Evaluation method of human-machine fusion operation of industrial equipment with variable load based on identification model

Abstract

This invention discloses a method for evaluating human-machine fusion operation during load-variable operation of industrial equipment based on an identification model. This method first obtains various operational data and bit number data from the industrial equipment's load-variable operation. It then obtains the operational data of the integrated mobile control (IMPC) as a skill evaluation benchmark. Finally, the operator's skills are evaluated based on four performance indicators during the load-variable process: safety constraints, product quality constraints, task completion time, and energy consumption. This method uses load-variable operation data from a two-layer industrial model predictive control algorithm as the skill evaluation benchmark. This method can objectively reflect and differentiate the operational skill levels of different operators from multiple dimensions.

Description

Technical Field

[0001] The present invention relates to the technical field of process operation training for industrial devices, and in particular to an evaluation method for variable load human-machine fusion operation of industrial devices based on an identification model. Background Art

[0002] Load adjustment is a common operational task in process plants, particularly industrial gas production plants, nuclear power plants, and oil refining plants. These plants all require safe, stable, and efficient load adjustment to meet the urgent needs of downstream operations. For large-scale process production plants, dynamic load adjustment often involves the coordinated adjustment of multiple operating variables. Furthermore, large-scale load adjustment involves complex nonlinear operations, requiring operators of process plants without automatic load-variation systems to possess proficient manual load-variation skills.

[0003] In recent years, most chemical industry process plants have deployed variable-load operator training systems (OTS). However, existing industrial OTS systems have limited operator evaluation functionality, mostly assessing operator skills from the perspectives of completeness of operating steps and correctness of operating sequences. They lack the ability to objectively evaluate and guide operators from multiple dimensions, leveraging human-machine integration. Summary of the Invention

[0004] The purpose of the present invention is to address the deficiencies of the existing technology and provide an evaluation method for variable load human-machine fusion operation of industrial equipment based on an identification model.

[0005] The object of the present invention is achieved through the following technical solution: a method for evaluating the variable load human-machine fusion operation of an industrial device based on an identification model, comprising the following steps:

[0006] (1) Obtaining various operation data and bit number data during the variable load operation of industrial equipment: Connecting to the DCS platform of the industrial equipment through the industrial communication protocol, obtaining various production data during the production process of the industrial equipment in real time, and automatically recording various operation data of the operator;

[0007] (2) Obtaining the operational data of the IMPC as a benchmark for skill assessment: A two-layer IMPC is designed to complete the switching task between any two working conditions in several typical working conditions. The two-layer IMPC consists of an upper-layer steady-state optimization calculation module and a lower-layer dynamic prediction control module. The steady-state optimization calculation module provides the lower layer with the optimal tracking target during the process, and the dynamic prediction control module smoothly and quickly controls the load adjustment process to the set target given by the upper layer without violating the constraints. After all tasks are completed, the variable load operation data of the IMPC are collected.

[0008] (3) Evaluate the operator's skills based on four performance indicators during the variable load process: safety constraints, product quality constraints, task completion time, and energy consumption;

[0009] Furthermore, the step (3) is divided into the following sub-steps.

[0010] (3.1) Determine DCS alarm constraints in industrial production processes.

[0011] (3.2) Determine four evaluation indicators for variable load operation skills: safety constraints, product quality constraints, task completion time and energy consumption;

[0012] (3.3) Design safety constraint indicators: Safety constraints are key indicators reflecting production safety. The safety constraint score is recorded as T1. The initial score of T1 is 100. When the operator's manual operation triggers the advanced alarm constraint, the T1 score is gradually deducted.

[0013] (3.4) Design product quality constraint indicators: Product quality constraint indicators consist of two parts: hard product quality constraints and soft product quality constraints. Hard product quality constraints are related to the quality of the industrial plant's products and are determined by whether the values ​​of key process variables trigger low-level alarm constraints. Soft product quality constraints are related to whether the relationships between several important materials match each other at the end of manual operation. The following steps are used to obtain a steady-state material matching relationship. First, actual data is collected and the steady-state operating condition data is identified using a steady-state detection algorithm. Second, a second-order polynomial model is fitted to the steady-state operating condition data to obtain a steady-state material matching relationship. The product quality constraint score is denoted as T2, and the initial score of T2 is 100. If the operator triggers the low-level alarm constraint of the industrial plant during manual operation, or if the material relationship does not meet the constraint conditions at the end of manual operation, the T2 score will be gradually deducted.

