One of the most important factors which is used to evaluate the performance of a combine harvester during harvest is the amount of grain loss. Inappropriate settings and calibration of a combine harvester operational factors could result in significant grain loss during machine operation. The performance factors that contribute to grain loss must be determined to effectively reduce the loss and increase the yield. The measurement and monitoring of grain loss from a combine harvester are very important; however, this method is usually time-consuming, tedious, and labour-intensive. Therefore, the operating settings of a combine harvester must be optimized to achieve the best configuration for minimizing grain loss. The experiment was conducted using a central composite design (CCD) with different speeds, header heights, cleaning fan speeds, and feed rates over 30 runs. An artificial neural network (ANN) model was developed to predict the grain loss and yielded an excellent R2 value of 0.9892, MAE of 1.7291, and RMSE of 3.6794. The ANN model was then used to optimize the operating parameters of the combine harvester using a genetic algorithm (GA) to determine the predicted minimum grain loss during harvest. The optimized parameters using the ANN-GA resulted in a minimum grain loss of 17.12 kg/ha. The results showed that optimization using ANN-GA could yield a better prediction accuracy for grain loss during harvest.
Figure 1: Experimental run for measuring grain loss
Figure 2: Structure of the neural network used in the study
Muhammad Isa Bomoi, Nazmi Mat Nawi, Samsuzana Abd Aziz, Muhamad Saufi Mohd Kassim (2023). Application of Artificial Neural Networks and Genetic Algorithm for the Prediction of Grain Loss from a Medium-sized Combine Harvester. In: Barbosa, J.C., Silva, L.L., Rico, J.C., Coelho, D., Sousa, A., Silva, J.R.M., Baptista, F., Cruz, V.F., (Eds.) Proceedings of the XL CIOSTA and CIGR Section V International Conference. Évora, Universidade de Évora, pp. 129-136.
Artikel penuh: http://surl.li/sfxtj
Tarikh Input: 29/03/2024 | Kemaskini: 04/04/2024 | ainzubaidah
PEJABAT PENTADBIRAN
UNIVERSITI PUTRA MALAYSIA
43400 UPM SERDANG