On-site special operation risk index assessment method and system for thermal power plant based on AI risk matrix model
By automatically calculating the risk index of on-site operations in thermal power plants using an AI risk matrix model, the problems of lag and misjudgment in traditional manual assessments have been solved, enabling rapid and accurate risk assessment and early warning, and improving the efficiency and accuracy of safety management in thermal power plants.
Patent Information
- Application Number
- CN202511806561.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-03
- Publication Date
- 2026-03-03
AI Technical Summary
Traditional safety management in thermal power plants relies on manual experience, which leads to delayed risk assessment, heavy workload, and a high risk of error. It also lacks the ability to perceive dynamic changes in risks and cannot achieve rapid and accurate risk assessment and early warning.
By employing an AI risk matrix model, through data acquisition, model building, and risk coefficient correction, the risk index of special on-site operations is automatically calculated, and a five-color heat map is generated for visual early warning.
It enables rapid and accurate assessment of on-site operational risks in thermal power plants, reduces manual statistical time, improves the efficiency and accuracy of safety management, and allows for timely detection and early warning of potential risks.
Smart Images
Figure CN121599480A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of safety production management technology for thermal power plants, specifically to a method and system for assessing the risk index of special operations on-site in thermal power plants based on an AI risk matrix model. Background Technology
[0002] Power safety is a crucial guarantee for the stable operation of the national economy. Thermal power plants operate in complex environments, involving various special operations such as high temperature, high pressure, high altitude, and hot work. Safety risks are multi-source, cumulative, and dynamic. Traditional safety management relies primarily on manual experience and operational risk assessments based on regulations such as the "Power Safety Work Regulations," which has the following significant shortcomings: First, risk assessment is delayed. The daily task list relies on manual extraction from the "two tickets" (work ticket and operation ticket) system, which makes it difficult to quickly and accurately match with the risk matrix, resulting in a lag in safety supervision. Secondly, the workload is large and error-prone. Using manual methods for job safety analysis (JSA) and risk level assessment is time-consuming, labor-intensive, highly subjective, and prone to oversights and misjudgments.
[0003] Secondly, there are management "blind spots." Existing methods fail to effectively utilize multi-source information such as historical accident cases and real-time meteorological data, lack the ability to perceive dynamic changes in risks, and cannot form data-driven proactive early warnings.
[0004] Therefore, there is an urgent need for a method and system that can automate and intelligently assess risks in order to improve the efficiency and accuracy of safety management in thermal power plants. Summary of the Invention
[0005] The purpose of this invention is to provide a method and system for assessing the risk index of special operations in thermal power plants based on an AI risk matrix model, so as to solve the problems mentioned in the background art.
[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for assessing the risk index of special operations at thermal power plants based on an AI risk matrix model, comprising the following steps: S1: Data acquisition steps: Obtain the daily work task list, work risk list, and safety measure list from the two-ticket system and the safety, health, and environmental management system of the thermal power plant; S2: Model building steps, including building the MES risk matrix calculation model, risk level evaluation model, and operational risk coefficient calculation model; S3: Risk coefficient correction step, inputting the acquired historical accident case data and real-time meteorological data into the operation risk coefficient calculation model, dynamically calculating and outputting the risk coefficient factor β; S4: Task risk calculation step: Input the task list, risk list and safety measure list into the MES risk matrix calculation model. According to the predetermined risk index calculation formula R=β×(M×E×S), automatically calculate the risk index R of each task, where M represents the control measure status parameter for a specific hazard, E represents the frequency parameter of the hazard, and S represents the possible consequences parameter of the accident. S5: Global risk assessment step, inputting the risk indices of all work tasks obtained in S4 into the risk level evaluation model, performing aggregate calculations to generate the overall plant operation risk index, and generating a five-color heat map of plant operation risk based on the geographical location information of the work tasks.
[0007] Furthermore, in S1, the task list includes extended tags, which are used to mark tasks from multiple dimensions such as task type, task characteristics, and environmental sensitivity.
[0008] Furthermore, in S3, the operational risk coefficient calculation model employs a machine learning algorithm to establish a mapping relationship between accident causes and risk coefficient factor β by analyzing historical accident case data, thereby achieving dynamic correction of β.
[0009] Furthermore, in S4, the values of parameters M, E, and S are predefined by the operation risk list and the safety measures list, and parameter M is downgraded and adjusted according to the effectiveness of the measures in the safety measures list.
[0010] Furthermore, in S5, the risk level evaluation model aggregates the risk indices of all work tasks by weighted average or by selecting the maximum value of the risk index, thus obtaining the overall plant operation risk index.
[0011] A risk index assessment system for special operations at thermal power plants to implement the method, comprising: The data acquisition module is used to integrate and obtain the work task list, work risk list, and safety measure list from the two-ticket system and the health, safety and environmental management system; The data processing and storage module is used to store and manage the acquired list data, historical accident case data, and real-time meteorological data; The risk model calculation engine, which embeds the MES risk matrix calculation model, risk level evaluation model, and operational risk coefficient calculation model, is used to perform risk coefficient correction, task risk calculation, and global risk assessment. The visualization and early warning module is used to display the plant-wide operational risk index and the five-color heat map of operational risks in the plant area, and to send early warning information to management personnel when the risk level exceeds the threshold.
