The stochastic nature of renewable energy sources has increased the need for intraday trading in electricity markets. Intradaymarkets provide the possibility to the market participants to modify their market positions based on their updated forecasts. In this paper, we propose a multistage stochastic programming approach to model the trading of a Virtual Power Plant (VPP), comprising thermal, wind and hydro power plants, in the Continuous Intraday (CID) electricity market. The order clearing in the CID market is enabled by the two presented models, namely the Immediate Order Clearing (IOC) and the Partial Order Clearing (POC). We tackle the proposed problem with a modified version of Stochastic Dual Dynamic Programming (SDDP) algorithm. The functionality of our model is demonstrated by performing illustrative and large scale case studies and comparing the performance with a benchmark model.

The integration of continuously varying and not easily predictable wind power generation is affecting the stability of the power system and leads to increasing demand for balancing services. In this study, a short-term operation model of a district heating system is proposed to optimally schedule the production of both heat and power in a system with high wind power penetration. The application of the model in a case study system shows the increased flexibility offered by the coordination of power generation, consumption and heat storage units which are available in district heating systems.

Simulations of production costs, flows, and prices are crucial inputs to generation and transmission planning studies. To calculate average system performance for many alternatives over long time periods, it is necessary to simulate large numbers of hourly combinations of renewable production and loads across large regions. As this is usually impractical for full network representations of such systems, aggregation of buses and lines is desirable. We propose an improved aggregation method for creating multi-area representations of power systems that yields more accurate estimates of the quantities required by planners. The method is based on partitioning the original large system into smaller areas and making a reduced equivalent for each area. The partitioning is based on available transfer capability (ATC) between each pair of network buses. Because ATC depends on net load conditions, separate partitions are defined for subsets of similar load and wind conditions, significantly enhancing the accuracy of optimal power flow solutions. We test the method on the IEEE 118-bus test system and the Polish 3120-bus system considering 150 load/wind scenarios, comparing the results to those of admittance-based partitioning methods. Accuracy is improved with only a negligible increase in simulation time.

Many countries are planning for a large-scale expansion of wind power. This development will have a significant impact on power system operation and economics. One of the challenges is that the difficulty to forecast wind power generation will increase the need for real-time balancing. This paper presents a study of how the impact of wind power forecast errors can be reduced by changes in the market design. The study is based on the conditions in the Nordic electricity market. A characteristic of this market is that there is a large share of flexible hydro generation; hence, ramp and unit commitment constraints rarely constrain dispatch. The need for regulation during real-time is provided in a voluntary real-time balancing market, where players can be compensated for their redispatch costs. Case studies are presented which show that a shift from day-ahead to intraday trading and increased demand response can improve the performance when the share of wind power is increasing.