Comprehensive evaluation of the cutting performance of sugarcane harvester based on fuzzy theory and neural network

The cutting performance of the sugarcane harvester is affected by many factors. However, the influences of these factors are difficult to calculate accurately, and the qualitative performance evaluation has some fuzziness. Therefore, the fuzzy comprehensive evaluation is integrated with the neural network to assess the cutting performance. The ratio of ragged root of sugarcane affected by many indexes is evaluated with the fuzzy comprehensive evaluation, and the evaluated results are taken as the training samples of the BP neural network. After it is trained, the BP neural network is used to estimate and forecast the cutting performance to reduce the minute and complicated computing processes of the fuzzy comprehensive evaluation and to lessen the effects of the error of the training samples on the forecast performance. Thus, it is beneficial to improve the efficiency of solving problem and the self-learning ability of the evaluation model. The results indicate that the method integrated the fuzzy comprehensive evaluation with the BP neural network can obtain better application effects