Electronic apparatus and method to train neural network to determine defective air conditioner
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Smart building systems currently determine air conditioner defects without considering weather information or building characteristics, leading to reduced accuracy and user inconvenience.
Innovation Solution
An electronic apparatus that uses a neural network model to acquire weather and space information, predicts temperature, and determines defects in air conditioners by comparing predicted and measured temperatures, generating notifications when defects are detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a neural network model is used to predict temperature and determine defects, then measurement precision and reliability improve, but device complexity increases
Solution Approach 1:
The neural network model is trained in advance using historical weather data, space information, and air conditioner operation data to learn the relationship between environmental factors and temperature variations. This pre-training allows the system to make accurate defect predictions without requiring complex real-time calculations, thereby improving measurement precision while managing device complexity.
Solution Approach 2:
The neural network model acts as an intermediary between raw input data (weather information, space characteristics, operation data) and defect determination. It processes and integrates multiple data sources to produce predicted temperature information, which is then compared with actual temperature to identify defects. This intermediary approach simplifies the overall system architecture while maintaining high prediction accuracy.
2Reliability
If weather information and space characteristics are considered in defect determination, then reliability improves, but loss of information increases due to additional data processing requirements
Solution Approach 1:
The input data is segmented into distinct categories: weather information (external temperature, humidity), space information (volume, insulation characteristics, occupancy), and air conditioner operation data (running time, set temperature). This segmentation allows the neural network to process each type of information separately and efficiently, reducing data processing overhead while maintaining comprehensive analysis for reliable defect determination.
Solution Approach 2:
The system transforms raw weather information and space characteristics into standardized parameters that the neural network can process efficiently. By converting diverse data types into uniform parameter formats, the system reduces information loss during processing while maintaining the reliability needed for accurate defect detection.
Data Source
AI summary
An electronic apparatus is provided. The electronic apparatus includes a communication interface, a memory, and a processor. The processor according to the disclosure is configured to acquire weather information and information on a space where an air conditioner is installed, train a neural network model based on the weather information and the information on the space, based on acquiring driving information of the air conditioner and a measured temperature of the space through the communication interface, input the measured temperature of the space and the external temperature into the neural network model and acquire predicted temperature information per time for the space, determine whether a defect exists in the air conditioner based on the predicted temperature information and the measured temperature of the space, and based on determining that a defect exists in the air conditioner, generate a notification signal.