Скачать 82.05 Kb.
DEVELOPMENT OF A LINEAR TEST RIG FOR ELECTRICAL POWER TAKE OFF FROM WAVES
N. J. Baker1, M. A. Mueller2, L. Ran3 , P.J. Tavner3, S. McDonald4
1Lancaster University Renewable Energy Group, Engineering Department, Lancaster, UK
2Institute for Energy Systems, School of Engineering, University of Edinburgh, Edinburgh, UK
3New and Renewable Energy Group, School of Engineering, University of Durham, Durham, UK
4New and Renewbale Energy Centre, Blyth, Northumberland, UK
This paper describes the design build and test of a Linear Test Rig for demonstrating novel topology linear machines. The work is aimed at aiding the development of power take off for wave and tidal energy converters. The rig consists of three parts: Prime Mover, Test Machine and Converter. Experience and method for the design manufacture and initial testing of each of these componnts is detailed.
Dr Baker is an Engineering lecturer within the Lancaster University Renewable Energy Group pursuing research interests in various aspects of power capture and conversion in wave energy devices.
The technically accessible wave resource is capable of providing 11-15% of the UK’s electricity supply . Work on exploiting the resource commenced in the 1970s, but only since the end of the 1990s has there been significant growth in marine technology developers and successful offshore prototype deployments. According to the Carbon Trust, currently there are in excess of 50 devices in development. Although there appears to be no single technology that is considered optimum, some developers are at a more advanced stage than others. For example, Ocean Power Delivery have tested a full scale prototype at the European Marine Energy Centre and have signed a contract for a small farm of devices off the Portuguese coast. AWS Ocean Power demonstrated a full scale prototype in Portugal in 2004. Wavedragon are currently undergoing consents for a device off the coast of Wales. These devices have hydraulic, linear generator and water turbine power take off respectively and illustrate the diverse ways in which wave energy can be captured. High pressure hydraulic power take off schemes have long been proposed for wave energy devices, and the technology in the Wavedragon can be found in any low head hydro scheme, yet linear generator power take off is new technology with numerous electrical and mechanical engineering design challenges.
In this paper the authors focus on linear generators, and present the evolution of an electrical generator concept which has been developed to overcome some of the problems encountered when using conventional electrical generator technology for direct drive systems.
Electrical Power Take Off Systems in Wave Energy
Although the power take off method varies in alternative devices, the electrical power conversion is very similar for each, namely conventional high speed rotary electrical generators. Wave energy devices reciprocate at a typical peak velocity of 1-2m/s. In order to convert this low speed into high speed rotary motion, hydraulic or pneumatic power conversion is used. The use of hydraulics operating at 400 Bar can give compact actuators which is a distinct advantage in some types of Wave Energy Converter (WEC) where size and weight are an issue. However, standard hydraulic systems are expensive and their seals are designed to operate at velocities lower than a typical WEC. Sea water ingress is one potential problem  and oil leakage into the sea is another concern. The interface required to convert the low speed reciprocating motion to rotary motion is considered as an area requiring further research in the Ove Arup report . Current development work is being undertaken to exploit digital displacement hydraulics for wave energy devices . This technology demonstrates high conversion efficiency at both full and part load, which is important for WECs as they are likely to spend a large proportion of time operating at part load.
Reliance on conventional high speed rotary electrical generators has forced the engineer to design complex power take-off schemes. If the electrical generator could be directly coupled to the wave energy device there would be fewer stages. So-called direct drive energy systems have been successfully implemented in wind energy converters (Enercon). In a WEC the generator would reciprocate at the low speeds of the actual device, one or two orders of magnitude lower than the velocities experienced in conventional high speed rotary generators. At such low velocity the force reacted by the generator will be very large requiring a physically large machine. Rare earth permanent magnets made from Neodymium Iron Boron (Nd-Fe-B) represent one method of getting relatively high flux densities into the magnetic circuit of electrical machines utilising small pole widths. One family of machines, known as Variable Reluctance Permanent Magnet (VRPM) machines, uses the interaction of small pole pitched magnets with iron teeth. Prototypes of two members of this family have been built, demonstrated and reported on for this application . Other comparisons have also been undertaken, e.g. . VRPM machines exhibit high shear stress but also poor power factor . An overrated power converter is required to supply the reactive power to the high inherent inductance .
