Petri network-based power production operation safety supervision method

Abstract

The invention relates to the technical field of geological monitoring, in particular to an electric power production operation safety supervision method based on a Petri network. According to the invention, the problems of low efficiency, incomplete coverage, data dispersion and response lag of an existing supervision mode can be solved, and the comprehensiveness, accuracy and real-time performance of power production operation safety supervision are improved.

Description

Technical Field

[0001] This invention relates to the field of geological monitoring technology, and in particular to a method for safety supervision of power production operations based on Petri nets. Background Technology

[0002] Power production operations are a core component in ensuring a stable energy supply for society. These operations encompass a variety of complex scenarios, including working at heights, operating on high-voltage lines, and equipment maintenance, and face multiple safety risks such as improper personnel operation, aging equipment, and sudden environmental changes. Once a safety accident occurs, it can not only cause equipment damage and power outages but also potentially lead to casualties, resulting in serious losses to the social economy and the safety of people's lives and property. Therefore, safety supervision of power production operations is of paramount importance.

[0003] Current safety supervision of power production operations mainly relies on a combination of traditional manual inspections and simple automated monitoring, which has many shortcomings: First, manual inspections are inefficient and have limited coverage, failing to comprehensively and in real-time monitor potential risks throughout the entire operation process. They also lack the ability to dynamically analyze risk evolution in complex environments, leading to inaccurate risk identification. Second, various data in the current operation process are often scattered across different systems or storage media, with inconsistent data formats and standards, resulting in difficulties in data integration and low utilization, thus affecting the accuracy of overall risk assessment. Third, existing methods lack real-time response capabilities, failing to quickly trigger alarms and provide effective risk response plans in the early stages of a risk, often missing the best opportunity for intervention and causing serious consequences. Summary of the Invention

[0004] To address the aforementioned problems, the present invention aims to provide a Petri net-based method for supervising the safety of power production operations, thereby resolving the issues of low efficiency, incomplete coverage, fragmented data, and delayed response in existing supervision methods, and improving the comprehensiveness, accuracy, and real-time performance of power production operation safety supervision.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A Petri net-based method for safety supervision of power production operations includes the following steps: Step S1: Collect on-site data and transmit it to a server. Remove outliers using a cleaning algorithm, and then store the standardized data in a unified database. Step S2: Decompose the power operation process into sub-processes and stages. Define storage locations representing operation status and transitions representing events. Connect storage locations and transitions using directed arcs based on operation logic. Combine historical accident data and expert experience to set triggering conditions and risk weights for transitions, and complete the construction of a Petri net model for power operation safety supervision. Step S3: Run the Petri net model in real time, monitor storage location tokens and transition triggering. When a transition is triggered, calculate the risk level based on risk weights and current operation data, and analyze the risk propagation path and impact range. Step S4: Automatically generate supervision tasks based on risk levels, monitor violations and safety hazards in the operation, trigger an alarm once an anomaly is detected, and retrieve risk response strategies from the Petri net model.

[0007] More preferably, the specific steps of the standardization process in step S1 are as follows: S11, determine whether the data type is numerical or video; S12, when the data type is numerical, use a normalization method to map the data to the [0,1] interval; when the data type is video, first use a cleaning algorithm to remove noise, and then use the H.264 encoding standard for compression, with a compression ratio set to 50:1.

[0008] More preferably, the normalization method formula is as follows: Where Xnorm is the normalized data, X is the original data, Xmin is the minimum value in the original data, and Xmax is the maximum value in the original data.

[0009] More preferably, in step S1, partitioning table technology is used to partition and store the aligned data according to the data acquisition time to form a job dataset.

[0010] More preferably, the risk weight determination in step S2 adopts the weight analysis method, in which experts in the field of power safety score the risk factors of each change, construct a judgment matrix, and calculate the risk weight value of each change.

[0011] More preferably, in step S3, the risk level calculation formula is: Where R represents the risk level of the current operation. Let i be the risk weight of the i-th transition. Let be the real-time risk value of the i-th transition (ranging from 0 to 1), and n be the number of transitions triggered in the current workflow. The risk level is divided into four levels: low, medium, high, and very high, with corresponding numerical ranges of 0-0.3, 0.3-0.6, 0.6-0.8, and 0.8-1.0, respectively.

