Server-Based Dishwasher Defect Diagnosis Using Neural Network Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing dishwashers face challenges in accurately determining the cause of defects, leading to prolonged detection and repair times, especially when defects are caused by user misuse, and current systems lack efficient methods for identifying and addressing these issues.

Innovation Solution

A server communicates with a dishwasher to train a neural network using operation information, enabling it to identify the cause of defects and provide guidance to users or service providers, reducing the need for on-site visits and preparing necessary repairs in advance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a repairer directly operates the dishwasher on the spot to determine the cause of defects, then the cause can be identified, but the customer has to wait a long time and the repair process is prolonged

Engineering Contradiction:
Improvedefect cause identification accuracyVSAvoidrepair time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary defect diagnosis by automatically collecting operation information from the dishwasher and comparing it with training data before the repairer arrives. This preliminary action identifies the defect cause in advance, so when the repairer arrives, they already have the diagnosis result and can proceed directly to repair, eliminating the need for on-site testing and significantly reducing repair time while maintaining accurate defect cause identification

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The manual mechanical process of on-site defect diagnosis by repairers is replaced with an automated information processing system. The server automatically collects operation data, compares it with training data using pattern recognition algorithms, and outputs defect cause information electronically. This substitution of mechanical diagnostic procedures with automated information processing maintains high diagnostic accuracy while dramatically reducing the time required

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Loss of information

If the customer explains symptoms to the service center, then some information is provided, but it is difficult to determine the specific cause of the defect

Engineering Contradiction:
Improvedefect information completenessVSAvoiddefect cause determination accuracy
Core Design Contradiction:
Loss of informationVSMeasurement precision

Solution Approach 1:

The system implements a feedback mechanism where the server continuously collects operation information from the dishwasher, compares it with training data, and generates defect cause information that is fed back to both the customer and repairer. This feedback loop ensures that complete and accurate defect information is provided automatically, eliminating the information loss that occurs when relying solely on customer descriptions and enabling precise defect cause determination

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The server acts as an intermediary between the dishwasher and the customer/repairer. It collects operation information directly from the dishwasher's control unit, processes this information through pattern recognition algorithms, and transmits the processed defect cause information to the user terminal and repairer. This intermediary role ensures that complete, accurate, and objectively measured defect information is provided, bridging the gap between the dishwasher's actual state and the customer's understanding

Inventive Principle:
Principle #24Intermediary (Mediator)

3Adaptability or versatility

If multiple types of defects occur in the dishwasher, then comprehensive coverage is achieved, but it becomes difficult to grasp the specific cause when misuse is involved

Engineering Contradiction:
Improvedefect type coverageVSAvoidmisuse defect detection difficulty
Core Design Contradiction:
Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The system uses a universal pattern recognition approach that can handle multiple types of defects through a single integrated framework. The server collects various operation information parameters and compares them comprehensively with training data that covers different defect scenarios including misuse cases. This universal method automatically adapts to different defect types without requiring separate detection mechanisms, achieving both comprehensive defect type coverage and accurate identification even for subtle misuse-related defects

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system detects defects by monitoring changes in operation parameters and comparing them against expected parameter ranges in the training data. When misuse occurs, it causes characteristic parameter changes that the system can identify through pattern recognition. By continuously monitoring parameter variations and comparing them with learned patterns of normal and abnormal operation, the system can detect even subtle misuse-related defects while maintaining the ability to handle multiple defect types

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3920125B1Server communicating with dishwasher
Publication Date: 2025.11.26 LG ELECTRONICS INC
  • EP3920125B1 patent drawingFigure 1
  • EP3920125B1 patent drawingFigure 2~3
  • EP3920125B1 patent drawingFigure 4

AI summary

A server communicating with a dishwasher is provided. A server communicating with a dishwasher includes a communication unit configured to communicate with one or more dishwashers, a memory configured to store a training result acquired by using a plurality of causes of defect occurrence and operation information corresponding to each of the plurality of causes of defect occurrence, and a processor configured to receive operation information of a specific dishwasher from the specific dishwasher among the one or more dishwashers and acquire the cause of defect occurrence of the specific dishwasher by using the operation information of the specific dishwasher and the training result. The training result includes a relationship parameter corresponding to a relationship between the plurality of causes of defect occurrence and the operation information.