Method and system for improving HSE comprehensive management efficiency of gas collection warehouse station
Through intelligent sensors and big data analysis technology, the information islands and data inconsistency in HSE management of gas storage stations were solved, and a decision support system was established, which realized the unified management and sharing of HSE data of gas storage stations, improving management efficiency and security.
Patent Information
- Application Number
- CN202411052601.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The HSE comprehensive management method of traditional gas storage stations has problems such as information islands, inconsistent data, and inefficient management, making it difficult to achieve comprehensive data analysis and scientific decision-making.
Intelligent sensors are used for real-time data acquisition, Hadoop distribution processing technology is used for data storage and calculation, data mining is carried out through data cleaning, clustering analysis and random forest regression models, decision support systems are established, and scientific suggestions are provided to improve management efficiency.
It realizes unified management and sharing of HSE data at the gas storage station, breaks information silos, improves the scientificity and security of management, promptly detects abnormal situations and warns, and improves the level of safety and environmental protection.
Smart Images

Figure CN118940959A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of gas-gathering depot station management, and in particular relates to a method and a system for improving the HSE comprehensive management efficiency of a gas-gathering depot station. Background Art
[0002] With the rapid development of China's energy industry, natural gas, as an important component of clean energy, has an increasing demand; as a key link in the transportation and storage of natural gas, gas gathering stations bear important functions and responsibilities. Since gas gathering stations involve flammable and explosive substances and are highly risky, they have high requirements for safety, environmental protection and occupational health management (HSE). However, the traditional HSE comprehensive management method for gas gathering stations has problems, such as information islands, inconsistent data, and low management efficiency.
[0003] First, because the HSE data of gas-gathering depots are scattered in different departments and systems, the phenomenon of information islands is serious, making it difficult to conduct comprehensive data analysis and management; second, due to the manual operation of data collection and collation, the accuracy and timeliness of the data cannot be guaranteed, which affects the scientificity and accuracy of management decisions; in addition, the traditional HSE management method relies on paper documents and manual feedback, which is inefficient and unable to respond to and handle emergencies in a timely manner. Therefore, a method and system for improving the comprehensive management efficiency of HSE in gas-gathering depots is proposed. Summary of the invention
[0004] In view of the defects of the prior art, the present invention provides a method and system for improving the comprehensive HSE management efficiency of gas gathering stations. By collecting and analyzing data from intelligent sensors, abnormal situations and potential risks can be discovered in a timely manner, and an alarm can be issued through an early warning system to remind relevant personnel to take corresponding measures, thereby avoiding accidents.
[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows: A method for improving the HSE comprehensive management efficiency of a gas gathering station comprises the following steps:
[0006] (2) Data collection: Use sensors and monitoring equipment to collect real-time data at the gas gathering station, including gas concentration, pressure, temperature, and liquid level elevation after gas liquefaction;
[0007] (2) Data storage and processing: The amount of sensor data is large, so Hadoop’s distributed processing technology is used for the storage and calculation of big data;
[0008] (3) Data cleaning and organization: Clean and organize the collected data, remove outliers and duplicate data, ensure the accuracy and consistency of the data, and use convolutional neural networks to output feature data;
[0009] (4) Data mining and modeling: Cluster analysis is used to establish a random forest regression model; characteristic data from gas collection stations are obtained to predict various characteristic data, which are then compared with real data. When abnormal data is found, early warning is performed to identify problems and make improvements in a timely manner;
[0010] (5) Decision support analysis: Based on the data analysis results, a decision support system is established to provide scientific basis and optimization suggestions for managers.
[0011] Furthermore, in step (3), linear interpolation is used to repair missing data with smaller intervals.
[0012] Where: x t and x t+j are the storage station parameters sampled at time t and time t+j respectively; x t+1 The missing parameter at t+i.
[0013] Furthermore, in step (3), the mean smoothing method is used to process the distorted data.
[0014] or
[0015] In the formula, and are the thresholds of adjacent sampling data errors.
[0016] Furthermore, in step (3), the convolutional neural network input x is characteristic data of gas concentration, pressure, temperature and liquid level elevation after gas liquefaction, and the output y is input characteristic data for clustering analysis to establish a random forest regression model.
[0017] Furthermore, let the network input x be M-dimensional: the hidden layer has K nodes in total, the output is L-dimensional, and the output of the output layer node k is:
[0018]
[0019] Where: jk is the adjustment weight from the hidden layer to the output layer; θ k is the bias of the output layer node k; y k As a response to the corresponding input signal, it is output to the workspace as an important regulatory variable according to different backgrounds.
