An unattended system for a sulphur-burning plant and a method of operation

Abstract

The application discloses an unattended system and a running method of a sulfur-burning sulfuric acid plant, which comprises five subsystems, namely, an intelligent acquisition system, an intelligent identification system, an intelligent evaluation system, an intelligent decision system and an intelligent execution system. By further integrating various technologies such as equipment, operation, process, automation and information, the self-automation capability of links such as acquisition, identification, evaluation, decision and execution is established, the system can autonomously respond to the requirements and environmental changes of external quality, quantity and time dimensions, unreasonable operation and market fluctuations, and can also self-manage the integrity and stability of the system, so that the plant can reduce human intervention to the maximum extent and transition to unmanned.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent control of chemical production processes, and in particular relates to an unattended system and an operating method for a sulfuric acid plant. Background Art

[0002] With the recent development of automation technology, sulfuric acid plants, as a representative of traditional chemical industry, mostly use a "two-conversion, two-absorption" process. Through sulfur incineration, conversion, and absorption, they can produce a variety of products, including 98.5% sulfuric acid, 20% nicotinic acid, 65% nicotinic acid, and liquid SO3. Sulfur-based acid plants are characterized by a wide range of products, a long process flow, multiple upstream and downstream interferences, and diverse production models. Their operation is inevitably more complex than that of chemical plants with a single product. The entire industry faces practical challenges such as insufficient production efficiency and difficulty in controlling the production process. Furthermore, existing technologies are limited and insufficient to support automation, or even intelligent production. Therefore, research on intelligent control of sulfuric acid plants is of great practical significance. On the one hand, it can serve as a model and guide for traditional chemical industry, improving the automation level of traditional industry plants, and on the other hand, it can lay a solid foundation for the intelligent development of these plants. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an unmanned system and operation method for a sulfuric acid plant, overcome the shortcomings of the existing technology, integrate and comprehensively utilize existing technical resources to produce a set of overall solutions, which can effectively reduce operator intervention, improve the overall production efficiency of the plant, and thus realize autonomous operation of the plant.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] An unmanned system for a sulfuric acid plant includes an intelligent collection system, which is electrically connected to an intelligent identification system, an intelligent evaluation system, an intelligent decision-making system, and an intelligent execution system in sequence; wherein:

[0006] The intelligent collection system is used to collect production data and scene data;

[0007] The intelligent recognition system is used to pre-process the collected production data and scene data, identify abnormal points, and provide feedback on abnormal data;

[0008] The intelligent evaluation system is used to conduct multi-dimensional analysis and evaluation of the production data and scenario data identified by the intelligent recognition system;

[0009] The intelligent decision-making system establishes an analysis, prediction, decision-making and disposal model based on the results of the intelligent evaluation system. The analysis, prediction, decision-making and disposal model combines big data, process knowledge and expert experience, and includes different production modes such as environmental interference, working condition switching, output adjustment and stable operation.

[0010] The intelligent execution system uses the results provided by the intelligent decision-making system to judge the execution conditions, select the optimal execution plan and execution system for processing, and feed back the execution results to the intelligent collection system through the control system. The control system then confirms the execution results as a whole.

[0011] Preferably, the production data is mainly provided by the control system, and the collection of production data can be completed through communication integration with the control system; the scene data is provided by a fixed or mobile camera, and the collection of scene data is completed through a video monitoring platform.

[0012] Preferably, the intelligent recognition system is used to realize data quality perception, valve instrument abnormality perception, and process parameter abnormality warning perception, and can timely identify data anomalies and feed back the identification results to the intelligent evaluation system.

[0013] Preferably, the intelligent evaluation system is used to evaluate the device's operating stability, asset reliability, control performance, dynamic risk, external response capability, and production performance, providing data support for the intelligent decision-making system under various operating conditions.

[0014] Preferably, scene data in the intelligent acquisition system is divided into two categories:

[0015] One type is production process control video data, which is analyzed by cameras and the visual intelligent analysis system, and the corresponding data results are obtained and presented in the form of data;

[0016] The other category is security management data, which is mainly provided by the intelligent patrol system and the intelligent security management platform equipped with law enforcement recorders.

[0017] Preferably, the intelligent identification system constructs a multi-identification system for identifying the credibility of production data, the reliability of actuators, and abnormal process indicators through data preprocessing.

