Decentralized Building Energy Management via Model Predictive Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The current electricity market is inelastic, leading to instability and load management difficulties, with direct load control causing consumer dissatisfaction and price-based methods being less effective due to lack of smart price control, resulting in inefficient energy usage and high costs.
Innovation Solution
A decentralized control system using model predictive control (MPC) is introduced for major consumers in residential buildings, incorporating occupancy profiles and device usage patterns to optimize energy usage, with mixed integer linear programming and flexible appliance scheduling, allowing appliances like AC units, EVs, and water heaters to shift consumption based on grid price signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If direct load control is used to manage electricity demand, then utility companies can shut down devices during high demand periods, but consumers experience dissatisfaction and the method cannot be performed frequently
Solution Approach 1:
The system enables consumers to automatically manage their own electricity demand through smart appliances and devices that self-regulate based on pricing signals and user-defined preferences, eliminating the need for utility-imposed shutdowns while maintaining demand management effectiveness
Solution Approach 2:
The system implements real-time feedback loops where smart meters communicate pricing signals to consumer devices, which then automatically adjust their operation accordingly, creating a responsive demand management system that respects consumer preferences while achieving load management goals
2Adaptability or versatility
If price-based control methods are used to encourage consumption shifting, then consumers can adjust usage based on electricity prices, but the methods are less effective due to lack of smart price-based control
Solution Approach 1:
The system allows consumers to pre-configure their preferences, constraints, and priorities for different appliances before peak pricing periods occur, enabling automatic optimal decision-making that achieves effective demand response without requiring real-time consumer intervention
Solution Approach 2:
The system dynamically adjusts appliance operation based on real-time pricing signals while respecting pre-configured consumer preferences, creating a flexible yet effective demand response mechanism that adapts to changing price conditions while maintaining consumer satisfaction
3Ease of operation
If traditional on/off controllers are used for appliances, then operation is simple, but cost savings during dynamic pricing periods are limited
Solution Approach 1:
The system replaces simple mechanical on/off controllers with intelligent control algorithms that use pricing signals and predictive models to automatically optimize appliance operation, achieving significant cost savings while maintaining ease of use through automation
Solution Approach 2:
The system changes the control parameters from binary on/off states to continuous optimization based on pricing signals, thermal models, and appliance-specific constraints, enabling cost-effective operation while maintaining simplicity through automated decision-making
Data Source
AI summary
Measurements of energy usage including details of power consumption can be stored for power usage devices. The measurements of energy usage can be used to predict future consumption for each of the power usage devices. A power consumption can be modified using the prediction. The energy cost can be optimized based using the prediction, such as, for example, by modifying a power consumption of one of the power usage devices.