A model based on the non-linear artificial neural network （ANN） is established to predict the thickness of the water film on road surfaces. The weight and the threshold can be determined by training test data, and t...A model based on the non-linear artificial neural network （ANN） is established to predict the thickness of the water film on road surfaces. The weight and the threshold can be determined by training test data, and the water film thickness on the road surface can be accurately predicted by the empirical verification based on sample data. Results show that the proposed ANN model is feasible to predict the water film thickness of the road surface.展开更多
Unpailt roads are generally subject to erosion, when they need to be bitumen, civil engineers need to know the geotechnical capabilities of the soil layers to be used as support, among these capabilities, for example,...Unpailt roads are generally subject to erosion, when they need to be bitumen, civil engineers need to know the geotechnical capabilities of the soil layers to be used as support, among these capabilities, for example, soil characteristics to withstand erosion. CBR has often been used to classify these soils according to their compaction. In this article, we propose a correlation between CBR and eroded soil mass through a simulator. Indeed, in this article we show that using a simulator, soils can be classified according to their ability to withstand water erosion, whether internal or external. Indeed it is shown that the mass of eroded soil is related to the compaction capacity of the soil just as the CBR also has. We study the effects and influence of soil compaction on the ability of an unpaved road to resist erosion caused by falling raindrops. To do this, lateritic soil is submitted to different compaction pressures. The compacted soil is then submitted to CBR test and rain fall through a mini rain simulator. Correlations between eroded soil masse and compaction pressure as well as CBR are derived. The study shows that the compaction reduces the erodibility and increases the bearing capacity of soil. The formula obtained is significant because we have a new way of evaluating soils in the laboratory.展开更多
文摘A model based on the non-linear artificial neural network （ANN） is established to predict the thickness of the water film on road surfaces. The weight and the threshold can be determined by training test data, and the water film thickness on the road surface can be accurately predicted by the empirical verification based on sample data. Results show that the proposed ANN model is feasible to predict the water film thickness of the road surface.
文摘Unpailt roads are generally subject to erosion, when they need to be bitumen, civil engineers need to know the geotechnical capabilities of the soil layers to be used as support, among these capabilities, for example, soil characteristics to withstand erosion. CBR has often been used to classify these soils according to their compaction. In this article, we propose a correlation between CBR and eroded soil mass through a simulator. Indeed, in this article we show that using a simulator, soils can be classified according to their ability to withstand water erosion, whether internal or external. Indeed it is shown that the mass of eroded soil is related to the compaction capacity of the soil just as the CBR also has. We study the effects and influence of soil compaction on the ability of an unpaved road to resist erosion caused by falling raindrops. To do this, lateritic soil is submitted to different compaction pressures. The compacted soil is then submitted to CBR test and rain fall through a mini rain simulator. Correlations between eroded soil masse and compaction pressure as well as CBR are derived. The study shows that the compaction reduces the erodibility and increases the bearing capacity of soil. The formula obtained is significant because we have a new way of evaluating soils in the laboratory.
