Building Comfort Forecasting via Dynamic Model Calibration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

It is challenging for non-experts to predict the impact of controlling active elements like heating and shading on interior comfort in buildings due to their combined or antagonistic influences and the complex interactions with building thermal characteristics and external heat transfers.

Innovation Solution

A method for predicting the temporal evolution of interior comfort parameters like temperature, luminosity, or hygrometry, which involves acquiring initial measurement values, forecasting external climatic conditions, and accounting for thermal contributions and exchanges within the building, allowing for dynamic calibration and scenario comparison to optimize comfort management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a user controls active elements (heating, shading) to influence interior comfort, then the interior comfort parameter can be adjusted, but it becomes difficult to predict the impact due to combined or antagonistic influences and complex thermal interactions

Engineering Contradiction:
Improveprediction accuracy of interior comfortVSAvoidcomplexity of thermal interactions
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by forecasting external climatic conditions (temperature, humidity, solar radiation) in advance and pre-calculating their impact on interior comfort parameters. This allows the system to proactively adjust active elements before the actual thermal effects occur, improving prediction accuracy while managing complexity through advance planning

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by continuously monitoring actual interior comfort measurements and comparing them with predicted values. The difference between measured and predicted values is used to dynamically adjust the model parameters and improve future predictions, resolving the contradiction between prediction accuracy and system complexity through adaptive learning

Inventive Principle:
Principle #23Feedback

2Adaptability or versatility

If multiple active elements are controlled simultaneously to achieve desired interior comfort, then comfort levels can be optimized, but the combined and antagonistic influences make prediction difficult

Engineering Contradiction:
Improveability to optimize comfort under various conditionsVSAvoiddifficulty in tracking thermal contributions
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The system segments the thermal contributions by separately identifying and quantifying the impact of each active element (heating, shading, ventilation) and each external factor (solar radiation, ambient temperature, humidity). This segmentation allows the system to track individual contributions while managing multiple simultaneous controls, reducing information loss through systematic decomposition of complex thermal interactions

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If real-time measurements and forecasts are continuously updated to improve prediction accuracy, then adaptive comfort management is achieved, but computational requirements and system complexity increase

Engineering Contradiction:
Improveaccuracy of interior comfort forecastVSAvoidcomputational energy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system applies partial action by updating only the most critical model parameters and forecasts that have the greatest impact on prediction accuracy, rather than continuously recalculating all parameters. This selective updating approach maintains adequate prediction accuracy while reducing computational energy consumption by focusing resources on the most influential variables

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3087447B1Method for forecasting an interior comfort parameter and associated terminal
Publication Date: 2019.12.25 SOMFY ACTIVITES SA
  • EP3087447B1 patent drawingFigure 1~2
  • EP3087447B1 patent drawingFigure 3
  • EP3087447B1 patent drawingFigure 4

AI summary

Terminal and method for forecasting the behaviour of a quantity representative of the interior comfort of a building such as a temperature, a degree of luminosity or a degree of humidity. The method comprises a step (E1) consisting in employing a forecast model (M). The method comprises a step (E2) consisting in acquiring at least one first measurement value (MGI) at a first given instant (t1). The method comprises a step (E3) consisting in employing a first forecast (PGE) of the temporal evolution of a quantity representative of the exterior climatic conditions. The method comprises a step (E4) consisting in establishing a second forecast (PGI) of the temporal evolution by applying the forecast model (M), said second forecast (PGI) comprising an estimation of the value of the quantity representative of the interior comfort for at least one second given instant (t2) subsequent to the first given instant (t1).