scholarly journals Capturing Electricity Market Dynamics in the Optimal Trading of Strategic Agents using Neural Network Constrained Optimization

2021 ◽  
Author(s):  
Mihály Dolányi ◽  
Kenneth Bruninx ◽  
Jean-François Toubeau ◽  
Erik Delarue
Keyword(s):  
Neural Network ◽  
Electricity Markets ◽  
Electricity Market ◽  
Mathematical Program ◽  
Market Dynamics ◽  
Equilibrium Constraints ◽  
Market Outcomes ◽  
Optimal Trading ◽  
Optimal Bidding ◽  
Upper Level

In competitive electricity markets the optimal trading problem of an electricity market agent is commonly formulated as a bi-level program, and solved as mathematical program with equilibrium constraints (MPEC). In this paper, an alternative paradigm, labeled as mathematical program with neural network constraint (MPNNC), is developed to incorporate complex market dynamics in the optimal bidding strategy. This method uses input-convex neural networks (ICNNs) to represent the mapping between the upper-level (agent) decisions and the lower-level (market) outcomes, i.e., to replace the lower-level problem by a neural network. In a comparative analysis, the optimal bidding problem of a load agent is formulated via the proposed MPNNC and via the classical bi-level programming method, and compared against each other.

Capturing Electricity Market Dynamics in the Optimal Trading of Strategic Agents using Neural Network Constrained Optimization

2021 ◽  
Author(s):  
Mihály Dolányi ◽  
Kenneth Bruninx ◽  
Jean-François Toubeau ◽  
Erik Delarue
Keyword(s):  
Neural Network ◽  
Electricity Markets ◽  
Electricity Market ◽  
Mathematical Program ◽  
Market Dynamics ◽  
Equilibrium Constraints ◽  
Market Outcomes ◽  
Optimal Trading ◽  
Optimal Bidding ◽  
Upper Level

In competitive electricity markets the optimal trading problem of an electricity market agent is commonly formulated as a bi-level program, and solved as mathematical program with equilibrium constraints (MPEC). In this paper, an alternative paradigm, labeled as mathematical program with neural network constraint (MPNNC), is developed to incorporate complex market dynamics in the optimal bidding strategy. This method uses input-convex neural networks (ICNNs) to represent the mapping between the upper-level (agent) decisions and the lower-level (market) outcomes, i.e., to replace the lower-level problem by a neural network. In a comparative analysis, the optimal bidding problem of a load agent is formulated via the proposed MPNNC and via the classical bi-level programming method, and compared against each other.

scholarly journals Capturing Electricity Market Dynamics in the Optimal Trading of Strategic Agents using Neural Network Constrained Optimization

2021 ◽  
Author(s):  
Mihály Dolányi ◽  
Kenneth Bruninx ◽  
Jean-François Toubeau ◽  
Erik Delarue
Keyword(s):  
Neural Network ◽  
Electricity Markets ◽  
Electricity Market ◽  
Mathematical Program ◽  
Market Dynamics ◽  
Equilibrium Constraints ◽  
Market Outcomes ◽  
Optimal Trading ◽  
Optimal Bidding ◽  
Upper Level

In competitive electricity markets the optimal trading problem of an electricity market agent is commonly formulated as a bi-level program, and solved as mathematical program with equilibrium constraints (MPEC). In this paper, an alternative paradigm, labeled as mathematical program with neural network constraint (MPNNC), is developed to incorporate complex market dynamics in the optimal bidding strategy. This method uses input-convex neural networks (ICNNs) to represent the mapping between the upper-level (agent) decisions and the lower-level (market) outcomes, i.e., to replace the lower-level problem by a neural network. In a comparative analysis, the optimal bidding problem of a load agent is formulated via the proposed MPNNC and via the classical bi-level programming method, and compared against each other.

