Fertilizer Strength Prediction Model Based on Shape Characteristics

Abstract

The accurate prediction of fertilizer strength can reduce the crushing rate in the process of transportation and utilization, ensure the efficient utilization of fertilizer, so as to realize the sustainable and clean production of crops. To achieve this goal, for the first time, a fertilizer strength prediction model based on the shape characteristics is proposed to this paper by the support vector machine and the improved differential evolution algorithm. Firstly, the fertilizer shape characteristics and strength are measured by the independently developed agricultural material shape analyzer and the digital strength tester. Secondly, a fertilizer strength prediction model based on a support vector machine is constructed, in which the optimal combined kernel function is proposed according to the Pearson correlation coefficient and the normal distribution law. Finally, a differential evolution algorithm is improved to optimize the internal parameters of the fertilizer strength prediction model. The experiment results show that the maximum error rate of the fertilizer strength prediction model is between −5% and 5%. The proposed method of this paper can lay a solid foundation for fertilizer production and quality inspection, which will help reduce fertilizer crushing, improve fertilizer utilization to realize the sustainable and clean production of crops.

Keywords

• Fertilizer strength prediction model
• shape characteristics
• support vector machine
• combined kernel function
• improved differential evolution algorithm