4198 — Experimental Results of a Predictive Neural Network HVAC Controller

Click here to purchase

Proportional, integral, and derivative (PID) control is widely used in many HVAC control processes and requires constant attention for optimal control. Artificial neural networks offer the potential for improved control of processes through predictive techniques. This paper introduces and shows experimental results of a predictive neural network (PNN) controller applied to an unstable hot water system in an air-handling unit. Actual laboratory testing of the PNN and PID controllers show favorable results for the PNN controller.

Units: Dual

Citation: ASHRAE Transactions, 1998, Vol. 104, Part 2, Toronto, ON

Product Details

Published:
1998
File Size:
1 file , 110 KB
Product Code(s):
D-7665