Research On Modeling Method Of Grain Storage Quantity Detection Model Based On Janssen Principle

At present,the practical method for detecting the quantity of grain storage grain in China is mainly the weighing method.This method is not only time-consuming and labor-intensive,but also causes a lot of waste.Therefore,it is urgent to study the detection methods that can grasp the quantity of grain stored in each place in real time,efficiently and accurately.This paper mainly studies the quantitative detection model of polynomial grain storage based on Janssen principle,analyzes the data preprocessing method based on sensor median mean and standard deviation,and studies the modeling of Janssen neural network detection model based on improved cuckoo search algorithm.method.The main research work of this thesis is as follows:1.The theoretical model of the Janssen principle and its polynomial grain storage quantity detection is studied.The internal relationship and model characteristics of the grain are analyzed.According to the pressure distribution characteristics of the bottom surface of the grain pile,a median mean value and standard of the output value of the sensor based on the inner and outer ring are proposed.Poor data preprocessing method.2.By introducing the initialization method of the nest in the quasi-Monte Carlo optimization cuckoo algorithm,the variable discovery probability Pa,the vain-based algorithm based on the dimensional update strategy,the non-uniform mutation operator and the variable search step,The improved cuckoo search algorithm improves the search speed of the cuckoo algorithm and enhances the local search ability of the cuckoo algorithm.The improved cuckoo optimization algorithm is used to optimize the detection model parameters based on the Janssen principle,and finally based on the improved crest.The Janssen principle detection model optimized by the bird optimization algorithm.3.The BP neural network Janssen principle detection model based on quadratic polynomial expansion is proposed.The improved CS algorithm is used to realize BP neural network modeling optimization.The superior BP neural network Janssen principle detection model is established.The actual modeling results are verified.The validity and reliability of the detection model.