With the increasing integration of power plants into the frequency-regulation markets, the importance of optimal trading has grown substantially. This paper conducts an in-depth analysis of their optimal trading behavior in sequential day-ahead, intraday, and frequency-regulation markets. We introduce a probabilistic multi-product optimization model, derived through a series of transformation techniques. Additionally, we present two reformulations that re-frame the problem as a mixed-integer linear programming problem with uncertain parameters. Various aspects of the model are thoroughly examined to observe the optimal multi-product trading behavior of hydro power plant assets, along with numerous case studies. Leveraging historical data from Nordic electricity markets, we construct realistic scenarios for the uncertain parameters. Furthermore, we then proposed an algorithm based on the No-U-Turn sampler to provide probability distribution functions of cleared prices in frequency-regulation and day-ahead markets. These distribution functions offer valuable statistical insights into temporal price risks for informed multi-product optimal-trading decisions.

The increasing importance of cross-border trade has brought the topic of allocating cross-border transmission capacity under the limelight. In this paper, we focus on two different transmission capacity calculation methods, namely, Available Transfer Capacity and Flow-based Capacity allocation, in the continuous intraday electricity market. The effect of these two methods is studied on the continuous intraday electricity market through an agent-based model comprising various types of trading agents. It also models market operator agent and transmission system operator agent inspired by the Single Intraday Coupling project in Europe. The price-volume decisions for the orders posted by the market participants are determined by two different strategies where one of them adapts to changing market conditions and new information while the other is naive. The model accounts for ramping constraints which enables the analysis of the trading behavior and interaction of agents across multiple delivery products simultaneously. A switch parameter is availed in the trading timeline for the thermal and storage agents to change from a less conservative approach of ignoring the ramping and charging/discharging rate constraints respectively to considering them. An interplay between different switch parameters and cross-border transmission capacity calculation methods is studied.

This thesis investigates the role of different short-term electricity market design aspects that can facilitate better coordination of resources within the power system. The work also emphasizes on better cross-border integration of the short-term markets to improve the market liquidity, competition, social welfare, and flexibility in the system, which is essential for facilitating the integration of renewable sources. Apart from the policy design issues, the thesis also focuses on developing several mathematical models to support algorithmic trading decisions in short-term markets with various sources of stochasticities. The models proposed in this thesis enable improved trading decisions closer to real-time both for the production and consumption portfolio. 

The work presented in this thesis encompasses the short-term electricity markets with a primary focus on intraday electricity markets. The contributions of this work are in two directions, one is from the perspective of the trader in the intraday market. In this direction of the work, the aim was to develop optimization models to support algorithmic decision-making for generation and consumption portfolios. To this end, multistage stochastic programming problems have been developed to model cross-border continuous intraday (CID) trading for a price-taking virtual power plant with hydropower, wind power, and thermal power assets. The order clearing in the CID market is enabled by the two presented models, namely the Immediate Order Clearing and the Partial Order Clearing models. Further advancements in the multistage model has been achieved by proposing a bilevel model to tackle the problem of unit commitment of thermal power plant in the continuous intraday market. The lower level of the model accounts for the continuous market clearing considering for the minimum generation level of the thermal power plant. In this model, the virtual power plant is able to post its own prices in the intraday market. The resulting multistage stochastic programming problem with integer variables is tackled by Stochastic Dual Dynamic integer Programming algorithm.

For the consumption portfolio, the participation of an electric vehicle aggregator in the intraday and balancing market is modeled as a multistage stochastic programming problem. Intraday markets allow the electric vehicle aggregator to trade furtherbased on their updated forecasts of consumption and price development in the market. The response of the electric vehicle aggregator to the prices in the market helps the power system tobetter manage the imbalances that might be injected by the intermittent generation sources. The algorithmic contribution in this work includes the deployment of randomized progressive hedging which was found to be faster than the conventional progressive hedging algorithm. Furthermore, the performance was accelerated by a parallel randomized progressive hedging algorithm and an asynchronous version of the randomized progressive hedging algorithm is leveraged to speed up the multistage model of electric vehicle aggregator trading.

