Research Projects

Hydrokinetic energy conversion offers a promising solution for tapping into the kinetic energy of flowing water in rivers, tidal currents, and artificial channels, making it an attractive renewable energy source. This technology holds significant potential for addressing energy needs in remote or rural regions with access to flowing water. However, realizing its full potential requires overcoming key challenges, particularly optimizing the performance of hydrokinetic turbines to ensure efficient and reliable energy generation. Among the various turbine configurations, the Darrieus turbine — characterized by its vertical axis and high-power coefficient at elevated tip speed ratios — shows great promise. However, issues like self-starting difficulties and torque pulsations due to changing blade angles present significant obstacles to its widespread adoption. The primary goal of this proposed research is to develop an Artificial Neural Network (ANN)--based model to optimize the performance of hydrokinetic turbines by addressing these challenges. The ANN model will incorporate a wide range of turbine design and operating parameters, such as solidity, blade profile, helicity, aspect ratio, pitch angle, number of blades, and the presence of struts or end plates. Additionally, the model will account for the variation in water velocity typically found in Indian canals, which ranges from 0.5 to 5 m/s, ensuring that turbine performance is optimized for different flow conditions. By training the model on a comprehensive dataset drawn from computational simulations (using tools like QBlade), experimentation, and existing literature, it will be possible to predict key performance metrics such as power coefficient, torque coefficient, and tip-speed ratio. This approach offers a robust pathway to improving turbine efficiency across a wide range of operating conditions, including the flow variability typically found in Indian canal systems, where water velocities range from 0.5 to 5 m/s. The aim is to integrate diverse design parameters into an ANN-based model, enabling accurate prediction and optimization of Darrieus hydrokinetic turbine performance while addressing challenges such as torque pulsations and self-starting failures. A combination of computational simulations (QBlade) and experimental validation will be employed. Simulations will be conducted to explore the effects of different design parameters under varying flow conditions, including the fluctuating velocities of canal systems, and the results will be validated through experimental data from hydrodynamic testing facilities. Once the ANN is trained, it will be capable of predicting turbine performance across a broad range of scenarios, offering a powerful tool for optimizing turbine designs. This research will significantly advance both the theoretical understanding of hydrokinetic turbine dynamics and their practical application in renewable energy systems, particularly in Indian canal networks.