High flux density in the air-gap gives rise to a normal magnetic force of attraction between the steel translator and steel stator. These ‘air gap closing forces’, referred to as the Maxwell stress force, can present the mechanical engineer with difficulties when considering lubrication of the machine. In some machines, the air gap closing force can be in the region of twice the useful force of the machine . The Maxwell Stress force can be eliminated by removing one of the iron surfaces. It is most usual to remove the stationary iron, so that the coils are supported in a non-magnetic material. Under such circumstances the machine is then referred to as air-cored. Tubular air-cored topologies have been investigated by these  and other authors . Elsewhere it has been considered as a generator for free piston engines , where the small amplitude (76 mm) , high frequency (50 Hz) and high velocity (11 m/s) of oscillation result in a much smaller machine size. Size of machines covered here are around two orders of magnitude greater and hence require radically different mechanical structures.
Hydrodynamic parameters of wave energy devices required for numerical modelling, such as the radiation coefficient and added mass, are complex non linear frequency dependent variables. Predicting behaviour in realistic mixed frequency sea states thus often relies heavily on scaled tank testing for parameter derivation and model verification. Initial development of wave energy devices has traditionally been done at small scale (1/100th-1/50th), allowing easy and cheap modification whilst providing accurate replication of device performance. Preservation of the Froude number, i.e. the ratio of inertial to gravity forces, is required to re-produce full scale behaviour in a small model. In practice this means the velocity of the device is scaled by the square root of the geometry scale factor, whereas the value of acceleration is unaltered. Hence a quarter scale model will oscillate at half the speed and twice the frequency of full scale. Taking a typical scaling factor for laboratory wave energy research as 1/100th scale, a 1.5 m/s full scale velocity would be reduced by the square root of the scaling factor and give a corresponding velocity of 0.15 m/s. For power take off systems, and in particular for electrical machines, research at this scale is unfavourable. Scaled testing using velocities an order of magnitude lower than those present at full scale introduces problems to the designer which offset the motivation for prototyping at small scale. When looking to develop scaled prototypes of power take off systems, there are two more appropriate scaling techniques. Firstly the displacement amplitude and time step may be scaled down linearly, such that the device’s velocity is preserved thus a 0.1 Hz 2 m amplitude full scale wave is represented by a 0.4 Hz 0.5 m amplitude quarter scale model, both having a peak velocity of 1.25 m/s.
Secondly the displacement profile can be unaltered and the reactive force scaled down, so the prototype follows the displacement pattern of a full scale device but only extracts a proportion of full power. The rig introduced here is capable of both these methods of scaling, although the specification was drafted on the former.
There is no one central facility where developers can test power take off mechanisms at a reasonable scale and under typical reciprocating conditions as would be experienced offshore. In the development of the Pelamis device OPD performed a number of tank tests at different scales, but it was not until the 1/7th scale device that the power take-off mechanism was tested and ultimately the company built a full-scale power take off test rig in order to develop the control strategies for the hydraulic power take off mechanism under typical wave conditions . Such an approach has the advantage that the company is able to do all testing in-house, but it is costly. In this paper the authors describe the development of a facility for testing alternative power take off systems for WECs. Although the focus of the project was for testing electrical machines, the test rig is equally applicable to hydraulic power take off. Fig 1 shows the outline of the rig consisting of three parts: Prime Mover, Test Machine and Converter.
Figure 1: Generic layout of Linear Test Rig (LTR)
Тат (Thermally Activated Technologies). Тат consider, that: to 2020th 5% of total energy consumed in the usa will fall to utilized...