| China is the country that has the largest output of peanut,Peanut planting area and output has been growing.60%of peanut production in our country is used in the oil in China.Extrusion technology has been widely applied to various fields,especially in the field of food.It is used for conveying,mixing,shearing,cooking,forming the raw material and has wide applicability with many raw materials,we minimize the destruction of nutrients through the pretreatment of high temperature and short time and improving oil output.It can also effectively grease anti nutrients inactivated,improve production capacity and reduce power consumption etc.This paper studies the pretreatment of oil producing based on single screw.The research involves the influence of material moisture,barrel temperature,and die diameter,spindle speed,the four main parameters on the extrusion production quality.The quality of the product including the expansion degree,oil rate and per kw yield,productivity and other factors.Combined with the characteristics of nonlinear mapping precision of the neural network,the neural network is trained so that the trained neural network has the strong ability to map the relationship between input parameters and output parameters,so as to explore the influence of process parameters on the production quality,and tap the optimum parameters to provide a reference for the production process.In addition,the paper also studies the change principle of the material state in the extrusion process,which provides a basis for the further analysis of the neural network.(1)Qualitative conclusions:the spindle speed influences the most and the die nozzle diameter the least on the puffing degree;while the die nozzle diameter and the barrel temperature influence the barrel extracting oil rate the most and the least,respectively;besides,the spindle speed and the moisture content influence the residual oil rate the most and the least,respectively.(2)Quantitative conclusions:According to the trained neural network,the best parameter is[10,95,12,60](the four parameters represent the moisture content,the temperature of the sleeve,the die nozzle diameter and the spindle speed,the same below)for the smallest residual oil rate(1.05%)and[10.100,14,65]for the largest barrel extracting oi(53.33%),and[10,95,12.60]for the minimum residual oil rate(1.05%),and[9,90,14,65]for the largest productivity(16.21kg/h),and[9,100,10,65]for the highest productivity per kW(5.43kg/kw·h).Further considering the multi-objective optimization of various parameters,with the electricity output and productivity as the goal,the best combination of parameters is[9,100,10,65],and the per kilowatt yield is 5.43kg/kw·h,the productivity is 14.87kg/h.And considering the residual oil rate the lowest,the highest per kilowatt yield and productivity as multi-objective optimization,the optimal parameter combination is[10.95,12.60],the degree of electricity production is 4.53kg/kw·h,the productivity was 13.54kg/h,residual oil rate was 1.05%.(3)Along the screw axis,material is gradually being squeezed,the oil gradually is being squeezed,so the material oil rate gradually decreased;at the same time,due to the reduced material total quality,moisture content have increased;the microstructure of materials is studied and it is found that as the material is gradually being squeezed out of oil,the cell wall is destroyed gradually,and the boundaries between the cells gradually disappear.The neural network model proposed in this paper can functionally map the relationship between process parameters and process quality.Therefore,we can excavate the inner link,explore the optimal combination of process parameters based on the neural network model,which provides the best combination of parameters for different quality indicators for us then,and the corresponding demand in combination during the practice of the process is used to determine the most suitable parameters,providing guidance and reference for the extrusion process of peanut oil. |
PDF Full Text Request