Physics-informed neural networks (PINNs) are increasingly employed for surrogate modelling of soil behaviour. Existing surrogate models for unsaturated soil only account for seepage in rigid soil, neglecting the complex coupling between deformation and seepage in unsaturated soil. This study develops a new surrogate model for hydro-mechanical coupling in unsaturated soil using the PINN approach. Dimensionless governing equations, including mass balance and force balance equations, are derived and adopted for physical constraints. With absence of explicit constitutive relations, this new surrogate model utilises sparse measured data to identify pore water pressure, effective stress and deformation in unsaturated soil. Separate neural networks are employed to facilitate efficient back-propagation for coupled problem involving multiple outputs. The newly developed model is then applied to simulate two cases with sparse measurements in unsaturated soil. The results illustrate that the newly developed surrogate model successfully learns the elasto-plastic constitutive relation of suction-induced volume change from experimental data. Meanwhile, model predictions regarding both water flow and stress distribution align within the 95 % confidence interval of theoretical values, demonstrating interpretability of PINN model. Furthermore, by adhering to physical constraints, the relative error in predicting soil deformation from neural networks significantly reduces from 49 % to less than 10 %. These findings suggest PINN model with separate networks is capable to simulate unsaturated soil considering both deformation and seepage, even with sparse measured data and incomplete physical constraints.

Okra (Abelmoschus esculentus) is an important vegetable in Ethiopia due to its nutritional value and culinary uses. However, its production is hindered by several challenges. Key issues include diseases like powdery mildew, fusarium wilt, and viral infections, which significantly reduce yields, and pests such as aphids, whiteflies, and fruit borers that further damage crops. The scarcity of improved okra varieties and insufficient drought management exacerbate these challenges. Farmers' perceptions of okra as a low-value crop affect investment and cultivation practices. Additionally, drought, compounded by poor irrigation infrastructure, poses a severe threat to okra production. Despite these challenges, Ethiopia's diverse agro-climatic conditions and fertile soils in regions like Amhara and Oromia offer favorable environments for cultivating okra, with potential yields reaching up to 20 tons per hectare under optimal management. To overcome these constraints, it is essential to improve disease and pest management, develop and distribute drought-resistant varieties, and educate farmers on better practices. Changing perceptions through awareness and community engagement, coupled with supportive government policies, are crucial for enhancing okra production, thereby improving food security and economic stability for Ethiopian farmers in the future.

Thermo -mechanical (TM) behaviour of energy pile (EP) is strongly affected by the pile constraints. The influence mechanism of different constraints on TM behaviour of EP should be revealed. In this paper, the model experiments in laboratory are carried out to research effects of pile end constraints and sand compactness on TM properties of EP. The research results display that the strain of end bearing pile increases gradually along pile depth due to the larger constraint at the pile tip, while the strain distribution of friction pile is large at both ends and small in the middle. Compared with friction pile, the end bearing pile has larger pile tip soil pressure and smaller side friction of pile. The displacement change of EP decreases during the operation in summer mode after applying 0.5 kN load on the pile top. The displacement of end bearing pile can restore to 0 mm, while the friction pile has a settlement of -0.028 mm. In winter mode, both the end bearing pile and friction pile produce unrecoverable settlement, and applying load on pile top will further aggravate the settlement amplitude. As for influence of sand compactness, compared with loose sand, the dense sand has greater thermal conductivity which benefits to the thermal efficiency of EP. This results in smaller pile excess temperature and recovery rate of surrounding soi1 temperature. Accordingly, the pile top displacement and strain also decrease due to small temperature variation, but the pile side friction and side soil pressure increase. The mechanical properties of EP under the condition of different pile -soil parameters are simulated by a 3-D numerical model, which is validated by experiments. It is shown that when the Poisson's ratio of the soil increases from 0.2 to 0.4, the top displacement of the pile increases from -7.35 mm to -8.76 mm, which attributes to the increase in the lateral deformation of the soil, resulting in a decrease in the soil constraint on the pile. Although the distributions of pile side friction along the pile depth are basically the same for different poisson's ratios, there are differences at the pile tip. Also, increasing the pile -soil interface friction coefficient will result in the enhancement of constraint and pile side friction and the decrease of pile body displacement and thus improve the mechanical properties of EP. The study helps to reveal the influence mechanism of various constraints on the TM properties of EP and can provide reference for the safe and efficient operation of EP.