Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
O-1 APPENDIX O. VALIDATION OF ARTIFICIAL NEURAL NETWORK APPROACH FOR PREDICTING GEOSYNTHETIC-REINFORCED PAVEMENT PERFORMANCE The performance of geosynthetic-reinforced flexible pavements includes fatigue cracking, permanent deformation, and international roughness index (IRI). In this study, the artificial neural network (ANN) model was used to predict the critical responses of geosynthetic- reinforced pavements. A geosynthetic-reinforced pavement with any given material properties was then equivalent to an unreinforced pavement with the modified material properties to obtain the identical pavement responses. The process of validating this approach is illustrated in Figure O-1 and involves the following steps: 1. Identify the in-service geosynthetic-reinforced pavement sections from the long-term pavement performance (LTPP) database and Texas Pavement Management Information System (PMIS). This study focused on the in-service pavement sections with the placement of geosynthetics in conjunction with the unbound base courses. 2. Collect the pavement structure data, including layer thickness, construction dates, material design information, and falling weight deflectometer data. 3. Collect the traffic data from the identified pavement sections, which should be compatible with the input of traffic module in the Pavement ME Design software. 4. Collect the climatic data or weather station information from the identified pavement sections. 5. Collect the performance data from the identified pavement sections, including fatigue cracking, rutting, and IRI. 6. Employ the proposed ANN approach to determine the modified material properties of an unreinforced pavement. 7. Input the unreinforced pavement structure data, the collected traffic data and climatic data, and the determined modified material properties into the Pavement ME Design software to predict the pavement performance (i.e., fatigue cracking, rutting, and IRI). 8. Compare the predicted pavement performance with that measured from the field.
O-2 Figure O-1. Flow Chart of the Process of Validating the Proposed ANN Approach After a thorough review of the in-service pavement sections in the LTPP database and PMIS, a total of 74 pavement sections containing geosynthetics were found in the LTPP database, and a total of 51 pavement sections containing geosynthetics were found in the PMIS. A full list of the identified pavement sections is presented in Appendix P. Of these identified pavement sections, most had stabilized base courses, which were not under consideration in this study. In the remaining pavement sections, a total of three qualified pavement sections were from the LTPP database, and a total of three qualified pavement sections were from the PMIS. Some Identify In-service Geosynthetic-reinforced Pavement Sections Collect Field Data Structure Data Material Information Traffic Data Performance Data Climate Data Determine Modified Material Properties Predict Pavement Performance Using Pavement ME Design Does Predicted Pavement Performance Match Field Measurement? Finish Yes Modify Proposed ANN Approach No
O-3 of these identified pavement sections had the stabilized soil over the untreated subgrade, as illustrated in Figure O-2. Figure O-2. Conversion of Resilient Moduli of Two-Layer Structure to Single Resilient Modulus of One-Layer Structure Based on the Odemarkâs method, the resilient moduli of the two-layer structure are equivalent to the single resilient modulus of the one-layer structure using Equation O-1. ( ) ( ) 31/3 1/31 1 2 2 0 1 2 H E H E E H H ï© ï¹+ = ïª ïº +ïª ïºï« ï» (O-1) where 0E is the resilient modulus of the one-layer structure, 0H is the thickness of the one-layer structure, 1E is the resilient modulus of the stabilized soil, 1H is the thickness of the stabilized soil, 2E is the resilient modulus of the subgrade, and 2H is the thickness of the subgrade. The comparisons of the geosynthetic-reinforced pavement performances between the ANN approach predictions and the field measurements for these identified pavement sections are presented below. LTPP Section 16-9032 The pavement section 16-9032 consists of a 6-inch hot-mixed and dense-graded asphalt concrete, a 23.2-inch crushed gravel unbound base, and a semi-infinite subgrade, which is classified as AASHTO 7-5 soil. A 0.1-inch woven geotextile is placed at the interface between the unbound base and subgrade. The comparisons of geosynthetic-reinforced pavement performance between the predictions by the proposed ANN approach and the field measurements are presented in Figures O-3âO-5. The predicted rutting depth and IRI results were in good agreement with the field measurements. The fatigue cracking of the geosynthetic-reinforced pavement was slightly overestimated by the proposed ANN approach. These findings indicate that the proposed ANN approach is capable of accurately predicting the performance of geosynthetic-reinforced pavements. Figures O-3âO-5 also present the predicted performance of the control pavement. It is demonstrated that the geotextile placed at the base/subgrade interface H1 H2 Resilient Modulus: E1 Resilient Modulus: E2 Stabilized Soil Subgrade H0 Equivalent Resilient Modulus: E0 Two-Layer Structure One-Layer Structure
