Forward and reverse mapping of extreme learning machine model of a precision planter

The extreme learning machine (ELM) regression model was developed to predict the performance parameters of a pneumatic planter. The diameter of the metering disc hole, suction and peripheral speed of the disc were the independent parameters, and miss index, multiple index and precision is seed spacing were the dependent parameters. Single hidden layer of 15 nodes with sigmoid activation function in the hidden layer and linear function in the output layer was taken. The ELM predicted the performance of the pneumatic planter with mean square error of 0.708. The ELM developed was reverse mapped to identify the optimum levels of independent parameters to achieve the lowest miss and multiple indices, and precision in seed spacing. The reverse mapping of the ELM resulted in 3 solutions, and they can be adopted in the actual precision planter. The forward and reverse mapping of the ELM regression model was found to be effective in prediction of performance and optimization of the design and operational parameters of the planting equipment.