[0012] More preferably, the specific steps for automatically generating the inspection task in step S4 are as follows: S41, read the risk level calculated in step S3 and judge it; S42, if the risk level is low, conduct routine inspections within 24 hours using remote video monitoring and regular on-site spot checks to check the equipment operating status and the basic operating procedures of the operators; if the risk level is medium, conduct key inspections of the work links with potential risks within 8 hours using remote video monitoring and on-site inspections every 2 hours to verify the risk situation; if the risk level is high, conduct a comprehensive inspection of high-risk links immediately within 2 hours using real-time remote monitoring and immediate on-site inspections, and stop the operation with serious risks; if the risk level is extremely high, conduct a thorough investigation of all work links within 30 minutes using real-time remote monitoring and immediate on-site investigations, and stop all operations and evacuate the operators to a safe area.

[0013] The present invention has the following beneficial effects:

[0014] 1. This invention replaces some manual inspection work by automating data collection and real-time operation of Petri net models, achieving full-process monitoring without blind spots and greatly improving supervision efficiency;

[0015] 2. This invention solves the problems of data dispersion and inconsistent formats by standardizing data processing and centralizing storage, and determines risk weights by combining expert experience and historical data, making risk level calculation more in line with actual operation scenarios;

[0016] 3. The present invention is based on the dynamic risk analysis of the Petri net model, which can quickly identify risks and trigger supervision tasks in a graded manner, thus buying time for risk disposal and reducing the probability of accidents and losses.

[0017] 4. By analyzing the risk propagation path and scope of impact, this invention can not only address current risks in a timely manner, but also provide data support for predicting risks in subsequent operations and optimize safety supervision strategies. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for safety supervision of power production operations based on Petri nets. Detailed Implementation

[0019] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] A Petri net-based method for safety supervision of power production operations includes the following steps: Step S1: Collect on-site data, including equipment operating parameters (such as numerical data like voltage, current, and temperature), operator operation videos (video data), and environmental data (such as numerical data like wind speed and humidity), and transmit them to a server. Outliers are removed using a cleaning algorithm to ensure data quality, and the standardized data is then stored in a unified database. Step S2: Decompose the power operation process into sub-processes and stages, define locations representing the operation status (representing the specific state of the operation) and transitions representing events (representing events that trigger changes in the operation status). Connect locations and transitions using directed arcs based on the operation logic. Combine historical accident data and expert experience to set triggering conditions and risk weights for transitions, completing the Petri net model construction for power operation safety supervision. Step S3: Input the real-time operation data from the unified database in Step S1 into the constructed Petri net model. The Petri net model is started and runs in real time, monitoring the storage tokens (the existence of the token indicates that the corresponding operation status is effective) and the triggering of transitions. When a transition is triggered, the risk level is calculated by combining the risk weight and the current operation data, and the risk propagation path and impact range are analyzed. Step S4: Automatically generate supervision tasks according to the risk level, monitor violations and safety hazards in the operation, and trigger an alarm once an anomaly is determined, while retrieving the risk response strategy from the Petri net model.

[0021] The specific steps of the standardization process in step S1 are as follows: S11, determine whether the data type is numerical or video; S12, when the data type is numerical, use a normalization method to map the data to the [0,1] interval; when the data type is video, first use a cleaning algorithm to remove noise, and then use the H.264 encoding standard for compression, with a compression ratio set to 50:1, to reduce storage space usage while ensuring video quality.

[0022] The specific formula for the normalization method is as follows: Where Xnorm is the normalized data, X is the original data, Xmin is the minimum value in the original data, and Xmax is the maximum value in the original data.

[0023] In step S1, partitioning table technology is used to partition and store the aligned data according to the data acquisition time to form a job dataset. This operation can improve the efficiency of data query and access.

[0024] In step S2, the risk weights are determined using a weighted analysis method. Experts in the field of power safety score the risk factors of each change and construct a judgment matrix to calculate the risk weight value of each change.

[0025] In step S3, the risk level calculation formula is as follows: Where R represents the risk level of the current operation. Let i be the risk weight of the i-th transition. Let be the real-time risk value of the i-th transition (calculated based on the actual situation of the corresponding work process, such as equipment operating parameters and personnel operation behavior, through preset rules or models, with a value range of 0-1), and n be the number of transitions triggered in the current work process; the risk level is divided into four levels: low, medium, high, and extremely high, with corresponding value ranges of 0-0.3, 0.3-0.6, 0.6-0.8, and 0.8-1.0, respectively.

