Manne Janga Reddy

A significant volume of water is lost in water distribution networks due to undetected underground leaks. Traditional leak detection methods are often inefficient, labor-intensive, and time-consuming. To address these challenges, this study investigates the utility of machine learning (ML) techniques for leak localization. However, the effectiveness of ML-based methods is highly dependent on the strategic placement of sensors. The proposed methodology employs spectral clustering to form clusters and integrates network topology with sensitivity-based centrality metrics to identify sensor locations within each cluster based on centrality scores. A multi-layer perceptron neural network framework is trained using pressure and flow data from these sensor locations to localize leaks. Incorporating 3D coordinates led to more spatially diverse and topographically optimized sensor placements, improving network coverage compared to 2D-based clustering. Leak localization accuracy varied significantly, from 31.62 to 84.79% at zero tolerance and from 47.27 to 92.82% at 100 m tolerance, highlighting the critical role of optimal sensor placement and data quality in determining the effectiveness of ML-based leak localization. The study concludes that zone-based leak localization maintains high accuracy even under noisy conditions, highlighting its field applicability.

This study proposed a D-vine copula-based methodology to compute the four-variate joint return period of drought events, incorporating severity, duration, peak, and interarrival time. The VIC-3L hydrological model, calibrated using differential evolution, was employed to simulate gridded runoff in the Upper Bhima basin using climate projections from five CMIP6 GCMs under the SSP2-4.5 and SSP5-8.5 scenarios. Future runoff projections reveal substantial seasonal shifts relative to a historical period. In the future, early monsoon runoff (June–August) is expected to decline, while post-monsoon runoff (September–December) might increase. INM-CM5-0 and MIROC6 models have projected higher runoff in post-monsoon months, whereas IPSL-CM6A-LR consistently simulates lower runoff. To characterize the hydrological droughts, the study employed a 3-month Standardized Runoff Index (SRI) and run theory. Spatiotemporal analysis of extreme hydrological droughts over the Upper Bhima basin indicates a decline in drought duration and severity but an increase in peak and reduced interarrival time under future scenarios. Under the SSP2-4.5 scenario for extreme droughts, the basin was projected to experience critically decreasing return periods, about 17.7% (19%) of the study area in the mid-(late-) century.

The climate change brings increasingly serious challenges for society, one of which is the increase in extreme events. Extremes such as droughts have led to substantial impacts on the livelihood of humans and animals. Understanding the spatial extent of drought‐prone regions is important for its assessment and effective management. The Reliability–Resilience–Vulnerability (RRV) integrated index map approach gives a fair idea about the drought‐prone regions. This study conducted an extensive RRV analysis of droughts, using the drought properties duration and severity. These properties were assessed using the Standardised Precipitation Index (SPI‐3). The investigation is conducted across India, with precipitation data collected at a spatial resolution of 0.25°, utilising a dataset spanning 200 years (1901–2100). Historical data for 121 years (1901–2021) collected from the India Meteorological Department, and data for the future period (2022–2100) were collected from the NEX‐GDDP, ACCESS CM_2 climate model for the SSP585 scenario. The analysis is carried out for five periods 1901–1940 (1st), 1941–1980 (2nd), 1981–2021 (3rd), 2022–2060 (4th), and 2061–2100 (5th). The spatial distribution of drought properties, namely duration and severity at the 50th, 70th, and 90th percentile levels, revealed the places with significant and prolonged drought conditions. From the integrated RRV index map, an increase in drought severity was observed until the 4th period, followed by a decline in the 5th period. Noticeable changes are observed in the distribution of the spatial extent of the map, indicating areas prone to drought (integrated RRV map) in India for the 2nd and 3rd periods compared to the 1st period. The results indicate a shift in drought‐prone areas from traditionally arid and semi‐arid regions to more humid regions. The time series analysis of percentage area of India under drought shows that there is a substantial increase up to the 4th period and then decreases in the 5th period. These findings can assist decision‐makers and government bodies identify areas requiring urgent attention for drought management, enabling them to develop region‐specific action plans.

A fuzzy multi-objective optimisation model was investigated for water quality management in a river under uncertain conditions. In this study, to deal with multiple pollutants simultaneously, the National Sanitation Foundation's water quality index was considered as one of the model's objective functions based on fuzzy set theory. This made it possible to investigate the overall effect of uncertainties on simultaneous changes. Another objective function was the total treatment costs for wastewater discharged into the river. A water quality simulation model and a non-dominated-archiving ant colony optimisation algorithm were used to determine the values of water quality parameters and the model's optimal solutions, respectively. Furthermore, a simulation–optimisation approach was adopted for facilitating the problem-solving process and applied to a hypothetical case study resembling a river system in Iran. The results show that the proposed model significantly reduced the total wastewater treatment costs compared with a similar single-objective model with a more cautious and a cost-effective approach. Although the treatment costs were increased compared with the similar deterministic model, more feasible approach was taken by considering the uncertainties associated with the objectives. With suitable modifications, the model could be easily adapted for other river systems.