[0020] Furthermore, in step (4), the method of the gas collection station characteristic data prediction process is as follows:
[0021] A. After the original data is cleaned of abnormal data and operation process data, N samples are selected;
[0022] B. Using cluster analysis to extract features from samples to obtain characteristic parameters, the characteristic parameters include storage station capacity, pressure, storage station inlet and outlet pipe diameters, gas flow, and inlet and outlet gas efficiency;
[0023] C. Use a standardized method to non-dimensionalize the feature parameters and obtain a data set for the random forest regression model;
[0024] D. randomly split the data set into a training set and a test set, wherein: the training set (70% of the data set) is used to train the random forest regression model, and the test set (30% of the data set) is used to evaluate the prediction effect of the random forest regression model;
[0025] E. Use the test set to test the random forest regression model;
[0026] F. Use the random forest regression model to make early warning judgments based on the real-time operation data of the gas collection station.
[0027] Furthermore, the random forest regression model was tested using the mean determination coefficient (R 2 ), root mean square error (RMSE), and relative root mean square error (rRMSE) were used as evaluation indicators to evaluate the prediction effect of the random forest regression model, and the minimum number of leaves and the number of decision trees in the random forest were reasonably selected until the R of the random forest regression model was 2 Greater than 0.999.
[0028] A system comprising a method for improving the comprehensive HSE management efficiency of a gas gathering depot station comprises a sensor module, a monitoring module, a data acquisition module, a data processing module, a decision support module and a controller module, wherein the sensor module, the monitoring module, the data acquisition module, the data processing module, the decision support module and the controller module are electrically connected.
[0029] Furthermore, the data acquisition module is responsible for receiving the real-time environmental parameter data of the gas collection station acquired by the sensor module and the monitoring module, and transmitting it to the data processing module; the data processing module receives and processes the environmental parameter data transmitted by the data acquisition module.
[0030] Furthermore, the data processing module includes a data storage module and a data analysis module. The data storage module adopts Hadoop's distributed processing technology, and the data analysis module analyzes and mines the data stored in the data storage module.
[0031] Compared with the prior art, the present invention has the following beneficial effects:
[0032] The present invention utilizes big data analysis technology to conduct in-depth analysis and mining of the HSE data of gas-gathering depots; through data cleaning, mining and modeling, potential problems and room for improvement can be discovered; a decision support system is established to provide scientific basis and optimization suggestions for managers, helping them to formulate reasonable management strategies and measures to improve the safety and environmental protection level of gas-gathering depots; and by establishing an integrated data platform, unified management and sharing of HSE data of gas-gathering depots can be achieved, breaking the traditional information islands and enabling seamless connection and intercommunication of data between departments and systems. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 A flow chart of the method for improving the HSE comprehensive management efficiency of a gas gathering station according to the present invention;
[0034] Figure 2 This is a system structure diagram for improving the HSE comprehensive management efficiency of gas-gathering storage stations in the present invention;
[0035] Figure 3 This is a structural diagram of the radial basis function neural network tracking model of the present invention. DETAILED DESCRIPTION
[0036] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.
[0037] like Figure 1 As shown, a method for improving the HSE comprehensive management efficiency of a gas gathering station includes the following steps:
[0038] (1) Data collection: Use sensors and monitoring equipment to collect real-time data at the gas gathering station, including gas concentration, pressure, temperature, and liquid level after gas liquefaction;
[0039] (2) Data storage and processing: The amount of sensor data is large, so Hadoop’s distributed processing technology is used for the storage and calculation of big data;
[0040] The data of gas collection stations are mainly collected by sensors. As time goes by, the amount of data will become larger and larger. For the storage and calculation of big data, Hadoop-based distributed processing technology is preferred. Hadoop is a parallel programming model and computing framework for processing massive data. It is used for parallel computing of large-scale data sets, and it can easily write distributed parallel programs. Hadoop consists of many elements, the bottom of which is Hadoop Distributed File System (HDFS), which stores files on all storage nodes in the Hadoop cluster. The upper layer is the MapReduce engine, which consists of JobTrackers and Task'Trackers.
[0041] (3) Data cleaning and organization: Clean and organize the collected data, remove outliers and duplicate data, ensure the accuracy and consistency of the data, and use convolutional neural networks to output feature data;
[0042] During the data collection process, aging or performance degradation of sensor equipment, line failure or interference during data transmission may cause data loss or distortion. During the data collection process, interference such as uneven distribution of user demand and insufficient operation time of the warehouse station will inevitably cause abnormal fluctuations in the monitored data. If this data is used directly for processing, early warning and prediction, it will seriously affect the reliability of early warning and prediction, leading to misjudgment.
[0043] Therefore, in the process of sampling data initialization, it is very important to perform noise reduction and repair processing on the data to eliminate interference data. For missing data, linear interpolation can be used to repair missing data with a small time interval:
[0044]
[0045] Where: x t and x t+j are the storage station parameters sampled at time t and time t+j respectively; x t+1 The missing parameter at t+i.
