On-site grid code compliance testing for wind farms



 

Before a wind energy farm (WEF) is connected to the grid it must be tested to certify that it complies with both performance requirements and the grid code. Testing and verification is made difficult because the dependence on primary source conditions, i.e., sufficient wind and sun of the required level and stability. For this reason on-site grid simulation equipment has been developed which allows testing without interference with the grid.

One of the problems with testing WEF for compliance is the availability of sufficient steady wind to ensure full test of the performance and compliance of the WEF to the grid code under representative conditions. This does not apply to solar energy farms as most days will ensure steady conditions, and passing clouds may co-operate to test ramp up and down rates. Testing ramp rates for wind and solar can be difficult as they are dependent on the prevailing weather conditions, and finding a day that satisfies the test conditions could be difficult.

In addition, certain performance in response to grid conditions cannot usually be tested on site directly as the grid cannot be modified to produce these conditions, and thus modelling is conventionally used to certify compliance. On-site testing methods have been developed to test individual units in wind farms, using the remaining units to simulate the grid.

One of the most important tests is the low voltage ride through ability. WEFs were originally allowed to disconnect during fault conditions but current grid codes require ride-through capability. The term ride-through has an indeterminate meaning, depending on which administration is involved.

Fig. 1: Transformer based voltage sag generator [3].

It describes an intermediate condition between “on” and “off’”. In this particular context, “ride-through” describes the ability of an inverter or turbine to remain on line during a faulted condition (in this case, low or zero voltage) and resume operation directly after clearance of the faulted condition without any mechanical switching operation. It does not necessarily imply that active power is being supplied during the fault, nor that the system voltage is being supported by the wind turbine. It simply means that the turbine does not trip for the required duration of the fault.

SA grid code testing

The South African grid code compliance test for wind energy farms requires the following tests/verification for each installation:

  • Active power management
  • Transmission system (TS) or distribution system (DS) voltage requirements
  • Signals, communications and control
  • Dynamic properties (simulations) [1]

Active power management, system voltage requirements and signals, communications and controls are tested/verified on site. These tests are carried out by applying the relevant controls to the system and observing the response. The grid is assumed to remain normal under these test conditions, i.e., the response of the unit is not triggered by the grid. For example the frequency control test will require to be simulated by means of injection of a frequency signal into the WEF controller to simulate appropriate changes of frequency.

The following are conventionally confirmed by simulation and modelling in accordance with South African grid code guideline:

  • Fault ride through capability
  • Fast acting reactive and/or active power contribution during faults
  • Post fault active power recovery
  • Power oscillations damping capability
  • Synthetic inertial capability

Fig. 2: Reactor based fault simulation network [2].

The first three of these tests involve the reaction of the WEF to faults on the grid, and rely on simulations, as it is normally not desirable or possible to place fault conditions on the live grid for the purposes of testing, especially if repeat tests are required.

Simulation, although useful, does not always provide reliable results, as models tend to use average and typical conditions and values rather than the real on-site values, and do not provide verification of the capabilities of equipment as built. For these reasons, methods have been developed which allow simulation of grid fault conditions on site, without affecting the grid. The ability to perform tests on real wind turbine equipment to analyse LVRT and fault response capability has been made possible by suitable grid fault simulation systems. There are several test systems used as simulators [3].

Generator plant

In the generator based approach a diesel powered synchronous generator produces controlled symmetrical voltage sags by changing the field excitation. Hardware costs are high due to the weight and scale of the diesel engine and the synchronous generator. Only symmetrical faults can be emulated with this test method. Besides, ramp-up and ramp-down times are within several cycles of mains frequency, which is too slow to emulate realistic grid faults.

Fig. 3: Various grid code LVRT requirements [4].

Transformer based plant

A combination of a step-down auto-transformer and switches is another implementation option of a voltage sag generator. The structure of a typical transformer based VSG is shown in
Fig. 1. Appropriate control sequences can be applied to the switch S2 in order to produce a voltage sag with desirable depth. By opening the switch S1, a 100% voltage dip is generated. Operate times of the switches can be a limitation and some applications of this technique use electronic switches.