[0012] Furthermore, the data acquisition module interconnects with the source system via an enterprise service bus or API interface.
[0013] Furthermore, the visualization and early warning module supports dynamically displaying the risk levels of different areas in the form of heat maps on the factory's electronic map.
[0014] This invention provides a method for assessing the risk index of special operations in thermal power plants based on an AI risk matrix model, which has the following beneficial effects: 1. When compiling the daily work task list for the power plant for the first time, this invention used parallel data aggregation technology across disciplines and designed and developed a compilation platform. Through interconnection with the two ticketing systems, the work task list was aggregated in seconds, which greatly reduced the statistical editing time of safety, health and environmental personnel. In addition, the system automatically obtains the work risk list and safety measure list from the safety, health and environmental system.
[0015] 2. In addition to the default tags for work tasks in the system module of this invention, an "Extended Tags" option is also set. The "Extended Tags" option can be selected in multiple ways, such as defining work tasks from multiple dimensions. For example, according to the execution of specific tasks, it can be customized as meetings, inspections, reports, evaluations, etc.; according to task characteristics, it can be customized as high-altitude operations, hot work operations, etc. The flexible setting of various parameters can export the corresponding work task list by value, which facilitates subsequent calculation and analysis functions.
[0016] 3. The system of this invention realizes automatic scoring of each operation task through a calculation model. The scoring strategy takes into account the different focuses of company-level and department-level operation risks within the company. Each score is "verifiable". The integration of the evaluation subject and the evaluation object can promptly identify and improve the problems existing in the current power plant operation risk index assessment. It provides a typical case for realizing "quantitative evaluation" and "dynamic evaluation". Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the workflow of a method for assessing the risk index of special operations in thermal power plants based on an AI risk matrix model, according to the present invention. Detailed Implementation
[0018] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.
[0019] like Figure 1 As shown, a method for assessing the risk index of special operations at thermal power plants based on an AI risk matrix model includes the following steps: S1: Data acquisition steps: Obtain the daily work task list, work risk list, and safety measure list from the two-ticket system and the safety, health, and environmental management system of the thermal power plant. In S1, the work task list contains extended tags, which are used to mark the work tasks from multiple dimensions such as work type, task characteristics, and environmental sensitivity. S2: Model building steps, including building the MES risk matrix calculation model, risk level evaluation model, and operational risk coefficient calculation model; S3: Risk coefficient correction step. The acquired historical accident case data and real-time meteorological data are input into the operation risk coefficient calculation model to dynamically calculate and output the risk coefficient factor β. In S3, the operation risk coefficient calculation model adopts a machine learning algorithm to establish a mapping relationship between accident causes and risk coefficient factor β by analyzing historical accident case data, so as to realize the dynamic correction of β. S4: Task Risk Calculation Step. Input the task list, task risk list, and safety measure list into the MES risk matrix calculation model. According to the predetermined risk index calculation formula R=β×(M×E×S), the risk index R of each task is automatically calculated. Here, M represents the control measure status parameter for a specific hazard, E represents the frequency parameter of the hazard occurrence, and S represents the possible consequences parameter of the accident. In S4, the values of parameters M, E, and S are predefined by the task risk list and the safety measure list, and parameter M is downgraded and adjusted according to the effectiveness of the measures in the safety measure list. S5: Global risk assessment step. Input the risk indices of all tasks obtained in S4 into the risk level evaluation model, perform aggregation calculation, generate the plant-wide operation risk index, and generate a five-color heat map of plant operation risk based on the geographical location information of the tasks. In S5, the risk level evaluation model aggregates the risk indices of all tasks by weighted average or by selecting the maximum value of the risk index to obtain the plant-wide operation risk index.
[0020] A risk index assessment system for special operations at thermal power plants for implementing the method includes: The data acquisition module is used to integrate and obtain the work task list, work risk list, and safety measure list from the two-ticket system and the health, safety and environmental management system; The data processing and storage module is used to store and manage the acquired list data, historical accident case data, and real-time meteorological data; The risk model calculation engine has embedded an MES risk matrix calculation model, a risk level evaluation model, and an operational risk coefficient calculation model, which are used to perform risk coefficient correction, task risk calculation, and global risk assessment. The visualization and early warning module is used to display the plant-wide operational risk index and the five-color heat map of operational risks in the plant area, and to send early warning information to management personnel when the risk level exceeds the threshold.
[0021] The data acquisition module interconnects with the source system via enterprise service bus or API interface, while the visualization and early warning module supports dynamic display of risk levels in different areas on the factory electronic map in the form of heat map.
[0022] Specifically, in this embodiment, the thermal power plant deploys this assessment system. The system's data acquisition module connects with the power plant's two-ticket management system and safety, health, and environmental (SHE) system through an API interface. It automatically retrieves the daily work task list at regular intervals. At the same time, it obtains a predefined work risk library (containing E and S standard values for various hazards) and a safety measure library from the SHE system.
