Building Control System Prediction Model Update Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing building equipment control systems face challenges in maintaining accurate prediction models over time, leading to suboptimal performance in minimizing resource consumption costs while ensuring occupant comfort, as prediction model accuracy decays, necessitating ongoing updates to maintain effective optimization.
Innovation Solution
A building system with a control system that generates predictions of load and resource prices, solves optimization problems to minimize costs, monitors error metrics, detects trigger conditions, and updates prediction models to adjust control inputs, ensuring ongoing accuracy and optimal operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If prediction models are used to optimize building equipment control, then cost minimization is improved, but prediction model accuracy decays over time
Solution Approach 1:
The system continuously monitors prediction errors by comparing actual building equipment performance and utility prices against model predictions. When the error metric exceeds a threshold, the system triggers a model update, creating a closed-loop feedback mechanism that maintains prediction accuracy over time without requiring continuous manual intervention
Solution Approach 2:
The system implements periodic model updates based on monitored error metrics rather than continuous updates. The update frequency is dynamically determined by actual model performance degradation, allowing the system to update models only when necessary to maintain accuracy, thus balancing reliability with time efficiency
2Measurement precision
If prediction models are updated frequently to maintain accuracy, then prediction accuracy is improved, but system complexity increases
Solution Approach 1:
The system performs partial updates only when prediction accuracy degrades below a threshold, rather than implementing continuous full model updates. This selective updating approach maintains sufficient prediction accuracy while reducing the computational complexity and resource requirements of the update mechanism
Solution Approach 2:
The system changes the update frequency parameter dynamically based on monitored prediction errors. When errors are low, updates are deferred; when errors exceed thresholds, updates are triggered. This adaptive parameter adjustment simplifies the update system by removing the need for complex scheduling algorithms while maintaining prediction accuracy
3Reliability
If continuous monitoring of prediction errors is implemented, then prediction accuracy is maintained, but computational resources are consumed
Solution Approach 1:
The system uses simple, computationally inexpensive error metrics that can be calculated quickly from existing operational data. Rather than employing complex continuous validation algorithms, the system uses straightforward comparisons between predicted and actual values, maintaining accuracy while minimizing computational energy consumption
Data Source
AI summary
A building system includes building equipment operable to consume one or more resources and a control system configured to generate, based on a prediction model, predictions of a load on the building equipment or a price of the one or more resources for a plurality of time steps in an optimization period, solve, based on the predictions, an optimization problem to generate control inputs for the equipment that minimize a predicted cost of consuming the resources over the optimization period, control the building equipment to operate in accordance with the control inputs, monitor an error metric that characterizes an error between the predictions and actual values of the at least one of the load on the building equipment or the price of the one or more resources during the optimization period, detect an occurrence of a trigger condition, and in response to detecting the trigger condition, update the prediction model.