Anatolijs Mahnitko

Modeling of the generating company behavior in energy and reserve market

In the market conditions generating companies put more attention to the question of energy and reserve distribution or purchasing. Each generating company tries to reduce its costs of energy production for increasing its profit from participation in the market. Thus, the question of energy and reserve distribution between generating stations and purchased amount is one of the key aspects of generating company behavior. In article the algorithm of generating company costs minimization, considering spot market and bilateral trading, is proposed. As criteria of optimization task cost minimization function is used. As variables in the model are used: energy, reserve, participation of the power plant in the market (both for reserve providing and for energy production). The algorithm operability was approved on example.

A model for maximizing the profit of a HPP cascade considering hydraulic link via reservoirs

At the conditions of market relationships and economic and financial independence of power company (GENCO), these are guided by the principle of achieving maximum efficiency in their actions. For a hypothetical cascade of hydropower plants (HPPs) consisting of three HPPs, a problem formulation aimed at optimum loading of each individual HPP is used, ensuring maximum profit for the whole cascade from the sales of the generated electric power at the day-ahead market. The structural diagram of a problem solution algorithm is presented, which uses the method of statistical modeling. The results of the modeling of cascade operating regime are illustrated graphically. The results have been used to calculate the summary profit of the HPP cascade. The structural diagram of the algorithm can be basically used for an arbitrary mix of HPPs in the cascade, with a weekly regulation cycle.

Modelling of wind, hydroelectric and thermal power plant coordinated dispatch in Latvian power system

Liberalization and restructuring of power sector changes the task of power system control substantially. Efficient control of operating regimes becomes especially important in competitive markets. This paper provides a formulation of optimal control task, describes a simplified model for simulation of four different energy sources, consider a solution for optimization task in the numerical example and evaluates the impact of key parameters for selection of optimal regime. For determination of optimal solution, the following methods of mathematical programming are applied: dynamic programming - for efficient use of hydroelectric resources, linear programming - for optimization of other power plants and import. Utilization of mentioned methods guarantees the acquisition of global optimum.

An analytical approach for calculations of power system generation adequacy indices

Generation adequacy indices (GAI) calculations are performed not only for electrical power system development problems but also for on-line control problems. Liberalization of power and energy market along with increased focus on renewable energy sources, which is motivated by both ecological and economical incentives, has led to comprehensive changes in power systems. Practically, the main mathematical method for calculating GAI of an electrical grid with complex structure is Monte-Carlo simulation (MCS) technique. It can provide sufficiently accurate results, but the simulation consumes really huge amount of computational time, especially for large power systems. Generation capacity that has to be installed in a power system for meeting reliability criteria depends not only on system demand but also on generation mix. The problem at hand is to improve computational efficiency of the GAI calculating technique without decrease in calculations accuracy. In order to solve the problem, this paper proposes an analytical method for GAI calculations, that is referred to as supplied demand algorithm (SDA). The paper compares the MCS technique and the SDA. The analyzed methods have been developed using the C# and VBA programming language. Input data for the methods includes generation mix data, generation and load probability indices. The comparison results allow to select the proper method for the GAI calculations and show the main factors that affect the resulting GAI.

Algorithmization in the task of optimization of cascade hydro power plants

Our work focuses on the problem of optimizing a cascade of three small hydro power plants. The optimization is based on deciding how much electricity to produce dependent on the day-ahead hourly electricity prices. The problem is reduced to controlling the water volumes in the cascades reservoirs. We examine two approaches — linear programming based on Matlabs Simplex method (the Linprog function) and a Dynamic programming method. Linear programming is used for the limiting case when electricity prices exhibit daily cyclicality and Dynamic programming works for a finite time period. Two numerical examples are considered.

Modification of supplied demand algorithm for generation adequacy indices calculations

In order to determine the indicators of the adequacy of meshed united electric power systems (EPSs), an algorithm of accelerated calculations is proposed, which is based on probabilistic equivalenting of intersystem power flows at the conditions of probabilistic demand and supply from adjacent EPSs. The conditions are considered and the fluctuant character is pointed out of the convergence of the iteration process of the formation of the resulting indicators of the nodal power imbalances. The recurrent relation is reflected algorithmically. In order to improve the convergence of the iteration process, a modification of the guaranteed demand algorithm is proposed, which takes into account the fluctuation character of the convergence by means of parabolic interpolation of the mathematical expectation of the power flow. The calculation procedure is illustrated on the basis of a three-node closed circuit of a united EPS. The algorithm has been appraised on the basis of a 42-node circuit of a meshed EPS. Ways for further refinement of the algorithm are proposed.