Soil moisture prediction is a key parameter for effective irrigation scheduling and water use efficiency. However, accurate long-term prediction remains challenging, as most existing models excel in short- to medium-term prediction but struggle to capture the complex temporal dependencies and non-linear interactions of soil moisture variables over extended horizons. This study proposes a hybrid soil moisture prediction method, integrating a long short-term memory (LSTM) network and extreme gradient boosting (XGBoost) model for multistep soil moisture prediction at 24 h, 72 h, and 168 h horizons. The LSTM captures temporal dependencies and extracts high-level features from the dataset, which are then used by XGBoost for final predictions. The study uses real-world data from the D.A.T.A (Demonstrating Applied Technology in Agriculture) research farm at ABAC (Abraham Baldwin Agricultural College) Tifton, GA, USA, utilizing watermark soil moisture sensors and weather station’s data installed on the farm. Results show that the proposed method outperforms other hybrid models, achieving R2 values of 98.67%, 98.54%, and 98.56% for 24, 72, and 168 h predictions, respectively. The study findings highlight that LSTM-XGBoost offers a precise long-term soil moisture prediction, making it a practical tool for real-time irrigation scheduling, enhancing water use efficiency in precision agriculture.
30 September 2024 in Journal of Smart Science and Technology
This study is conducted to enhance the performance of bitumen at high temperatures. To do so, styrene-butadiene-styrene (SBS) and Nano-silica polymer additives were used. Initially, pure bitumen versus bitumen 60/70 samples with 2.5%, 3.5%, 4.5%, and 5.5% SBS polymer, and modified bitumen samples with 4.5% SBS polymer and 1%, 2%, 3%, and 4% Nano-silica were prepared using a high shear mixer. Polymer separation tests, rolling thin film oven tests, dynamic shear rheometer tests, and rotational viscosity tests were performed on the modified bitumen samples. Preliminary results of the dynamic shear rheometer test showed that SBS polymer improved the high-temperature performance of bitumen and increased its resistance to permanent deformation. However, Nano-silica had a more significant effect than SBS in improving the high-temperature performance of bitumen. The combination of 4.5% SBS and 4% Nano-silica enhanced the high-temperature performance of bitumen. Additionally, the results of the bitumen separation test showed that combining bitumen with SBS polymer caused the polymer to separate from the bitumen. Nevertheless, adding 1% Nano-silica resolved this issue. Asphalt mixture samples for the above compounds were then prepared based on the Superpave mix design method, and a dynamic creep test was performed. The results of this test showed that the combination of SBS and Nano-silica polymers significantly increases the resistance of the mixture at high temperatures. The combination of 4.5% SBS polymer and 4% Nano-silica, compared to the control sample with pure bitumen at 50 °C and 60 °C, reduces the cumulative strain by 51% and 41%, respectively, demonstrating the best performance. Therefore, the combination of Nano-silica with SBS polymer can be widely used in places with hot weather.
Foamed asphalt mixtures, created using reclaimed asphalt pavement (RAP) and foamed bitumen, offer energy savings, reduced use of virgin materials, and lower transportation costs, combining the characteristics of rigid and flexible pavements. This study evaluated the mechanical performance of foamed asphalt mixtures with varying bitumen content (1–3 %) and cement content (0–2 %) to identify the optimal combination for pavement applications. Samples were tested for uniaxial compressive strength (UCS), indirect tensile strength (ITS), resilient modulus (RM), and tensile strength ratio (TSR) under laboratory conditions. To predict the results, a nonlinear regression model and an artificial neural network (ANN) were employed. The ANN model demonstrated greater accuracy with significantly lower prediction errors compared to the nonlinear regression model. The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method was then used to select the optimal combination of materials. TOPSIS prioritizes mixtures with the shortest geometric distance to the positive ideal solution (best values for all attributes) and the longest distance from the negative ideal solution. The results showed that UCS and RM increased as the bitumen content increased from 1 % to 2 %, but these properties decreased when the bitumen content exceeded 2 %. In contrast, ITS (dry and saturated) showed continuous improvement with an increase in bitumen content from 1 % to 3 %. TOPSIS analysis identified the mixture with 3 % bitumen and 2 % cement as the optimal combination, achieving the best overall performance in the UCS, ITS, RM, and TSR tests. This study highlights the utility of foamed asphalt mixtures for sustainable construction, demonstrating that ANN predictions and TOPSIS can effectively guide material selection to achieve superior mechanical performance while reducing environmental impact.
In Portland Cement Concrete (PCC) pavement of the roads, dowels bar transfers vehicle loading to the unloaded slab. Load Transfer Efficiency (LTE) is used to evaluate dowel bars in PCC pavement. This parameter is defined as the vertical displacement ratio of the loaded slab versus the unloaded slab. In this study, the impact of effective factors (friction coefficient between dowel and concrete slab, wheel loading, dowel diameter and dowel spacing) on load transfer efficiency was studied with modeling by using the ABAQUS finite element software. The verification process was presented to increase confidence in model results and the response data from the numerical simulation agrees well with analytical solution. The results show that with increasing the friction coefficient between slab and dowel, load transfer efficiency increases but the failure of concrete around dowel bars was found to initiate at the face of joint. Furthermore, if strains remain in elastic range, increasing in wheel loading magnitude does not lead to reduce load transfer efficiency but dowel diameter or its spacing have an important role on load transfer by dowels.
Soil moisture is a key parameter in agriculture. Its accurate prediction is essential for effective irrigation scheduling and water use efficiency. This study introduces a hybrid approach integrating Long Short-Term Memory (LSTM) network and Extreme Gradient Boosting (XGBoost) model for multistep soil moisture prediction for (24 hours, 72 hours, and 168 hours) ahead. The LSTM captures temporal dependencies and extracts high-level features from the dataset, while XGBoost uses these features to make final predictions. The proposed method was trained and evaluated on a real-world data from the D.A.T.A (Demonstrating Applied Technology in Agriculture) research farm at ABAC (Abraham Baldwin Agricultural College) Tifton, GA, USA, utilizing watermark soil moisture sensors and weather station’s data installed in the farm. Experimental results show that, the pro-posed method outperforms standalone models like LSTM, XGBoost, Gradient Boosting (GB), Extra Trees (ET) and others. It achieved R² values of 0.9867, 0.9854, and 0.9856 for 24, 72 and 168-hour predictions respectively. These results demonstrate the effectiveness of LSTM in extracting high-level temporal features, and XGBoost in making final predictions. The proposed hybrid model offers precise soil moisture prediction, making it a practical tool for real-time irrigation scheduling and enhancing water use efficiency in precision agriculture.