Brendan Fox

Control of an LCC HVDC system for connecting large offshore wind farms with special consideration of grid fault

This paper describes the control and operation of an HVDC system comprising a line-commuted converter (LCC) HVDC and a STATCOM for connecting offshore wind farms based on DFIGs. During fault on the main grid, fast communications have previously been relied upon to make the wind farm aware of the condition and reduce its power output. Here, an alternative method is examined which enables automatic power balancing during fault. This is achieved through frequency modulation on the offshore network via the STATCOM. Several methods of fault detection using frequency threshold, rate of change of frequency (ROCOF) and rate of change of AC voltage (ROCOVac) are used to indicate when the wind farm power output should be reduced to achieve power balancing, and are compared with results using direct communications. PSCAD/EMTDC simulations show the effectiveness of the proposed control, which allows for faster fault identification. As a result the STATCOM DC over-voltage can be significantly reduced, requiring small DC capacitor, and tripping of the wind farm can be avoided.

Study of fault ride-through for DFIG based wind turbines

Wind power generator and total capacity has increased dramatically in the last 10 years. Most generators now being installed use doubly-fed induction machines (DFIGs). These allow active and reactive power control through the rotor side converter. Therefore todays wind turbines have a significant impact on the power system. To ensure power quality, several utilities have introduced special grid connection codes for wind farm developers. The requirements range from reactive power control, frequency response and, last but not least, fault ride-through. All these requirements, especially fault ride-through, are a challenge for wind turbine producers. New control strategies and hardware are needed which utilize the flexibility provided by the DFIG converters. This paper outlines the proposed grid codes. It then describes a detailed DFIG model and control strategy. The performance of the new control through severe fault conditions is demonstrated.

Validation of Fixed Speed Induction Generator Models for Inertial Response Using Wind Farm Measurements

This paper outlines the use of phasor measurement unit (PMU) records to validate models of fixed speed induction generator (FSIG)-based wind farms during frequency transients. Wind turbine manufacturers usually create their own proprietary models which they can supply to power system utilities for stability studies, subject to confidentiality agreements. However, it is desirable to confirm the accuracy of supplied models with measurements from the particular installation, in order to assess their validity under real field conditions. This is prudent due to possible changes in control algorithms and design retrofits, not accurately reflected or omitted in the supplied model. One important aspect of such models, especially for smaller power systems with limited inertia, is their accuracy during system frequency transients. This paper, therefore, assesses the accuracy of FSIG models with regard to frequency stability, and hence validates a subset of the model dynamics. Such models can then be used with confidence to assess wider system stability implications. The measured and simulated response of a wind farm using doubly fed induction generator (DFIG) technology is also assessed.

Grid Integration of Wind Farms Using SVC and STATCOM

This paper considers the use of the static VAr compensator (SVC) and static synchronous compensator (STATCOM) for wind farm integration. Wind farm models based on fixed speed induction generators (FSIG), using AC connection and equipped with either SVC or STATCOM, are developed. Stability problems with the FSIG are described. An investigation is conducted on the impact of STATCOM/SVC ratings and network strength on system stability after network faults, and comparison is also made between the performances of the two devices. It was found that the SVC and STATCOM considerably improve the system stability during and after disturbances, especially when the network is weak. It showed that the STATCOM gave a much better dynamic performance, and provided better reactive power support to the network, as its maximum reactive current output was virtually independent of the PCC voltage.

Integrated governor control for a diesel-generating set

This paper is concerned with the detailed implementation of real-time fuzzy logic speed control for a standby diesel-generating set. The implementation platform is that of the Mathworks xPC Target. This rapid prototyping scheme permits the automatic cross-compiling of the nonreal-time Simulink control system model into real-time C code, which is executable on the xPC target PC. The digital governor xPC target hardware consists of a desktop PC with a National Instrument Input/Ouput card and a Softing Controller Area Network Card. The paper details the fuzzy control model and the methods with which to communicate with the engine control module. Tests were conducted on a 50-kVA diesel-generating set. The results show that the fuzzy controller is superior to the variable gain PID-type governor used by the conventional engine control module.