The other direction of the work is from the point-of-view of policymakers where an open-source agent-based model is proposed to model the cross-border continuous intraday electricity market. In this model, the trading behavior of different agents, including renewable, thermal power producers, storage, and consumers, in a cross-border  Continuous Intraday market is simulated. The continuous market clearing is performed by a market operator agent. Two capacity calculation methods, Available Transfer Capacity, and Flow-based Market Coupling are availed to compute the cross-border transmission capacities. The agents are enabled to trade for multiple delivery products simultaneously in the CID market. Furthermore, the agents can choose between two different trading strategies, naive and modified trader adaptive aggressiveness, to decide the price-volume curves they post in the CID market. To simulate a realistic trading behavior, a user-defined parameter, switch, is introduced for the storage and thermal agents to allow them to choose the time-instant in the trading timeline when the traders can switch from trading towards increasing their profits to considering their ramping constraints in the CID market. The role of better forecasts of imbalance prices in deciding order prices and thereby the transaction prices in the CID market is also discussed.

Ultimately, a forward-looking mechanism is proposed for the system operator to dispatch generators to activate their up- and down-regulation bids on the basis of having lowest expected costs, taking into account both their production costs and potential deviations from nominated outputs. The proposed method has been mathematically applied to compare three imbalance pricing models.

I denna avhandling undersöks vilken roll som olika aspekter i utformningen av korttidselmarknader har när det kommer till att möjliggöra bättre koordination av resurser i elkraftsystem. Arbetet belyser även hur förbättrad integration av korttidsmarknader över nationsgränser ökar marknadslikviditeten, konkurrensen, det sociala välståndet samt flexibiliteten i systemet. Dessa aspekter är grundläggande för integrationen av förnybara energikällor. Utöver frågor relaterade till utformning av policys så fokuserar avhandlingen även på framtagandet av ett flertal matematiska modeller som ska stödja algoritmiska handelsbeslut på korttidsmarknader med ett flertal olika stokasticiteter. De modeller som föreslås i denna avhandling möjliggör förbättrade handelsbeslut närmre realtid för både produktions- och konsumtionsportföljer.  Arbetet som presenteras i denna avhandling omfattar korttidselmarknaderna med ett huvudsakligt fokus på intradagmarknaden. De vetenskapliga bidragen är av två inriktningar. I den ena inriktningen tas en deltagare på intradagmarknadens perspektiv. I denna del av arbetet var syftet att utveckla optimeringsmodeller för att stödja algoritmiskt beslutsfattande för produktions- och konsumtionsportfolios. I detta syfte utvecklades stokastiska flerstegsproblem för att modellera den kontinuerliga intradaghandeln för ett pristagande virtuellt kraftverk med tillgångar i form av vattenkraftverk, vindkraftverk och värmekraftverk. Budklareringen på den kontinuerliga intradagmarknaden behandlas genom de två presenterade modellerna ‘Direkt Budsklarering’ och ‘Partiell Budsklarering’. Ytterligare förbättringar av flerstegsmodellen har uppnåtts genom att föreslå en modell i två nivåer för att hantera driftplaneringsproblemet för värmekraftverk på den kontinuerliga intradagmarknaden. Modellens undre nivå svarar för den kontinuerliga marknadsklareringen och tar hänsyn till minsta möjliga produktionsnivå för värmekraftverket. I denna modell kan det virtuella kraftverket informera intradagmarknaden om sina egna priser. Det motsvarande stokastiska problemet med heltalsvariabler hanteras genom en stokastisk dual dynamisk heltalsprogrammeringsalgoritm. Avseende konsumtionsportföljer så modelleras en elbilsaggregators deltagande på intradag- och balanseringsmarknaden som ett stokastiskt flerstegsproblem. 

