Controlling vapor compression system using probabilistic surrogate model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vapor compression systems, such as heat pumps and air-conditioning systems, face inefficiencies due to the complexity of mathematical models and slow convergence of extremum-seeking controllers, which hinder real-time optimization and energy efficiency.
Innovation Solution
A data-driven approach using a probabilistic surrogate model and Bayesian optimization to determine optimal setpoints for vapor compression systems, separating optimization from control and incorporating uncertainties to reduce the number of samples needed for model construction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If mathematical models are used to predict optimal control inputs, then energy efficiency can be improved, but the models become difficult to derive, calibrate, and update
Solution Approach 1:
The patent creates a simplified surrogate model that copies the essential input-output behavior of the complex vapor compression system without replicating its full physical complexity. This surrogate model uses measured operational data to establish relationships between control inputs and energy consumption, avoiding the need to derive and calibrate complex first-principles mathematical models while still enabling energy optimization.
Solution Approach 2:
The patent replaces traditional physics-based mathematical modeling with a data-driven surrogate model approach. Instead of using complex thermodynamic equations and physical principles that require extensive calibration, the system uses empirical data from actual system operation to build a simplified predictive model that captures the essential energy consumption patterns.
2Loss of energy
If extremum-seeking controllers are used to optimize control inputs, then energy consumption can be minimized, but the convergence time becomes several hours
Solution Approach 1:
The patent performs preliminary optimization using the surrogate model to predict and identify near-optimal control settings before applying them to the actual system. This pre-computation step uses the fast-evaluating surrogate model to explore the control space and converge to optimal settings, avoiding the slow real-time convergence requirements of traditional extremum-seeking controllers.
Solution Approach 2:
The surrogate model serves as an intermediary between the control system and the actual vapor compression system. It acts as a fast, simplified proxy that enables rapid optimization calculations, allowing the system to quickly identify optimal control inputs without requiring slow, iterative real-time convergence of complex controllers.
3Loss of energy
If traditional optimization methods are used, then optimal setpoints can be determined, but the computational burden increases and real-time optimization is hindered
Solution Approach 1:
The patent uses a computationally inexpensive surrogate model that can be rapidly evaluated many times during optimization. This lightweight model sacrifices some fidelity compared to full physics-based models but enables fast, real-time optimization calculations that can be performed repeatedly without excessive computational burden.
Data Source
AI summary
A controller for controlling a vapor compression system is provided. The controller is configured to control an operation of the VCS with different combinations of setpoints for different actuators of the VCS to estimate a cost of operation of the VCS for each of the different combinations of setpoints, and compute, using a Bayesian optimization of the combinations of setpoints and their corresponding estimated costs of operation, a probabilistic surrogate model, wherein the probabilistic surrogate model defines at least first two order moments of the cost of operation in the probabilistic mapping. The controller is further configured to select an optimal combination of setpoints having the largest likelihood of being a global minimum at the surrogate model according to an acquisition function of the first two order moments of the cost of operation.