Hydropower projects are usually very large, designed to last for decades and then to be refurbished. Owing to the huge costs involved and the need to achieve the best possible results and efficiency, extensive testing of hydropower performance is required, although the sheer size of these projects precludes testing at full scale. An independent facility from EPFL, the Swiss Federal Institute of Technology in Lausanne carries out reduced scale tests of hydraulic machinery from all over the world, based on IEC industry standards.
Hydraulic power plants, which account for about 16% of global electricity generation, are engineered to fit the specific hydraulic conditions of the hydropower plant site, i.e. head (the difference in vertical elevation between the head [reservoir] water level and the level where the generating turbines are installed) and yearly flow duration curves.
Every project undergoes a competitive tendering process designed to identify the maximum value that can be generated from the available hydropower resource. This helps hydro equipment suppliers shorten the development, design, manufacturing and installation time of turbine units as much as possible. Computer-aided tools are used systematically to design the turbines, while experimental tests on reduced-scale physical models of the turbines are carried out to verify that the design will meet or even exceed expected turbine performance.
This model testing stage, which is a key milestone in the development phase and is subject to stringent contractual conditions, relies on an International Standard developed by IEC TC 4: Hydraulic turbines.
Model testing helps optimize hydropower turbines
One feature of hydropower turbine technology that often surprises is that the hydraulic design of each turbine unit – the shape of the runner blades or the hydraulic layout – is unique. It is optimized and tailored for the operating conditions, head and flow, of the specific project. For any of the hydropower plants listed below, a mere 0.1% deviation in efficiency represents at least 500 MW, which is equivalent to the entire capacity of the Swiss Mühleberg nuclear power plant!
The risk analysis of such large projects leads to the performance of reduced scale physical model tests. These allow the consequences of further unexpected problems during the operation time of the prototype to be mitigated – for instance a lower efficiency than the one guaranteed by the turbine supplier, high levels of pressure fluctuation or cavitation that limits the operating range of the turbines. Tests also allow for verification that the turbine runaway speed [i.e.,maximum speed at full flow] does not exceed the maximum speed limit for the generator.
When a unit is modernized, the risk of unexpected outage and lack of generation increases. Modernization consists of upgrading key components of the turbine or the pump-turbine, such as runner, guide vanes and ancillary equipment. This is achieved by optimizing these components within the aged layout of the existing base line machine. This is likely to have been designed and engineered anywhere between 30 and 90 years ago and the drawings for it may no longer even exist! As a result, early on in the modernization process of a hydropower plant, most often at the time of the water-use licensing renewal process, a so-called base line model test is performed to assess accurately both the efficiency of existing units and the value of the modernized units that are being proposed by manufacturers competing for the work.
Renowned lab key work relies on IEC Standard
In this context, the Laboratory for Hydraulic Machines (LMH-EPFL) of the Swiss Federal Institute of Technology in Lausanne (EPFL) has played a key role. It has been involved in the modernization process for more than 25 years.
Its “main activities are teaching, research and service in the field of hydrodynamics of rotating machines such as hydraulic turbines, pumps, pump-turbines, etc.”
LMH-EPFL is equipped with three “universal type” test rigs complying with IEC 60193:1999. This is a standard specification for performing reduced-scale model acceptance tests for all types of hydraulic machinery such as turbines, storage pumps and reversible pump-turbines, with either vertical or horizontal axis.
The three rigs are equipped with high-precision measuring instruments. As such they are suitable for performing development and acceptance tests with an accuracy of better than 0,3% measured efficiency, as prescribed by IEC 60193.
They are generally used in closed circuit mode for performing efficiency and cavitation tests. Cavitation flows are commonly observed in hydropower plants when the generating unit is operating in off-design conditions: i.e., with the output power lower or higher than the nominal value. They might represent a risk in terms of the stability of the hydromechanical system by inducing pressure pulsations in the hydraulic circuit as well as erosion of the runner blades.
For a given project, a 1/10 to 1/14 reduced-scale model featuring geometrical homology with the prototype is manufactured and installed on LMH-EPFL test rigs between the high-pressure pipe (headwater) and the low-pressure tank (tailwater) for efficiency tests. The low-pressure tank features a free surface where the reference pressure can be adjusted with a vacuum pump to investigate cavitation development.
An efficiency hill chart is commonly used to illustrate measurements made over an enhanced operating range. The “hydraulic” power is derived from measurements of the differential pressure between the machine inlet and outlet sections and of the mass flow discharge while the “mechanical” power is derived from measurements of the rotation speed and of the resulting torque acting on the DC electrical machine. The efficiency of the machine is then calculated and the maximum error should be less than 0,3%, according to IEC 60193.
Discharge is the factor most likely to display the highest error. Every EPFL test rig is equipped with an electromagnetic flow meter measuring discharge values ranging from 0,05 to 1,35 m3/s. The inline calibration of the flow meter is performed by comparing the flow meter output values with the corresponding discharge value given by the time required to fill the 150 m3 tank.
Measurements other than the classic efficiency and cavitation tests can also be performed in model scale, ranging from air injection for pressure fluctuation mitigation to four quadrant measurements:
- hill chart efficiency measurements over an enhanced operating range
- Winter-Kennedy and/or ultrasonic flow meter index tests
- cavitation mapping and observations such as visualization of cavitation development in the draft tube cone of a reduced-scale physical model of a Francis turbine, according to the operating points of the efficiency hill chart
- pressure fluctuation survey
- air injection for mitigating pressure fluctuations
- axial thrust measurements
- guide vane torque measurements
- runaway speed and, for reversible pump-turbine, four quadrant measurements
- dimensional checks of the runner and guide vane blades
The list of model tests performed by EPFL Laboratory for Hydraulic Machines is available at http://lmh.epfl.ch/expertise
It is estimated that an additional hydro capacity of 1 000 GW will be installed by 2050, mainly in Africa, Asia and South America. This represents a massive potential contract volume for hydro equipment suppliers. In addition, a total capacity of 1 000 GW will be refurbished over the next few decades. All of this hydropower work, together with the development of renewable energy sources, offers significant research and testing activity potential for hydropower electrical generation in general and for LMH-EPFL in particular.
By Dr Loïc Andolfatto, Mech. Eng. MSc. Ecole normale supérieure de Cachan, France, Head of EPFL LMH Testing Group, Expert for IEC TC 4 / Working Group 33 and Prof. François Avellan, Eng. Dr Université d’Aix Marseille 2, Director of EPFL LMH, Convenor of IEC TC 4 Maintenance Team 32