Hybrid grey wolf optimizer-ANN for predicting wheel energy consumption in off-road vehicles and enhancing resource management

This study uses a hybrid artificial neural network (ANN) with the Gray Wolf optimization algorithm (GWO) to predict wheel energy consumption in off-road vehicles. The main objective is to improve resource management and reduce the energy consumed due to wheel-soil interaction. Experimental data were collected through a Bevameter device in a controlled environment. Key parameters such as penetration depth, penetration velocity, vertical load, plate size, and number of passes were considered as inputs to the neural network. The neural network was trained using two trial-and-error methods and the GWO algorithm, and its performance was evaluated using MSE and R² metrics. The results showed that the GWO method performed better than the trial-and-error method, with a lower MSE of 0.5123 and a higher coefficient of determination of 0.9812. Data analysis showed that increasing speed and vertical load led to increased energy consumption while increasing the number of passes due to soil compaction reduced the energy consumption. This study shows that a hybrid neural network with the GWO algorithm can effectively predict the energy consumption in the indentation of plates in the soil, which is a kind of representative of the wheel.