[0046] For distorted data, since the data is continuously sampled and transmitted back in sequence, the data sampled in adjacent periods will not change suddenly. If the range of sampled data at a certain moment is greater than or less than 10% of the monitored value before and after, the sampled data is considered distorted. The mean smoothing method can be used:
[0047] or
[0048] In the formula, and are the thresholds of adjacent sampling data errors.
[0049] like Figure 3 As shown, assume that the network input x is M-dimensional: x = [x 1 ,x 2 ,...,x M ] T The hidden layer has a total of K nodes, the output y is L-dimensional, the length of the input and output sample pairs is N, and the action function of the radial basis network hidden layer nodes is the Gaussian basis function:
[0050]
[0051] The input data vector is mapped to the hidden layer, and the output of the hidden layer node j is:
[0052]
[0053] Where: δ is the standardization constant of the hidden layer nodes, c is the center vector of the Gaussian function of the hidden layer nodes, c j =[c j1 ,c j2 ,...,c jM ] T
[0054] The hidden layer to output layer of the RBF network realizes a linear mapping of the conversion dimension, that is, the output of the output layer node k is:
[0055]
[0056] Where: jk is the adjustment weight from the hidden layer to the output layer; θ k is the bias of the output layer node k. k As a response to the corresponding input signal, it is output to the workspace as an important regulatory variable according to different backgrounds.
[0057] (4) Data mining and modeling: Cluster analysis is used to establish a random forest regression model; characteristic data from gas collection stations are obtained to predict various characteristic data, which are then compared with real data. When abnormal data is found, early warning is performed to identify problems and make improvements in a timely manner;
[0058] The prediction process of gas collection station characteristic data based on cluster analysis and random forest regression model can be divided into the following steps:
[0059] A. Obtain historical operation data: including gas concentration, temperature, pressure and liquid level elevation after gas liquefaction in the gas gathering station. After cleaning abnormal data and operation process data (such as data generated when the station is shut down), select N samples;
[0060] B. Cluster analysis is used to extract features from historical operation data to obtain characteristic parameters. The main characteristic parameters are: storage station capacity, pressure, storage station inlet and outlet pipeline diameter, gas flow, and inlet and outlet gas efficiency.
[0061] C. Due to the differences in units and orders of magnitude between parameters, a standardized method was used to non-dimensionalize the samples to obtain data sets for the random forest regression model. These data sets were randomly split into training sets and test sets, where: the training set (70% of the data set) was used to train the random forest regression model, and the test set (30% of the data set) was used to evaluate the prediction effect of the random forest regression model.
[0062] D. Optimize the random forest regression model: Use the test set to test the random forest regression model, using the mean determination coefficient (R 2 ), root mean square error (RMSE), and relative root mean square error (rRMSE) were used as evaluation indicators to evaluate the prediction effect of the random forest regression model. The minimum number of leaves and the number of decision trees of the random forest were reasonably selected until the R2 of the random forest regression model was greater than 0.999.
[0063] E. Use the random forest regression model to make early warning judgments based on the real-time operation data of the gas collection station.
[0064] Coefficient of determination (R 2 ), root mean square error (RMSE), and relative root mean square error (rRMSE) formulas are as follows:
[0065]
[0066] (5) Decision support analysis: Based on the data analysis results, a decision support system is established to provide scientific basis and optimization suggestions for managers.
[0067] Establish a decision support system to provide managers with scientific basis and optimization suggestions, help them formulate reasonable management strategies and measures, and improve the safety and environmental protection level of gas gathering stations.
[0068] like Figure 2 As shown, a system for improving the comprehensive HSE management efficiency of a gas gathering station is included, including a sensor module, a monitoring module, a data acquisition module, a data processing module, a decision support module and a controller module. The sensor module, the monitoring module, the data acquisition module, the data processing module, the decision support module and the controller module are electrically connected.
[0069] The data acquisition module is responsible for receiving the real-time environmental parameter data of the gas collection station obtained by the sensor module and the monitoring module, and transmitting it to the data processing module; the data processing module receives and processes the environmental parameter data transmitted by the data acquisition module.
[0070] The data processing module includes a data storage module and a data analysis module. The data storage module adopts Hadoop's distributed processing technology, and the data analysis module analyzes and mines the data stored in the data storage module.
[0071] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.
