Abstract:Hydropower plants operating in fragile mountainous regions face severe hydroabrasive erosion of hydraulic components, which results in drop in efficiency, frequent interruptions in power generation and downtime for repair/replacement. Due to high flow velocity in components like nozzles and buckets, a Pelton turbine is highly susceptible to erosion. The effect of erosion is observed as cumulative loss of material and study of its gradual progress required frequent measurement. In this study, in-situ images of Pelton buckets were captured throughout a sediment period at a hydropower plant located in Himalayan region, India and analysed for characteristics of progressive erosion damages. A multi-layered paint method was used to visualize the initiation of erosion in different parts of bucket. Removal of colours due to erosion provided the information on erosion initiation and various erosion zones in bucket. The erosion of outer side of bucket was analysed with single layer paint removal and images from past records of the plant. Further, the effect of initial condition of a bucket on cumulative erosion after a sediment season is also presented for both coated and uncoated buckets. The in-situ photographs during the silted water period suggested that the outlet portion inside the bucket faces significant erosion. The technique of image capturing throughout sediment season was successful in obtaining progressive erosion. Such method with a reference scale in the image is an effective method to quantify erosion like width increase of the splitter more frequently. The paint in the buckets was found capable of providing information on erosion initiation and erosion zones inside Pelton buckets. The erosion characteristics of outer side of the bucket confirms the presence of a Coanda effect. This study will facilitate turbine designers and manufacturers in better coating and/or surface treatment to specific zones of bucket to minimize efficiency loss due to distorted profile of the bucket, as more prone surfaces can be coated with thicker coatings to protect it longer from erosion.

Effective monitoring of Pelton turbine based hydropower plants using data-driven approach

Performance monitoring of kaplan turbine based hydropower plant under variable operating conditions using machine learning approach

A Hybrid Method for the Run-Of-The-River Hydroelectric Power Plant Energy Forecast: HYPE Hydrological Model and Neural Network

Condition monitoring and early diagnostics methodologies for hydropower plants

Hydro-abrasive erosion in Pelton turbines: Comprehensive review and future outlook

Smart hydropower management: utilizing machine learning and deep learning method to enhance dam’s energy generation efficiency

A Data-Driven Framework for Small Hydroelectric Plant Prognosis Using Tsfresh and Machine Learning Survival Models

A Machine Learning Approach to Forecasting Hydropower Generation

On the use of artificial neural networks for condition monitoring of pump-turbines with extended operation

A Fault Prognostic System for the Turbine Guide Bearings of a Hydropower Plant Using Long-Short Term Memory (LSTM)

Hydropower Station Status Prediction Using RNN and LSTM Algorithms for Fault Detection

Condition Monitoring for Nuclear Turbines with Improved Dynamic Partial Least Squares and Local Information Increment

Numerical prediction of sediment erosion in reference Francis turbine for complete operating range

Selection and Performance Prediction of a Pump as a Turbine for Power Generation Applications

Application of Artificial Neural Networks for Monitoring and Optimum Operation Prediction of Solar Hybrid MGT Systems

Sediment Erosion on Pelton Turbines: A Review

A Health Monitoring Model for Circulation Water Pumps in a Nuclear Power Plant Based on Graph Neural Network Observer

Analysis of hydro-abrasive erosion in a high-head Pelton turbine injector using a Eulerian-Lagrangian approach

Analysis and Visualization of Progressive Erosion in Pelton Buckets

Real-time reservoir operation using data mining techniques

Real-time power prediction approach for turbine using deep learning techniques