[0018] Preferably, the credibility of production data is mainly identified through dynamic changes in the data. For monitored process variable data, if it remains unchanged for a long time or changes in a state for a short time, it can be identified as abnormal.

[0019] Preferably, the reliability of the actuator is identified through two aspects: on the one hand, the valve opening and feedback deviation and change trend in the production data are confirmed; on the other hand, the built-in parameters of the valve positioner in the intelligent device management system are used for judgment.

[0020] Preferably, the identification of abnormal process indicators is the identification of abnormal conditions in the overall process operation. The theoretical production data corresponding to the system under the current working conditions is calculated by combining process simulation software with core process parameters. The theoretical production data is compared with the actual production data, and a score is given according to the data deviation range. If the score is lower than the corresponding set score, it is identified as an abnormality.

[0021] The core process parameters include device load; the theoretical production data include temperature and pressure.

[0022] Preferably, the system also includes a system stability assessment query evaluation system, which includes IO abnormal records within a period of time, and compares the IO abnormal records within a period of time with the trend on the DCS screen. For the same time interval, the overall stable state of the entire IO group is determined; for different time intervals, the stable state of a single IO point is determined.

[0023] A method for operating an unmanned system for a sulfuric acid plant, comprising the following steps:

[0024] Asset Reliability Assessment focuses on comprehensive health monitoring of operating equipment. Detailed single-parameter analysis is performed on all critical measurement points on the equipment, such as current and vibration. Based on process indicators, measurement point status is categorized into three categories: normal, warning, and alarm. For each status category, a corresponding health score range is established: Normal: A health score of 70-100 indicates that the equipment is in good operating condition. Warning: A health score of 30-70 indicates potential equipment issues requiring attention. Alarm: A health score of 0-30 indicates a significant equipment failure or anomaly requiring immediate action. The specific health score is calculated by linear interpolation based on the position of the measurement point value within the corresponding range. Furthermore, considering the varying importance of different measurement points to the overall operation of the equipment, each measurement point is assigned a weighting factor (1-5). Using the entropy weighting method, the health score and weighting factors of each measurement point are combined to calculate the overall equipment health (0-100). This overall health score is then used to determine the overall status of the operating equipment (normal, warning, or alarm).

[0025] Control performance evaluation is an important indicator to measure the performance of control systems, focusing on testing the three core performances of the loop: accuracy, stability and speed. By comparing the actual value with the given value, the accuracy of loop control is evaluated. The higher the accuracy rate, the more accurate the loop control. The stability of the loop is evaluated by the proportion of the actual value deviating from the limit value. The higher the stability rate, the more stable the loop operation. The response ability of the loop is tested when the given value changes. The shorter the response time, the faster the loop response. By comprehensively evaluating the performance of these three aspects, the control performance of the loop can be fully understood, providing a basis for optimizing the control strategy.

[0026] Dynamic risk assessment is based on process risk and carefully constructs an accident chain model. The initial cause of the accident, risk level, development steps and final result that may lead to the accident are identified to construct a complete accident chain. The development steps and risk levels (1-5 levels) of the accident chain are combined to construct a safety risk matrix, and each element in the matrix represents a specific risk situation. Risk assessment is carried out according to the safety risk matrix, and the corresponding operation prompt is given according to the risk level. When in the risk-free stage (1-2 levels), the operator is prompted to intervene; when there is a certain risk (3 levels), the operator is prompted to intervene at the right moment; when there is a greater risk (4-5 levels), the operator is prompted to intervene immediately.

[0027] Production performance evaluation involves the comprehensive scoring of multiple indicators. According to the production process indicators, energy consumption indicators, resource comprehensive utilization indicators, pollutant emission indicators and product analysis, etc., the scoring interval of each indicator is set, and the score of each indicator is calculated according to the actual production situation. According to the importance of each indicator, different weight coefficients are allocated, and the final production performance score is calculated by using the weighted summation method. The allocation of weight coefficients should reflect the influence degree of each indicator on production performance. By comparing the production performance scores of different time periods or different production batches, the advantages and disadvantages in the production process are analyzed, and a basis for optimizing the production strategy is provided.

[0028] The intelligent execution system contains multiple control schemes, aiming to adapt to the control needs in different scenarios. As a traditional control method, PID control adjusts the proportional, integral and differential parameters to achieve stable control of the system, suitable for scenarios with high control accuracy requirements. APC advanced control uses advanced algorithms and models to accurately control complex processes, suitable for industrial processes with high nonlinearity and uncertainty. Program control executes control tasks according to pre-set programs and logical sequences, suitable for scenarios that need to operate according to fixed steps. Big data intelligent control uses big data technology and machine learning algorithms to analyze and predict massive data, achieving intelligent control. Suitable for scenarios with high real-time and predictive requirements.