Claims
1. A method for improving the HSE comprehensive management efficiency of a gas gathering station, characterized in that: The following steps are involved: (1) Data collection: Use sensors and monitoring equipment to collect real-time data at the gas gathering station, including gas concentration, pressure, temperature, and liquid level after gas liquefaction; (2) Data storage and processing: The amount of sensor data is large, so Hadoop’s distributed processing technology is used for the storage and calculation of big data; (3) Data cleaning and organization: Clean and organize the collected data, remove outliers and duplicate data, ensure the accuracy and consistency of the data, and use convolutional neural networks to output feature data; (4) Data mining and modeling: Cluster analysis is used to establish a random forest regression model; the characteristic data of the gas collection station is obtained to predict the characteristic data, and then compared with the real data. When abnormal data is found, early warning and judgment are made, and problems are discovered and improved in a timely manner; (5) Decision support analysis: Based on the data analysis results, a decision support system is established to provide scientific basis and optimization suggestions for managers.
2. A method for improving the HSE comprehensive management efficiency of a gas gathering station according to claim 1, characterized in that: In step (3), linear interpolation is used to repair missing data with small intervals. Where: x t and x t+ j are the storage station parameters sampled at time t and time t+j respectively; x t+1 The missing parameter for t+i.
3. A method for improving the HSE comprehensive management efficiency of a gas gathering station according to claim 1, characterized in that: In step (3), the mean smoothing method is used to process the distorted data. In the formula, and are the thresholds of adjacent sampling data errors.
4. A method for improving the HSE comprehensive management efficiency of a gas gathering station according to claim 1, characterized in that: In step (3), the convolutional neural network input x is the characteristic data of gas concentration, pressure, temperature and liquid level elevation after gas liquefaction, and the output y is the input characteristic data for clustering analysis to establish a random forest regression model.
5. A method for improving the HSE comprehensive management efficiency of a gas gathering station according to claim 4, characterized in that: Assume that the network input x is M-dimensional: the hidden layer has K nodes, the output is L-dimensional, and the output of the output layer node k is: Where: jk is the adjustment weight from the hidden layer to the output layer; θ k is the bias of the output layer node k; y k As a response to the corresponding input signal, it is output to the workspace as an important regulatory variable according to different backgrounds.
6. A method for improving the HSE comprehensive management efficiency of a gas gathering station according to claim 1, characterized in that: In step (4), the method of the gas collection station characteristic data prediction process is as follows: A. After the original data is cleaned of abnormal data and operation process data, N samples are selected; B. Using cluster analysis to extract features from samples to obtain characteristic parameters, the characteristic parameters include storage station capacity, pressure, storage station inlet and outlet pipe diameters, gas flow, and inlet and outlet gas efficiency; C. Use a standardized method to non-dimensionalize the feature parameters and obtain a data set for the random forest regression model; D. randomly split the data set into a training set and a test set, wherein: the training set (70% of the data set) is used to train the random forest regression model, and the test set (30% of the data set) is used to evaluate the prediction effect of the random forest regression model; E. Use the test set to test the random forest regression model; F. Use the random forest regression model to make early warning judgments based on the real-time operation data of the gas collection station.
7. A method for improving the HSE comprehensive management efficiency of a gas gathering station according to claim 1, characterized in that: The random forest regression model was tested using the mean determination coefficient (R 2 ), root mean square error (RMSE), and relative root mean square error (rRMSE) were used as evaluation indicators to evaluate the prediction effect of the random forest regression model, and the minimum number of leaves and the number of decision trees in the random forest were reasonably selected until the R of the random forest regression model was 2 Greater than 0.
999.
8. A system for realizing the method for improving the HSE comprehensive management efficiency of a gas gathering station as described in any one of claims 1 to 7, characterized in that: It includes a sensor module, a monitoring module, a data acquisition module, a data processing module, a decision support module and a controller module. The sensor module, the monitoring module, the data acquisition module, the data processing module, the decision support module and the controller module are electrically connected.
9. A system for improving the HSE comprehensive management efficiency of a gas-gathering station according to claim 8, characterized in that: The data acquisition module is responsible for receiving the real-time environmental parameter data of the gas collection station acquired by the sensor module and the monitoring module, and transmitting it to the data processing module; the data processing module receives and processes the environmental parameter data transmitted by the data acquisition module.
10. A system for improving the HSE comprehensive management efficiency of a gas-gathering station according to claim 9, characterized in that: The data processing module includes a data storage module and a data analysis module. The data storage module adopts the distributed processing technology of Hadoop, and the data analysis module analyzes and mines the data stored in the data storage module.
Citation Information
Patent Citations
Working site safety check improvement method and system based on data mining
CN104820907A
Nuclear power unit electric power prediction method based on clustering analysis and random forest regression
CN111242345A
Intelligent data processing method and system applied to fishery management
CN117406593A
County garbage transfer station leachate treatment equipment management system based on big data
CN118297577A
Equipment modeling method suitable for discrete point location table
CN118395290A
Cited By
Evaluation system and method for group enterprise HSE management
CN119624368A
