摘要: |
Reliable and economical design of Portland cement concrete (PCC) pavement structural systems relies on various factors among which proper characterization of the expected permeability response. Permeability is an extremely important factor that strongly relates the durability of PCC pavement systems to prevailing environmental conditions. For this reason, the Kansas Department of transportation (KDOT) uses the Rapid Chloride Permeability testing protocol to determine the resistance of concrete to water, chlorides and other chemical intrusion. Typically, the Rapid Chloride Permeability test measures the number of coulombs that pass through a concrete sample over a period of six hours at a concrete age of 56 days. Results of this test make it possible to obtain the desired time-dependant concrete permeability response. During the past years, KDOT has performed numerous numbers of full six-hour permeability response tests. Unfortunately, KDOT has not been able to capitalize on the richness of the available full-response permeability database in order to shorten the testing duration. Accordingly, it is proposed herein that we utilize the statistical/artificial neural network (SANN) approach for the development of appropriate SANN-based models that can effectively project the full six-hour response from a shorter-duration permeability response. The proposed research is in direct response to KDOT's need to reduce the testing time of the currently used KDOT rapid chloride permeability test. These developed models will allow KDOT to reduce the duration of the Rapid Chloride Permeability test from 6 hours to 1 or 2 hours. Additionally, these models will allow KDOT to: i) increase the number of tests that can be performed in one day, and ii) minimize/eliminate damage to the testing cells, particularly, when excessively permeable concrete is tested. The proposed research will provide an excellent research opportunity to produce innovative prediction models that will impact Chloride Permeability testing procedure for KDOT and other DOT agencies. |