[0014] (3.5) Design operation speed index: Based on the IMPC operation data, the task completion time index score is calculated as follows:

[0015]

[0016] Among them, T MPC and T Operator are the time taken by IMPC and human operators to complete the same operation tasks.

[0017] (3.6) Design energy consumption index: Based on the IMPC operation data, the energy consumption index score is calculated as follows:

[0018]

[0019] Among them, E MPC and E Operatorare the energy consumption of IMPC and human operator to complete the same task.

[0020] (3.7) Design a multi-index weighting method.

[0021] The final operation evaluation result T is obtained by weighting the above four indicators. Final , as shown below:

[0022]

[0023] Among them, q i Represents the weighting coefficient of the i-th indicator.

[0024] The present invention provides a beneficial effect by designing a multi-dimensional skill assessment method to evaluate the dynamic variable load operation skills of industrial plant operators. This method assesses the variable load operation skills of industrial plant operators based on four performance indicators: safety constraints, product quality constraints, task completion time, and energy consumption. This method uses variable load operation data from a two-layer industrial model predictive control algorithm as a skill assessment benchmark. This method can objectively reflect and differentiate the operational skill levels of different operators from multiple dimensions. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is the final operation skill score radar chart. DETAILED DESCRIPTION

[0026] The present invention discloses a method for evaluating the variable load human-machine fusion operation of an industrial device based on an identification model, comprising the following steps:

[0027] Step 1: Obtain various operation data and bit number data during the variable load operation of the industrial device: Connect to the industrial device DCS platform through the industrial communication protocol to obtain various production data during the industrial device production process in real time and automatically record various operation data of the operator;

[0028] Step 2: Obtaining operational data from the IMPC as a benchmark for skill assessment: Based on previously developed training systems for variable-load operation of industrial plants, a two-layer IMPC was designed to complete the task of switching between any two typical operating conditions. The two-layer IMPC consists of an upper-layer steady-state optimization calculation module and a lower-layer dynamic predictive control module. The steady-state optimization calculation module provides the lower layer with the optimal tracking target during the process, while the dynamic predictive control module smoothly and quickly controls the load adjustment process to the target set by the upper layer without violating any constraints. After all tasks are completed, the IMPC's variable-load operation data is collected and used as a benchmark for evaluating human operator skills.

[0029] Step 3: Evaluate the operator's skills based on four performance indicators during the variable load process: safety constraints, product quality constraints, task completion time, and energy consumption;

[0030] This step is the core of the present invention and is divided into the following sub-steps.

[0031] 1) Determine DCS alarm constraints in industrial production processes.

[0032] Combine process and historical data to determine the quality requirements of the variable load production process of industrial equipment, including DCS low-level and high-level alarm constraints.

[0033] 2) Determine four evaluation indicators for variable load operation skills.

[0034] By combining process knowledge, the operator's variable load operation skills are generally evaluated from the following four perspectives: (1) operation safety; (2) product quality constraints; (3) operation speed index; (4) energy consumption index.

[0035] 3) Design safety constraint indicators: safety constraints.

[0036] Safety constraint indicators are related to the safety of industrial equipment and are determined by whether the values ​​of key process variables trigger advanced alarm constraints. Advanced alarms trigger the DCS's safety interlock system to protect industrial equipment. Safety constraints are key indicators of production safety, and the safety constraint score is denoted as T1. The initial T1 score is 100, and as operator manual actions trigger advanced alarm constraints, the T1 score is gradually deducted.

[0037] 4) Design product quality constraint indicators: product quality constraints.

[0038] Product quality constraints consist of hard and soft constraints. Hard constraints relate to the quality of the industrial plant's products and are determined by whether the values ​​of key process variables trigger low-level alarm constraints. Soft constraints are related to whether the relationships between several key materials match at the end of manual operation. If these relationships do not match at the end of manual operation, the plant's operating conditions will not remain stable in subsequent simulations.

[0039] The steady-state material matching relationship is obtained through the following steps. First, actual data is collected and the steady-state operating condition data is identified using a steady-state detection algorithm. Note that this steady-state operating condition data includes not only typical steady-state operating condition data but also some atypical steady-state operating condition data. Second, a second-order polynomial model is fitted to the steady-state operating condition data to obtain the steady-state material matching relationship.

