An Integrated Approach for Value-Oriented Energy Forecasting and Data-Driven Decision-Making Application to Renewable Energy Trading

Carrière, Thomas; Kariniotakis, Georges

doi:10.1109/tsg.2019.2914379

Cited by 52 publications

(19 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In [33], the framework put forward in [16] is extended by proposing linear decision rules to improve both the forecasting and trading performance of a WPP participating in a DA market. For a similar case study with photovoltaic (PV) plants, [34] describes an ERM formulation based on neural networks. Both of these works deal with variations of the newsvendor problem and proposed solutions do not guarantee feasibility of decisions for more complex problems.…”

Section: B Literature Reviewmentioning

confidence: 99%

“…Reduced modeling effort [28], [30], [31], [37] * --[29] * , † † - [33], [34] † -This work * , † † case studies of increasing complexity related to renewable trading. First, we examine trading in a DA market under different pricing mechanisms and propose strategies that balance trading cost and predictive accuracy.…”

Section: Multiple Types Of Uncertaintymentioning

confidence: 99%

See 1 more Smart Citation

Prescriptive Trees for Integrated Forecasting and Optimization Applied in Trading of Renewable Energy

Stratigakos

Camal

Michiorri

et al. 2022

IEEE Trans. Power Syst.

Self Cite

View full text Add to dashboard Cite

Deriving decisions from data typically involves a sequential process with two components, forecasting and optimization. Forecasting models learn by minimizing a loss function that stands as a proxy for task-specific costs (e.g., trading, scheduling) without considering the downstream optimization, which in practice creates a performance bottleneck and obscures the impact of data on decisions. This work proposes a single data-driven module that leverages the structure of the optimization component and directly learns a policy conditioned on explanatory data. For this purpose, we describe an algorithm to train ensembles of decision trees by directly minimizing taskspecific costs, and prescribe decisions via a weighted Sample Average Approximation of the original problem. We then develop a generic framework to assess the impact of explanatory data on prescriptive performance. To illustrate the efficacy of the proposed modeling approach, we consider two case studies related to trading renewable energy. First, we examine trading in a day-ahead market and propose strategies that balance optimal trading decisions and predictive accuracy. Next, we append a storage device and co-optimize the day-ahead offers and the operational policy, based on a tractable approximation using the linear decision rule approach. The empirical results demonstrate improved prescriptive performance compared to solutions derived under the standard stochastic optimization framework. Further, we provide valuable insights on how explanatory data impact optimization performance and how this impact evolves under different market designs.

show abstract

Section: B Literature Reviewmentioning

confidence: 99%

Section: Multiple Types Of Uncertaintymentioning

confidence: 99%

Prescriptive Trees for Integrated Forecasting and Optimization Applied in Trading of Renewable Energy

Stratigakos

Camal

Michiorri

et al. 2022

IEEE Trans. Power Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Training is performed without considering the subsequent optimization problem, which arises later on the model chain. Two valueoriented forecasting approaches are proposed in [20] for the case of a photovoltaic (PV) generator participating in the DA market. In the first, the individual energy and price forecasting models are tuned based on the decision costs, whereas in the second, a direct policy selection algorithm is applied, mapping the input data to decisions.…”

Section: Introductionmentioning

confidence: 99%

A Value-Oriented Price Forecasting Approach to Optimize Trading of Renewable Generation

Stratigakos

Michiorri

2021

2021 IEEE Madrid PowerTech

View full text Add to dashboard Cite

The participation of renewable generators in electricity markets involves employing a number of forecasting and decision-making tools. The standard approach consists in forecasting power output and market quantities, and then inputting the results into an optimization problem to derive optimal decisions. Typically, forecasting models are trained to optimize accuracy without considering the subsequent decisionmaking process. In this paper, we consider training forecasting models with a value-oriented approach that aims to minimize the suboptimality of decisions induced by a set of predicted inputs. We consider a risk-aware renewable generator participating in a day-ahead market subject to imbalance costs, and train ensembles of decision trees to forecast the imbalance penalty by directly minimizing trading costs for the provided strategy. The results indicate that our innovative approach leads to improved trading performance, compared to the standard method in which forecasting models are trained to minimize prediction errors.Index Terms-Decision trees, electricity markets, energy trading, value-oriented forecasting, alternative loss functions.

show abstract

“…In this context, the PV power output forecast becomes very important. With the same approach, but for a different purpose, the energy production forecast can be used to engage in energy trading (Carriere & Kariniotakis, 2019). The implementation of a forecasting system can reduce uncertainty over the energy price, leading to economic benefits (Alessandrini et al, 2014;Barthelmie et al, 2008;Kraas et al, 2013).…”

Section: Introductionmentioning

confidence: 99%

Solar Energy Production Forecast Using Standard Recurrent Neural Networks, Long Short-Term Memory, and Gated Recurrent Unit

Buturache

Stancu

2021

View full text Add to dashboard Cite

Solar radiation is among the renewable resources on which modern society relies to partially replace the existing fossil fuel-based energy resources. Awareness of how the energy is produced must complement awareness of how it is consumed. In the economic context, the gains derive from predictability across the entire supply chain. This paper represents a compressive study on how standard recurrent neural networks, long short-term memory, and gated recurrent units can be used to forecast power production of photovoltaic (PV) systems. This approach can be used for other use cases in solar or even wind power prediction since it provides solid fundamentals for working with weather data and recurrent artificial neural networks, being the core of any smart grid management system. Few studies have explored how these models should be implemented, and even fewer have compared the outcomes of different model types. The data used consist of weather and power production data with a one-hour resolution. The data were further pre-processed to unveil the maximum information. The most effective model parameters were selected to make the forecast. Solar energy plays a key role among other renewable energy sources in the European Union’s climate action and the European Green Deal. Under these initiatives, important regulations are implemented and financial resources made available for those who possess the capabilities required to solve the open points. The much-needed predictability that gives the flexibility and robustness needed for deploying and adopting more renewable technologies can be ensured by utilizing a neural-based predictive approach.

show abstract

An Integrated Approach for Value-Oriented Energy Forecasting and Data-Driven Decision-Making Application to Renewable Energy Trading

Cited by 52 publications

References 26 publications

Prescriptive Trees for Integrated Forecasting and Optimization Applied in Trading of Renewable Energy

Prescriptive Trees for Integrated Forecasting and Optimization Applied in Trading of Renewable Energy

A Value-Oriented Price Forecasting Approach to Optimize Trading of Renewable Generation

Solar Energy Production Forecast Using Standard Recurrent Neural Networks, Long Short-Term Memory, and Gated Recurrent Unit

Contact Info

Product

Resources

About