used.

[0034] Step 104: According to the first monitoring data sequence corresponding to each data collection point, the first predicted data corresponding to each data collection point is obtained by using the first recurrent neural network to predict.

[0035] Specifically, after obtaining the first monitoring data sequence corresponding to each data collection point, the first recurrent neural network can be further used to predict the data of each data collection point according to the sequence to obtain the first predicted data corresponding to each data collection point. The first recurrent neural network is a deep learning network, which can be obtained by training with historical monitoring data and corresponding labels (prediction data).

[0036] The method provided in this embodiment determines at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system, wherein the device list contains attribute information corresponding to at least one device of the rail transit system, and the class table contains attribute information corresponding to at least one device type of the rail transit system; then, feature extraction is performed on the first monitoring data of each data collection point in the rail transit system to obtain second monitoring data of each data collection point; further, differential processing is performed on the second monitoring data of each data collection point to obtain a first monitoring data sequence corresponding to each data collection point, and according to the first monitoring data sequence corresponding to each data collection point, a first recurrent neural network is used to predict and obtain first predicted data corresponding to each data collection point.

[0037] The present invention first determines at least one data collection point based on the equipment list of the rail transit system and the class table of the rail transit system, that is, determines the data collection points that the rail transit system needs to collect, and performs differential processing on the second monitoring data of each data collection point to obtain a first monitoring data sequence corresponding to each data collection point, and then uses the first recurrent neural network to predict the first predicted data corresponding to each data collection point, thereby improving the monitoring efficiency of the equipment operation status monitoring of the rail transit system.

[0038] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, both the equipment list and the class table contain an equipment class code field, and the equipment class code field is used to characterize the equipment type; based on the equipment list of the rail transit system and the class table of the rail transit system, at least one data collection point is determined, including: Cartesian product is performed on the attribute information corresponding to the device class code field in the device list and the attribute information corresponding to the device class code field in the class table to obtain a Cartesian product result; Based on the set corresponding to the Cartesian product result, each data collection point is determined.

[0039] Specifically, in some embodiments, both the device list and the class list contain a device class code field, and the device class code field is used to characterize the device type. For example, the device list is shown in Table 1 below: Table 1:

[0040] Among them, the main fields of the device list include device class code, device number, device description, parent device number, device Internet Protocol (IP) address and port, and device definition type. The above fields can also be understood as the basic attributes of the device.

[0041] For example, the class representation is as shown in Table 2 below: Table 2:

[0042] Among them, the class table includes the following main fields: device class code, attribute name, IO type, alarm description, alarm level, and sampling period of the drive software (milliseconds). The above are common attributes of the device and can be considered as attributes that each device has uniformly. The class table can be preset according to the attributes of the device in the system, or modified and maintained regularly.

[0043] Correspondingly, in some embodiments, the process of determining at least one data collection point in step 102 is implemented by the following steps, including: Perform a Cartesian product of the attribute information corresponding to the device class code field in the device list and the attribute information corresponding to the device class code field in the class table to obtain the Cartesian product result, and determine each data collection point based on the set corresponding to the Cartesian product result. It can be understood that by performing a Cartesian product with the data corresponding to the device class code in the device list and the device class code data in the class table, all the corresponding data collection point data in the point table can be obtained, which is convenient for generating point table data. Cartesian Product refers to the product of two sets, where the elements of each set are paired with each element of the other set. If there are two sets A and B, their Cartesian product is usually expressed as A×B and is defined as the set of all possible ordered pairs (a,b), where a belongs to A and b belongs to B. For example, assume there are two sets: A={1,2};B={x,y}.

[0044] The Cartesian product of A and B is A×B: A×B={(1,x),(1,y),(2,x),(2,y)} Exemplarily, the point table obtained by performing Cartesian product using the data corresponding to the device class code in the device list and the device class code data in the class table is as follows: Table 3: Table 3: Furthermore, the point table data is determined, that is, each data collection point is determined. The subsequent monitoring data collection, preprocessing, analysis, monitoring, and prediction can be based on the point table data. The points in the point table are considered to be transmitted to the server by the device through the transmission protocol. We can classify and process them according to the return value type. For example, VO / VI is a virtual point that can store some cache data according to the intention. AO / AI is a digital point that stores values ​​of Integer, Float, and Double types. DO / DI is a Boolean point, representing the on or off of the switch. The device has a specific device name, and the points (attributes) under the device also have the concept of belonging to a certain device. Through this means, we can define the device excel table, and we can generate the corresponding point excel table through the Cartesian product of the device and the class table by the device and class table. We only need to monitor and process the corresponding points to know the sensor value of the device, so as to determine whether the feedback value exceeds the standard and causes the device to alarm.

