Estudo, modelagem matemática e controle do processo de secagem de crambe (Crambe abyssinica) em secador de leito fixo de bancada

Abstract

In a continuing increase in demand for biodiesel scenario in Brazil has experienced great potential for production from oil. Between different cultures researched at present, the crambe (Crambe abyssinica) has special prominence, it is imperative to carry out studies aimed at the proper performance of this crop in the field, as well as maintaining seed quality and oil. It is known that seed quality is a result of several factors that need to be controlled, such as planting and harvesting characteristics, drying, processing and storage conditions. These drying factors are especially important since it is directly linked to product quality and energy efficiency of the process. The long industrial drying has undergone an automation process, is essential for better process control. For the design of a dryer on an industrial scale is essential to conduct studies of the kinetics and equilibrium moisture content. Accordingly assays were performed to obtain the isotherms Crambe abyssinica beans to temperatures of 30, 45 and 60 ° C and different relative humidity conditions. The equilibrium moisture content was conducted using the static method with the use of sulfuric acid solutions. Models Luikov, Keey, Henderson, Chung & Pfost, Henderson-Thompson, modified Halsey, modified Oswin and Smith modified were fitted to the experimental values and recommended those with the best coefficients of determination (R ²). The beans passed through a hydration process, carried out by the method of immersion in distilled water ultrathermostatic bath was given an empirical model to determine the grain hydration kinetics, enabling the evaluation of different grain moisture in the drying process. The drying kinetics was held in dryer fixed bench bed, experiments were conducted in order to examine the influence of three variables in the process: air temperature, air velocity and initial moisture content of the grain. The experiments were designed through a complete experimental design with three variables and two levels. The considered response variable was the elapsed time from the beginning of the process until no more significant weight change of the sample considered in this work as the drying time. Statistical analysis was performed to verify the effect of each variable in the process, using a p-value of 0.05. An empirical model was found and validated by the model coefficient of determination (R²). The experimental data of drying kinetics were fitted to literature models: Newton, Newton modified, modified Newton-power and the Forward. In the drying process was implemented a control loop PID (Proportional Integrative Derivative), tuning the parameters of the controller is given by Ziegler method and Nichols and Cohen and Coon, both open loop, by analyzing the curve reaction. The refinement of the adjustment was made manually. To evaluate the performance of controllers IAE criteria were used (error module Integral) and ISE (Integral errors squared). The hygroscopic behavior analysis showed that temperature is a parameter which has a significant influence in the studied groups. The equilibrium moisture content showed typical behavior, decreasing with increasing air temperature. The lowest equilibrium moisture content was achieved with a temperature of 60 ° C, the highest temperature among the three analyzed. The model modified Smith was the best described the sorption isotherms for the three temperatures investigated. The best conditions for crambe moisture removal were achieved at 60 ° C. A statistical analysis of experimental data was performed, demonstrating a negative influence of temperature on drying time, and positive influence of the air speed, the variable initial grain moisture had no significant effect. The empirical model was found to be a model of 1st order, which represents 72% of the data. The modified Newton model - Power 2 with two parameters set was the best described the drying kinetics for the three evaluated temperatures. Has implemented an automatic control system PID, to tune the controller parameters were used methods of Ziegler & Nichols and Cohen & Coon open loop through performance analysis parameters found by Cohen & Coon method achieved better results . an attempt to improve the performance was carried out through a manual fine-tuning, but there was no improvement. The controller implemented kept the air temperature close to the setpoint established in the presence of disturbances, and thus can ensure drying process in a more standardized product, better quality and efficiency.