Neural Networks Prediction of Cutter Load Pressure of Sugarcane Harvester

In order to realize cutter load pressure prediction and automatic control signal acquisition, combined with orthogonal test and BP neural network and regression analysis, respectively established cutter load pressure prediction mathematical model. After analyzing the results, it can be seen that the accurate fitting rate of the cutting load pressure mathematical model constructed by the BP neural network has reached 85.2%, while the accurate fitting rate of the regression model is only 33.3%. And the verification test also shows that the constructed cutter load pressure BP neural network model predicts the cutting load pressure obtained under the new test factors, and the relative error of the obtained cutting pressure is basically within 5%. It shows that the prediction model based on the relationship between cutting load pressure and factors established by BP neural network can better fit the data and has higher accuracy, and this BP neural network model can continuously and automatically generate a new knowledge base and reduce actual experiments. The number of times laid the foundation for the design and development of the automatic control system for the cutting depth of the cutting blade of the sugarcane harvester.