Neural modelling of electricity prices quoted on the Day-Ahead Market of TGE S.A. shaped by environmental and economic factors

Abstract

The paper contains the results of research on the impact of the number of factors used to build the Day-Ahead Market model at Polish Power Exchange S.A. Five models with a different number of factors influencing the model were tested. To test the quality of models according to the adopted evaluation criteria, i.e., mean square error and the coefficient of determination for the weighted average prices sold in a given hour of the day, the influence of weather factors, socio-economic factors and energy demand were adopted. The results obtained from the analysis show a relatively high correctness of the simplest of the adopted models, which differs slightly from the best model.

Neural modeling of prices on the Day-Ahead Market at the Polish Power Exchange supported by an evolutionary algorithm and inspired by quantum computing

“…the data for the period between the second half of 2002 and June 30, 2019. The work here reported is a continuation of previous research on DAM modeling in Poland, see Ruciński (2017Ruciński ( , 2018Ruciński ( , 2022, Tchórzewski and Ruciński (2016, 2019.…”

“…The coefficient As a result of the investigations carried out, it turned out that the Perceptron ANN is relatively the best neural model for the DAM system, although the MSE error and determination coefficient values are close to those obtained for the Recursive ANN. During the six-month period adopted for further research (see Ruciński, 2022), the MSE error for the Perceptron network ranged from 0.006723 to 0.001065. The MSE error for the Recursive network, which ranged from 0.006723 to 0.00025243, was also relatively good, although slightly worse than that for Perceptron.…”

Hourly identification and simulation of the TGE S.A. Day-Ahead Market system

“…• models with continuous time and models with discrete and even pulse time, • models with lumped parameters and models with distributed parameters, • deterministic models and stochastic models, as well as a number of other kinds of models (Box and Jenkins, 1983;Chodakowska, Halicka, Kononiuk andNazarko, 2005, Ejdys, Halicka andGodlewska, 2015;Halicka, 2006;Marlȩga, 2022;Pop lawski and Weżgowiec, 2015;Ruciński, 2022;Tchórzewski and Marlȩga, 2019a).…”

“…At the beginning of the 21 st century, the interest in modeling of the Electricity Market System (EMS) increased significantly (see, in particular, Chodakowska et al, 2005;Wnukowska, 2005), including the Polish Electricity Exchange System (PEE) as a subsystem operating on TGE S.A. 1 (see Marlȩga, 2019a, 2021;Ruciński, 2022), due to the very dynamically changing international conditions, including structural changes caused by COVID, and then by the energy crisis brought about by the international conflicts over energy resources (see the Report of the President of the Energy Regulation Board, Prezes. .…”

“…In the study of the DAM system model, which is reported in the present paper, two methods were used, i.e. the arx parametric method (see Ejdys, Halicka and Godlewska, 2015;Marlȩga, 2022;Nazarko, 2018;Tchórzewski, 2013;Zimmer and Englot, 2005) and the Perceptron Artificial Neural Network (see Vargas, Pedrycz and Hemerly, 2019;Osowski, 2013;Ruciński, 2022;Tadeusiewicz, 1993). The results of the literature review in this area are included in the paper by Marlȩga (2021), which shows the originality of the conducted research against the background of the available world literature in this field.…”