Sugarcane squeezing process prediction method based on depth feature recognition

Abstract

The invention discloses a sugarcane squeezing process prediction method based on depth feature recognition. The method comprises the steps of 1 collecting a plurality of groups of original data; 2 removing abnormal data from the original data acquired in the step 1, and performing normalization processing to obtain normalized data; 3 performing multi-stage screening on the standardized data obtained in the step 2 to obtain feature vectors which are highly correlated with the energy consumption and the extraction rate and are low in redundancy; 4 searching the effects of different feature combinations and model parameters on the single data-driven model from the feature vector candidate set obtained by screening in the step 3 by adopting a mixed chicken flock algorithm to obtain a parameter variable, energy consumption and an extraction rate under the optimal performance of the single model; 5 establishing a first layer of deterministic prediction output; and 6 establishing a multi-model combination model to realize deterministic prediction and probabilistic prediction of the extraction rate and the energy consumption. According to the method, the model fitting effect and the prediction precision are greatly improved, and the problems that the indexes are difficult to measure on line and the like are solved.

Description

technical field [0001] The invention relates to the technical field of sugarcane crushing process design optimization, in particular to a sugarcane crushing process prediction method based on deep feature recognition. Background technique [0002] Extraction of sugarcane juice is the first link of sugar production. The extraction rate of pressing and production energy consumption are two important indicators of this section. Whether they meet the standards will affect the smooth operation and economic benefits of the entire sugar production. Due to technical limitations, these indicators are currently calculated through offline laboratory experiments. This method has a lag, which makes it impossible to quickly adjust the system indicators in time. Therefore, being able to monitor these indicators in real time has positive significance for guiding the optimal control of the process operation. [0003] With the development of artificial intelligence technology, the "black box...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a method for predicting the sugarcane crushing process based on deep feature recognition, thereby overcoming the shortcomings of the existing sugarcane crushing process analysis system that is hysteresis and difficult to integrate multiple models for prediction

Images