作者: Amin Taheri-Garavand , Venkatesh Meda , Leila Naderloo
DOI: 10.1016/J.EAEF.2018.08.001
关键词:
摘要: Abstract In this study, the application of a versatile approach for modeling and prediction moisture content dried savory leaves using hybrid artificial neural network-genetic algorithm has been presented. Genetic Algorithm was used in order to find best Feed Forward Neural Network (FFNN) structure estimation drying process leaves. The experiments were performed at three air temperatures 40, 60 80 °C levels relative humidity 20%, 30% 40% velocity 1, 1.5 2.0 m/s forced conductive dryer. Optimized network by GA had two hidden layers with 9 17 neurons first second layers, respectively. Mean Square Error (MSE) value (0.000094606) correlation coefficient (0.9992) FFNN-GA showed that can be accurately predicted from input variables: temperature, airflow velocity, time. Moreover, results optimized topology could denote superior ability intelligent model on-line Savory different conditions.