Creating equipment control sequences from constraint data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building automation systems, particularly HVAC control systems, face challenges in efficiently managing energy costs and equipment wear due to their retroactive response mechanisms, which fail to anticipate and optimize building conditions proactively, leading to suboptimal performance in complex and diverse building environments.
Innovation Solution
A method utilizing machine learning engines to create equipment control sequences by accessing constraint state curves and thermodynamic models of controlled spaces, training heterogeneous neural networks with sensor data to optimize equipment operation, thereby determining optimal control states and sequences for building equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If retroactive HVAC control systems are used to respond to current building conditions, then equipment can be operated to maintain comfort parameters, but energy costs and equipment wear cannot be optimized proactively
Solution Approach 1:
The system performs preliminary actions by using machine learning models to predict future building conditions and pre-adjust equipment operation. Instead of reacting to current conditions, the system anticipates future temperature changes, occupancy patterns, and weather conditions, then proactively modifies HVAC operation to meet future comfort requirements while minimizing energy consumption and equipment wear.
2Measurement precision
If complex thermodynamic models are used to represent building systems, then modeling precision is improved, but computational resources and processing power requirements increase
Solution Approach 1:
The system transforms complex thermodynamic models into simplified neural network representations by changing the parameter form from continuous differential equations to discrete neural network weights and activations. This parameter transformation maintains the essential thermodynamic relationships while enabling efficient computation on standard hardware, reducing the computational burden while preserving modeling accuracy.
Data Source
AI summary
A structure thermodynamic model, which models the physical characteristics of a controlled space, inputs a constraint state curve which gives constraints, such as temperature, that a controlled space is to meet; and outputs a state injection time series which is the amount of state needed for the controlled space to optimize the constraint state curve. The state injection time series curve is then used as input into an equipment model, which models equipment behavior in the controlled space. The equipment model outputs equipment control actions per control time (a control sequence) which can be used to control the equipment in the controlled space. Some embodiments train the models using training data.