Laundry Machine Foreign Object Detection Using AI Noise Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Laundry machines lack effective detection methods for foreign objects, such as metals and hard plastics, which can cause damage to the tub during the washing or drying process, leading to noise issues and potential equipment damage.
Innovation Solution
A method and system utilizing a microphone and artificial intelligence (AI) to detect noise patterns generated by foreign objects, estimating their type, and stopping the laundry machine operation to prevent damage, with a server-based AI model learner generating a foreign object classifying engine for accurate classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a microphone is mounted to detect noise during washing/dewatering, then noise due to unbalance can be detected, but it becomes difficult to distinguish various noise patterns caused by foreign objects
Solution Approach 1:
The patent segments the noise detection task into multiple frequency bands using FFT (Fast Fourier Transform). By dividing the noise spectrum into different frequency ranges, the system can identify specific noise patterns characteristic of foreign objects versus unbalance conditions, thereby improving foreign object identification accuracy while maintaining reliable noise detection.
Solution Approach 2:
The patent transitions from analyzing noise in the time domain to the frequency domain using spectral analysis. This dimensional change allows the system to distinguish between different noise sources (foreign objects vs. unbalance) by their unique frequency signatures, resolving the contradiction between detecting overall noise and identifying specific foreign object patterns.
2Measurement precision
If AI technology is used to estimate foreign object types, then foreign objects can be identified at the initial stage, but computational burden increases
Solution Approach 1:
The patent implements preliminary action by training the AI model offline before deployment. The foreign object classification model is pre-trained on a dataset of noise patterns during the manufacturing or setup phase. During actual washing operations, the pre-trained model only needs to perform inference with minimal computational resources, achieving high detection accuracy while minimizing real-time energy consumption.
Solution Approach 2:
The patent uses a simplified version of the AI model for embedded deployment. Instead of running complex training algorithms on the washing machine's microprocessor, the system copies the trained model parameters to the device and uses only the inference portion, which requires significantly less computational power and energy while maintaining foreign object detection accuracy.
3Reliability
If the laundry machine stops operation to prevent damage, then tub and fabric protection is improved, but washing productivity decreases
Solution Approach 1:
The patent implements continuous feedback monitoring of noise patterns during the washing cycle. The system real-time analyzes frequency spectrum data and provides immediate feedback when foreign objects are detected. This allows the machine to stop only when necessary (when foreign objects are confirmed), rather than stopping for every minor anomaly, thus protecting the tub while minimizing impact on washing productivity.
Solution Approach 2:
The patent changes the operational parameters dynamically based on detected conditions. When foreign objects are detected, the system adjusts parameters such as drum speed, water flow, or heating to mitigate damage while continuing operation. This selective parameter adjustment provides tub protection without requiring a complete stop, thereby maintaining washing efficiency.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Minimizes tub and fabric damage by identifying foreign objects at the beginning of the washing/drying process, preventing noise and damage through AI-powered sound recognition and 5G network communication, distributing computational burdens for efficient noise data processing.
Implementation Method 1
collecting noise data generated when a tub rotates after putting a laundry in the tub through a microphone mounted in a laundry machine
Data Source
AI summary
A foreign object detecting system includes an apparatus of detecting foreign objects in a laundry machine and a server. The server includes an artificial intelligence model learner configured to generate a foreign object classifying engine trained with the collected noise data through an artificial neural network. the server transmits the trained foreign object classifying engine trained through the artificial intelligence model learner to the foreign object classifying apparatus, the foreign object type classifier classifies the type of foreign objects through the trained foreign object classifying engine transmitted from the server, and the communicator of the apparatus of detecting foreign objects in a laundry machine transmits information about the type of foreign objects to a foreign object appliance. According to the present disclosure, the foreign objects in the laundry machine can be detected using artificial intelligence (AI), an artificial intelligence-based foreign object type classifying technique, and a 5G network.