Assessment of the parameter importance

Based on the results of the uncertainty and sensitivity analyses, an inference about the importance of the examined parameters for modelling the outcomes of the Russian short-term wholesale energy market can be made. In the Table 5.14, the input parameters are listed according to the level of uncertainty and overall impact on the simulated prices. The impacts of parameter uncertainty and sensitivity are classified as

“Low” (L), “Medium-low” (ML), “Medium-high” (MH) and “High” (H). “Low”

uncertainty means that the parameter value does not significantly vary from the base value in the course of modelling, whereas “high” uncertainty means that the model parameter can deviate significantly from its base value. Similarly, “Low” impact means that the modelled market prices are little affected by changes of the model parameter. In contrast, “high” impact denotes that changes in the parameter in question may cause notable change of the modelled prices.

Table 5.14 Assessment of the model input parameters importance

Parameter Uncertainty level Impact on prices Overall parameter importance Jan Apr July Oct Jan Apr July Oct

GRES capacity L L ML L L L L MH Low

Operating reserves H H H H ML ML MH H Medium

Must-run CHP MH MH ML L H MH ML L High

Hydro production MH H ML L MH H H ML High

The findings suggest that the most critical model parameters that can strongly influence the model energy market price estimates are the must-run production of the CHP plants and the amounts of hydro generation. The overall impacts of these parameters are marked as “high” due to their large impacts on the model prices in most of the representative months. The results also suggest that the reserve capacity requirements

may also significantly contribute to the overall model uncertainty, especially during the mid-autumn periods. At the same time, the overall importance of the parameter of the availability of condensing power plants is defined as “low” due to particularly low impacts of the assumptions about parameter value on the model prices.

In addition to the above, the results of the uncertainty and sensitivity analysis of the model of the Russian wholesale energy market allow the following important implications on potential model improvements to be made:

• Modelling of the minimum electric output constraints of CHP on per plant of per region basis can increase the accuracy of the market simulations, especially for the winter and spring time periods.

• Operation of the water system in the first price sub-area of the market must be performed on a per plant or per region basis taking into account the forecasts of water inflows and reservoir levels. The impacts of spring flooding during April-July need to be considered in a more sophisticated manner.

• The amounts of tertiary operating reserves in the first price sub-area can play an essential role in modelling of the wholesale market price levels, especially during the summer and autumn seasons. Careful evaluation of the amounts of primary, secondary and tertiary reserves can contribute to reduction of uncertainty of the obtained energy price estimates.

6 Case study: application of the model to analysis of changes in the regulatory environment of the Russian wholesale energy market

This chapter focuses on practical application of the developed mathematical model of the Russian short-term electricity market. The use of the model is demonstrated through its application to analysis of changes in the regulatory environment of the short-term electricity market of Russia. The validity of the model for examining the impacts of regulation on the wholesale energy market outcomes in Russia is explored using an example of regulatory modifications of the design of the unit commitment procedure temporary adopted in the Russian wholesale electricity market at the end of 2015.

6.1

Validation of a simulation model is an important process, which helps to ensure that the model has a satisfactory level of accuracy within the required area of application.

Ideally, the model validation should be performed each time a model is applied for the examination of a specific question of interest and if the model is designed for several purposes, the validity of the model should be determined with respect to each purpose (Sargent, 2009). By performing the model validation, a confidence that the produced model estimates are correct and the model can operate reasonably well within its intended area of application is acquired.

The present chapter pursuit two main objectives. First, the usefulness of the developed mathematical model of the Russian short-term electricity market is demonstrated by an example of modelling the impacts of one of the recent modifications of the rules of the unit commitment procedure in the actual market on the day-ahead energy market prices and revenues of power producers. Second, using information about the anticipated effects of the proposed regulatory changes on the behavior of the actual short-term energy market outcomes, a conclusion about the validity of the developed model is made by comparing the directions of these changes with those produced by the model.

environment of the Russian wholesale energy market Following the description of the actual modifications to the UC procedure applied in the market, changes in the formulation of the thermal unit commitment problem in the model are introduced. The model price and revenue estimates obtained under the alternative UC problem formulation are compared with those obtained in the base case simulations and the difference between the original and modified model outcomes is used to evaluate the degree of impacts that the proposed regulatory changes could have on the actual market prices and profits of the electric producers in Russia. The directions of the resulting changes in the model price and producers’ revenues are compared with that expected in the actual market in result of the new regulations adoption and the validity of the developed model for studying the impacts of changes in the regulatory environment on functioning of the Russian wholesale electricity market is assessed.

This chapter is organized are follows. First, a description of the actual UC procedure modifications introduced by the Market Council at the end of 2015 is given and the reasons for these modifications are explained. After that, changes in the mathematical formulation of the thermal UC problem in the model required to reflect the alternative conditions of generators participation the actual UC procedure in the market are described. The results of the price and producers’ revenue simulations under alternative framework of the generators’ commitment procedure are then presented and the impacts of the examined UC rules modifications on the modelled market outcomes are analysed.

The chapter closes with conclusions about the validity of the model for analysing the impacts of changes in the actual procedure of the centralized generators’ commitment on the price outcomes of short-term energy market in Russia.