On-Line Implementation of a Fuzzy Neural Ramp Metering Algorithm

摘要

A Fuzzy Logic Ramp Metering Algorithm will address the needs of Seattle's freeway211u001esystem and overcome limitations of the existing ramp metering algorithm. This 211u001eproject progressed toward implementing and testing a fuzzy neural ramp metering 211u001ealgorithm on-line at the Traffic Systems Management Center (TSMC) for the 211u001eWashington State Department of Transportation's Northwest Region. Improvements 211u001ewere made to neural network predictors to allow better generalization. Code was 211u001ewritten for the fuzzy ramp metering algorithm and its interface with the pre-211u001eexisting TSMC code. Of the new code written, approximately 95 percent of it was 211u001efor the interface, and only 5 percent of it was for the ramp metering algorithm 211u001eitself. Interfacing the fuzzy controller with the existing TSMC software required 211u001emodification of 15 pre-existing files related to the ramp metering database, real-211u001etime skeleton, and ramp metering and data collector communications.