[0045] The method provided in this embodiment first performs a Cartesian product on the attribute information corresponding to the device class code field in the device list and the attribute information corresponding to the device class code field in the class table to obtain a Cartesian product result, and then determines each data collection point based on the set corresponding to the Cartesian product result. Furthermore, comprehensive monitoring of the equipment operation status can be achieved by only performing monitoring and processing on the corresponding data collection points, thereby improving the monitoring efficiency of the equipment operation status monitoring of the rail transit system.

[0046] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, feature extraction is performed on first monitoring data of each data collection point in the rail transit system to obtain second monitoring data corresponding to each data collection point, including: Determine preset features; the preset features are features used to predict abnormal operation of the equipment; The first monitoring data of each data collection point is extracted according to the preset characteristics to obtain the second monitoring data corresponding to each data collection point.

[0047] Specifically, in some embodiments, the process of extracting features from the first monitoring data of each data collection point in the rail transit system in step 102 to obtain the second monitoring data corresponding to each data collection point can be implemented in the following manner: First, the preset features are determined. The preset features are features used to predict abnormal operation of equipment. It can be understood that the first monitoring data obtained from each data collection point is the original data of each data collection point. The original data collected in a large-scale rail transit system may contain hundreds of indicators, but the features that can often predict abnormal operation of equipment or play an important role in alarms are concentrated on specific features (i.e. indicators).

[0048] Furthermore, after the preset features for predicting abnormal operation of the equipment are determined, the preset features can be extracted from the original monitoring data to obtain monitoring data corresponding to the preset features, that is, the monitoring data corresponding to the preset features are extracted from the first monitoring data of each data collection point, and the monitoring data corresponding to the preset features of each data collection point are determined as the second monitoring data corresponding to each data collection point. The second monitoring data corresponding to each data collection point is used for subsequent indicator prediction, abnormal condition alarm, etc.

[0049] In the method provided in this embodiment, preset features for predicting abnormal operating conditions of equipment are first determined, and then the first monitoring data of each data collection point is extracted according to the preset features to obtain second monitoring data corresponding to each data collection point, so as to facilitate subsequent indicator prediction based on the second monitoring data corresponding to each data collection point, as well as abnormal condition alarm, thereby improving the monitoring efficiency of equipment operating condition monitoring of the rail transit system.

[0050] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the first recurrent neural network is a long short-term memory network, including a forget gate, an input gate, and an output gate; according to a first monitoring data sequence corresponding to each data collection point, the first recurrent neural network is used to predict and obtain first prediction data corresponding to each data collection point, including: Performing sequence windowing on the first monitoring data sequence corresponding to each data collection point to obtain monitoring data corresponding to a preset number of time steps corresponding to each data collection point; For each data collection point, the forget gate is used to determine the old state that needs to be forgotten in the monitoring data corresponding to the current time step; Using the input gate, determine the candidate state of the unit in the monitoring data corresponding to the current time step; Determine the state of the unit in the current time step according to the old state of the unit to be forgotten in the monitoring data corresponding to the current time step, the candidate state of the unit in the monitoring data corresponding to the current time step, and the state of the unit in the previous time step; Using the output gate, determine the hidden state of the current time step based on the cell state in the current time step; The first prediction data corresponding to each data collection point is determined according to the hidden state corresponding to the last time step in the preset number of time steps.

[0051] Specifically, in some embodiments, the first recurrent neural network is a long short-term memory network (Long Short-Term Memory, LSTM). The design of LSTM allows the network's memory unit to decide whether to store or forget information, and it can learn long-term dependencies. The LSTM network consists of three main gates: 1. Forget Gate: Determines which information should be discarded at the current time step. 2. Input Gate: Determines which new information should be added to the unit state at the current time step. 3. Output Gate: Determines what information to output at the current time step. For each time step, the LSTM unit updates its internal state, which can carry information about the previous time step. This structure of LSTM makes it more effective in processing long sequence data because it can learn which information is important and needs to be retained, and which information is irrelevant and can be forgotten.

