Odor detecting method for refrigerating appliance and refrigerating appliance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing refrigerating appliances lack an effective method to accurately detect and remove odors, leading to unpleasant smells that can affect user experience.
Innovation Solution
An odor detecting method that acquires environmental data, preprocesses it by scaling, standardizing, and eliminating abnormal data, and inputs it into a classification model to predict odor properties, triggering a deodorization module for timely odor removal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If environmental data is collected and processed using a unified preprocessing method, then the processing system is simple, but the classification accuracy of odor properties deteriorates due to different data characteristics
Solution Approach 1:
The patent applies different preprocessing methods to different dimensions of environmental data based on their specific characteristics. Gas data undergoes logarithmic transformation to handle skewed distributions, while temperature and humidity data use standard normalization. This localized preprocessing approach improves odor classification accuracy by treating each data type appropriately rather than applying a uniform method.
2Measurement precision
If gas data is included in environmental data for comprehensive odor detection, then the odor detection capability is improved, but the data processing complexity increases
Solution Approach 1:
The patent transforms gas data using logarithmic transformation to change its parameter distribution from skewed to more uniform, making it compatible with the classification model. This parameter change reduces the overall data processing complexity by making all environmental data dimensions more homogeneous in distribution, while still maintaining comprehensive odor detection capability.
3Reliability
If abnormal data is eliminated through strict preprocessing, then the classification model reliability is improved, but the loss of potentially useful information increases
Solution Approach 1:
The patent performs preliminary standardization and normalization of environmental data before feeding it to the classification model. By pre-processing the data to remove obvious abnormalities and standardize distributions in advance, the classification model can focus on learning meaningful patterns rather than being distracted by data quality issues, thus improving reliability without excessive information loss.
Data Source
Figure 1
Figure 2~3
AI summary
An odor detecting method for a refrigerating appliance and a refrigerating appliance are provided. The method includes: acquiring environmental data in the refrigerating appliance; preprocessing the environmental data, where a preprocessing method used for data in at least one dimension of the environmental data is different from a preprocessing method used for data in other dimensions; and inputting the preprocessed environmental data into a preset classification model and acquiring a classification result, where the preset classification model is configured to predict a current odor property in the refrigerating appliance according to the preprocessed environmental data. Through the solution, the obtained environmental data can be efficiently processed, and abnormal data therein is eliminated, which can help improve the classification accuracy of the odor properties in the appliance, so as to timely find and remove the peculiar smell and keep the air in the refrigerating appliance fresh more desirably.