Impacts of Emission Trading on Optimal Bidding Strategies of Generation Companies in Day-Ahead Electricity Markets

2014 ◽  
Vol 521 ◽  
pp. 476-479 ◽  
Author(s):  
Guo Zhong Liu
Keyword(s):  
Electricity Markets ◽  
Electricity Market ◽  
Emissions Trading ◽  
Emission Trading ◽  
Distribution Functions ◽  
Bidding Strategy ◽  
Chance Constrained Programming ◽  
Optimization Approach ◽  
Generation Company ◽  
Optimal Bidding

The impacts of Emission trading on building the optimal bidding strategy for a generation company participating in a day-ahead electricity market is investigated. The CO2 emission price in an emissions trading market is evaluated by using an optimization approach similar to the well-developed probabilistic production simulation method. Then upon the assumption that the probability distribution functions of rivals bidding are known, a stochastic optimization model for building the risk-constrained optimal bidding strategy for the generation company in the framework of the chance-constrained programming is presented. Finally, a numerical example is served for demonstrating the feasibility of the developed model and method, and the optimal bidding results are compared for the two situations with and without the CO2 emissions trading.

Emissions Trading, Electricity Restructuring, and Investment in Pollution Abatement

2010 ◽  
Vol 100 (3) ◽  
pp. 837-869 ◽  
Author(s):  
Meredith Fowlie
Keyword(s):  
Electricity Markets ◽  
Electricity Market ◽  
Emissions Trading ◽  
Pollution Abatement ◽  
State Level ◽  
Economic Regulation ◽  
Market Outcomes ◽  
Capital Intensive ◽  
Trading Program ◽  
Stationary Sources

This paper analyzes an emissions trading program that was introduced to reduce smog-causing pollution from large stationary sources. Using variation in state level electricity industry restructuring activity, I identify the effect of economic regulation on pollution permit market outcomes. There are two main findings. First, deregulated plants in restructured electricity markets were less likely to adopt more capital intensive environmental compliance options as compared to regulated or publicly owned plants. Second, as a consequence of heterogeneity in electricity market regulations, a larger share of the permitted pollution is being emitted in states where air quality problems tend to be more severe. (JEL L51, L94, L98, Q53, Q58)

scholarly journals Strategic generation investment using a stochastic rolling-horizon MPEC approach

2020 ◽  
Author(s):  
Thomas Kallabis ◽  
Steven A. Gabriel ◽  
Christoph Weber
Keyword(s):  
Imperfect Information ◽  
Bilevel Optimization ◽  
Rolling Horizon ◽  
Equilibrium Constraints ◽  
Market Outcomes ◽  
Producer Price ◽  
Electricity Spot Markets ◽  
Realistic Representation ◽  
Upper Level ◽  
Investment Process

AbstractInvestments in power generation assets are multi-year projects with high costs and multi-decade lifetimes. Since market circumstances can significantly change over time, investments into such assets are risky and require structured decision-support systems. Investment decisions and dispatch in electricity spot markets are connected, thus requiring anticipation of expected market outcomes. This strategic situation can be described as a bilevel optimization problem. At the upper level, an investor decides on investments while anticipating the market results. At the lower level, a market operator maximizes welfare given consumer demand and installed generation assets as well as producer price bids. In this paper, we formulate this problem as a mathematical program with equilibrium constraints (MPEC). We consider this model to include a dynamic, rolling-horizon optimization. This structure splits the investment process into multiple stages, allowing the modification of wait-and-see decisions. This is a realistic representation of actors making their decision under imperfect information and has the advantage of allowing the players to adjust their data in between rolls. This more closely models real-world decision-making and allows for learning and other feedback in between rolls. The rolling-horizon formulation also has the beneficial byproduct of computational advantage over a fixed-horizon stochastic optimization formulation since smaller problems are solved and we provide supporting numerical results to this point.

scholarly journals Day-Ahead Electricity Price And Spike Forecasting Using Machine Learning Techniques

2021 ◽  
Author(s):  
Harmanjot Singh Sandhu
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Electricity Markets ◽  
Electricity Market ◽  
Numerical Experiments ◽  
Status Report ◽  
Electricity Prices ◽  
Forecasting Accuracy ◽  
Using Data ◽  
The Impact