[0052] Correspondingly, the specific implementation process of obtaining the first predicted data corresponding to each data collection point by using the first recurrent neural network prediction in step 104 includes the following steps: First, the first monitoring data sequence corresponding to each data collection point is windowed to obtain monitoring data corresponding to a preset number of time steps corresponding to each data collection point. For example, the first monitoring data sequence is divided into monitoring data corresponding to 10 time steps, and then the three gates of the first recurrent neural network are used to learn which information is important time by time step, so as to predict and obtain the first prediction data corresponding to the future moment.

[0053] Then, similar operations are performed for each time step to determine the hidden state of the current time step. It can be understood that the hidden state of the current time step can represent the predicted data at a certain moment in the future. Specifically, for each data collection point, the forget gate is used to determine the old state of the unit in the monitoring data corresponding to the current time step that needs to be forgotten, and the input gate is used to determine the candidate state of the unit in the monitoring data corresponding to the current time step. According to the old state of the unit in the monitoring data corresponding to the current time step that needs to be forgotten, the candidate state of the unit in the monitoring data corresponding to the current time step, and the unit state in the previous time step, the unit state in the current time step is determined; then, the output gate is used to determine the hidden state of the current time step according to the unit state in the current time step.

[0054] Furthermore, first prediction data corresponding to each data collection point is determined according to a hidden state corresponding to a last time step among a preset number of time steps.

[0055] Optionally, the first recurrent neural network is obtained by training the initial recurrent neural network using the historical monitoring data of each data collection point and the prediction data corresponding to the historical monitoring data of each data collection point. The prediction accuracy of the trained first recurrent neural network LSTM is higher.

[0056] The training process of the first recurrent neural network is as follows: Sample data acquisition: store the feature data within the preset time period of each data collection point into the IoTDB time series database; Data segmentation: The feature data within the preset time period of each data collection point is divided into training set and test set.

[0057] Model training: Perform model training on the initial LSTM based on the data in the training set.

[0058] Cross-validation and performance evaluation: Perform cross-validation in the training set to evaluate the performance and generalization ability of the model. Use the test set to evaluate the performance of the model, and the evaluation indicators include Mean Square Error (MSE), Root Mean Square Error (RMSE), etc. Select the best model based on the validation and evaluation results.

[0059] Model parameter adjustment: According to the prediction results, the model parameters are adjusted by using random search for the best model to obtain the first recurrent neural network (that is, the first recurrent neural network LSTM after training).

[0060] The method provided in this embodiment uses a long short-term memory network to predict the first predicted data corresponding to each data collection point based on the first monitoring data sequence corresponding to each data collection point, which can improve the prediction accuracy of the operating status of rail transit system equipment.

[0061] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the method further comprises: For any data collection point, determine whether to trigger an alarm for the data collection point according to the first prediction data corresponding to the data collection point and a preset alarm threshold; After determining the alarm that triggers the data collection point, use the alarm management component to issue an alarm.

[0062] Specifically, in some embodiments, the method further includes an alarm step, and the specific implementation process of the alarm step is as follows: First, for any data collection point, determine whether to trigger an alarm at the data collection point based on the first predicted data corresponding to the data collection point and the preset alarm threshold. The preset alarm threshold can be set based on historical alarm information or determined based on the performance parameters of each device. For example, an alarm is triggered when the device temperature corresponding to data collection point 1 exceeds the preset temperature, and an alarm is triggered when the device pressure value corresponding to data collection point 2 exceeds the preset pressure value.

[0063] Furthermore, after determining the alarm that triggers the data collection point, the alarm management component is used to issue an alarm. Among them, the alarm management component usually refers to a module used to monitor, record and handle abnormal situations in a software system or hardware device. The alarm management component is an example of the AlertManager component. AlertManager is a component in the Prometheus ecosystem that is specifically used to process alarm information from Prometheus. Its core functions include alarm deduplication, grouping, routing, notification, and suppression. The workflow of the AlertManager component is as follows: Alert generation: Prometheus evaluates indicators according to defined rules, generates alerts when conditions are met, and sends them to AlertManager.

[0064] Alert reception: AlertManager receives alerts from Prometheus through the HTTP API.

[0065] Alarm deduplication: Deduplication of received alarms.

[0066] Alarm grouping: Aggregate related alarms into a group based on configured rules.

[0067] Alarm routing: Alarms are sent to different receivers based on the alarm content and predefined routing rules.

[0068] Notification sending: Send alarm notification according to the configuration of the receiver.

[0069] Suppression judgment: If the alarm suppression rule is configured, it will check whether the alarm meets the suppression conditions.

