• Energy Research

Conventional models to support policymaking for the energy sector have been largely based on deterministic or static settings that focus on planning welfare- maximising investment pathways. But, in a liberalised market, since investments are made by competitive, profit-maximising companies, the increased intervention of government policy in the trading arrangements creates uncertain responses to incentives. Industry may perceive policy risks to be high, and major companies may choose to act more cautiously than governments expect. This presents a modelling challenge, and we propose an extension to the use of real options in this context. We model several features of the low-carbon investment context, viz., irreversibility, delay, and competition, which impinge upon the radical policy imperatives for structural change in electricity markets to meet ambitious sustainability targets.

With the rise of distributed energy resources and the increasing activation of flexibility resources by Distribution Systems Operators (DSOs), the Transmission System Operators (TSOs) need to co-ordinate their actions with those of the DSOs. This research uses a look-ahead multi-interval (LA-MI) framework for analyzing this coordination and explores two formulations. Firstly in the exogenous DSO model, a mixedinteger linear program is developed to reflect the pragmatic approach in many real situations whereby the TSO can only anticipate statistically the actions of the DSO. In the embedded DSO model, as a comparator, we propose a new organizational setup for the TSO-DSO operational coordination mechanism. In the resulting bilevel decomposition, a new method to calculate Benders cuts is developed and tested on a modified IEEE 118-bus test system as a transmission network and two modified IEEE 33-bus test systems as distribution networks. The benefits of the LAMI coordination framework are substantial in comparison with the current Look-Ahead Single-Interval (LA-SI) coordination framework widely used in Europe.

This paper applies a multi-factor, stochastic latent moment model to predicting the imbalance volumes in the Austrian zone of the German/Austrian electricity market. This provides a density forecast whose shape is determined by the flexible skew-t distribution, the first three moments of which are estimated as linear functions of lagged imbalance and forecast errors for load, wind and solar production. The evaluation of this density predictor is compared to an expected value obtained from OLS regression model, using the same regressors, through an out-of-sample backtest of a flexible generator seeking to optimize its imbalance positions on the intraday market. This research contributes to forecasting methodology and imbalance prediction, and most significantly it provides a case study in the evaluation of density forecasts through decision-making performance. The main finding is that the use of the density forecasts substantially increased trading profitability and reduced risk compared to the more conventional use of mean value regressions.

Motivated by the events following a natural experiment in 2015, when the market rules for electricity spot trading were changed in Britain, we analyse the operational effects of market participants responding to price incentives for spillage and shortage positions in a single price, real-time market. We develop an analytical model for optimal real-time decisions by generators and speculators based upon forecasts of the conditional distribution of the total system imbalance between instantaneous supply and demand. From this, we examine the effects of time delays in information transparency for the consequent statistical arbitrage positions. We backtested this model empirically to the Austrian system imbalance settlements process within the German/Austrian integrated market. Results suggest that permitting additional intraday flexibility from a physical generator or a non-physical trader can be beneficial for the agents themselves, the system operator and market efficiency.

This paper provides the first EU wide analysis of the variation in Capacity Remuneration Requirements throughout Europe which aim to resolve the “missing money” problems in various member states. The findings of this analysis point to an asymmetric investment case for gas-fired peaking power plants throughout the EU. Under the assumptions of the European Commission Reference Scenario, pan-European power optimisation and investment models are specified for 2030. The results show that future investment in gas generators will depend on the availability of capacity payments. Capacity remuneration mechanisms can provide this “missing money,” but we show that capacity remuneration requirements vary considerably across countries. We consider and model the impacts of country specific climate policy targets, sovereign risk, capital allowances, corporate taxes and future gas network tariffs on investor returns and therefore remuneration requirements. In the context of harmonised energy trading, this raises questions of how generation adequacy should be achieved, particularly in the context of higher penetrations of renewables.

Increasing variable renewable power generation (e.g., wind) is expected to reduce wholesale electricity prices by virtue of its low marginal production cost. This merit-order effect of renewables displacing incumbent conventional (e.g., gas) generation forms the theoretical underpinning for investment decisions and policy in the power industry. This paper uses a game-theoretic market model to investigate how intermittently available wind generation affects electricity prices in the presence of forward markets, which are widely used by power companies to hedge against revenue variability ahead of near-real-time spot trading. We find that in addition to the established merit-order effect, renewable generation affects power prices through forward-market hedging. This forward effect reinforces the merit-order effect in reducing prices for moderate amounts of wind generation capacity but mitigates or even reverses it for higher capacities. For moderate wind capacity, uncertainty over its output increases hedging, and these higher forward sales lead to lower prices. For higher capacities, however, wind variability conversely causes power producers to behave less aggressively in forward trading for fear of unfavorable spot-market positions. The lower sales counteract the merit-order effect, and prices may then paradoxically increase with wind capacity despite its lower production cost. We confirm the potential for such reversals in a numerical study, suggesting new empirical questions while providing potential explanations for previously contradictory observed effects of market fundamentals. We conclude that considering the conventional merit-order effect alone is insufficient for evaluating the price impacts of variable renewable generation in the presence of forward markets. This paper was accepted by Vishal Gaur, operations management.

Forecasting quantile and value-at-risk levels for commodity prices is methodologically challenging because of the distinctive stochastic properties of the price density functions, volatility clustering and the importance of exogenous factors. Despite this, accurate risk measures have considerable value in trading and risk management with the topic being actively researched for better techniques. We approach the problem by using a multifactor, dynamic, quantile regression formulation, extended to include GARCH properties, and applied to both in-sample estimation and out-of-sample forecasting of traded electricity prices. This captures the specification effects of mean reversion, spikes, time varying volatility and demonstrates how the prices of gas, coal and carbon, forecasts of demand and reserve margin in addition to price volatility influence the electricity price quantiles. We show how the price coefficients for these factors vary substantially across the quantiles and offer a new, useful synthesis of GARCH effects within quantile regression. We also show that a linear quantile regression model outperforms skewed GARCH-t and CAViaR models, as specified on the shocks to conditional expectations, regarding the accuracy of out-of-sample forecasts of value-at-risk.