Assessment of cyclone risk and case study of Gaja cyclone using GIS techniques and machine learning algorithms in coastal zone of Tamil Nadu, India.

Cyclones can cause devastating impacts, including strong winds, heavy rainfall, storm surges, and flooding. The aftermath includes infrastructure damage, loss of life, displacement of communities, and ecological disruptions. Timely response and recovery efforts are crucial to minimize the socio-economic and environmental consequences of cyclones. To accelerate the time-consuming risk assessment process, particularly in geographically diverse regions, a blend of multi-criteria decision-making and machine learning models was utilized. This novel approach swiftly assessed cyclone risk and the impact of the Gaja cyclone in Nagapattinam, India. The method involved assigning weights to distinct criteria, unveiling notable vulnerability aspects like elevation, slope, proximity to the coast, distance from cyclone tracts, Lu/Lc, population density, proximity to cyclone shelters, household density, accessibility to healthcare facilities, NDVI, and levels of awareness. Daddavari, Ettugudi, Kodikarai, Vedharanyam, Velankanni, and Thirupoondi face high/extreme cyclone risk. Nagore, Nagapattinam, Pillai, Enangudi, and Sannanllur have low/no threat. To further enhance the precision of the study, machine learning algorithms like SVM, SAM, and MLC were deployed. These models were instrumental in generating pre- and post-cyclone land use maps. The influence of Gaja cyclones effects shows decreasing of agriculture land from 34% to 30%, aquaculture increase 1%, barren land decrease from 8% to 6%, Built-up land decrease from 15% to 13%, land with scrub and salt pan also decrease from 21% to 17% and 10%-8%. Mostly effect of Gaja cyclone is dramatic increase of water body from 8% to 21%. Conducting cyclone risk zone analysis and pre/post-cyclone Land Use Land Cover (LULC) detection in Nagapattinam offers valuable insights for disaster preparedness, infrastructure planning, and climate resilience. This study can enhance understanding of vulnerability and aid in formulating strategies to mitigate cyclone impacts, ensuring sustainable development in the region.

Performance assessment of a solar powered hydrogen production system and its ANFIS model.

Apart from many limitations, the usage of hydrogen in different day-to-day applications have been increasing drastically in recent years. However, numerous techniques available to produce hydrogen, electrolysis of water is one of the simplest and cost-effective hydrogen production techniques. In this method, water is split into hydrogen and oxygen by using external electric current. In this research, a novel hydrogen production system incorporated with Photovoltaic - Thermal (PVT) solar collector is developed. The influence of different parameters like solar collector tilt angle, thermal collector design and type of heat transfer fluid on the performance of PVT system and hydrogen production system are also discussed. Finally, thermal efficiency, electrical efficiency, and hydrogen production rate have been predicted by using the Adaptive Neuro-Fuzzy Inference System (ANFIS) technique. Based on this study results, it can be inferred that the solar collector tilt angle plays a significant role to improve the performance of the electrical and thermal performance of PVT solar system and Hydrogen yield rate. On the other side, the spiral-shaped thermal collector with water exhibited better end result than the other hydrogen production systems. The predicted results ANFIS techniques represent an excellent agreement with the experimental results. In consequence, it is suggested that the developed ANFIS model can be adopted for further studies to predict the performance of the hydrogen production system.