[0070] Logging: Record detailed logs of alarm processing for auditing and troubleshooting.

[0071] In the method provided in this embodiment, for any data collection point, whether to trigger the alarm of the data collection point is determined based on the first prediction data corresponding to the data collection point and the preset alarm threshold. When it is determined that the alarm of the data collection point is triggered, the alarm management component is used to issue an alarm, which facilitates the operation and maintenance personnel to respond in time, thereby improving the monitoring efficiency of the equipment operation status monitoring of the rail transit system.

[0072] According to a method for monitoring the equipment operation status of a rail transit system provided by the present invention, the method further comprises: Create a data panel in the data visualization component; the data panel contains at least two monitoring indicators; Determine the first monitoring data sequence corresponding to each data collection point as the data source in the data panel; According to each monitoring indicator and the data source in the data panel, an indicator prediction curve of the rail transit system is drawn in the data panel.

[0073] Specifically, in some embodiments, the method further includes visually displaying the monitoring data and the prediction results. An example of a process for visually displaying the monitoring data and the prediction results is as follows: First, create a data panel in the data visualization component. The data visualization component is a tool or software that displays data in the form of graphics or charts, making complex data information easier to understand and analyze. Data visualization components such as Grafana, Grafana is an open source data visualization and monitoring platform that is widely used to display time series data. The data visualization component Grafana supports a variety of data sources, such as Prometheus, Elasticsearch, InfluxDB, etc. Deploy Prometheus and Grafana in the system environment, and then call the data visualization component to display monitoring data and prediction results. The type of displayed chart can be selected as needed, such as line charts, bar charts, pie charts, scatter charts, dashboards, etc.

[0074] Furthermore, the first monitoring data sequence corresponding to each data collection point is determined as the data source in the data panel DashBord. It should be noted that the created data panel DashBord contains at least two monitoring indicators, that is, the monitoring data to be obtained, such as multiple indicators corresponding to the first monitoring data sequence corresponding to each data collection point, are displayed to achieve intuitive display of multi-indicator prediction data.

[0075] Furthermore, according to the data sources in each monitoring indicator and data panel, the indicator prediction curve of the rail transit system is drawn in the data panel. By integrating the characteristics of multiple indicators, it is possible to capture the changes in the system status more comprehensively, improve the ability to identify and respond to potential problems in a timely manner, overcome the limitations of the existing rail transit operation status monitoring and prediction system that relies on a single indicator, and further improve the accuracy and reliability of the prediction.

[0076] In the method provided in this embodiment, first, a data panel is created in a data visualization component; the data panel contains at least two monitoring indicators, and then, the first monitoring data sequence corresponding to each data collection point is determined as the data source in the data panel; further, according to each monitoring indicator and the data source in the data panel, an indicator prediction curve of the rail transit system is drawn in the data panel, which overcomes the limitation of the existing rail transit operation status monitoring and prediction system relying on a single indicator, and further improves the accuracy and reliability of the indicator prediction of the rail transit system operation status.

[0077] Figure 2 FIG. 2 is a flow chart of the method for monitoring the equipment operation status of the rail transit system provided by the present invention. Figure 2 As shown, the method includes: Step 201: Collect monitoring data using a monitoring component; Step 202: data preprocessing; Step 203: feature extraction and processing; Step 204: The IO TDB time series database stores the characteristic data; Step 205: dividing the feature data into a training set and a validation set; Step 206: training the long short-term memory network and the spatial state model; Step 207, cross-validation and performance evaluation of the long short-term memory network and the spatial state model; Step 208: Adjust model parameters; Step 209: monitoring prediction and alarm; Step 210: Visually display the prediction results; Step 211: Model maintenance.

[0078] This embodiment provides a multi-index monitoring and trend prediction method for the equipment operation status of a rail transit system that integrates a long short-term memory network LSTM, a spatial state model, and cross-validation technology. Through refined multi-index feature extraction, combined with deep learning and time series analysis, accurate modeling and prediction of the future operation status of the equipment are achieved. In addition, timely alarms are achieved through monitoring thresholds to indicate possible health problems so that operation and maintenance personnel can identify and avoid potential faults in a timely manner.

[0079] The following is a description of the device for monitoring the equipment operating status of a rail transit system provided by the present invention. The device for monitoring the equipment operating status of a rail transit system described below and the method for monitoring the equipment operating status of a rail transit system described above can be referenced to each other.