Intradagmarknader tillåter elbilsaggregatorn att handla baserat på uppdaterade konsumtionsprognoser samt prisutvecklingen på marknaden. Elbilsaggregatorns reaktion på marknadens priser bidrar till att hantera de obalanser som intermittenta förnybara energikällor kan skapa. Det algoritmiska bidraget från detta arbete inkluderar delandet av en randomiserad progressiv hedging vilken bevisades vara snabbare än konventionella progressiva hedgingalgoritmer. Vidare förbättrades prestandan genom en parallell randomiserad progressiva hedgingalgoritmer, och en asynkron version av den randomiserade progressiva hedgingalgoritmer nyttjas för att snabba upp multiskedesmodellen för en elbilsaggregators handel.

Den andra inriktningen på arbetet är från en beslutsfattares perspektiv, där en öppet tillgänglig agentbaserad modell har föreslagit för att modellera den gränsöverskridande kontinuerliga intradagmarknaden. I denna modell simuleras handelsbeteendet på en kontinuerlig gränsöverskridande intradagmarknaden för olika agenter, inklusive producenter av förnybar kraft och värmekraft, energilager och konsumenter. Den kontinuerliga marknadsklareringen genomförs av en marknadsoperatörsagent. Två olika metoder för beräkning av överföringskapacitet finns tillgängliga, dels ‘Tillgänglig Överföri-\\ngskapacitetsmetoden’ och dels ‘Flödesbaserad  Marknadskopplingsmetoden’. Agenterna har möjlighet att handla med ett flertal leveransprodukter samtidigt på den kontinuerliga intradagmarknaden. Vidare kan agenterna även välja mellan två olika handelsstrategier, antingen att använda en naiv och modifierad adapativ handelsagressivitet eller att bestämma de pris-volymkurvor som de skickar in till den kontinuerliga intradagmarknaden. För att simulera ett realistiskt handelsbeteende för agenter med energilager och termisk kraft har en användardefinierad parameter, switch, tagits fram. Denna parameter möjliggör att agenterna kan välja den tidpunkt där de går från inkomstökande handel till att ta hänsyn till sina rampbegränsningar på den kontinuerliga intradagmarknaden. Den påverkan förbättrad prognostisering av obalanspriser har på bestämmandet av budpriser och därmed transaktionspriset på den kontinuerliga intradagmarknaden diskuteras också.

Slutligen föreslås en framåtblickande mekanism som systemoperatörer kan nyttja för att beordra generatorer att aktivera sina upp- och nedregleringsbud baserat på lägsta förväntade kostnad. Mekanismen tar hänsyn till både generatorers produktionskostnader och potentiella avvikelser från planerad produktion. Metoden har applicerats matematiskt för att jämföra tre olika modeller för prissättning av obalanser.

The growing prominence of electric vehicle (EV) aggregators in the modern power system is drawing more attention towards modeling their behavior in the short-term electricity markets. The demand-side flexibility offered by the EVs can be leveraged to reduce their charging costs. In this paper, the participation of an EV aggregator in the intraday and balancing market is modeled as a multistage stochastic programming problem. The computational complexity introduced by the peculiarities of the intraday market is solved by a  progressive hedging algorithm (PHA), a scenario-based decomposition technique. A randomized scenario sampling approach is implemented to accelerate the PHA which is further improved with a parallel randomized PHA. Finally, an asynchronous version of the parallel randomized PHA is leveraged to speed up the multistage model of EV aggregator trading. We compare the computation time of the modified versions of the PHA algorithm with the conventional PHA for the proposed EV aggregator model. Furthermore, we also show the value of EV aggregator trading in the intraday and balancing markets by comparing its cost to baseline models. 

