Search

Even though weather forecasting techniques have been constantly improved, wind power prediction is still a hard task. An improved statistical description of forecasting characteristics can quantify the probability of a certain forecast error. In order to deal with imperfect wind power prediction, system operators have extra cost as a result of increasing reserve level. Application of storage system composed of batteries is prior option to minimize the consequence of the forecast errors. In this paper, the statistical distribution of forecast errors is used to generate forecast uncertainty. An algorithm is proposed to calculate the optimal capacity and operation strategy of a battery system taking into account the short-term power exchange and the expected imbalance penalties of a wind farm, in order to reduce the cost as a result of forecast uncertainty. [ABSTRACT FROM AUTHOR]

Renewable resources are becoming more and more obtainable and affordable due to the development of technology and the enactment of government policies. Electric utilities are required to deliver reliable power to customers and must operate utility grids within strict voltage limits. As renewable energy becomes a larger player amongst the resources supplying energy to these grids, issues begin arising due to the intermittent nature of these resources. The output from solar power aligns reasonably well daytime consumption on the electricity grid, reducing the need for new coal power stations. However, high penetration photovoltaic (PV), can lead to voltage instability due to intermittencies related to cloud cover. If the PV power is injected into a power system directly on a large scale, it may produce issues related to dispatchability, reliability and stability. It is desirable to select a smoothing storage algorithm that would filter out the highest rate transitions, but would still be fast enough to avoid significant lag with respect to current power production. For traditional testing, a moving average algorithm was used. For comparison, the author has tested two other algorithms. The results are being compared, showing that the dual moving average smoothing algorithm has merits in improving smoothness. [ABSTRACT FROM PUBLISHER]

Hybrid electric vehicles, featuring both a gasoline engine and a battery storage system became commercially available from two major auto manufacturers this year. While these hybrids offer better fuel economy and lower emissions than conventional cars, their battery packs are relatively small and are charged by the onboard gasoline engine-generator. In an effort to achieve efficiency and emissions levels closer to those of a true electric vehicle, EPRI is developing a grid-connected hybrid that operates the majority of the time in an all-electric mode, using its small gasoline engine only to extend driving range or to provide extra power. The development initiative-being pursued in alliance with General Electric, General Motors, Ford Motor, EPRI member utilities, and other key technology stakeholders-is at first focused on producing midsize hybrid buses and delivery vans and demonstrating their potential for cost savings to fleet vehicle operators. Expanded in the past year to put more emphasis on systems and components development, the initiative has as its ultimate goal the commercialization of grid-connected hybrid buses, vans, trucks, and automobiles [ABSTRACT FROM PUBLISHER]