Publication Date


Document Type

Honors Project

Degree Name

Bachelor of Arts




William Williams


PID controller, Feedback, Electronics, Neural networks, Deep learning, Machine learning, Neural networks (Neurobiology)


This thesis project is the first in a total of two thesis projects. The main focus of this project was the design and creation of a noise eater circuit which utilizes a PI (Proportional, Integral) controller for its control element. In this paper, the design and testing of the circuit are discussed. We were able to successfully build our own functioning noise eater, which relied on a system of several op amps for the controller portion. The second part of this project, which will trail into the next theses, is training a deep neural network controller as a replacement for the PI controller. This portion of the project is inspired by a proposed deep learning circuit by Cheon et al. [8]. By the end of this project, we had completed several testing procedures of the deep learning controller, but have faced problems with ’memory’ that have yet to be resolved. The next thesis project will focus on improving the training procedure of the deep learning controller, with the goal of training it to the point where it can mimic the same process of a proportional, integral, derivative (PID) controller.


2018 Isabelle Elise Bunge. Access limited to the Smith College community and other researchers while on campus. Smith College community members also may access from off-campus using a Smith College log-in. Other off-campus researchers may request a copy through Interlibrary Loan for personal use.




xvii, 60 pages : illustrations (some color) Includes bibliographical references (pages 57-60)