The growing prominence of electric vehicle (EV) aggregators in the modern power system is drawing more attention towards modeling their behavior in the short-term electricity markets. The demand-side flexibility offered by the EVs can be leveraged to reduce their charging costs. In this paper, the participation of an EV aggregator in the intraday and balancing market is modeled as a multistage stochastic programming problem. The computational complexity introduced by the peculiarities of the intraday market is solved by a progressive hedging algorithm (PHA), a scenario-based decomposition technique. A randomized scenario sampling approach is implemented to accelerate the PHA which is further improved with a parallel randomized PHA. Finally, an asynchronous version of the parallel randomized PHA is leveraged to speed up the multistage model of EV aggregator trading. We compare the computation time of the modified versions of the PHA algorithm with the conventional PHA for the proposed EV aggregator model. Furthermore, we also show the value of EV aggregator trading in the intraday and balancing markets by comparing its cost to baseline models.

The increasing importance of cross-border trade has brought the topic of allocating cross-border transmission capacity under the limelight. In this paper, we focus on two different transmission capacity calculation methods, namely, Available Transfer Capacity and Flow-based Capacity allocation, in the continuous intraday electricity market. The effect of these two methods is studied on the continuous intraday electricity market through an agent-based model. The model comprises market participants, including renewable producers, consumers, storage, and thermal power producers. The market operator agent is responsible for managing the shared order book and clearing orders while the transmission system operator agent calculates and updates the cross-border transmission capacities in the continuous intraday electricity market. The price-volume decisions for the orders posted by the market participants are determined by two different strategies where one of them adapts to changing market conditions and new information while the other is naive. The model accounts for ramping constraints which enables the analysis of the trading behavior and interaction of agents across multiple delivery products simultaneously. A switch parameter is availed in the trading timeline for the thermal and storage agents to change from a less conservative approach of ignoring the ramping and charging/discharging rate constraints respectively to considering them. An interplay between different switch parameters and cross-border transmission capacity calculation methods is studied.

The increasing variable renewable energy sourcesin the power system have led to rise in the trading volumesin the electricity markets. In this paper, we extend an existingopen source agent-based model for simulating the behavior andinteractions of the renewable, consumer, thermal, and marketoperator agents in the continuous intraday (CID) electricitymarket by introducing the energy storage agents. Furthermore,the limitation of the earlier model to account for the CID tradeof a single delivery product is relaxed by extending the modelto consider the simultaneous trading of all the possible deliveryproducts in a day. A realistic trading behavior is enabled byan user-defined parameter for the thermal and storage agentsto choose the switching point in the trading timeline when thestrategy of the trading agent navigates from increasing theirprofits by trading in the CID market to avoiding any imbalancesby considering their physical constraints. Comparative casestudies are performed to demonstrate the evolution of the tradingbehavior of the agents throughout the trading horizons formultiple delivery products with different switching instances.

The surge in intermittent renewable energy sources has led to larger volumes traded in the short-term electricity markets. As a result, the problem of optimally trading in the cross-border continuous intraday (CID) electricity market has come under the limelight. In this paper, we propose a multistage stochastic programming model to determine the participation of a virtual power plant (VPP), comprising wind power, hydropower, and thermal power portfolio, in a cross-border CID market. The VPP determines the order volumes to be submitted at the market price by assessing the available orders in other bidding areas. We also present an order clearing model that accounts for the cross-border transmission capacities along with the liquidity in the CID market. The risk-averse behavior of the VPP is captured by the nested conditional value at risk (CVaR) formulation. In order to demonstrate the functionality of the proposed model, several case studies are performed by constructing data models based on historical trading data obtained from the Nord Pool.

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 intraday electricity market provides the possibility to trade closer to real-time, aiming at assisting the market participants in improving their market position, based on their updated forecasts. The growing popularity of the intraday markets coupled with the single intraday coupling (SIDC) project in Europe drives certain trends and trading behaviours, that are analysed in this paper. Further, we present different approaches to transform the price data of the historical trades into price time series. The price time series obtained is then leveraged to derive insights on the correlation of the prices in the day-ahead and intraday market; wind power and hydropower generation, and demand time series.