Artificial intelligence washing machine and controlling method therefor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional washing machines fail to accurately detect laundry amount and material, leading to insufficient washing performance and potential damage to laundry, as they only consider the amount rather than the type of laundry during the washing process.
Innovation Solution
A washing machine utilizing machine learning and an artificial neural network to detect laundry amount and material by analyzing current values from the motor during accelerated rotation, allowing for adaptive washing modes based on the detected amount and state of the laundry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional washing machines only detect laundry amount without considering laundry material, then the detection process is simple, but washing performance is insufficient and laundry may be damaged
Solution Approach 1:
The detection process is divided into two distinct phases: a learning phase where the system collects and stores current patterns corresponding to different laundry amounts and materials, and a detection phase where the stored patterns are compared against real-time current values. This segmentation allows the system to achieve high detection accuracy without requiring complex real-time analysis, as the complex learning process is performed offline during the learning phase.
Solution Approach 2:
The system performs preliminary learning and pattern storage before actual washing operations. During the learning phase, the system pre-processes data by collecting current values at various rotation speeds, identifying characteristic points, and storing reference patterns in memory. This preliminary action enables the detection phase to operate efficiently with simple pattern matching, resolving the contradiction between accuracy and complexity.
2Reliability
If washing machines use simple laundry amount detection methods, then the operation is straightforward, but washing performance is insufficient when laundry material varies
Solution Approach 1:
The washing machine performs self-learning by automatically collecting current patterns during operation and storing them in memory without requiring external intervention. The system autonomously identifies characteristic points, creates reference patterns for different laundry types, and uses these patterns for automatic detection. This self-service capability ensures reliable washing performance across various laundry materials while maintaining ease of operation, as the system adapts automatically without user complexity.
3Measurement precision
If washing machines detect both laundry amount and material using machine learning, then washing performance is optimized, but the detection time increases
Solution Approach 1:
The detection process is divided into a learning phase (performed offline) and a detection phase (performed in real-time). During the learning phase, the system collects current patterns and stores them in memory. During the detection phase, the system only needs to compare real-time current values against the pre-stored patterns, which is a much faster operation. This segmentation eliminates time loss by performing the computationally intensive learning process offline while maintaining fast real-time detection.
Solution Approach 2:
The system performs preliminary pattern recognition and storage before actual washing. By pre-processing the data during the learning phase and storing reference patterns in memory, the system avoids performing complex analysis during the actual washing operation. This preliminary action significantly reduces detection time while maintaining high accuracy, as the real-time detection only requires simple pattern matching against pre-computed references.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The present invention relates to a washing machine and a control method of washing machine. The control method of washing machine includes: a first detection step of acquiring amount of laundry accommodated in a washing tub (4); a first washing step of performing washing based on a first laundry amount, when the first laundry amount acquired in the first detection step is equal to or larger than a preset first threshold value; a second detection step of, after the first washing step, acquiring the laundry amount accommodated in the washing tub by an output of an output layer of an artificial neural network while using a current value input to a motor (9) for rotating the washing tub during an accelerated rotation of the washing tub as an input data of an input layer of the artificial neural network previously learned by machine learning; and a second washing step of performing washing based on a second laundry amount, when the second laundry amount acquired in the second detection step is smaller than the first threshold value.