[0029] The present application can achieve the following beneficial effects:

[0030] 1. The collected data is more comprehensive and complete, including not only production data but also relevant video stream data. By analyzing the video stream data, corresponding data is obtained, and the video stream data and production data are combined for analysis to achieve control over the production status of the device.

[0031] 2. The data for identification and evaluation is more scientific, including not only the basic information of simple process parameters, but also the results of overall analysis of related parameters. The availability of data has been greatly improved, and abnormal conditions of the device can be discovered in advance by monitoring the data.

[0032] 3. The execution process is more reliable and complete. The latest model analysis and fitting technology are used to evaluate the working conditions. At the same time, a multi-angle execution system is used to achieve full coverage of system control and improve problems in the control process.

[0033] 4. Since human intervention is reduced and the system is adjusted in real time, the stability of the system is enhanced, the production efficiency of the device is improved, and at the same time, the reduction of personnel costs and production costs will produce certain economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be further described below with reference to the accompanying drawings and examples:

[0035] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0036] Figure 2 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0037] Example 1:

[0038] The preferred solution is Figures 1 to 2 The figure shows an unmanned system for a sulfuric acid plant. The intelligent data collection system is responsible for collecting production data and scene data. Production data is primarily provided by the control system and can be collected through communication integration with the control system. Scene data is primarily provided by fixed or mobile cameras and is collected through a video surveillance platform.

[0039] Since the control system in the intelligent acquisition system is a DCS control system and there are basically no other heterogeneous systems on site, the data collected by the system mainly include real-time production data corresponding to the DCS side, internal parameters of the system modules, virtual calculation data, etc.

[0040] The scene data in the intelligent acquisition system is divided into two categories: one is production process control video data, which is analyzed by cameras and the onboard visual intelligent analysis system to obtain corresponding data results and present the results in the form of data. Some of this data is used for on-site equipment control in dangerous scenarios such as those without instrument installation and measurement conditions and areas with human activity. Another part of this data is used for comparison and early warning between the DCS system and on-site measurements. The other category is safety management data, which is mainly provided by the intelligent inspection system and the intelligent safety management platform equipped with law enforcement recorders. This data monitors the on-site environmental conditions in real time and provides a visual basis for on-site environmental risk assessment.

[0041] The intelligent recognition system pre-processes the collected data, identifies anomalies, and provides feedback on these anomalies. By establishing intelligent recognition systems for data quality, valve and instrument anomalies, equipment anomalies, and process parameter anomaly warnings, data anomalies can be identified promptly and the results fed back to the evaluation system.

[0042] Through data preprocessing, the intelligent identification system builds multiple identification systems such as production data credibility, actuator reliability (valves), and process indicator anomalies.

[0043] The credibility of production data is primarily determined by dynamic changes in the data. For monitored process variables, prolonged periods of invariance or brief jumps can be identified as abnormalities, triggering early warnings. Data credibility verification primarily determines the usability of collected bit numbers, including dead value determination, trend analysis, and out-of-limit detection. In the future, this determination can be made based on data characteristics.

[0044] Actuator reliability is primarily determined through two factors. First, the deviation and trend of valve opening and feedback data in production data are verified. When the deviation between valve opening and feedback is within a certain range, the valve positioner's built-in parameters in the intelligent device management system are used for evaluation. This evaluation determines the followability (stroke deviation) between the valve's manipulated value (MV attribute bit) and the controller output (OP attribute bit). Strong followability (small stroke deviation) indicates a normal valve; weak followability indicates a valve warning or fault. Valve stroke deviation can be observed on the DCS diagram, thereby determining the valve's status. An evaluation system is considered excellent if it matches actual results. "Good" indicates excellent performance with occasional missed or false positives. "Poor" indicates both missed and false positives.

[0045] The identification of process indicators is the identification of overall process operation abnormal conditions, which is different from the traditional index identification method. Theoretical production data (temperature, pressure, etc.) corresponding to the system under the current working condition are calculated by combining the core process parameters (device load, etc.) through process simulation software, and the theoretical production data are compared with the actual production data. According to the data deviation interval, the score is evaluated. If the score is lower than the corresponding set score, it can be identified as abnormal.

