Abstract
The implementations of residential demand response (DR) based on heating, ventilation, and air conditioning (HVAC) are inseparable from effective control algorithms for coordinating the operating schedules of multiple HVAC devices. In this work, both model-based and data-driven HVAC control strategies are developed to determine the optimal control actions for HVAC systems. The control objectives are to minimize customers' electricity costs, customers' discomfort, and the utility-level load violation. In the model-based approach, a thermal resistance-capacitance (RC) HVAC model is formulated to capture buildings' thermodynamic behaviors, and a distributed solution algorithm (i.e., alternating direction method of multipliers) is applied to determine the day-ahead HVAC operation schedules. In the data-driven approach, the neural networks continuously interact with the environment during the training process to learn what control actions to take under certain circumstances and then are used for online decision-making. The case study is performed on a utility system with one hundred houses. Simulation results demonstrate that the model-based approach can save 22% of the total cost compared to the data-driven approach, while the data-driven approach does not require outdoor temperature forecast information and its computational speed is 46 times faster than that of the model-based approach.
Highlights
A S ELECTRICAL load continues to grow, it is highly interesting for power utilities to reduce the system peak demand and increase the utilization of electricity infrastructure with minimal investment in power generation and delivery systems
In this paper, both model-based and data-driven HVAC control strategies are developed for residential demand response (DR) management programs
In the data-driven approach, it is assumed that detailed building models and/or day-ahead outdoor temperature forecasting are not available, and the agents continuously interact with the environment to learn competitive HVAC control actions
Summary
A S ELECTRICAL load continues to grow, it is highly interesting for power utilities to reduce the system peak demand and increase the utilization of electricity infrastructure with minimal investment in power generation and delivery systems. Most of the proposed methods, either model-based or data-driven, are only deployed for single households and cannot be extended to large-scale residential demand control and optimization. Our work aims to fill the above research gaps by comparing the performance and application scenarios of different approaches Both the model-based and the data-driven HVAC control strategies are developed for residential DR programs in this work. Our contributions are summarized as follows: 1) applying a distributed system architecture to decrease the computational complexity and the action space dimension of the central controller, 2) describing a distributed model-based solution algorithm for large-scale residential DR, 3) presenting a data-driven HVAC control strategy for residential DR without the knowledge of building models and weather forecasting data, and.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
