Demand response and renewable energy integration in the chilean electricity market

Abstract

In the Chilean electricity market, renewable energy sources had a participation rate of 43% of total electric generation. In this context, the target for renewable energy participation levels is 60% for 2035 during 2017. Recent studies suggest that this target is achievable. However, there are some concerns about the impact that renewable energy imposes on the grid, which could lead to the inefficient incorporation of these technologies. One reason for this potential inefficiency is that electricity demand is not directly exposed to the spot price, so it is not very responsive to market conditions. The current literature on demand-side flexibility in the Chilean electricity system is limited and there are no studies that analyse the effects of the introduction of demand flexibility on the efficient use of renewable energy technologies in the Chilean electricity market. This thesis develops a deterministic Demand Response Unit Commitment (DR-UC) model. This model is formulated as a mixed-integer nonlinear program (MINLP) using a two-part quadratic disutility function. This formulation allows us to assess the impact of demand-side flexibility when demand can be shifted or adjusted in an electricity system using five different generation technologies (solar, wind, hydro reservoirs, run-of-the-river and thermal power plants). Unlike most of the literature, which considers that only a fraction of the total demand is flexible, I place no restriction regarding the amount of demand flexibility. I simulated the operation of the Chilean electricity system for different scenarios in the year 2035. I use the results of these simulations to assess the benefits provided by demand flexibility in the Chilean electricity market and the impact on renewable energy generation and on the operation of other technologies. The Demand Response benefits found in this study corresponds to a lower bound of its possible benefits to the system. The results of this thesis suggest that demand-side flexibility integration increases the efficient use of renewable energy investment for the year 2035. This is due to the reduction of both renewable energy curtailment and cycling operation of low-cost thermal power plants, as well as lower spot price variability. When incorporating demand-side flexibility, diesel generation is reduced by up to 42% due to the shift in consumption from peak to non-peak hours. An important result of the study is to show that a Demand Response would increase CO2 emissions because coal power plants operate at higher capacity. The software code that implements the model and assesses its impact is offered as free software, and is one of the major contributions of this work. It is available in the Git repository of the project.