Optimizing the seed-cell filling performance of an inclined plate seed metering device using integrated ANN-PSO approach

TLDR: In this article, a 3-5-1 artificial neural network (ANN) model was developed for predicting the cell fill of an inclined plate seed metering device, and the particle swarm optimization (PSO) algorithm was applied to obtain the optimum values of the operating parameters corresponding to 100% cell fill.

Abstract: Uniform seed distribution within the row is the prime objective of precision planters for better crop growth and yield. Inclined plate planters are generally used for sowing bold seeds like maize, groundnut, chickpea, and their operating parameters like the forward speed of operation, the seed metering plate inclination, and the seed level in the hopper affect the cell fill and subsequently the uniform seed distribution. Therefore, to achieve precise seed distribution, these parameters need to be optimized. In the present study, out of the different optimization techniques, a new intelligent optimization technique based on the integrated ANN-PSO approach has been used to achieve the set goal. A 3–5-1 artificial neural network (ANN) model was developed for predicting the cell fill of inclined plate seed metering device, and the particle swarm optimization (PSO) algorithm was applied to obtain the optimum values of the operating parameters corresponding to 100% cell fill. The most appropriate optimal values of the forward speed of operation, the seed metering plate inclination, and the seed level in the hopper for achieving 100% cell fill were found to be 3 km/h, 50-degree, and 75% of total height, respectively. The proposed integrated ANN-PSO approach was capable of predicting the optimal values of operating parameters with a maximum deviation of 2% compared to the experimental results, thus confirmed the reliability of the proposed optimization technique.