Research On Electronic Data System Of Grain Fine Drying Process Parameters And Its Intelligent Use Strategy

Grain drying is an important means to ensure grain security.Reasonable drying process can maintain grain quality,improve storage characteristics,and then improve economic value.At present,the automation level of domestic grain drying operation is relatively low,and its process parameter setting and adjustment are lack of professional knowledge and data guidance,which largely rely on manual experience.On the one hand,it is difficult to determine accurate and reliable drying process parameters according to actual conditions and drying indicators at the beginning of drying operation,and it is also difficult to determine accurate and reliable drying process parameters according to actual grain moisture Timely and reasonable adjustment of process parameters for changes in environmental conditions and equipment operation.On the other hand,when quality problems occur after the drying operation,it is difficult to accurately diagnose the causes,resulting in unstable control and quality deterioration in the actual drying operation.In this paper,the electronic data system of grain fine drying process parameters is established on The Basic Data Manual of Grain Fine Drying formed by our research group after years of experimental research on grain drying.At the same time,combined with the artificial intelligence method,the strategies and algorithms such as the optimization query of basic drying data,the setting and adjustment of drying operation parameters,and the diagnosis of drying problems are studied,and the application programs are compiled to realize the corresponding functions,to provide data and decision support for the actual grain drying fine operation.The main research work of this paper is as follows:(1)Construction of electronic data system for basic data of drying process and design of original data query strategy.Analyze and sort out the data structure of The Basic Data Manual of Grain Fine Drying,design the table structure of My SQL database,and input 675 data tables and more than 450000 drying data into the database.By establishing the index table of each drying index,the basic drying data is retrieved hierarchically to query the parameter table and parameter value of each drying index under different drying conditions,to improve the query efficiency and convenient for practical use.(2)Research on optimization method of drying process parameters based on drying characteristics and quality requirements.By selecting NSGA-Ⅱ multi-objective optimization algorithm,the basic parameters are optimized according to the drying indexes and quality indexes in The Basic Data Manual of Grain Fine Drying,so as to ensure the drying quality and maximize the drying efficiency.By querying and optimizing the basic data in the data book,the basic optimization parameters that meet the drying requirements are obtained for initial parameter setting and real-time parameter adjustment in the actual drying operation.(3)Research on the setting and adjustment of process parameters in practical drying operation.The LSTM neural network model is trained by using the actual drying data,and the basic data is transformed into the actual drying parameters.After training with 2209 groups of actual data,the overall fitting degree of the network model reaches 0.998.The environmental temperature and humidity,grain temperature,relative humidity and drying accumulated temperature are used as network inputs to predict the actual air temperature and the rotation speed frequency of grain discharge.Before drying,query the data manual and output the initial setting parameters through LSTM in combination with the drying index and climatic conditions.During drying,according to the change of climate conditions,the LSTM neural network outputs real-time drying operation parameters,and adjusts the current operation parameters to ensure that the expected drying indicators are achieved.(4)Research on the diagnosis method of practical drying operation problems.On the basis of analyzing the mechanism of grain drying,this paper takes the drying accumulated temperature as a comprehensive index to reflect the drying process and the essential cause of drying problems.Based on this,the concept of drying accumulated temperature section is proposed to describe the drying state.A pattern recognition model is built through LSTM neural network to identify the drying accumulated temperature section of drying operation parameters.The overall recognition accuracy reaches 92.0%,and to judge whether the drying accumulated temperature setting is correct.Compare the actual operation parameters with the drying parameters retrieved in The Basic Data Manual of Grain Fine Drying in the drying accumulated temperature section,and determine the cause of the drying problem by checking the abnormal parameters.(5)Development of electronic data system for grain fine drying process parameters.The Qt development framework is selected to compile the user interface and logic program.By calling the basic data in the My SQL database and using the DLL Dynamic link library method to call the genetic algorithm and neural network model,the above intelligent use strategy for parameter query,parameter setting and problem diagnosis is realized.