[0080] Figure 3 Schematic diagram of the structure of the equipment operation status monitoring device for the rail transit system provided by the present invention. Figure 3 As shown, the equipment operation status monitoring device 300 of the rail transit system includes: A determination module 310 is used to determine at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system; the device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; A monitoring module 320, configured to extract features from the first monitoring data of each of the data collection points in the rail transit system to obtain second monitoring data of each of the data collection points; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

[0081] The device provided in this embodiment determines at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system, wherein the device list contains attribute information corresponding to at least one device of the rail transit system, and the class table contains attribute information corresponding to at least one device type of the rail transit system; then, feature extraction is performed on the first monitoring data of each data collection point in the rail transit system to obtain second monitoring data of each data collection point; further, differential processing is performed on the second monitoring data of each data collection point to obtain a first monitoring data sequence corresponding to each data collection point, and according to the first monitoring data sequence corresponding to each data collection point, a first recurrent neural network is used to predict and obtain first predicted data corresponding to each data collection point.

[0082] The present invention first determines at least one data collection point based on the equipment list of the rail transit system and the class table of the rail transit system, that is, determines the data collection points that the rail transit system needs to collect, and performs differential processing on the second monitoring data of each data collection point to obtain a first monitoring data sequence corresponding to each data collection point, and then uses the first recurrent neural network to predict the first predicted data corresponding to each data collection point, thereby improving the monitoring efficiency of the equipment operation status monitoring of the rail transit system.

[0083] According to a device 300 for monitoring equipment operation status of a rail transit system provided by the present invention, both the equipment list and the class table contain equipment class code fields, and the determination module 310 is specifically used to: Perform a Cartesian product of the attribute information corresponding to the device class code field in the device list and the attribute information corresponding to the device class code field in the class table to obtain a Cartesian product result; Based on the set corresponding to the Cartesian product result, each of the data collection points is determined.

[0084] According to a device 300 for monitoring equipment operation status of a rail transit system provided by the present invention, the monitoring module 320 is specifically used for: Determine a preset feature; the preset feature is a feature used to predict abnormal operation of the equipment; The first monitoring data of each data collection point is extracted according to the preset characteristics to obtain the second monitoring data corresponding to each data collection point.

[0085] According to a device 300 for monitoring equipment operation status of a rail transit system provided by the present invention, the first recurrent neural network is a long short-term memory network, including a forget gate, an input gate and an output gate; the monitoring module 320 is further used for: Performing sequence windowing on the first monitoring data sequence corresponding to each of the data collection points to obtain monitoring data corresponding to a preset number of time steps corresponding to each of the data collection points; For each of the data collection points, the forget gate is used to determine the old state of the unit to be forgotten in the monitoring data corresponding to the current time step; Determining, using the input gate, a candidate state of the unit in the monitoring data corresponding to the current time step; Determine the state of the unit in the current time step according to the old state of the unit to be forgotten in the monitoring data corresponding to the current time step, the candidate state of the unit in the monitoring data corresponding to the current time step, and the state of the unit in the previous time step; Determining the hidden state of the current time step according to the cell state in the current time step using the output gate; The first prediction data corresponding to each of the data collection points is determined according to the hidden state corresponding to the last time step of the preset number of time steps.

[0086] According to a device 300 for monitoring equipment operation status of a rail transit system provided by the present invention, the device further includes an alarm module: The alarm module is used to: For any of the data collection points, determining whether to trigger an alarm for the data collection point according to the first prediction data corresponding to the data collection point and a preset alarm threshold; After determining that an alarm of the data collection point is triggered, an alarm management component is used to generate an alarm.

[0087] According to a device 300 for monitoring equipment operation status of a rail transit system provided by the present invention, the device further includes a display module; The display module is used for: Creating a data panel in a data visualization component; the data panel includes at least two monitoring indicators; Determine the first monitoring data sequence corresponding to each of the data collection points as the data source in the data panel; According to each of the monitoring indicators and the data source in the data panel, an indicator prediction curve of the rail transit system is drawn in the data panel.