[0026] In step S4, the specific steps for automatically generating the supervision task are as follows: S41, read the risk level calculated in step S3 and judge it; S42, if the risk level is low, conduct routine inspections within 24 hours using remote video monitoring and regular on-site spot checks to check the equipment operating status and the basic operating procedures of the operators; if the risk level is medium, conduct key inspections of the work links with potential risks within 8 hours using remote video monitoring and on-site inspections every 2 hours to verify the risk situation; if the risk level is high, conduct a comprehensive inspection of high-risk links immediately within 2 hours using real-time remote monitoring and immediate on-site inspections, and at the same time stop the operation with serious risks; if the risk level is extremely high, conduct a thorough investigation of all work links within 30 minutes using real-time remote monitoring and immediate on-site investigations, and at the same time completely stop the operation and evacuate the operators to a safe area.

[0027] This invention automates the entire process from data collection to final decision support, breaking the traditional isolation and lack of information sharing among different stages of safety supervision. By establishing a scientific risk weight calculation model and risk level assessment formula, it achieves precise quantification of operational risks and clearly understands the risk propagation process by tracking token flow paths. Multi-source data collection ensures the comprehensiveness and accuracy of the data, while image recognition and video analysis technologies improve the efficiency of identifying violations and safety hazards. The collaborative management model combining offline and online methods enables comprehensive and thorough supervision of the work site.

[0028] The above description is merely a specific embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for safety supervision of power production operations based on Petri nets, characterized in that: Includes the following steps: Step S1: Collect on-site data and transmit it to the server. Remove outliers using a cleaning algorithm, and then store the standardized data in a unified database. Step S2: Decompose the power operation process into sub-processes and links, define the storage locations representing the operation status and the transitions representing the events, connect the storage locations and transitions with directed arcs according to the operation logic, and set the triggering conditions and risk weights of the transitions by combining historical accident data and expert experience, and complete the construction of the Petri net model for power operation safety supervision. Step S3: Run the Petri net model in real time to monitor the token and transition triggering status. When a transition is triggered, calculate the risk level by combining the risk weight and the current operation data, and analyze the risk propagation path and impact range. Step S4: Automatically generate inspection tasks based on risk level, monitor violations and safety hazards in operations, trigger alarms once an anomaly is detected, and retrieve risk response strategies from the Petri net model.

2. The method for safety supervision of power production operations based on Petri nets according to claim 1, characterized in that: The specific steps of the standardization process in step S1 are as follows: S11. Determine whether the data type is numeric or video. S12. When the data type is numeric, the normalization method is used to map the data to the [0,1] interval. When the data type is video, the cleaning algorithm is first used to remove noise, and then the H.264 encoding standard is used for compression, with the compression ratio set to 50:

1.

3. The method for safety supervision of power production operations based on Petri nets according to claim 2, characterized in that: The specific formula for the normalization method is as follows: ； Where Xnorm is the normalized data, X is the original data, Xmin is the minimum value in the original data, and Xmax is the maximum value in the original data.

4. The method for safety supervision of power production operations based on Petri nets according to claim 1, characterized in that: In step S1, partitioning table technology is used to partition and store the aligned data according to the data acquisition time to form a job dataset.

5. The method for safety supervision of power production operations based on Petri nets according to claim 1, characterized in that: In step S2, the risk weights are determined using a weighted analysis method. Experts in the field of power safety score the risk factors of each change and construct a judgment matrix to calculate the risk weight value of each change.

6. The method for safety supervision of power production operations based on Petri nets according to claim 1, characterized in that: In step S3, the risk level calculation formula is as follows: ； Where R represents the risk level of the current operation. Let i be the risk weight of the i-th transition. Let n be the real-time risk value of the i-th transition (ranging from 0 to 1), and n be the number of transitions that have been triggered in the current workflow. The risk level is divided into four levels: low, medium, high, and very high, with corresponding numerical ranges of 0-0.3, 0.3-0.6, 0.6-0.8, and 0.8-1.0, respectively.

7. The method for safety supervision of power production operations based on Petri nets according to claim 6, characterized in that: In step S4, the specific steps for automatically generating the inspection task are as follows: S41. Read the risk level calculated in step S3 and make a judgment on it; S42. If the risk level is low, routine inspections shall be conducted within 24 hours by means of remote video monitoring and regular on-site spot checks to check the equipment operating status and the basic operating procedures of the operators. If the risk level is medium, remote video monitoring and on-site inspections every 2 hours will be conducted within 8 hours to focus on checking the operational processes with potential risks and verify the risk situation. If the risk level is high, a comprehensive inspection of the high-risk process will be carried out immediately within 2 hours by means of real-time remote monitoring and immediate on-site inspection, and operations with serious risks will be stopped. If the risk level is extremely high, all operational processes will be thoroughly investigated within 30 minutes through real-time remote monitoring and immediate on-site inspection. At the same time, all operations will be stopped and personnel will be evacuated to a safe area.

Images