Fluid Transport Control Using Self-Learning Signal Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing control methods for fluid transport systems, such as HVAC and district heating networks, face challenges due to varying installation conditions and require extensive manual configuration, leading to inefficiencies and high operational costs.
Innovation Solution
A self-learning control process that automatically selects a subset of input signals based on received values, updating itself to optimize performance indicators, reducing the need for manual configuration and enabling efficient control across different systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If Model Predictive Control (MPC) framework is used to improve control performance, then energy efficiency is improved, but device complexity increases due to requiring manual configuration of model structure for each specific system
Solution Approach 1:
The control system automatically performs model structure selection and parameter identification without requiring manual expert configuration. The system self-adjusts by selecting relevant input signals from available sensors and automatically determining model parameters through system identification algorithms, enabling plug-and-play deployment across different HVAC systems.
Solution Approach 2:
The system dynamically adjusts model parameters and structure based on operating conditions. Instead of using a fixed model structure, the system changes parameters such as selected input signals, model order, and parameter values according to the specific system characteristics and operating regime, achieving adaptability without manual reconfiguration.
2Ease of operation
If reinforcement learning is applied to reduce manual configuration, then ease of operation is improved, but training time increases significantly
Solution Approach 1:
The system performs preliminary system identification and model structure selection during a short initial phase before normal operation begins. This preliminary action captures essential system characteristics, allowing the main control algorithm to start with pre-configured parameters and achieve effective control without requiring extensive online training.
Solution Approach 2:
Instead of using all available input signals, the system selectively identifies and uses only the most relevant subset of inputs for control. This partial action approach reduces the dimensionality of the learning problem and training data requirements, significantly decreasing training time while maintaining control performance.
3Measurement precision
If all available input signals are used for control, then measurement precision is improved, but device complexity increases due to processing large number of signals
Solution Approach 1:
The system extracts and selects only the most relevant input signals from the full set of available sensors. Through automatic relevance detection and feature selection algorithms, the system identifies the subset of inputs that most significantly impact control performance, discarding redundant or less important signals to simplify processing.
Solution Approach 2:
The system segments the large set of input signals into relevant and irrelevant groups. By dividing the full signal set and selectively processing only the relevant segment for control decisions, the system maintains measurement precision while reducing computational complexity and processing requirements.
Data Source
AI summary
A method for controlling operation of a fluid transport system by applying a self-learning control process. The method includes: receiving obtained values of input signals during operation of the system during a first period of time, which is controlled by a predetermined control process, automatically selecting a subset of the input signals based on the received obtained values of the input signals, receiving obtained values of at least the selected subset of input signals during a second period of time, which is controlled by applying the self-learning control process, which is configured to control operation based only on the selected subset of input signals, and wherein applying the self-learning control process includes updating the self-learning control process based on the received obtained values of the selected subset of the input signals and based on at least an approximation of a performance indicator function.