Abstract – This paper reviewed some of the technical aspect of the utilization of battery energy storage system to solve several power systems problems particularly when connected to renewable energy systems. This paper will propose a novel design of a three-phase battery energy storage system as an interface between the supply system and the load. The proposed three-phase multi-purpose Battery Energy Storage System will provide active and reactive power independent of the supply voltage with excellent power quality in terms of its waveform. The paper will discuss the hardware configuration and software technologies currently being used to implement the proposed design. The simulation results and the hardware experimental results of the three-phase battery energy storage system will be presented and discussed.
INTRODUCTION
Renewable energy sources are increasingly being exploited for electricity generation. In the case of wind
generation, the present installed capacity world-wide totals around 6000MW. In the UK, there are more than 50 wind farms with over 700 operational turbines, with total capacity of more than 300 MW installed [1]. This is only around 0.5% of the UK total electricity generation capacity, but it is likely that the proportion will continue to increase, because the resource is plentiful and the generation cost at around 3.5p/kWh is already competitive with gas. Denmark has at present a total installed wind generation capacity of more than 1500 MW, over 10 % of the total generation capacity, and has an ambitious target of 4000 MW offshore wind generation capacity by 2030 [2].
The utilization of wind energy for electricity generation has already been shown to be attractive in both environmental and economic terms, and is likely to extend offshore in the near future. The intermittent nature of wind means that power is not always available, and although accurate wind speed forecasting can improve the predictability of wind power from each generator, an energy storage system is usually required to maintain network stability and reliability. Other renewable energy sources with intermittent or cyclic behavior include photovoltaics (PV). PV generators have daily cycles where the theoretical maximum output is within a clearly defined envelope. However, there is considerable seasonal variation. The energy from day to day can vary by a factor of ten, and the actual power at any particular time of day may be much less than the theoretical maximum, and subject to abrupt changes. PV generators are still considered expensive for grid connected application, but may soon become economically attractive, especially when they have a secondary purpose as a building cladding material, or where the generated power partly matches a building load demand, e.g. for airconditioning. The economic exploitation of renewable energy sources such as wind, wave, and solar for electrical power generation can be limited by the variable and intermittent nature of the supply.
Even in the normal power system, the consumer expects an electricity supply which is available at all times and provides a tight control of voltage and frequency for all appliances. The fact that all consumers are free to alter their demand at any time, coupled with the inability to store AC power, creates the underlying power system control task. This is a factor associated with the following power system operation and control problems which all relate to uncertainty:
- It is impossible to precisely forecast demand even from hour to hour,
- Power system planning hinges on forecast annual peak growth that is not known until afterwards,
- It is not possible to predict the numerous disturbances that do occur,
- Generator availability can vary unpredictably from day to day,
- System configuration constantly alters with outages of lines, cables and substation equipment.
The present method of handling uncertainty is very costly because it requires redundancy of system equipment and operation of additional generation to allow for unexpected outages or higher than expected demand. The greatest difficulties are posed by the daily peak periods, when not only the highest demand occurs but also the fastest rates of load change. The largest demand forecasting errors are at peak periods, a serious problem as the power system becoming more susceptable to disturbances as loadings increase.
Power systems throughout the world have also been affected by growing financial stringencies (due to the worldwide economic downturn) as well as being constrained by the community's environmental expectations. Both these factors have impeded the establishment of new power stations and extra transmission lines which are needed to maintain the existing levels of power system reliability. The constant threat to the reliability of electricity is posed by the uncertainty associated with power system operation, namely, the continually varying loads and the numerous perturbations and disturbances to the power system, sometimes even lead to serious interruptions. As an example, the extensive system collapse which interrupted New York in 1977, was costed at $US350 million, 20% of the value of New York’s electrical network [3].
The conventional capital intensive approach for achieving reliability of supply had required long term financial decisions that are no longer possible with the uncertain rates of electricity demand growth coupled to the prevailing world economic uncertainties. The present situation undermines system security as well as reliability of supply to consumers so there is need for a new approach which would be able to reduce uncertainty. Although there are different aspects to power system uncertainty, the main problem is set by the difficulty of demand forecasting. Demand forecasting has two aspects, short term for day to day operation and the annual peak demand on which the major cost commitments, both capital and operational, are focussed in the continuing development of the power system. Daily peak periods, because of their short duration, are associated with only a small amount of the daily energy consumption, even on the day of annual peak demand.
This paper explores the ramifications of introducing fast responding battery energy storage systems into in the power systems containing renewable distributed energy resources which, with sufficient capacity and rating, could remove the uncertainty in the intermittent nature of the renewable energy resources as well as the uncertainty due to forecasting the annual peak demand. Energy storage can address these problems, by smoothing the net power supplied to the grid, or by enabling the energy to be stored and dispatched later (for example to supply peak demands), thus giving a higher value to the generated power.
They would also benefit the day to day operation by curtailing the fastest demand variations, particularly at the daily peak periods. The paper will also discuss the hardware configuration and software technologies currently being used to implement some of the above objectives, in particular, the fast control of active and reactive power demand and generation when a Battery Energy Storage System is connected to the grid.
The proposed three-phase multi-purpose Battery Energy Storage System will provide active and reactive power independent of the supply voltage with excellent power quality in terms of its waveform.
D. Sutanto.
Department of Electrical Engineering
The Hong Kong Polytechnic University
Hong Kong
E-mail: eesutant@polyu.edu.hk
No comments:
Post a Comment