Related Experiment Video
Updated: Jul 29, 2026
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.1K
Optimization of Position and Number of Hotspot Detectors Using Artificial Neural Network and Genetic Algorithm to
Jeong Hoon Rhee1, Sang Il Kim2, Kang Min Lee3
1School of Civil and Environmental Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Korea.
Sensors (Basel, Switzerland)
|July 2, 2021
Summary
This study uses silo hotspot detectors and artificial neural networks (ANN) to accurately predict internal storage levels. Optimization with a genetic algorithm (GA) reduced sensor needs, improving economic feasibility.
Area of Science:
- Agricultural Engineering
- Data Science
Background:
- Efficient silo operation relies on accurate internal storage level management.
- Existing measurement methods face limitations.
- Silo hotspot detectors, typically for temperature monitoring, offer a novel approach.
Purpose of the Study:
- To develop a novel method for predicting silo internal storage levels using temperature data.
- To optimize sensor placement and quantity for improved accuracy and cost-effectiveness.
Main Methods:
- Utilized internal temperature data from silo hotspot detectors.
- Trained an artificial neural network (ANN) algorithm for level prediction.
- Combined ANN with a genetic algorithm (GA) for optimization of sensor configuration.
Main Results:
- Achieved high prediction accuracy (up to 97%) with an optimal ANN structure (9-19-19-1).
- The ANN-GA technique reduced the required number of sensors from seven to four or five.
- Demonstrated economic feasibility through sensor reduction.
Conclusions:
- Artificial neural networks combined with genetic algorithms offer a highly accurate and efficient solution for silo level management.
- This approach overcomes limitations of traditional methods and enhances economic viability.
