Learning device and inference device for maintenance of air conditioner
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the maintenance of air filters in air conditioners fail to consider the maintenance of air filters in air conditioners to consider the maintenance of air filters in air conditioners to consider the maintenance of air filters in air conditioning systems are not comprehensive, leading to inefficient and costly maintenance.
Innovation Solution
A learning device and inference device that utilize trained models to estimate the degree of clogging, air-conditioning power, and maintenance cost of air filters, incorporating neural networks for supervised learning to determine the optimal maintenance timing based on various factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a rule-based method using current value comparison is used to detect air filter clogging, then the detection reliability is improved, but the maintenance cost cannot be optimized because other factors are not considered
Solution Approach 1:
The patent segments the maintenance determination process into three separate machine learning models: a first model for estimating clogging degree from current values, a second model for estimating power consumption based on temporal information, and a third model for determining optimal maintenance timing by integrating outputs from the first two models. This segmentation allows each model to specialize in specific aspects while maintaining overall system reliability and comprehensiveness.
Solution Approach 2:
The patent transitions from a single-parameter rule-based method (current value comparison) to a multi-parameter machine learning approach that incorporates current values, temporal information, power consumption data, and maintenance cost considerations. This parameter expansion enables comprehensive maintenance timing determination while preserving detection reliability through the first model's specialized clogging estimation.
2Productivity
If maintenance is performed frequently to ensure air filter cleanliness, then the air-conditioning system efficiency is improved, but the maintenance cost increases
Solution Approach 1:
The patent implements a dynamic maintenance timing determination system that adapts to varying operating conditions. The third model integrates temporal information processed by the second model to determine optimal maintenance timing based on actual system state, power consumption patterns, and clogging degree estimates. This dynamic approach replaces static frequent maintenance schedules with adaptive timing that maintains efficiency while reducing unnecessary maintenance costs.
Solution Approach 2:
The patent employs feedback mechanisms where the machine learning models continuously process operating data, power consumption information, and maintenance history to refine maintenance timing recommendations. The system learns from past maintenance outcomes and operational patterns to optimize future maintenance schedules, balancing efficiency maintenance with cost reduction through data-driven feedback loops.
3Reliability
If maintenance is performed based on fixed schedules regardless of actual condition, then the system operation reliability is maintained, but the maintenance cost increases and efficiency is reduced
Solution Approach 1:
The patent uses the first machine learning model to preliminarily estimate the clogging degree of the air filter based on current sensor values before determining maintenance timing. This preliminary assessment allows the system to proactively identify when maintenance is actually needed rather than following fixed schedules, maintaining reliability by detecting clogging early while avoiding unnecessary maintenance actions that waste resources.
Solution Approach 2:
The patent implements a self-service maintenance determination system where machine learning models automatically analyze operating data, power consumption patterns, and clogging indicators to determine optimal maintenance timing without requiring external intervention or fixed schedule adherence. The system serves itself by learning from its own operational data to optimize maintenance decisions, reducing both cost and efficiency loss while maintaining reliability.
Data Source
AI summary
A model generation unit converts each of a first model, a second model, and a third model into a trained model. First training data includes a first parameter representing the degree of clogging of an air filter, a second parameter pertaining to the air-conditioning power of an air-conditioning system, and a third parameter representing an increased amount of electric power cost of the air-conditioning system due to the first parameter during operation of the second parameter. Second training data includes a fourth parameter representing a first date and time and a fifth parameter pertaining to air-conditioning power of the air-conditioning system assumed on the first date and time. Third training data includes a sixth parameter representing a second date and time and a seventh parameter representing a maintenance cost of the air filter on the second date and time.