Complexity Analysis And Process Control For Grain Drying

Drying is playing an important role in the post-harvesting process of grain,and is involved with the theories and techniques such as heat and mass transfer,mechanics and information science. Researching the advanced equipments andtechniques of grain drying plays a most important role in developing graineconomy and promoting the development of drying theory and technique. Thecontents of this paper are parts of the agicultural innovation found item "intelligentsensor of grain moisture content and its cotrol system" and the item of"Development on varied temperature methods and automatic control system ofmaize drying" which cooperated with Jilin Foodstuff Group. The content of actualresearch are mainly about complexity control strategy for grain drying process.In this paper, complexity description and analysis on grain drying process arecarried out for the first time in China based on the description and analysis on graindrying process and quality debasement. A control strategy of complex systembased on microenvironment of grain was put forward. The control system wasdivided into several subsystems according to the hiberarchy of complex system.The grain drying process was analyzed with principal component analysis (PCA)and a "Four Parameter Intelligent Control Method (FPICM)" based onmicroenvironment control strategy for drying process and neural network techniquewas designed. The details are as follows:(1) Grain drying process and its speciality of quality variety were analyzedaccording to the concept of objective complexity and the general character ofcomplex system. It is assumed that the grain drying process was a complex systemwith multi MIMO, different time dimensions, and high coupling, informationredundance.(2) Drying crack is the most obvious exterior character of grain quality varietyduring drying process. The mechanism of crack generation was inestigated, thedistribution of thermo-and hydro-stress inside maize kernel were studied.Employing the general Maxwell model a stress modelwas proposed, which wasthen used to simulate the distributions of principal stress and shear stress inside themaize kernel. Simulation results show that both temperature gradient and moisturegradient significantly influence the magnitude and distribution of the stresses.Either principal stress or shear stress exceeds the yield stress wil result in cracking.Results also show that the maximal principal stress occurs inside the cutinendosperm, while the maximal shear stress occurs on the surface of the cutinendosperm.(3) Germination is the synthetical criterion of grain quality. The thin-layerdrying experiment showed that the key factors influencing the grain drying qualitywere drying temperature, drying time and input moisture content. The relationshipsbetween grain quality and drying temperature or drying time obeys the first orderdynamic equation. The variety of grain quality experienced two phase, one forholding that the germination has little attenuation, and the other for acuteattenuation that the germination declined rapidly with increasing the dryingtemperature and drying time. There is a critical quality point between the phases.The critical quality point is not only affected by drying temperature but also bygrain intial moisture content.(4) According to the related influence factors of grain drying process, thesimilarity rule for grain drying process was deduced using dimension analysis. Thefunction expression of the similarity rule and the similarity empirical formula forgrain drying process were established according to the experiment. The accuracy ofthis empirical formula was validated using experiment. The results of experimentalvalidation show that the empirical model can predict the moisture content of grainaccurately within initialized parameters scope. An empirical model of auger speedwas deduced for controlling drying process. The results showed that the controllerworks well while the inlet moisture content keeps stability, but woks bad while theinlet moisture content are fluctant.(5) A control strategy for complex system based on microenvironment was putforward in the first time to deal with the control problem for complex system. Thecomplexity of grain drying process was analyzed according to the hiberarchy ofcomplex system. The system was divided into primary sub-state, directenvironment sub-state and indirect environment sub-state, and an environmentcontrol model consists of this parts was presented. The system was decoupled asthree components included quality controller, environment controller and localcontroller through the decoupling of time scales.(6) The parameters of drying process were analyzed by PCA and the controlparameters were optimized. It is concluded that the first critical point T4 and thesecond critical point T2 are the control object in the microenvironment controlprocess for grain drying. A model about the grain temperature and grain qualitywas established according to the results of PCA using nonlinear partial least square(NPLS). Grain quality during drying process were predicted by the model, theresult shows that the predicted values were more precise.(7) Based on the above-mentioned work and the complexity analysis fordrying process, according to the demand of maize drying process as well as dryingprocess quality control, FPICM for quality control of maize drying process was putforward based on microenvironment control strategy and neural network technique.The control system hardware for maize drying process was designed and thesoftware was developed using virtual instrument and neural network technique.Practice shows that this system has advantages of simple structure, high controllingaccuracy, simple operation, and strong anti-interference power etc.This paper describes a new method and principle of quality control in thegrain drying process using complexity science and technique. A control strategy ofcomplex system based on microenvironment of grain and a "Four ParameterIntelligent Control Method" for quality control of maize drying process was putforward. The quality intelligent control system for maize drying process wasdeveloped based on virtual instrument and neural network technology. The resultsof this study have laid the foundation of automatic control in the process of graindrying, and made the contribution to realize automation in the process of graindrying. The results of experiment proved the intelligent control system was able tocontrol the process of grain drying automatically and successfully instead ofmanual. It will take huge role in increase the production quality and productivity ofdrying workshop. The achievements in our study will offer strong technologysupport for raising the quality of the reserve grain, the technical level ofstate-owned grain depot in our country, and the competitive ability of our grain oninternational market. Also it will provide a kind of method that has practical valueto speed up the automation and modernization of the grain reservation andprocessing profession in our country, and it is valuable to apply extensivelyinteriorly.