Building HVAC Control Using Digital Twins and Particle Swarm Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Commercial buildings face challenges in maintaining optimal energy efficiency due to aging equipment, changing occupant behavior, and external factors like weather conditions, leading to significant energy consumption and CO2 emissions.
Innovation Solution
A computer-implemented control system using a model-based learning system with particle swarm optimization, which performs a zero-touch energy audit and optimization by deploying sensors to monitor energy usage and adjust settings of energy-consuming devices for improved efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional energy audit methods are used with on-site engineer visits, then detailed energy loss assessment can be obtained, but high time consumption and operational complexity occur
Solution Approach 1:
The patent creates a virtual copy of the building's energy system through digital twins and simulation models. These virtual replicas allow engineers to assess energy losses and optimize performance without physically visiting the site, thereby maintaining measurement precision while eliminating time consumption associated with on-site audits.
Solution Approach 2:
The patent replaces the mechanical system of on-site engineer visits with an automated remote monitoring and analysis system. Sensors, communication networks, and computational algorithms substitute for physical presence, enabling detailed energy assessment to be performed remotely and automatically, thus resolving the contradiction between assessment accuracy and time efficiency.
2Reliability
If building equipment operates continuously to maintain comfort, then occupant comfort is ensured, but energy consumption increases
Solution Approach 1:
The patent implements dynamic control strategies where HVAC equipment operation is continuously adjusted based on real-time environmental conditions, occupancy patterns, and predictive analytics. This allows the system to maintain occupant comfort reliability by adapting equipment operation to actual needs rather than running continuously, thereby reducing energy consumption.
Solution Approach 2:
The patent employs feedback mechanisms where sensors continuously monitor indoor environmental conditions and occupancy, and this information feeds back to the control system to adjust equipment operation. This closed-loop control ensures comfort is maintained only when and where needed, preventing unnecessary energy consumption from continuous operation.
3Ease of operation
If fixed thermostat settings are used, then system operation is simple, but energy efficiency deteriorates due to changing conditions
Solution Approach 1:
The patent implements intelligent thermostat systems that automatically adapt to changing conditions without requiring manual user adjustment. These systems use sensors, machine learning algorithms, and predictive models to self-optimize temperature settings based on occupancy patterns, weather forecasts, and historical data, thereby maintaining energy efficiency while preserving operational simplicity for users.
Solution Approach 2:
The patent uses predictive analytics and weather forecasting to pre-adjust thermostat settings before changes in environmental conditions or occupancy patterns occur. This preliminary action allows the system to proactively optimize energy efficiency in response to anticipated changes, rather than reacting to changes after they occur, thereby maintaining both simplicity and efficiency.
Data Source
AI summary
A simulation processor generates and stores a simulation model based on conditions associated with a physical structure, such as a building. A neural network processor implements a neural network, having an input layer coupled to receive sensor data from the structure and having an output layer coupled to supply control signals to the at least one electrically operable environmental control device. The neural network is trained using the simulation model. A particle swarm optimization processor programmed to receive the simulation results and perform particle swarm optimization, ascertains optimal parameters for controlling the at least one electrically operable environmental control device and supplies these optimal parameters to the neural network processor. The neural network processor uses the optimal parameters supplied by the particle swarm optimization processor to further train the neural network.