Optimization of a Low Loss Strategy for Combine Harvesters Based on Bayesian Network

In this study, the Bayesian network was used to optimize the harvesting strategy of the combine harvester for low loss rate. The range analysis and variance analysis of the orthogonal field experiments showed that the feed rate, concave clearance and threshing drum speed were main factors affecting the loss rate. Taken the main factors and loss rate as Bayesian network nodes, the optimal network structure was obtained through scoring search, in which the four scoring functions were respectively combined with hill-climbing method. An optimized harvesting strategy for low loss rate was inferred through the constructed network, and the strategy was implemented in field experiment. The results showed that the average loss rate was reduced by 0.8% compared to that normal harvesting strategy. Copyright (C) 2022 The Authors.