اندازه گیری pH محلول بیوراکتور حاوی شناساگر فنولرد با استفاده از پردازش تصویر و شبکه های عصبی RBF و ANFIS

نوع مقاله: علمی-پژوهشی

نویسندگان

1 گروه مهندسی سیتم و مکاترونیک دانشکده علوم و فنون نوین دانشگاه تهران

2 استادیار گروه مهندسی سیستم و مکاترونیک دانشکده علوم و فنون نوین دانشگاه تهران

3 دانشگاه تهران دانشکده علوم و فنون نوین گروه فناوری بین رشته ای مهندسی مکاترونیک

چکیده

In this paper, the pH of the culture solution is measured and the embedded system is designed to automatically and without human intervention to measure the pH instantaneously and help control the pH at a higher rate. For this purpose, solvent-based color processing with the camera and the use of artificial neural network algorithms are used and algorithms are implemented on a processor. The grid was taught using solute-like laboratory data. The number of RBF network layers and inputs were assigned to the network in three RGB classes, and compared to the standard RBF error solution at 0.35 error and the test error reached 0.1 and in the ANFIS network the error rate was less than 0.06 and the test error was less From 0.01, indicating a higher accuracy than the RBF network. This error rate is obtained with 37 different solutions and can be reduced by increasing the number of errors.

کلیدواژه‌ها

موضوعات