[0046] The intelligent evaluation system analyzes and evaluates the abnormal data identified by the identification system in multiple dimensions. By building a comprehensive evaluation capability in terms of device running stability, asset reliability, control performance, dynamic risk, external response capability (ability to resist external maximum interference), and production performance, the system provides scientific data support for the pre-judgment decision system under various working conditions.

[0047] System stability evaluation queries the abnormal records of IO in the evaluation system within a period of time, compares them with the trends on the DCS screen, and determines the overall stability of the IO group as a whole for the same time interval. For different time intervals, the stability of individual IO points is determined.

[0048] Asset reliability evaluation performs single-parameter analysis on all measurement points (current, vibration, etc.) included in the running equipment. The state of the measurement points is divided into three categories (warning, alarm, normal) according to the process indicators. Each category is given a corresponding health score (0-100 points), as follows: normal state corresponds to a health score of (70-100 points), warning state corresponds to a health score of (30-70 points), and alarm state corresponds to a health score of (0-30 points). The specific health degree is calculated according to the position of the value in the interval. Different measurement points have different weight coefficients (1-5 levels). The overall health degree of the equipment (0-100 points) and the overall state of the running equipment (warning, alarm, normal) are obtained through the entropy weight method.

[0049] Control performance evaluation tests mainly test the accuracy (accuracy rate), stability (stability rate), and speed (response time) of the loop. Accuracy reflects the accuracy of loop control through the actual value and the given value. Stability is evaluated by the proportion of actual value deviation from the limit value. Speed requires the ability of the loop to respond when the given value changes.

[0050] Dynamic risk assessment divides the process risk into corresponding accident chains. Each accident chain includes the initial cause of the accident, the risk level, the accident development steps, and the final result of the accident chain. The development steps of the accident chain and the risk level (levels 1-5) are combined to form a safety risk matrix. Risk assessment is performed based on the risk matrix and operators are reminded to perform appropriate safety operations. When the risk is zero (levels 1-2), the operator is prompted not to intervene; when there is a certain risk (level 3), the operator is prompted to intervene at an appropriate time; when there is a significant risk (levels 4-5), the operator is prompted to intervene immediately.

[0051] The production performance evaluation scores multiple indicators such as production process indicators, energy consumption indicators, comprehensive resource utilization indicators, pollutant emission indicators and product analysis. The indicator score of each indicator is calculated based on the scoring range determined by production. The final production performance score is obtained by weighted calculation based on the indicator score and the importance of the indicator.

[0052] Based on the results of the identification and evaluation system, the intelligent decision-making system establishes an analysis and prediction decision-making and disposal model that combines big data, process knowledge, and expert experience for different production modes such as environmental interference, working condition switching, output adjustment, and stable operation. Due to the large number of production modes and abnormal interferences, it is impossible to simply use one or several unified decision-making models to realize all decision-making scenarios. It is necessary to adopt the idea of ​​"specific analysis of specific problems, from specific to general" to build an intelligent decision-making system.

[0053] The intelligent execution system uses the results provided by the decision-making system to judge the execution conditions, select the optimal execution plan and execution system for processing, and feed back the execution results to the intelligent acquisition system through the control system. The system confirms the execution results as a whole. The intelligent execution system mainly includes intelligent PID control, APC control, intelligent program control, and big data intelligent control.

[0054] Intelligent PID control is based on the traditional PID algorithm to perform PID control of single loops and complex loops. By collecting PID parameters and analyzing previous data, it provides the optimal PID parameters suitable for the current working conditions, optimizes and improves the PID basic algorithm as the working conditions change, and provides a complete PID execution system.

[0055] Intelligent APC control performs corresponding multivariable model identification based on the data collected by the system, decouples the system through the model, and realizes edge-optimization control of the device. At the same time, it combines with the upstream RTO system to solve the optimal parameters of the device operating conditions.

[0056] Intelligent program control establishes an anthropomorphic operating mode based on the operator's operating experience and methods, programs and standardizes the operator's operating ideas, realizes dynamic control of switch quantity participation, and solves related control execution problems.

[0057] Big data intelligent control is based on big data. Through various superposition algorithms, it provides relevant optimization and adjustment directions based on data models and actual working conditions.

