Industrial Plant Command Recommendation Using Historical-State Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Operating industrial plants can be complex, especially in critical or alarm situations, where operators may benefit from recommendations for operational commands to manage the plant effectively, but existing methods lack a systematic approach to provide such recommendations based on historical data and simulations.
Innovation Solution
A computer-implemented method that receives alarms from sensors or operators, compares the current plant state to historic states, runs simulations based on variations of historic operational commands, and recommends the command with the highest quality value using a simulation model, thereby providing a data-driven operational command for actuator control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If operators rely on their understanding and experience to control the plant, then operational flexibility is maintained, but decision accuracy in critical situations deteriorates
Solution Approach 1:
The patent introduces a simulation-based recommendation system as an intermediary between the operator and the plant control system. This intermediary provides data-driven operational command recommendations by comparing current plant states with historical states and simulating potential outcomes, thereby improving decision accuracy without replacing operator judgment
Solution Approach 2:
The system performs preliminary actions by pre-computing simulation results for various operational commands based on historical data before critical situations occur. When an alarm is detected, the system quickly retrieves and presents pre-analyzed recommendations, enabling faster and more accurate decisions without real-time computational delays
2Reliability
If a recommendation system is introduced to improve decision-making, then operational command accuracy improves, but system complexity increases
Solution Approach 1:
The patent uses copying by creating virtual replicas of the plant through simulation models. These digital copies allow the system to test and evaluate operational commands in a virtual environment before recommending them to the operator, improving reliability without requiring physical modifications to the plant
Solution Approach 2:
The system performs preliminary analysis by pre-processing historical operational data and pre-computing simulation results for various scenarios. This preliminary action reduces the computational burden during critical situations, maintaining system reliability while managing complexity through advance preparation
3Measurement precision
If simulations are run for multiple operational command variations, then command recommendation quality improves, but computational time increases
Solution Approach 1:
The patent applies partial action by running simulations for a selected subset of the most promising operational command variations rather than exhaustively simulating all possible commands. The system identifies and focuses computational resources on the most relevant scenarios, maintaining high recommendation quality while reducing overall computational time
Solution Approach 2:
The system performs preliminary filtering of operational command options based on historical data analysis before running simulations. By pre-identifying the most relevant command variations to simulate, the system reduces the number of full simulations needed, thereby improving recommendation quality within acceptable timeframes
4Reliability
If historical data is extensively analyzed to improve recommendation accuracy, then decision reliability improves, but data processing complexity increases
Solution Approach 1:
The patent extracts only the most relevant features and parameters from extensive historical operational data rather than processing the complete dataset. By identifying and extracting key discriminative features that are most predictive of successful operational commands, the system improves decision reliability while reducing data processing complexity
Data Source
Figure 1
Figure 2
AI summary
The invention relates to the field of controlling industrial plants, particularly for controlling industrial plants by an operational command. The invention discloses a computer-implemented method for recommending an operational command (32), which is able to control at least one actuator (16) of the industrial plant (10). The method comprises the steps of: receiving an alarm (45) from at least one sensor and/or from an operator (20) of the plant (10), wherein the alarm (45) is related to a current state (40) of the plant (10); obtaining the current state (40) of the plant (10), the current state (40) comprising at least one current process value and/or at least one current operational command (22) related to the plant (10); comparing the current state (40) of the plant (10) to a list of historic states (30) of the plant (10), each historic state (30) comprising a plurality of historic process values (34) and/or at least one historic operational command (32) related to the plant (10); if the current state (40) matches to a subset of at least one of the historic states (30), starting a simulation, based on a simulation model (18) of the plant (10) and the matching historic state (30) as starting state; running a plurality of simulations, each simulation based on a variation of at least one of the historic operational commands (32); determining, for each simulation of the plurality of simulations, a quality value, based on at least one quality criterion; and recommending the variation of the operational command (32), which resulted in the simulation with the highest quality value.