A metaheuristic-optimization-based neural network for icing prediction on transmission lines

• Published in: Cold regions science and technology 2024-08, Vol.224, p.104249, Article 104249
• Main Authors: Snaiki, Reda, Jamali, Abdeslam, Rahem, Ahmed, Shabani, Mehdi, Barjenbruch, Brian L.
• Format: Article
• Language: English
• Subjects: Feature selection; Ice accretion; Ice-to-liquid ratio; Machine learning; Metaheuristic optimizer
• ISSN: 0165-232X; 1872-7441
• DOI: 10.1016/j.coldregions.2024.104249

Ice accretion on overhead transmission line systems is a leading cause of power outages and can lead to substantial economic losses in northern regions. Therefore, accurately and rapidly predicting ice accretion on power lines is crucial for ensuring the safe operation of the power grid. This study introduces a machine learning method for predicting the ice-to-liquid ratio (ILR), an important parameter for assessing ice accretion efficiency. While estimating ILR is vital for operational forecasting, many existing ice accretion models do not include this capability. A feedforward neural network (FFNN) trained with stochastic gradient descent and various metaheuristic optimizers - specifically particle swarm optimization, grey wolf optimizer, whale optimizer, and slime mold optimizer - is employed to forecast hourly ILR. Environmental data required for training and testing the FFNN model were obtained from the Automated Surface Observing System (ASOS). A global sensitivity analysis using the Sobol index, evaluated via the coefficients of a polynomial chaos expansion, was conducted to identify the most influential input parameters. The results indicate that only four input parameters significantly contribute to the variance in the response: precipitation, temperature, dew point temperature, and wind speed. Furthermore, the FFNN model trained with metaheuristic optimizers outperformed the stochastic gradient descent approach. With the predicted ILR, ice accumulation can be easily calculated as the product of ILR and the amount of liquid precipitation depth. •A novel machine learning method is proposed for predicting the ice-to-liquid ratio on power lines•Feedforward Neural Networks (FFNNs) were trained using various optimization techniques to improve prediction accuracy•Environmental data from the ASOS was used to train and test the model•A global sensitivity analysis revealed that precipitation, temperature, dew point, and wind speed are the most influential factors on ILR•The model trained with metaheuristics algorithms achieved better performance compared to the traditional approach