[0058] A method for operating an unmanned system for a sulfuric acid plant, comprising the following steps:

[0059] Asset Reliability Assessment focuses on comprehensive health monitoring of operating equipment. Detailed single-parameter analysis is performed on all critical measurement points on the equipment, such as current and vibration. Based on process indicators, measurement point status is categorized into three categories: normal, warning, and alarm. For each status category, a corresponding health score range is established: Normal: A health score of 70-100 indicates that the equipment is in good operating condition. Warning: A health score of 30-70 indicates potential equipment issues requiring attention. Alarm: A health score of 0-30 indicates a significant equipment failure or anomaly requiring immediate action. The specific health score is calculated by linear interpolation based on the position of the measurement point value within the corresponding range. Furthermore, considering the varying importance of different measurement points to the overall operation of the equipment, each measurement point is assigned a weighting factor (1-5). Using the entropy weighting method, the health score and weighting factors of each measurement point are combined to calculate the overall equipment health (0-100). This overall health score is then used to determine the overall status of the operating equipment (normal, warning, or alarm).

[0060] Control performance evaluation is a key metric for measuring control system performance, focusing on three core loop characteristics: accuracy, smoothness, and speed. By comparing actual values ​​with setpoints, the precision of loop control is assessed. A higher accuracy indicates more accurate loop control. The percentage of actual values ​​deviating from setpoints is evaluated to determine loop stability. A higher smoothness indicates smoother loop operation. The loop's ability to respond to changes in setpoints is tested. A shorter response time indicates a faster loop response. By comprehensively evaluating these three performance aspects, a comprehensive understanding of the loop's control performance can be achieved, providing a basis for optimizing control strategies.

[0061] The dynamic risk assessment is based on the process risk, and an accident chain model is carefully constructed. The initial cause that may lead to an accident, the risk level, the development steps and the final result are identified to construct a complete accident chain. The development steps and the risk level (1-5 level) of the accident chain are combined to construct a safety risk matrix, and each element in the matrix represents a specific risk situation. Risk assessment is carried out according to the safety risk matrix, and the corresponding operation prompt is given according to the risk level. When in the risk-free stage (1-2 level), the operator is prompted to take no action; when there is a certain risk (3 level), the operator is prompted to intervene at the right moment; when there is a greater risk (4-5 level), the operator is prompted to intervene immediately.

[0062] The production performance evaluation involves the comprehensive scoring of multiple indicators. According to the production process indicators, energy consumption indicators, resource comprehensive utilization indicators, pollutant emission indicators and product analysis, etc., the scoring interval is set for each indicator, and the score of each indicator is calculated according to the actual production situation. According to the importance of each indicator, different weight coefficients are allocated, and the final production performance score is calculated by using the weighted summation method. The allocation of weight coefficients should reflect the influence degree of each indicator on production performance. By comparing the production performance scores of different time periods or different production batches, the advantages and disadvantages in the production process are analyzed, and the basis for optimizing the production strategy is provided.

[0063] The intelligent execution system contains multiple control schemes, aiming to adapt to the control requirements in different scenarios. As a traditional control method, PID control realizes stable control of the system by adjusting the proportional, integral and differential parameters, and is suitable for scenarios with high control accuracy requirements. APC advanced control uses advanced algorithms and models to accurately control complex processes, and is suitable for industrial processes with high nonlinearity and uncertainty. Program control executes control tasks according to the preset program and logical sequence, and is suitable for scenarios that need to operate according to fixed steps. Big data intelligent control uses big data technology and machine learning algorithms to analyze and predict massive data, and realizes intelligent control. It is suitable for scenarios with high real-time and predictive requirements.

[0064] The above embodiments are only preferred technical solutions of the present application, and should not be regarded as limitations of the present application. The protection scope of the present application should be based on the technical solutions recited in the claims, including equivalent replacement solutions of the technical features recited in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present application.