Effects of Large Scale Wind Power on Total System Variability and Operation: Case Study of Northern Ireland

The paper simulates the potential impact of significant wind power capacity on key operational aspects of a medium-sized grid-power system, viz. generator loading levels, system reserve availability and generator ramping requirements. The measured data, from Northern Ireland, consist of three years of 1/2 hourly metered records of (i) total energy generation and (ii) five wind farms, each of 5 MW capacity. These wind power data were scaled-up to represent a 10% annual energy contribution, taking account of diversity on the specific variability of total wind power output. The wind power generation reduced the system non-wind peak-generation. This reduction equalled 20% of the installed wind power capacity. There was also a reduction in the minimum non-wind generation, which equalled 43% of the wind power capacity. The analysis also showed that the spinning-reserve requirement depended on the accuracy of forecasting wind power ahead of scheduling, i.e. on the operational mode. When wind power was predicted accurately, (i) it was possible to reduce non-wind generation without over-commitment, but, (ii) the spinning-reserve non-wind conventional generation would usually have to be increased by 25% of the wind power capacity, unless quick-start gas generation was available. However, with unpredicted wind power generation, (i) despite reductions in non-wind generation, there was frequent over-commitment of conventional generation, but (ii) usually the spinning-reserve margin could be reduced by 10% of the wind power capacity with the same degree of risk. Finally, it was shown that wind power generation did not significantly increase the ramping duty on the system. For accurately predicted and unpredicted wind power the increases were only 4% and 5% respectively.

Modelling the Impact of Wind Power Fluctuations on the Load following Capability of an Isolated Thermal Power System

Wind power is a clean and commercially competitive renewable energy technology that affords many utilities the opportunity to diversify and reduce their dependence on fossil fuels. However, the wind is also an intermittent energy source. Hence, many small and isolated utilities are concerned that, as the number and capacity of wind power plants increases, the resulting fluctuations in wind power output will impose excessive load following duty on their conventional units, leading to grid frequency control problems. This paper investigates the potential impact of wind energy development on the load following capability of a representative medium-sized, (2000 MW) thermal power system. Recorded system demand and wind power production data from the Northern Ireland system are statistically analysed, and the impact of expanded wind farm operation on net wind power and system demand fluctuations modelled and predicted for various time-scales of interest. The results demonstrate that the magnitudes of power output fluctuations from well-dispersed wind farms are small compared to system demand variations. Consequently, wind power expansion will not impose significant additional load following duty on the power system. Statistical analysis of net system demand and scaled wind farm time series data empirically verifies the validity of these findings.

Measurement-based estimation of wind farm inertia

The inertia of fixed-speed wind turbine generators (WTGs) helps to mitigate under-frequency transients, promotes fault ride-through and damps inter-area oscillations. It is therefore important to quantify this inertia. The authors use measured wind farm responses during under-frequency transients to provide this information. They discuss the extent of the data and the criteria used to select certain events for further analysis. The estimation of WTG inertia is based on a induction generator model. The basis of the model will be described. The manner in which the model is applied to estimate the inertia from the measured data is then explained. Finally, the implications of the results for power system operation are assessed.

Multiple Input Governor Control for a Diesel Generating Set

The paper presents a multiple-input single-output fuzzy logic governor algorithm that can be used to improve the transient response of a diesel generating set, when supplying an islanded load. The proposed governor uses the traditional speed input in addition to voltage and power factor to modify the fueling requirements during various load disturbances. The use of fuzzy logic control allows the use of proportional-integral-derivative (PID) type structures that can provide variable gain strategies to account for nonlinearities in the system. Fuzzy logic also provides a means of processing other input information by linguistic reasoning and a logical control output to aid the governor action during transient disturbance. The test results were obtained using a 50 kVA naturally aspirated diesel generator testing facility. Both real and reactive load tests were conducted. The complex load test results demonstrate that, by using additional inputs to the governor algorithm, enhanced generator transient speed recovery response can be obtained.

Flicker mitigation strategy for DFIGs during variable wind conditions

This paper presents a flicker mitigation scheme for the doubly-fed induction generator (DFIG) during variable wind conditions. The flicker mitigation strategy was developed based on the distribution line X/R ratio and the active power deviation from the average active power during variable wind conditions. Flicker emission was analyzed using a flicker meter based on the IEC standards. Both short-term and long-term flicker severities were analyzed during the time period of study. The flicker mitigation strategy was evaluated under different system conditions such as X/R ratio, distribution line length, short-circuit capacity (SCC), and wind variability. It is shown that the proposed control scheme mitigates flicker severity with different X/R ratios, distribution line lengths and different wind conditions. However, the proposed strategy is less effective with relatively low SCCs at the point of grid connection, due to large fluctuations of the voltage at the point of grid connection.