Odor detecting method for refrigerating appliance and refrigerating appliance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing refrigerating appliances lack an effective method to accurately detect and manage odors, which negatively impact user experience due to the deterioration of perishable foods stored within.

Innovation Solution

An odor detecting method utilizing machine learning models trained with data from multiple rounds of odor detecting experiments on various ingredients, combining sample data to improve prediction accuracy, and integrating a deodorization module to address unpleasant odors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional odor detection methods are used in refrigerating appliances, then the device complexity is low, but the measurement precision of odor properties is insufficient

Engineering Contradiction:
Improveodor detection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical/chemical odor detection methods with machine learning models that process environmental data (temperature, humidity, gas composition) to predict odor properties. This substitution enables high-precision odor detection without requiring complex specialized sensing hardware, thus improving measurement precision while controlling device complexity.

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

Solution Approach 2:

The patent introduces environmental data (temperature, humidity, gas composition) as intermediary parameters that correlate with odor properties. Instead of directly detecting odor, the system measures these intermediary environmental factors and uses machine learning models to infer odor characteristics, achieving accurate odor detection through indirect measurement.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If multiple machine learning models are trained with extensive sample data to improve odor detection accuracy, then the measurement precision improves, but the loss of time for data collection and model training increases

Engineering Contradiction:
Improveodor prediction accuracyVSAvoiddata collection time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs comprehensive data collection and machine learning model training in advance, before actual odor detection is needed. Extensive sample data from Q rounds of experiments on M groups of ingredients is used to pre-train N machine learning models. This preliminary action stores learned patterns in the models, enabling fast real-time odor prediction without requiring additional data collection time during actual use.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent collects and processes more sample data than minimally required (M groups of ingredients across Q rounds of experiments) to ensure the machine learning models are thoroughly trained. This excessive data collection during the preparation phase improves model robustness and generalization, allowing accurate odor prediction with minimal processing time during actual deployment.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4266048A1Odor detecting method for refrigerating appliance and refrigerating appliance
Publication Date: 2023.10.25 BSH HAUSGERATE GMBH
  • EP4266048A1 patent drawingFigure 1
  • EP4266048A1 patent drawingFigure 2
  • EP4266048A1 patent drawing

AI summary

An odor detecting method for a refrigerating appliance and a refrigerating appliance are provided. The method includes: acquiring environmental data of gas in a refrigerating appliance (S11); inputting the environmental data to N preset machine learning models (S12) to obtain N odor detecting results; and determining odor properties of the gas in the refrigerating appliance according to the odor properties in the N odor detecting results (S13). The above solution can improve accuracy of determining the odor properties of the gas in the refrigerating appliance.