End-to-end multi-agent deep reinforcement learning for cooperative multi-UAV tasks often involves complex policy learning, slow convergence, and difficulties in satisfying physical constraints. This paper proposes a hierarchical cooperative control method that combines multi-agent soft actor-critic (MASAC) with model predictive control (MPC). The method separates planning from execution: the high-level MASAC generates reference states for cooperative guidance, while the low-level soft-constrained MPC tracks these references subject to dynamic constraints. An inter-level slack-feedback gate uses normalized slack magnitudes in the position, velocity, and camera-attitude channels to adjust the next reference state. A composite dense reward provides continuous feedback on task progress, safety, terminal performance, and decision continuity. Simulations are conducted in a multi-UAV inspection scenario with randomly generated obstacles. The proposed method achieves a task success rate of 93.3
This study presents a strategic approach for sustainable turquoise hydrogen production in NEOM City, Saudi Arabia, by integrating a methane pyrolysis unit with a hybrid renewable energy system to produce carbon-free hydrogen and solid carbon. This system incorporates photovoltaic panels (PV), wind turbines (WT), fuel cells (FCs), a reformer (RF), an electrolyzer (EL), a hydrogen tank, and batteries (BT). HOMER software was employed for the techno-economic assessment of the hybrid renewable system, aiming to identify the optimal configuration for powering the methane pyrolysis unit. The most efficient setup, featuring 231,523.8 kW PV, 9 WTs, 40,000 kW FCs, and an RF producing 2400 kg/h, achieved the lowest cost of energy (COE) at 0.0945/kWh and a hydrogen production cost (COH) of1.884/kg. In contrast, a simpler PV-WT-BT system resulted in a higher COE (0.178/kWh) and COH (2.706/kg). Interestingly, when hydrogen from the pyrolysis unit fueled the FCs in a PV-WT-FC-BT system, the COE was 0.136/kWh, and COH was2.295/kg. These findings highlight methane pyrolysis integrated with renewable energy as a promising solution for cost-effective and low-emission hydrogen production, aiding the clean energy transition.
Reduced-switch-count (RSC) multilevel inverters (MLIs) are attractive for photovoltaic (PV) power conversion, lowering device count, cost, and control complexity while retaining multilevel power-quality benefits. This paper proposes a reconfigurable RSC-MLI topology using only 12 switches and four DC sources, generating either 21 levels (source ratio 1:2:3:4) or 31 levels (binary ratio 1:2:4:8), suited for PV arrays interfaced through individual MPPT-controlled DC–DC converters. To compute switching angles for selective harmonic minimization under low-switching-frequency PWM, the nonlinear transcendental equations are reformulated into vector–matrix form and solved using an improved Jacobian-based iterative update shown analytically to converge quadratically. Compared with Newton–Raphson, the method reduces average iterations from 35 to 20, computation time from 12.5 to 5.2 ms, and improves solving success rate from 91.4 to 100 ^2 ), temperature variation, and uniform DC-source fluctuations, confirming its robustness for grid-connected and standalone PV applications.
The growing demand for sustainable construction materials has increased the use of industrial by-products and mineral additives in cementitious composites. The compatibility among these additives directly affects the strength and durability of cement-based systems. Therefore, in this study, cementitious composites were produced using fly ash hybridized with colemanite and colemanite waste. The synergistic effect between the amorphous aluminosilicate structure of fly ash and the boron-rich composition of colemanite-based materials was examined. Flowability, compressive strength, ultrasonic pulse velocity (UPV), high-temperature resistance, and microstructural analyses were performed. Although the rough surfaces of colemanite and its waste reduced workability, the spherical morphology of fly ash mitigated this negative effect. While mineral additives generally reduced early-age strength, the combined use of fly ash and colemanite enhanced long-term compressive strength. In contrast, colemanite waste markedly reduced early-age strength due to its limited pozzolanic reactivity. Although hybridization with fly ash also resulted in early-age strength reductions, the hybrid systems exhibited substantial strength recovery over time, achieving 90-day strength gains of up to 27.39
This study presents a GIS-supported spatiotemporal analysis of long-term landslide occurrence across 81 provinces of Türkiye for 1950–2023 using normalized annual landslide densities, expressed as events per 1,000 km2. A national landslide inventory was structured within a GIS environment, linked to provincial and regional spatial units, normalized by province area, and analyzed as a standardized spatiotemporal geohazard database. Four nonparametric trend detection methods were comparatively applied: Mann–Kendall, Sen’s slope estimator, Spearman’s rank correlation test, and Innovative Trend Analysis. Statistical significance was evaluated at the 1, 5, and 10
