Optimal operation of concentrated solar power generation in electricity markets

Abstract

The growing concern regarding environmental issues has made it necessary for the high penetration of renewable energy sources (RES) in energy systems. A concentrated solar power (CSP) generator is a renewable energy-based unit equipped with thermal energy storage. The use of such generation has been widely advocated as a promising alternative to conventional fossil-fuel based generation units. The associated thermal energy storage has equipped CSP generation with flexibility, which makes CSP generation one of the most important technologies for driving the transition of power systems to high-renewable-penetration systems. To promote the development of CSP generation, one of the most important things is to increase the payoff from the investment. Moreover, with growing capacity, the participation of CSP generation in electricity markets may significantly affect the market equilibrium and profit distribution. There appear to be many questions to study about these impacts. To this end, this thesis aims to propose a framework for the modelling of strategic behaviors of CSP generation in electricity markets. The impact of strategic CSP generation on market equilibrium and profit distribution is investigated. Alongside grid level renewable-based generation another class of renewable-based sources of energy are so-called prosumers who also affect market outcomes in a more widely distributed way. This is also investigated in the thesis but most emphasis is on CSP generation. Firstly, the optimal participation of a small scale CSP generator in electricity markets is considered. A novel two-stage stochastic risk-constrained model is proposed for the optimal bidding schedule of a CSP generator in day-ahead (DA) energy markets and auxiliary markets. Given the inherent variability of solar irradiance and market prices, there can be inevitable predictive error. As a result, it is of great importance to optimally re-adjust the output of a CSP generator during RT operation. To this end, a model-predictive-control (MPC) based optimization model is developed for a CSP generator to reschedule its RT operational point to minimize the operational costs against predictive error. Secondly, with growing capacity of a CSP generator, the market power of a CSP generator will be accumulated to a point where it can affect the market clearing process. It is necessary to take this factor into consideration when a CSP generator bids into electricity markets. To this end, we develop a novel three-stage stochastic bi-level model for the optimal participation of a CSP generator in energy markets and auxiliary markets. By using the proposed model, the impact of the strategic behavior of a CSP generator on the equilibrium of electricity markets is investigated. Moreover, we further relax the assumption that other participants are non-strategic by using an equilibrium-problem-with-equilibrium-constraints method. That is, other participants are assumed to be strategic and self-interested, i.e., try to maximize their own profit considering the behavior of rival participants. Lastly, more and more consumers nowadays are equipped with photovoltaic (PV) and smart loads. This kind of consumer is named prosumer. A bi-level model is proposed for the participation of the aggregator of large-scale prosumers in joint energy and secondary reserve markets.