Mould Growth Prediction Using Time-Series Humidity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting and determining the cause of mould growth in enclosed spaces are inaccurate and fail to account for the dynamic and fluctuating nature of indoor environments, leading to unreliable predictions and late warnings, which can result in costly repairs and health risks.
Innovation Solution
A computer-implemented method that analyzes time-series sensor data of air temperature and relative humidity, extracting specific data features such as cross-correlation values and duration of humidity above thresholds, to predict mould growth and identify its cause by performing computational analysis using machine learning models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If current temperature and humidity readings are used to predict mould growth, then the prediction process is simple, but the prediction accuracy is insufficient
Solution Approach 1:
The system performs preliminary analysis by continuously monitoring and storing historical temperature and humidity data before mould growth occurs. This allows the model to predict mould risk in advance by analyzing patterns in the accumulated data, rather than simply reacting to current conditions.
Solution Approach 2:
The prediction system transitions from static current readings to dynamic time-series analysis. The model evaluates how temperature and humidity change over time, capturing the dynamic nature of indoor environments and their relationship to mould growth conditions.
2Ease of manufacture
If threshold-based warning systems are used, then the system is easy to implement, but warnings occur too late when mould has already started growing
Solution Approach 1:
The system performs preliminary analysis by continuously monitoring and storing historical temperature and humidity data before mould growth occurs. This allows the model to predict mould risk in advance by analyzing patterns in the accumulated data, rather than simply reacting to current conditions.
Solution Approach 2:
The system implements a feedback mechanism where prediction results are continuously updated as new sensor data becomes available. The model learns from past predictions and actual outcomes, adjusting its parameters to improve accuracy over time and provide increasingly timely warnings.
3Use of energy by moving object
If simple threshold comparisons are used for prediction, then the computational requirements are low, but the system produces many false positive warnings
Solution Approach 1:
The prediction system transitions from static current readings to dynamic time-series analysis. The model evaluates how temperature and humidity change over time, capturing the dynamic nature of indoor environments and their relationship to mould growth conditions.
Solution Approach 2:
The system changes from using single threshold parameters to multiple dynamic parameters including time offsets, correlation coefficients, and trend analysis metrics. These parameter changes enable more nuanced discrimination between temporary humidity spikes and genuine mould risk conditions.
4Device complexity
If sensor data from specific locations is used, then the measurement setup is simplified, but the prediction results vary significantly based on sensor placement
Solution Approach 1:
The system designs a universal prediction model that can process data from various sensor placements. By using time-series analysis and pattern recognition, the model adapts to different locations and configurations, making the system universally applicable regardless of where sensors are positioned in the room.
Data Source
Figure 1A~2
Figure 3A~5B
AI summary
The present invention relates to the technical field of building maintenance. It relates in particular to a computer-implemented method for predicting mould growth in an enclosed space. It also relates to a computer-implemented method for determining at least one cause of mould growth in an enclosed space. It further relates to a data processing system, a computer program and a computer-readable medium.