Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.703262
Title: Prediction of permanent deformation in asphalt mixtures
Author: Al-Mosawe, Hasan
ISNI:       0000 0004 6060 9520
Awarding Body: University of Nottingham
Current Institution: University of Nottingham
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
An asphalt mixture is combined of different sizes of aggregate, filler, and bitumen for application on the most common road construction materials. In asphalt pavement material there are different types of distress such as permanent deformation (rutting), fatigue cracking, ravelling, potholes, stripping, etc. There are many reasons for these types of distress, some of them related to the pavement structure, e.g. whether the underlying layers are weak, others related to the mixture properties. Other causes could be related to external conditions such as high temperature, high axle load, long duration of load application, etc. This research has focused on the permanent deformation (rutting) as a function of aggregate gradation. The aggregate gradations of more than twenty asphalt mixtures, manufactured with different gradations, were analysed by using the Bailey method of gradation analysis. The analysis was performed in relation to Repeated Load Axial Test (RLAT) testing results to study the performance of each mixture. The results showed that the Bailey method is not capable on its own to define the differences between the gradations of each mixture. Therefore, three more packing ratios were introduced to adequately describe the aggregate gradation. The aggregate particle packing was extensively studied through these packing ratios and it was shown how the different particle sizes interact with each other. Images were taken for two mixtures to validate the theory behind the ratios. The five packing ratios (two of Bailey and three new ratios) were used in Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) techniques for all the mixtures as input data to predict the mixture performance (RLAT permanent deformation and Indirect Tensile Stiffness Modulus ITSM stiffness modulus) and they showed good prediction capability. After establishing the impact of aggregate packing on the performance, six mixtures were re-manufactured and re-tested with different variables; the selection of the mixtures was made to cover a range of different gradations (ratios). The aim of this step was to understand the effect of these variables on the asphalt mixture in the light of the packing ratios. The variables that were used were binder content, testing temperature and compaction effort. The binder content results showed an interesting effect on the permanent deformation and stiffness of the asphalt mixture. The packing of aggregate was very helpful in understanding the different mixture behaviour with different binder content. The effect of aggregate packing was not shown at relatively low testing temperature, but as the temperature rises the aggregate packing effect starts to appear. The effect of compaction which was represented by the number of gyrations in gyratory compactor was inconsistent; results show over-compaction can lead to poor performance. Finally, a linear viscous method was introduced aiming to predict the rutting in an asphalt mixture. The method was based on using a multilayer linear programme (BISAR) and using viscous parameters of the mixture as input. The non-linear properties of the material were incorporated by using the RLAT test. For this purpose, six mixtures were used and tested in a wheel tracking machine. The predicted results were compared with the wheel tracking rut depth in the laboratory and showed good agreement at different temperatures. However, at high temperature (50 °C) the material properties in the RLAT test did not behave as linear viscous, which resulted in a much poorer prediction. Trials were made to predict field rut but it was found that special requirements were needed for the approach which were not available at the time of the research. However, for the available field data, the method was found to be a good predictor.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.703262  DOI: Not available
Keywords: TE Highway engineering. Roads and pavements
Share: