Water loss from urban water distribution networks (WDNs) challenges sustainable water management and utility operations. Early leak detection and accurate localisation are crucial for reducing non-revenue water and maintaining network efficiency. This study proposes a machine learning-based framework integrating EPANET-generated static hydraulic simulations with an artificial neural network (ANN) optimised using particle swarm optimisation (PSO) for effective feature extraction. Pressure data under various leak and no-leak scenarios were simulated, and sensitive sensor nodes were strategically selected to enhance accuracy. By bridging simulation and reality, this method successfully integrates hydraulic modelling with data-driven learning, demonstrating the hybrid framework’s strong potential for scalable and accurate leak detection, localisation, and magnitude estimation in WDNs.
The paper establishes legal drafting as the framework for investigating the causes of insufficient clarity in spatial planning texts in the Republic of Croatia. It theoretically and empirically defines the fundamental characteristics of redundant spatial-planning content, the avoidance of which would make plan texts clearer and more transparent. An analysis of redundant content in the sections regulating the construction of infrastructure systems, based on a representative sample of local-level spatial plans, showed its systematic prevalence in the overall content, with a recognizable set of recurring content errors. Its presence can be explained by insufficient understanding of the subject and purpose of spatial-planning regulation within the Croatian spatial planning system.
This study aimed to evaluate the impact of façade design parameters in north-oriented residential buildings, where scenic views take priority over ideal climatic orientation, and develop design strategies that ensure optimal energy performance under these specific contextual limitations. A decision-support model is proposed to analyse the sensitivity of façade design parameters to heating and cooling loads as well as CO₂ emission values, and to identify optimal façade solutions through three main stages: creating the building model, applying the sensitivity analysis method, and applying the optimisation method.
This study develops a refined trigonometric shear deformation theory combined with Eringen’s nonlocal elasticity for the hygrothermal buckling analysis of functionally graded sandwich nanoscale plates on Winkler–Pasternak foundations. The formulation satisfies zero transverse shear stresses at the surfaces without correction factors. Analytical solutions are obtained using Navier’s method for simply supported plates subjected to temperature and moisture gradients. Parametric studies assess the effects of geometric slenderness, foundation stiffness, and the shear function parameter q. The results agree closely with benchmark results of the trigonometric shear deformation theory (TSDT), with errors below 2%. The proposed approach enables efficient and accurate stability prediction of advanced sandwich structures, supporting the design of lightweight and durable components for civil, marine, and aerospace engineering applications.
A vessel-bridge collision can result in severe loss of life and substantial damage. This study presents a numerical analysis of the vessel impact on a column of the Ilok-Bačka Palanka bridge in Croatia. Using finite element modelling, different barge–column interactions under multiple loading conditions were simulated. The numerical model includes realistic vessel and bridge-pier geometry, as well as material behaviour. The numerical simulation results were compared with the damage caused to the column after the impact. The column damage was evaluated using a concrete degradation modelled via an erosion-based algorithm driven by instantaneous geometric strain. The calibrated numerical model was also used in a parametric analysis to determine the influence of the impact direction and speed on the observed bridge pier.