[0023] The operation risk coefficient calculation model in the risk model calculation engine is preloaded with data on near misses and minor accidents at the power plant over the past five years, and is connected to the real-time temperature data interface of the local meteorological bureau. The model is trained using a linear regression algorithm, and when the real-time temperature exceeds 39 degrees Celsius, the risk coefficient factor β for outdoor operations will be automatically increased.
[0024] For a specific task of "high-altitude hot work on boiler #3", the system performs the following evaluation: The task was identified from the task list as "high-altitude operation", "hot work", and "affected by temperature". The corresponding standard parameters for the task were matched from the task risk list: E (frequency) = 6 (daily exposure), S (consequence) = 15 (very serious, may lead to death). The safety measures list confirmed that special operation approval, supervision and spark prevention measures have been implemented. Therefore, M (measure status) was rated as 2 (control measures are effective and risk is reduced). The ambient temperature that day was 40℃, and the operational risk coefficient calculation model calculated β=1.2 based on historical data. The MES risk matrix calculation model calculates the risk index of this task as: R = 1.2 × (2 × 6 × 15) = 216; The risk assessment model aggregates the risk indices of all tasks performed that day, resulting in a plant-wide risk index of 158 (high risk level). The boiler room area is marked in red (highest risk level) on the plant's electronic map in the visualization module, and the system automatically sends high-risk warning text messages to the safety director and project manager.
[0025] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.
Claims
1. A method for assessing the risk index of special operations at thermal power plants based on an AI risk matrix model, characterized in that: Includes the following steps: S1: Data acquisition steps: Obtain the daily work task list, work risk list, and safety measure list from the two-ticket system and the safety, health, and environmental management system of the thermal power plant; S2: Model building steps, including building the MES risk matrix calculation model, risk level evaluation model, and operational risk coefficient calculation model; S3: Risk coefficient correction step, inputting the acquired historical accident case data and real-time meteorological data into the operation risk coefficient calculation model, dynamically calculating and outputting the risk coefficient factor β; S4: Task risk calculation step: Input the task list, risk list and safety measure list into the MES risk matrix calculation model. According to the predetermined risk index calculation formula R=β×(M×E×S), automatically calculate the risk index R of each task, where M represents the control measure status parameter for a specific hazard, E represents the frequency parameter of the hazard, and S represents the possible consequences parameter of the accident. S5: Global risk assessment step, inputting the risk indices of all work tasks obtained in S4 into the risk level evaluation model, performing aggregate calculations to generate the overall plant operation risk index, and generating a five-color heat map of plant operation risk based on the geographical location information of the work tasks.
2. The method for assessing the risk index of special operations at thermal power plants based on an AI risk matrix model, as described in claim 1, is characterized in that... In S1, the task list includes extended tags, which are used to mark tasks from multiple dimensions such as task type, task characteristics, and environmental sensitivity.
3. The method for assessing the risk index of special operations at thermal power plants based on an AI risk matrix model according to claim 1, characterized in that, In S3, the operational risk coefficient calculation model uses a machine learning algorithm to establish a mapping relationship between accident causes and risk coefficient factor β by analyzing historical accident case data, so as to achieve dynamic correction of β.
4. The method for assessing the risk index of special operations at thermal power plants based on an AI risk matrix model according to claim 1, characterized in that, In S4, the values of parameters M, E, and S are predefined by the operation risk list and the safety measures list, and parameter M is downgraded and adjusted according to the effectiveness of the measures in the safety measures list.
5. The method for assessing the risk index of special operations at thermal power plants based on an AI risk matrix model according to claim 1, characterized in that, In S5, the risk level evaluation model aggregates the risk indices of all work tasks by weighted average or by selecting the maximum value of the risk index, and obtains the plant-wide work risk index.
6. A risk index assessment system for special operations at thermal power plants for implementing the method described in any one of claims 1-5, characterized in that, include: The data acquisition module is used to integrate and obtain the work task list, work risk list, and safety measure list from the two-ticket system and the health, safety and environmental management system; The data processing and storage module is used to store and manage the acquired list data, historical accident case data, and real-time meteorological data; The risk model calculation engine, which embeds the MES risk matrix calculation model, risk level evaluation model, and operational risk coefficient calculation model, is used to perform risk coefficient correction, task risk calculation, and global risk assessment. The visualization and early warning module is used to display the plant-wide operational risk index and the five-color heat map of operational risks in the plant area, and to send early warning information to management personnel when the risk level exceeds the threshold.
7. The risk index assessment system for special operations at thermal power plants according to claim 6, characterized in that, The data acquisition module interconnects with the source system via an enterprise service bus or API interface.
8. The risk index assessment system for special operations at thermal power plants according to claim 7, characterized in that, The visualization and early warning module supports dynamically displaying the risk levels of different areas in the form of heat maps on the factory's electronic map.