Air conditioner timing estimation controller using machine learning model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing air-conditioning control devices face challenges in accurately estimating startup times using machine learning models when insufficient training data is available, leading to inappropriate predictions.
Innovation Solution
An air-conditioning control device that employs a non-learning model for extrapolation in sparse data regions, combining dense and sparse training data to improve estimation accuracy by determining whether air-conditioning data is present in sparse regions and applying extrapolation techniques to estimate startup times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a machine learning model is used to estimate startup time, then estimation capability is improved, but estimation accuracy deteriorates when training data is insufficient
Solution Approach 1:
The patent applies different modeling approaches to different data density regions. In dense regions, a machine learning model is used for estimation, while in sparse regions, a non-learning model (linear extrapolation) is used. This local differentiation of modeling quality resolves the contradiction by matching the estimation method to the data availability in each region.
Solution Approach 2:
The patent segments the feature space into dense regions and sparse regions based on training data distribution. By dividing the estimation problem into two segments (dense region estimation and sparse region extrapolation), the system can apply appropriate methods to each segment, maintaining high accuracy overall while utilizing the automated ML approach where data permits.
2Ease of manufacture
If machine learning learning is performed with insufficient training data, then model creation is simplified, but prediction reliability deteriorates
Solution Approach 1:
The patent changes the estimation approach parameter based on data density. When training data is insufficient in certain regions, the system switches from ML-based estimation to linear extrapolation, effectively changing the operational parameter of the estimation system to maintain reliability without requiring more complex model creation processes.
3Device complexity
If a single machine learning model is used for all regions, then device complexity is reduced, but estimation accuracy in sparse regions deteriorates
Solution Approach 1:
The patent implements a dynamic estimation system that automatically selects between ML-based estimation and linear extrapolation based on the data density of the input features. This dynamic adaptation allows the system to maintain simplicity in data-rich regions while achieving high accuracy in data-sparse regions, resolving the contradiction between complexity and precision.
Data Source
AI summary
An air-conditioning control device includes: an air-conditioning data acquisition unit to acquire air-conditioning data; a determination unit to determine whether or not the air-conditioning data is present in a sparse region; an estimation unit to, in a case where the determination unit determines that the air-conditioning data is present in the sparse region, apply extrapolation using a non-learning model for associating the air-conditioning data with a startup time of an air conditioner to a machine learning model, and estimate the startup time of the air conditioner from the air-conditioning data by using the non-learning model; and a control unit to perform control to start up the air conditioner at the startup time estimated by the estimation unit.