Reactor networks

A typical reactor fault simulation network is shown in Fig. 2 [2].

According to the international standard for the measurement of power quality characteristics of wind turbines (IEC 61400-21), an inductive voltage divider is recommended, connected ahead of the plant to be tested.

This voltage divider consists of a longitudinal impedance (coil) L1 and a short-circuit impedance L2. The figure shows a simplified view of the test equipment. The impedances L1 and L2 can consist of several coils each (series and parallel connection). By changing the ratio L1 to L2 the depth of the voltage dip can be configured.

Depending on the respective grid code, different depths of voltage dips have to be simulated. For wind turbines, dips of <5%, 25%, 50% and 75% of the rated voltage are required (see Fig. 3).

Fig. 4: Back-to back converter-based grid code compliance test system [4].

The duration of the dip depends on the depth and ranges from several hundred milliseconds, to several seconds. In some cases the duration can also extended to several minutes. German and international guidelines demand the simulation of three-phase as well as two-phase faults.

In England, guidelines additionally demand single phase faults against earth. The test system is usually stored in specially equipped standard transport containers and contains the coils and switching devices. Large-size test systems (for generating plants in the multi-megawatt range), often require two or more 12 m containers. The mobile test system can thus be transported to the respective test site for free-field measurements.

Back to back converters

Impedance-based testing equipment is limited to voltage dips and swells. For this reason, many of the requirements remain unverified, except by simulation. The use of fully controllable converter systems operated as test equipment allows for a wide variety of tests that can be carried out on the generating unit.

A back-to-back converter is connected in between the grid and the load, as shown in Fig. 4. By controlling the load-side output voltage a wide range of grid faults can be emulated.

Fixed test units for this particular requirement have been established at several manufacturers and test laboratories [2, 3]. These units simulate the grid and fault behaviour but do not use complete turbines, but only the generator, which is motor driven, and the associated converters.

Motor control circuits are used to simulate rotor inertia on the turbine [2].

Latest developments have produced portable units which allow on site testing, as well as a method which uses active wind turbines on the site to simulate the grid.

On site test unit based on active wind turbines

The proposed method, which has been tested on a simulator, makes use of the installed wind turbine (WT) converters to emulate the grid and produce a given fault for the WT under test. To guarantee the safety of the grid, both the WT under test and the grid emulating WTs are disconnected from the grid and operate in stand-alone mode during the test.

Fig. 5: Test set up for onsite LVRT using active wind turbines [4].

The system configuration, testing procedure, control strategies were considered in the simulation. A wind farm with typical 2 MW full converter permanent magnetic synchronous generator WTs was modelled and simulated. Practical issues including the minimum converter capability for grid emulation, WT over-speed limitation, grid impedance and fault conditions emulation were studied.

Simulation results have shown that with only normal control algorithm modifications is it feasible to conduct on-site a series of LVRT tests for the full converter WT safely under the given grid capability and fault conditions [4]. Fig. 5 shows the test set-up.

References

[1] Nersa: “Grid Code compliance testing procedure for Wind Energy Facility (WEF) connected to the South African Distribution and Transmission systems.”
[2] J Dirksen: “Low voltage ride through”, Dewi Magazin, 23 August 2103.
[3] Yang, et al: “Benchmarking of voltage sag generators”, 38th Annual Conference of the IEEE Industrial Electronics Society.
[4] JC Turu and SM Muratel: “Novel On-site LVRT testing method for full converter wind Turbines”, Tongji University, 2014.
[5] NI Espinoza: “Grid code testing of wind turbines by voltage source converter based test equipment”, http://publications.lib.chalmers.se/records/fulltext/220144/220144.pdf

Send your comments to energize@ee.co.za

The post On-site grid code compliance testing for wind farms appeared first on EE Publishers.

Source: EE plublishers

More news