نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشجوی دکترا، گروه مهندسی آب، دانشگاه تبریز و کارشناس کنترل و پایداری، شرکت سهامی آب منطقهای گیلان
2 کارشناس برق – الکترونیک، سرپرست سد و نیروگاه سد سفیدرود، شرکت سهامی آب منطقهای گیلان
3 دانشجوی دکترا، مهندسی عمران-مدیریت منابع آب ، دانشگاه آزاد علوم و تحقیقات تهران و کارشناس کنترل و پایداری، شرکت سهامی آب منطقهای گیلان
4 کارشناس ارشد کامپیوتر – نرم افزار دانشگاه گیلان و کارشناس فناوری اطلاعات، شرکت سهامی آب منطقهای گیلان
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
The critical values of air temperature (minimum and maximum temperatures) are one of the factors that endanger agricultural and garden crops. The ability to model and predict changes in air temperature using precise methods is essential for many practical problems as well as many branches of science. This study used the models of adaptive neural-fuzzy inference system, artificial neural network, and gene expression programming to forecast the minimum, maximum, and average air temperature values at Rasht city station. The adaptive neural-fuzzy inference system, artificial neural network, and gene expression programming are ranked in the first through third priorities, respectively, despite there being a slight difference in the error accuracy of the mentioned models for temperature prediction. A further indication of the expression programming model's superiority to the other two models is the mathematical relationships it provided between the input and output variables. Based on the SI index, the minimum and daily temperature in the data training period is in the range of 0.1 to 0.2, which is good in terms of prediction accuracy, and the maximum temperature is in the range of 0.2 to 0.3. considered average. The findings demonstrated that the input values of modeling with three characteristics (input temperature one, two, and three days prior) have the best performance in predicting the air temperature of Rasht.
کلیدواژهها [English]