[0040] The product quality constraint score is denoted as T2, with an initial score of 100. If the operator triggers a low-level alarm constraint for the industrial unit during manual operation, or if the material relationship does not meet the constraint conditions at the end of the manual operation, the T2 score will be gradually deducted. Different tags contribute differently to the hard product quality constraint indicators because of their varying importance to the production of the industrial unit. Similarly, different product quality constraints are configured to contribute differently to the soft product quality constraint indicators.

[0041] 5) Design operation speed indicator: task completion time.

[0042] The task completion time metric reflects the speed of response to downstream demand. While ensuring plant safety and product quality, operators need to adjust industrial unit loads as quickly as possible to meet downstream demand. Based on IMPC's operational data, the task completion time metric score is calculated as follows:

[0043]

[0044] Among them, T MPC and T Operator are the time taken by IMPC and human operators to complete the same operation tasks.

[0045] 6) Design energy consumption index: energy consumption.

[0046] The energy consumption index reflects the economic efficiency of manual operation. Based on the operation data of IMPC, the energy consumption index score is calculated as follows:

[0047]

[0048] Among them, E MPC and E Operator are the energy consumption of IMPC and human operator to complete the same task.

[0049] 7) Design a multi-index weighting method.

[0050] This method obtains the final operation evaluation result T by weighting the above four indicators. Final , as shown below:

[0051]

[0052] Among them, q i Represents the weighting coefficient of the i-th indicator.

Claims

1. A method for evaluating variable load human-machine fusion operation of industrial equipment based on identification model, characterized in that: The following steps are involved: (1) Obtaining various operation data and bit number data during the variable load operation of industrial equipment: Connecting to the DCS platform of the industrial equipment through the industrial communication protocol, obtaining various production data during the production process of the industrial equipment in real time, and automatically recording various operation data of the operator; (2) Obtaining the operational data of the IMPC as a benchmark for skill assessment: Designing a two-layer IMPC to complete the switching task between any two working conditions in several typical working conditions, wherein the two-layer IMPC consists of an upper-layer steady-state optimization calculation module and a lower-layer dynamic predictive control module; the steady-state optimization calculation module provides the lower layer with the optimal tracking target during the process, and the dynamic predictive control module smoothly and quickly controls the load adjustment process to the set target given by the upper layer without violating the constraints; after all tasks are completed, collect the variable load operation data of the IMPC; (3) Evaluate the operator's skills based on four performance indicators during the variable load process: safety constraints, product quality constraints, task completion time, and energy consumption; The step (3) is divided into the following sub-steps: (3.1) Determine DCS alarm constraints in industrial production processes; (3.2) Determine four evaluation indicators for variable load operation skills: safety constraints, product quality constraints, task completion time and energy consumption; (3.3) Design safety constraint indicators: Safety constraints are key indicators that reflect production safety. The safety constraint score is recorded as T1. The initial score of T1 is 100. When the operator's manual operation triggers the advanced alarm constraint, the T1 score is gradually deducted. (3.4) Design product quality constraint indicators: Product quality constraint indicators are composed of hard product quality constraints and soft product quality constraints. Hard product quality constraints are related to the quality of industrial equipment products and are determined by whether the values ​​of key process variables trigger low-level alarm constraints. Soft product quality constraints are related to whether the relationships between several important materials at the end of manual operation are compatible. The steady-state material matching relationship is obtained through the following steps: First, actual data is collected and the steady-state operating condition data is identified using a steady-state detection algorithm. Second, a second-order polynomial model is fitted to the steady-state operating condition data to obtain the steady-state material matching relationship. The product quality constraint score is recorded as T2, and the initial score of T2 is 100. If the operator triggers a low-level alarm constraint of the industrial device during manual operation, or if the material relationship does not meet the constraint conditions at the end of the manual operation, the T2 score will be gradually deducted. (3.5) Design operation speed index: Based on the IMPC operation data, the task completion time index score is calculated as follows: Among them, T MPC and T Operator are the time taken by IMPC and human operator to complete the same operation task; (3.6) Design energy consumption index: Based on the IMPC operation data, the energy consumption index score is calculated as follows: Among them, E MPC and E Operator are the energy consumption of IMPC and human operator to complete the same task; (3.7) Design a multi-index weighting method; The final operation evaluation result T is obtained by weighting the above four indicators. Final , as shown below: Among them, q i Represents the weighting coefficient of the i-th indicator.

Images