Capacity Remuneration in the Nordic Electricity Markets

Sammanfattning

The electricity consumption is expected to increase in Sweden and Nordic countries in the coming years, which means we need to invest in new power plants to avoid adequacy problems. This project investigates six case studies to see the effect of line capacity, flexible demand, and four scenarios that may happen in Sweden in the near future, 2035 and 2045, according to the Svenska kraftnät, and to see which type of generation is profitable to invest in. The mixed complementary method is used to maximize social welfare and also investigate if the proposed model can deal with large-scale problems. The results indicate that without line capacity limitations, all the investments can happen in a single region. However, when line capacity is limited, nearly all regions contribute in the new investments in electricity production. This is due to the influence of the limited line capacity, which cause investments in power generation from technologies with low variable and investment costs. Additionally, peak power technologies generate power in response to demand fluctuations, which are attributed to the ramp rates of these power technologies. The results also indicate that we have invested more in renewable energy, which means that Sweden and Nordic countries need to continue with strategic reserves. This is due to the dependability of some renewable energy sources on the weather. Such countries can better control variations in the supply and demand of energy by increasing their reserves, especially during times of unexpectedly high demand or supply disruptions. The deterministic result demonstrate that as demand increases in one scenario investment in generation rises and it becomes profitable to invest in the most suitable power plant technology. Meanwhile, the stochastic results indicate that investment is profitable across all types of power generation technologies in a suitable biding area. This insight can help us avoid adequacy problems in the future. The results indicate that the proposed approach can be adopted in large-scale problems.