Industrial Setpoint Control With AI Selection and Safety Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional industrial process control systems, relying on hardcoded rulesets or mathematical models, fail to learn or improve over time, leading to performance limitations and safety risks, especially in mission-critical systems, and AI-based control systems introduce unpredictability and network vulnerabilities.
Innovation Solution
A multi-layered approach using an AI agent constrained by industrial system constraints, a setpoint source selector, and a limit module to ensure setpoint values remain within safety boundaries, preserving local control systems for fallback and maintaining deterministic control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If AI-based control systems are used to predict control parameters, then system adaptability and learning capability are improved, but determinism and safety are worsened due to unpredictable recommendations
Solution Approach 1:
The control system is segmented into multiple independent components: AI-based controllers for adaptability, traditional deterministic controllers for reliability, and a selector mechanism. Each component operates independently with defined responsibilities, allowing the system to leverage both AI adaptability and traditional determinism without compromising either.
Solution Approach 2:
A controller selector acts as an intermediary between AI-based controllers and traditional deterministic controllers. This mediator evaluates conditions and determines which controller should execute at any given time, ensuring that AI recommendations are only followed when they meet predetermined safety and performance criteria, thus maintaining determinism while enabling adaptability.
2Speed
If local AI-based control systems are deployed, then response time is improved, but maintenance complexity and operational difficulty are worsened
Solution Approach 1:
The controller selector is designed with universal functionality that works across different AI-based controllers and traditional controllers. It implements a standardized interface and evaluation framework that can handle various controller types, making the system easier to maintain and operate while preserving fast local response times.
Solution Approach 2:
The system allows dynamic adjustment of selector parameters such as performance thresholds, safety criteria, and controller priorities. This flexibility enables operators to optimize maintenance and operation based on specific industrial contexts without requiring deep AI expertise, reducing operational difficulty while maintaining fast response capability.
3Adaptability or versatility
If remote AI-based control systems are used, then network dependency is reduced, but control reliability is worsened due to network outages and latency
Solution Approach 1:
The system implements beforehand cushioning by preparing fallback mechanisms in advance. When network issues or latency problems occur, the controller selector automatically transitions to traditional deterministic controllers or local models that do not depend on remote AI systems, ensuring control continuity and reliability without sacrificing the ability to use remote AI when available.
Data Source
AI summary
In variants, a method for industrial process control can include: determining AI setpoint values using an AI agent; determining local setpoint values using a local control system; selecting a setpoint source from a set of candidate setpoint sources including the AI agent and the local control system; optionally determining a set of transition setpoint values based on setpoints provided by the setpoint source; and limiting the setpoint values; wherein the industrial system is operated based on the limited setpoint values.