Building Heat Load Estimation Using Deterministic and Stochastic Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current building environmental control systems face challenges in accurately predicting and managing heat disturbances, which are complex and nonlinear, leading to inefficiencies in heating and cooling processes and increased energy costs.
Innovation Solution
The implementation of a system that uses a processing circuit to perform system identification, augment the system model with a disturbance model, estimate historical heat disturbances, train heat disturbance models, and predict heat disturbances using deterministic and stochastic models to control building equipment effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional control systems are used for building environmental control, then the system structure is simple, but the accuracy of heat disturbance prediction is insufficient leading to energy inefficiency
Solution Approach 1:
The heat disturbance model is segmented into two distinct components: a deterministic model that captures periodic patterns (daily/weekly cycles) and a stochastic model that handles random fluctuations. This segmentation allows each model to specialize in capturing specific types of heat disturbances, improving overall prediction accuracy without requiring a single overly complex model
Solution Approach 2:
A disturbance model is introduced as an intermediary component between the building equipment and the control system. This disturbance model explicitly represents heat disturbances as a separate entity that can be estimated and compensated for, allowing the control system to account for thermal effects without directly modifying the building equipment
2Measurement precision
If deterministic and stochastic models are combined for heat disturbance prediction, then the prediction accuracy is improved, but the computational complexity increases
Solution Approach 1:
The system dynamically adapts the complexity of its models based on operating conditions. The deterministic model handles routine periodic patterns efficiently, while the stochastic model is activated when random fluctuations exceed threshold levels. This dynamic approach ensures high accuracy when needed while reducing computational overhead during stable periods
Solution Approach 2:
Rather than continuously running both deterministic and stochastic models at full complexity, the system applies partial action by using only the deterministic model during periods when periodic patterns dominate, and adding stochastic modeling only when thermal fluctuations exceed expected periodic variations. This reduces average computational power requirements while maintaining accuracy when disturbances are significant
3Measurement precision
If system identification is performed to identify model parameters, then the model accuracy is improved, but the data processing time is increased
Solution Approach 1:
System identification and model parameter calibration are performed in advance during commissioning and periodic re-commissioning phases, rather than in real-time during operation. Historical data is processed offline to identify deterministic model parameters (periods, amplitudes, phase shifts) and stochastic model characteristics, allowing the control system to use pre-characterized models during actual operation, thus avoiding real-time computational delays
Data Source
AI summary
An environmental control system for a building including building equipment operable to affect a variable state or condition of the building. The system includes a controller including a processing circuit. The processing circuit can obtain training data relating to operation of the building equipment and can perform a system identification process to identify parameters of a system model using the training data. The processing circuit can augment the system model with a disturbance model and estimate values of a historical heat disturbance in the training data based on the augmented system model. The processing circuit can train one or more heat disturbance models based on the training data and the estimated values. The processing circuit can predict a heat disturbance using the augmented system model along with the one or more heat disturbance models and can control the building equipment based on the predicted heat disturbance.