Smart thermostat with model predictive control and demand response integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional thermostats operate using fixed temperature setpoints, leading to suboptimal control of HVAC equipment and increased energy costs, as they do not account for time-varying energy prices or zone heat transfer characteristics.
Innovation Solution
A system comprising a network of thermostats and a computing system that uses model predictive control to determine optimal temperature setpoints by analyzing thermal behavior, performing system identification experiments, and adjusting based on utility provider data to minimize energy usage and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If conventional thermostats use fixed temperature setpoints, then the operation is simple and reliable, but energy costs increase and control optimality deteriorates
Solution Approach 1:
The system performs system identification experiments in advance to build thermal models of the building zone, capturing heat transfer characteristics and thermal mass effects. These pre-acquired models are then used by the model predictive controller to optimize future temperature setpoints, allowing the system to anticipate and prepare for upcoming thermal conditions rather than reacting to current deviations only.
Solution Approach 2:
The thermostat transitions from static fixed setpoints to dynamic optimized setpoints generated by model predictive control. The system continuously adjusts temperature setpoints based on predicted thermal behavior, time-varying energy prices, and building-specific thermal characteristics, enabling adaptive optimization while maintaining computational tractability through efficient MPC algorithms.
2Loss of energy
If model predictive control is implemented to optimize temperature setpoints, then energy efficiency improves, but system complexity and computational requirements increase
Solution Approach 1:
The system performs system identification experiments in advance to build thermal models of the building zone, capturing heat transfer characteristics and thermal mass effects. These pre-acquired models are then used by the model predictive controller to optimize future temperature setpoints, allowing the system to anticipate and prepare for upcoming thermal conditions rather than reacting to current deviations only.
Solution Approach 2:
The model predictive control system continuously monitors actual temperature deviations from setpoints and uses this feedback to update predictions and adjust future setpoints. The system incorporates real-time temperature measurements along with the thermal model to minimize prediction errors and optimize energy consumption dynamically, creating a closed-loop control system that adapts to changing conditions.
3Measurement precision
If system identification experiments are conducted to build thermal models, then control accuracy improves, but user comfort and operational simplicity may deteriorate
Solution Approach 1:
The system performs system identification experiments autonomously without requiring direct user intervention. The thermostat automatically implements the identification protocol, collects temperature data, and builds the thermal model independently. Users benefit from improved control accuracy without needing to understand or configure the complex identification process, as the system serves itself in acquiring and utilizing thermal characteristics.
Solution Approach 2:
The model predictive controller acts as an intermediary between the building's thermal dynamics and the HVAC system. It translates complex thermal models and energy price signals into simple, actionable temperature setpoints that the thermostat can directly implement. This intermediary layer shields users from complexity while delivering optimized performance through automated decision-making.
Data Source
AI summary
A system includes a plurality of thermostats corresponding to a plurality of HVAC systems that serve a plurality of spaces and a computing system communicable with the plurality of thermostats via a network. The computing system is configured to, for each space of the plurality of spaces, obtain a set of training data relating to thermal behavior of the space, identify a model of thermal behavior of the space based on the set of training data, perform a model predictive control process using the model of thermal behavior of the space to obtain a temperature setpoint for the space, and provide the temperature setpoint to the thermostat corresponding to the HVAC system serving the space. The plurality of thermostats are configured to control the plurality of HVAC systems in accordance with the temperature setpoints.