[0088] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communications interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the equipment operation status monitoring method of the rail transit system, and the method includes: Determine at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system; the device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; Performing feature extraction on the first monitoring data of each of the data collection points in the rail transit system to obtain second monitoring data of each of the data collection points; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

[0089] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0090] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the equipment operation status monitoring method of the rail transit system provided by the above methods, the method includes: Determine at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system; the device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; Performing feature extraction on the first monitoring data of each of the data collection points in the rail transit system to obtain second monitoring data of each of the data collection points; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

[0091] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the equipment operation status monitoring method of the rail transit system provided by the above methods, the method comprising: Determine at least one data collection point based on a device list of the rail transit system and a class table of the rail transit system; the device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; Performing feature extraction on the first monitoring data of each of the data collection points in the rail transit system to obtain second monitoring data of each of the data collection points; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

[0092] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0093] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring the equipment operation status of a rail transit system, characterized in that: include: Determining at least one data collection point based on the equipment list of the rail transit system and the class list of the rail transit system; The device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; Performing feature extraction on the first monitoring data of each of the data collection points in the rail transit system to obtain second monitoring data of each of the data collection points; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

2. The method for monitoring the equipment operation status of a rail transit system according to claim 1, characterized in that: The device list and the class table both contain a device class code field, and the device class code field is used to characterize the device type; The determining at least one data collection point based on the equipment list of the rail transit system and the class list of the rail transit system comprises: Perform a Cartesian product of the attribute information corresponding to the device class code field in the device list and the attribute information corresponding to the device class code field in the class table to obtain a Cartesian product result; Based on the set corresponding to the Cartesian product result, each of the data collection points is determined.

3. The method for monitoring the equipment operation status of a rail transit system according to claim 1, characterized in that: The extracting features of the first monitoring data of each data collection point in the rail transit system to obtain second monitoring data corresponding to each data collection point includes: Determine a preset feature; the preset feature is a feature used to predict abnormal operation of the equipment; The first monitoring data of each data collection point is extracted according to the preset characteristics to obtain the second monitoring data corresponding to each data collection point.

4. The method for monitoring the equipment operation status of a rail transit system according to claim 1, characterized in that: The first recurrent neural network is a long short-term memory network, including a forget gate, an input gate, and an output gate; the first prediction data corresponding to each data collection point is obtained by predicting the first monitoring data sequence corresponding to each data collection point using the first recurrent neural network, including: Performing sequence windowing on the first monitoring data sequence corresponding to each of the data collection points to obtain monitoring data corresponding to a preset number of time steps corresponding to each of the data collection points; For each of the data collection points, the forget gate is used to determine the old state of the unit to be forgotten in the monitoring data corresponding to the current time step; Determining, using the input gate, a candidate state of the unit in the monitoring data corresponding to the current time step; Determine the state of the unit in the current time step according to the old state of the unit to be forgotten in the monitoring data corresponding to the current time step, the candidate state of the unit in the monitoring data corresponding to the current time step, and the state of the unit in the previous time step; Determining the hidden state of the current time step according to the cell state in the current time step using the output gate; The first prediction data corresponding to each of the data collection points is determined according to the hidden state corresponding to the last time step of the preset number of time steps.

5. The method for monitoring the equipment operation status of a rail transit system according to any one of claims 1 to 4, characterized in that: The method further comprises: For any of the data collection points, determining whether to trigger an alarm for the data collection point according to the first prediction data corresponding to the data collection point and a preset alarm threshold; After determining that an alarm of the data collection point is triggered, an alarm management component is used to generate an alarm.

6. The method for monitoring the equipment operation status of a rail transit system according to any one of claims 1 to 4, characterized in that: The method further comprises: Creating a data panel in a data visualization component; the data panel includes at least two monitoring indicators; Determine the first monitoring data sequence corresponding to each of the data collection points as the data source in the data panel; According to each of the monitoring indicators and the data source in the data panel, an indicator prediction curve of the rail transit system is drawn in the data panel.

7. A device for monitoring the equipment operation status of a rail transit system, characterized in that: include: A determination module, configured to determine at least one data collection point based on a list of equipment of the rail transit system and a class list of the rail transit system; The device list includes attribute information corresponding to at least one device of the rail transit system, and the class table includes attribute information corresponding to at least one device type of the rail transit system; A monitoring module, used for extracting features from the first monitoring data of each data collection point in the rail transit system to obtain second monitoring data of each data collection point; Performing differential processing on the second monitoring data of each of the data collection points to obtain a first monitoring data sequence corresponding to each of the data collection points; According to the first monitoring data sequence corresponding to each of the data collection points, the first prediction data corresponding to each of the data collection points is obtained by using the first recurrent neural network prediction.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for monitoring the equipment operating status of the rail transit system as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for monitoring the equipment operating status of a rail transit system as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for monitoring the equipment operating status of a rail transit system as claimed in any one of claims 1 to 6 is implemented.

Images