Machine Learning Cable Fire Test Prediction From Small-Scale Results
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional large-scale product tests for cables, such as EN 50399, are costly and cumbersome, discouraging manufacturers from innovating new products due to the need for extensive cable production to meet fire safety standards.
Innovation Solution
Utilizing machine learning models trained on small-scale test results to predict the outcome of large-scale product tests, including classification and regression models to determine compliance with fire safety standards, reducing the need for extensive physical testing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If large-scale product tests (e.g., EN 50399) are performed to ensure fire safety compliance, then product reliability and safety are improved, but manufacturing cost and complexity increase significantly
Solution Approach 1:
The patent creates a virtual copy of the large-scale fire test through machine learning models. Instead of physically conducting extensive EN 50399 tests, the system trains ML models on historical test data to simulate and predict fire safety outcomes, thereby obtaining compliance information without the full physical testing complexity
Solution Approach 2:
The machine learning model acts as an intermediary between small-scale test results and large-scale test requirements. The model translates data from simpler small-scale tests into predictions about large-scale test outcomes, eliminating the need for direct complex large-scale testing while maintaining compliance assessment accuracy
2Reliability
If multiple cable specimens of sufficient length are manufactured to meet testing requirements, then test reliability is improved, but manufacturing cost and time increase
Solution Approach 1:
The system creates a digital representation of cable fire performance through machine learning models. Instead of physically producing multiple long cable specimens for testing, the model generates virtual test results based on input data, reducing material consumption while maintaining reliable compliance assessment
Solution Approach 2:
The patent transforms the testing approach by changing from physical cable length requirements to data input parameters. The machine learning model accepts various input parameters from small-scale tests and converts them into predictions of large-scale test outcomes, eliminating the need to manufacture specific quantities of cable for testing
3Reliability
If conventional large-scale fire tests are conducted to verify cable fire resistance, then product safety is ensured, but productivity and time efficiency deteriorate
Solution Approach 1:
The patent replaces the mechanical physical fire testing system with a computational machine learning system. Instead of conducting actual fire tests that require physical infrastructure, time, and resources, the system uses trained ML models to rapidly predict fire safety outcomes, dramatically increasing productivity while maintaining verification reliability
Solution Approach 2:
The machine learning models are pre-trained on historical fire test data before actual compliance assessments are needed. This preliminary training allows the system to rapidly predict outcomes for new products without requiring time-consuming physical tests, enabling quick compliance verification while maintaining accuracy
Data Source
AI summary
Systems and methods for using machine learning models to predict an outcome of a product test are described. According to certain aspects, an electronic device may calculate, based on a received set of small-scale results as a first input to a first machine learning model of a plurality of machine learning models, a first result predicting an outcome of the product tested according to the large-scale product test. The electronic device may then calculate, based on the set of small-scale results as a second input to at least one second machine learning model of the plurality of machine learning models, a second result predicting the outcome of the product tested according to the large-scale product test. The electronic device may then predict an outcome of the large-scale product test based at least on the first result and the second result.