Various machine learning-based methods and techniques are developed for forecasting day-ahead electricity prices and spikes in deregulated electricity markets. The wholesale electricity market in the Province of Ontario, Canada, which is one of the most volatile electricity markets in the world, is utilized as the case market to test and apply the methods developed. Factors affecting electricity prices and spikes are identified by using literature review, correlation tests, and data mining techniques. Forecasted prices can be utilized by market participants in deregulated electricity markets, including generators, consumers, and market operators. A novel methodology is developed to forecast day-ahead electricity prices and spikes. Prices are predicted by a neural network called the base model first and the forecasted prices are classified into the normal and spike prices using a threshold calculated from the previous year’s prices. The base model is trained using information from similar days and similar price days for a selected number of training days. The spike prices are re-forecasted by another neural network. Three spike forecasting neural networks are created to test the impact of input features. The overall forecasting is obtained by combining the results from the base model and a spike forecaster. Extensive numerical experiments are carried out using data from the Ontario electricity market, showing significant improvements in the forecasting accuracy in terms of various error measures. The performance of the methodology developed is further enhanced by improving the base model and one of the spike forecasters. The base model is improved by using multi-set canonical correlation analysis (MCCA), a popular technique used in data fusion, to select the optimal numbers of training days, similar days, and similar price days and by numerical experiments to determine the optimal number of neurons in the hidden layer. The spike forecaster is enhanced by having additional inputs including the predicted supply cushion, mined from information publicly available from the Ontario electricity market’s day-ahead System Status Report. The enhanced models are employed to conduct numerical experiments using data from the Ontario electricity market, which demonstrate significant improvements for forecasting accuracy.

scholarly journals Day-Ahead Electricity Price And Spike Forecasting Using Machine Learning Techniques

2021 ◽  
Author(s):  
Harmanjot Singh Sandhu
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Electricity Markets ◽  
Electricity Market ◽  
Numerical Experiments ◽  
Status Report ◽  
Electricity Prices ◽  
Forecasting Accuracy ◽  
Using Data ◽  
The Impact

Various machine learning-based methods and techniques are developed for forecasting day-ahead electricity prices and spikes in deregulated electricity markets. The wholesale electricity market in the Province of Ontario, Canada, which is one of the most volatile electricity markets in the world, is utilized as the case market to test and apply the methods developed. Factors affecting electricity prices and spikes are identified by using literature review, correlation tests, and data mining techniques. Forecasted prices can be utilized by market participants in deregulated electricity markets, including generators, consumers, and market operators. A novel methodology is developed to forecast day-ahead electricity prices and spikes. Prices are predicted by a neural network called the base model first and the forecasted prices are classified into the normal and spike prices using a threshold calculated from the previous year’s prices. The base model is trained using information from similar days and similar price days for a selected number of training days. The spike prices are re-forecasted by another neural network. Three spike forecasting neural networks are created to test the impact of input features. The overall forecasting is obtained by combining the results from the base model and a spike forecaster. Extensive numerical experiments are carried out using data from the Ontario electricity market, showing significant improvements in the forecasting accuracy in terms of various error measures. The performance of the methodology developed is further enhanced by improving the base model and one of the spike forecasters. The base model is improved by using multi-set canonical correlation analysis (MCCA), a popular technique used in data fusion, to select the optimal numbers of training days, similar days, and similar price days and by numerical experiments to determine the optimal number of neurons in the hidden layer. The spike forecaster is enhanced by having additional inputs including the predicted supply cushion, mined from information publicly available from the Ontario electricity market’s day-ahead System Status Report. The enhanced models are employed to conduct numerical experiments using data from the Ontario electricity market, which demonstrate significant improvements for forecasting accuracy.

scholarly journals A New Hybrid Wavelet-Neural Network Approach for Forecasting Electricity