O-4 has beneficial effects for reducing the rutting and international roughness index of flexible pavements. Figure O-3. Comparison of Rutting Depth between ANN Approach Prediction and Field Measurement for Pavement Section 16-9032 Figure O-4. Comparison of Fatigue Cracking between ANN Approach Prediction and Field Measurement for Pavement Section 16-9032 0 0.1 0.2 0.3 0.4 0.5 0.6 0 1 2 3 4 5 6 7 8 Ru tti ng D ep th (in ch ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced Predicted Control 0 4 8 12 16 20 0 1 2 3 4 5 6 7 8 Fa tig ue C ra ck ing (% L an e A rea ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced Predicted Control
O-5 Figure O-5. Comparison of IRI between ANN Approach Prediction and Field Measurement for Pavement Section 16-9032 LTPP Section 48-0167 The pavement section 48-0167 consists of a 5.6-inch hot-mixed and dense-graded asphalt concrete, a 13-inch crushed gravel unbound base, a 12-inch lime-treated soil, and a semi-infinite subgrade sandy soil. A 0.3-inch geogrid is placed at the interface between the unbound base and lime-treated soil. The comparisons of geosynthetic-reinforced pavement performance between the predictions by the proposed ANN approach and the field measurements are presented in Figures O-6âO-7. 70 80 90 100 110 120 130 140 0 1 2 3 4 5 6 7 8In ter na tio na l R ou gh ne ss In de x (in /m i) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced Predicted Control
O-6 Figure O-6. Comparison of Rutting Depth between ANN Approach Prediction and Field Measurement for Pavement Section 48-0167 Figure O-7. Comparison of Fatigue Cracking between ANN Approach Prediction and Field Measurement for Pavement Section 48-0167 LTPP Section 20-0160 The pavement section 20-0160 consists of a 5.6-inch hot-mixed and dense-graded asphalt concrete, a 7-inch crushed stone unbound base, a 6-inch treated subbase, and a semi-infinite silty clay subgrade. A 0.3-inch geogrid is placed at the interface between the unbound base and pozzolanic treated subbase. The comparisons of geosynthetic-reinforced pavement performance 0 0.1 0.2 0.3 0.4 0.5 0 1 2 3 4 5 6 Pe rm an en t D efo rm ati on (in ch ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced 0 2 4 6 8 10 0 1 2 3 4 5 6 Fa tig ue C ra ck ing (% L an e A rea ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced
O-7 between the predictions by the proposed ANN approach and the field measurements are presented in Figures O-8âO-9. Figure O-8. Comparison of Rutting Depth between ANN Approach Prediction and Field Measurement for Pavement Section 20-0160 Figure Q-9. Comparison of Fatigue Cracking between ANN Approach Prediction and Field Measurement for Pavement Section 20-0160 PMIS Section FM 02 Texas Farm-to-Market Road No. 02 (FM 02) consists of a 1-inch asphalt seal coat, a 7-inch new base course, a 10-inch old base course, and a clay subgrade. The geosynthetic is placed at the interface between the new base course and the old base course. Two types of geogrids with different sheet stiffness are used in the sections of FM 02-2 and FM 02-3, respectively. One type of geotextile is used in the section of FM 02-4. The pavement sections 0 0.2 0.4 0.6 0.8 0 1 2 3 4 5 6P erm an en t D efo rm ati on (in ch ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced 0 2 4 6 8 10 0 1 2 3 4 5 6 Fa tig ue C ra ck ing (% L an e A rea ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced
O-8 were constructed in January 2005. The average daily traffic is 800. The speed limit on FM 02 is 55 mph. The comparisons of geosynthetic-reinforced pavement performance between the predictions by the proposed ANN approach and the field measurements are presented in Figures O-10âO-15. Figure O-10. Comparison of Rutting Depth between ANN Approach Prediction and Field Measurement for Pavement Section FM 02-2 Figure O-11. Comparison of Fatigue Cracking between ANN Approach Prediction and Field Measurement for Pavement Section FM 02-2 0 0.1 0.2 0.3 0.4 0 1 2 3 4 5 6 Pe rm an en t D efo rm ati on (in ch ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced 0 2 4 6 8 10 0 1 2 3 4 5 6Fa tig ue C ra ck ing (% L an e A rea ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced
O-9 Figure O-12. Comparison of Rutting Depth between ANN Approach Prediction and Field Measurement for Pavement Section FM 02-3 Figure O-13. Comparison of Fatigue Cracking between ANN Approach Prediction and Field Measurement for Pavement Section FM 02-3 0 0.1 0.2 0.3 0.4 0 1 2 3 4 5 6 Pe rm an en t D efo rm ati on (in ch ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced 0 2 4 6 8 10 0 1 2 3 4 5 6F ati gu e C ra ck ing (% L an e A rea ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced
O-10 Figure O-14. Comparison of Rutting Depth between ANN Approach Prediction and Field Measurement for Pavement Section FM 02-4 Figure O-15. Comparison of Fatigue Cracking between ANN Approach Prediction and Field Measurement for Pavement Section FM 02-4 0 0.1 0.2 0.3 0.4 0 1 2 3 4 5 6 Pe rm an en t D efo rm ati on (in ch ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced 0 2 4 6 8 10 0 1 2 3 4 5 6Fa tig ue C ra ck ing (% L an e A rea ) Pavement Age (Year) Measured Predicted Geosynthetic-reinforced
