Air Conditioner Refrigerant Leak Detection Using Neural Network Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current refrigerant leak detection methods in air conditioners are poorly adaptive and have low accuracy due to subjective expert experience, leading to false detections and inadequate universality across different types of air conditioners.
Innovation Solution
A refrigerant leak detection method using a trained neural network model that acquires and processes operating parameters and environment information, including normalization and transformation, to accurately determine refrigerant leaks, improving adaptability and accuracy through machine learning algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If control rules based on expert experience are used for refrigerant leak detection, then the detection method is simple to implement, but the accuracy is low and there are many false detections
Solution Approach 1:
The patent replaces the mechanical system of expert-based control rules with an artificial neural network model. The neural network automatically learns optimal detection thresholds and patterns from training data, substituting human expert judgment with a data-driven intelligent system that achieves higher accuracy while maintaining ease of implementation through automated processing.
Solution Approach 2:
The patent transforms fixed control rules into dynamic detection parameters learned from data. The neural network model adapts its internal parameters (weights and biases) during training to optimize detection accuracy for different air conditioner types and operating conditions, replacing static expert rules with flexible, data-adaptive parameters.
2Adaptability or versatility
If control rules are designed for different types of air conditioners, then the detection can cover various models, but the control parameters are not unified and adaptability is poor
Solution Approach 1:
The patent creates a universal neural network-based detection system that can handle multiple air conditioner types through a single unified model. The neural network learns common patterns across different models during training, enabling one system to serve multiple functions and types without requiring separate control rules for each air conditioner model.
Solution Approach 2:
The neural network model performs self-adaptation by automatically learning optimal detection parameters from training data representing different air conditioner types. Instead of requiring manual configuration for each type, the system self-adjusts its internal parameters to achieve universal applicability across diverse models.
3Measurement precision
If more training data is used for the neural network model, then the detection accuracy increases, but the data processing time and complexity increase
Solution Approach 1:
The patent performs data preprocessing and model training in advance before actual detection operations. By pre-processing the training data and pre-training the neural network model offline, the system eliminates the need for time-consuming computations during real-time detection, achieving high accuracy without sacrificing operational speed.
Data Source
AI summary
Disclosed by the present disclosure is a refrigerant leak detection method and device for an air conditioner. The method includes: acquiring current operating parameters of an air conditioner and environment information of a surrounding environment of the air conditioner; inputting the current operating parameters and the environment information into a trained neural network model to obtain an amount of remaining refrigerant outputted by the neural network model; and determining, according to the amount of the remaining refrigerant, whether there is a refrigerant leak in the air conditioner. The present disclosure improves accuracy of detecting a refrigerant leak in the air conditioner by means of an artificial neural network algorithm.