2020 ◽  
Vol 24 (1) ◽  
Author(s):  
Heni BOUBAKER ◽  
SOUHIR BEN AMOR ◽  
Hichem Rezgui
Keyword(s):  
Neural Network ◽  
Electricity Markets ◽  
Electricity Market ◽  
Hybrid Approach ◽  
Wavelet Neural Network ◽  
Neural Network Approach ◽  
Predicting Performance ◽  
Empirical Wavelet Transform ◽  
Using Data ◽  
The Individual

This study investigates the performance of a novel neural network technique in the problem of price forecasting. To improve the prediction accuracy using each model’s unique features, this research proposes a hybrid approach that combines the -factor GARMA process, empirical wavelet transform and the local linear wavelet neural network (LLWNN) methods, to form the GARMA-WLLWNN process. In order to verify the validity of the model and the algorithm, the performance of the proposed model is evaluated using data from Polish electricity markets, and it is compared with the dual generalized long memory -factor GARMA-G-GARCH model and the individual WLLWNN. The empirical results demonstrated the proposed hybrid model can achieve a better predicting performance and prove that is the most suitable electricity market forecasting technique.

scholarly journals Modeling the Strategic Bidding of the Producers in Competitive Electricity Markets with the Watkins' Q (lambda) Reinforcement Learning

2006 ◽  
Vol 6 (2) ◽  
Author(s):  
Yuchao Ma ◽  
Ettore F. Bompard ◽  
Roberto Napoli ◽  
Jiang Chuanwen
Keyword(s):  
Decision Making ◽  
Reinforcement Learning ◽  
Electricity Markets ◽  
Complex Problem ◽  
Test System ◽  
Market Outcomes ◽  
Strategic Bidding ◽  
Optimal Bidding ◽  
Competitive Electricity Markets ◽  
Multi Agents

Competition has been introduced in the last decade into the electricity markets and is presently underway in many countries. A centralized approach for the dispatching of the generation units has been substituted by a market approach based on the biddings submitted by the supply side and, eventually, by the demand side. Each producer is a player in the market acting to maximize its utility. The decision making process of the producers and their interactions in the market are a typical complex problem that is difficult to be modeled explicitly, and can be studied with a multi agents approach. This paper proposes a model able to capture the decision making approach of the producers in submitting strategic biddings to the market and simulate the market outcomes resulting from those interactions. The model is based on the Watkins' Q (lambda) Reinforcement Learning and takes into account the network constraints that may pose considerable limitations to the electricity markets. The model can be used to define the optimal bidding strategy for each producer and, as well, to find the market equilibrium and assessing the market performances. The model proposed is applied to a standard IEEE 14-bus test system to illustrate its effectiveness.

Strategic Bidding and Risk Assessment Using Genetic Algorithm in Electricity Markets

2009 ◽  
Vol 10 (5) ◽  
Author(s):  
Arvind Kumar Jain ◽  
S.C. Srivastava
Keyword(s):  
Genetic Algorithm ◽  
Electricity Markets ◽  
Electricity Market ◽  
Value At Risk ◽  
Financial Risk ◽  
Congestion Management ◽  
Bidding Strategy ◽  
Strategic Bidding ◽  
Optimal Bidding ◽  
Sensitivity Factors

In an electricity market, suppliers are more concerned with maximizing their profit and minimizing the financial risk, which can be achieved through strategic bidding. In this paper, Equal Incremental Cost Criteria (EICC) has been used for developing the optimal bidding strategy. The rival's bidding behavior has been formulated using a stochastic optimization model. Genetic Algorithm (GA), along with ac sensitivity factors, has been used to decide the optimal bidding strategy including congestion management to maximize the profit of the suppliers, considering single sided as well as double sided bidding. Both pure as well as probabilistic strategies have been simulated. Results with Sequential Quadratic Programming (SQP), a classical optimization method, and dc sensitivity factors have also been obtained to compare and establish the effectiveness of proposed method. Value at Risk (VaR) has been calculated as a measure of financial risk.

Sign in / Sign up

Export Citation Format

Share Document