Claims

1. An unmanned system for a sulfuric acid plant, characterized by: It includes an intelligent collection system, which is electrically connected to the intelligent recognition system, the intelligent evaluation system, the intelligent decision-making system and the intelligent execution system in sequence; wherein: The intelligent collection system is used to collect production data and scene data; The intelligent recognition system is used to pre-process the collected production data and scene data, identify abnormal points, and provide feedback on abnormal data; The intelligent evaluation system is used to conduct multi-dimensional analysis and evaluation of the production data and scenario data identified by the intelligent recognition system; The intelligent decision-making system establishes an analysis, prediction, decision-making and disposal model based on the results of the intelligent evaluation system. The analysis, prediction, decision-making and disposal model combines big data, process knowledge and expert experience, and includes different production modes such as environmental interference, working condition switching, output adjustment and stable operation. The intelligent execution system uses the results provided by the intelligent decision-making system to determine the execution conditions, select the optimal execution plan and execution system for processing, and feed back the execution results to the intelligent acquisition system through the control system. The control system then confirms the execution results as a whole. The intelligent assessment system is used to evaluate the device's operational stability, asset reliability, control performance, dynamic risk, external response capability, and production performance, providing data support for the intelligent decision-making system under various operating conditions; The intelligent identification system builds a multi-recognition system through data preprocessing to identify the credibility of production data, the reliability of actuators, and abnormal process indicators; The reliability of production data is identified through the dynamic changes of the data. For monitored process variable data, if it remains unchanged for a long time or changes in a short time, it can be identified as an abnormality. The reliability of the actuator is identified from two aspects: on the one hand, the deviation and change trend of the valve opening and feedback in the production data are confirmed, and on the other hand, the built-in parameters of the valve positioner in the intelligent equipment management system are used for judgment; Identification of process indicator anomalies refers to the identification of abnormal conditions in the overall process operation. This is done by using process simulation software combined with core process parameters to calculate the theoretical production data corresponding to the system under the current working conditions. The theoretical production data is then compared with the actual production data, and a score is assigned based on the data deviation range. If the score is lower than the corresponding set score, it is identified as an anomaly. The core process parameters include device load; the theoretical production data include temperature and pressure.

2. The unmanned system for sulfuric acid plant according to claim 1, characterized in that: The production data is provided by the control system, and the collection of production data can be completed through communication integration with the control system; the scene data is provided by fixed or mobile cameras, and the collection of scene data can be completed through the video monitoring platform.

3. The unmanned system for sulfuric acid plant according to claim 1, characterized in that: The intelligent recognition system is used to realize data quality perception, valve instrument abnormality perception, and process parameter abnormality warning perception. It can identify data anomalies in a timely manner and feed back the identification results to the intelligent evaluation system.

4. The unmanned system for sulfuric acid plant according to claim 1, characterized in that: The scene data in the intelligent acquisition system is divided into two categories: One type is production process control video data, which is analyzed by cameras and the visual intelligent analysis system, and the corresponding data results are obtained and presented in the form of data; Another category is security management data, which is provided by the intelligent patrol system and the intelligent security management platform equipped with law enforcement recorders.

5. The unmanned system for sulfuric acid plant according to claim 1, characterized in that: It also includes a system stability assessment query evaluation system, which includes IO abnormal records within a period of time. The IO abnormal records within a period of time are compared with the trends on the DCS screen. For the same time interval, the overall stability of the entire IO group is determined; for different time intervals, the stability of a single IO point is determined.

6. The method for operating an unmanned system for a sulfuric acid plant according to any one of claims 1 to 5, characterized in that The following steps are involved: Single parameter analysis is performed on key measurement points of operating equipment, including current and vibration. The key measurement points are classified into three categories based on process indicators: normal, warning, and alarm. These are then assigned health score intervals of 70-100, 30-70, and 0-30. Assign a weight coefficient of 1-5 to each key measurement point, use the entropy weight method to calculate the overall comprehensive health of the equipment, and judge the overall status of the equipment based on this; Evaluate control system performance from three aspects: accuracy, stability, and speed. Accuracy is evaluated by comparing actual values ​​with given values. Stability is determined by evaluating the percentage of actual values ​​that deviate from the limit value. Speed ​​is measured by testing the loop's responsiveness when the given value changes. Build an accident chain model based on process risks, identify the initial cause, risk level, development steps and final results, and combine the development steps with risk levels of 1-5 to build a safety risk matrix; Risk assessment is performed based on a matrix: if there is no risk, no intervention is indicated; if there is a certain risk, intervention is indicated at an appropriate time; if there is a significant risk, immediate intervention is indicated; Set scoring intervals based on production processes, energy consumption, comprehensive resource utilization, pollutant emissions, and product analysis indicators, and calculate scores for each indicator based on actual production conditions; Assign weight coefficients based on the importance of each indicator, use the weighted summation method to calculate the final production performance score, and analyze the advantages and disadvantages of production by comparing the scores of different time periods or batches, providing a basis for optimizing production strategies.

Images