The project aims to utilize the advances in digitization for optimizing the maintenance of Pelton runners and thereby reducing the need for larger refurbishments. Pelton turbine will be used to develop monitoring methods and verify numerical analyses.

Visual investigations of the flow in Pelton turbines will be carried out at Waterpower Laboratory in NTNU. High-speed camera will be deployed to visualize the flow field in the Pelton runner and to monitor erosion in the Pelton turbine. Here, in-depth studies are required on how to interpret visual images from eroded surfaces in the Pelton turbine’s impeller.

Plan is to implement the research outcomes in Cheves hydropower plant (HPP), built and operated by Statkraft Peru. With an annual energy production of 825 GWh, it consists of two Pelton turbines of 84 MW each operating under 602 m of net head. Inflow to Cheves HPP is a combination of flows from Checras and Huaura river.

These rivers drain steep mountainous terrain towards the Pacific and are characterized with extremely high bed load. In terms of mineralogy, 81% of the suspended sediment is comprised of quartz, which is highly abrasive to the steel used in runners. Sediment-related challenges include: loss of runner efficiency, the cost of frequent change-out and repair of runner and the cost of abrasion-resistant tungsten carbide coatings of runners among others.

The aim is to develop an algorithm that will compare images of the turbine from multiple times in the erosion process, and thus estimate its condition. With such a system, maintenance can be planned in relation to the actual wear and power loss of the turbine, as well as to minimize downtime and optimize the timing of maintenance. 

Value Potential of the project:

There has been ongoing research in the condition of the hydraulic machineries due to changes in the operation pattern resulting from the introduction of new renewables in the energy system throughout the world. On the top of that, erosion is another unavoidable problem for the powerplants operating in sediment laden water such as in Peru and Nepal.

Optimized design of settling basin, sediment sluicing system, design of runner blades and guide vanes, coating of turbines and other different approaches have been applied to minimize the problem. But still some fine particles pass through the hydromechanical components and cause damage to them due to erosion which get worsened during wet season.

Such components need to be inspected and maintained on regular basis which will increase the down time as well as maintenance and operation cost of the powerplant itself. Real time monitoring techniques to inspect erosion condition from image processing can considerably reduce the downtime and unexpected unit shut down which will improve plants overall efficiency and reliability.