Josip Vasilj

Day-ahead scheduling and real-time Economic MPC of CHP unit in Microgrid with Smart buildings

This paper presents a model for day-ahead scheduling of the combined heat production heat and electric energy production for a residential microgrid taking into account the economic factors in a liberalized electricity markets, the technical factors in the safety/quality of supply, and the consumer preferences. This day-ahead scheduling model is complemented with a real-time economic model predictive control (MPC) model for a subsequent control with respect to the outcomes of the day-ahead scheduling. This combined scheduling and economic MPC provides a general set-up capable of overcoming several major difficulties encountered with a typical scheduling + tracking MPC set-up, e.g., the problems of connecting the economic objectives with different temporal resolution and different requirements in terms of delivery.

PV power forecast error simulation model

Presence of solar energy driven power sources has been increasing with remarkable trends in recent years. Such growth is mostly related to photovoltaic generation and it is a direct consequence of significant fall in costs for this technology. Such generation, together with load and wind power, requires forecasting in day-ahead operation planning. Furthermore, in power system studies, it is often necessary to simulate performance of such forecasts. This paper presents models for PV power forecast error simulation based on the stochastic process simulation methods. Specificities of PV power forecast error, autocorrelation, correlation and dependence of standard deviation on forecasting horizon, can be reproduced with this model. Furthermore, two distinct models are presented in this paper. First model simulates and superposes forecast error on known PV production time series. Second model is a further improvement of the first model incorporating PV production simulation for purpose of applications in the planning phase when actual data on production is unavailable. Proposed models provide an efficient solution for studies requiring PV power forecast error time series.

Application of Gaussian copulas in describing station impinging lightning overvoltages

This paper presents an application of the bivariate Gaussian copulas in describing - by means of a statistical distribution function - station impinging lightning overvoltages, which could be subsequently employed in the statistical and semi-statistical methods of station insulation coordination. The proposed method further employs a state-of-the-art transmission line model for lightning-surge transient analysis, constructed within the EMTP software. It also makes use of the electrogeometric model of lightning attachment to transmission lines in order to estimate the shielding failure and backflashover rates. The proposed method efficiently generates random lightning currents from these non-normal and correlated bivariate statistical probability distributions and employs them in the Monte-Carlo simulation for the purpose of station equipment insulation coordination.

Valuation of contract between power supplier and electric vehicle owner

The role of EVs can range from that of passive loads with no difference compared to the traditional ones to active components highly beneficial to the modern power system. An essential part in changing EV storages from passive to active components is the contractual framework. Therefore, we present several different agreements that can be formed between a power supplier and an EV owner. These agreements establish the terms for utilization of storages when an EV is parked and connected to a household charger. We show that these benefits are significant and present a valuable economic opportunity for both EV owners and power suppliers.

Optimal reconfiguration of distribution network using cycle-break/genetic algorithm

This paper presents novel approach for optimal distribution network reconfiguration using the combination of cycle-break algorithm and genetic algorithms. Significant improvements are introduced in the phases of initial population generation as well as other general operations inside genetic algorithm. These improvements lead to better convergence rate and computational time reduction. Even though genetic algorithms are widely used, problems related to inapplicability for real-size are often present. These problems are related to the high individual rejection rate due to violation of system constraints and distribution network radial structure requirements. Utilization of combined cycle-break algorithm and genetic algorithm solves these issues and allow real-size network application. Acknowledging this fact, algorithm described in the paper is used to find optimal distribution network topology while fulfilling system constraints and maintaining radial network requirements in all solution steps. The proposed algorithm for optimal distribution network reconfiguration is tested on several standard IEEE test cases. Optimal distribution network reconfiguration can be found under minimum network loss or optimal network loading framework.

Energy and reserve co-optimization in power system with wind and PV power

Due to increase in renewable generation presence in power systems worldwide, current practices in day-ahead or hour-ahead operation planning require revision. Both wind and solar energy governed power sources require forecasting similar to load forecasts performed in traditional power systems. When such stochastic generation reaches significant amounts, traditional day-ahead or hour-ahead operation planning is not suitable, and a novel approach is required in order to ensure reliable and economically efficient generation scheduling. This paper presents an optimization procedure for energy and reserve co-optimization accounting for uncertainty in both wind and solar power forecasting. Problem formulation is based on two-stage stochastic programming where energy and capacity procurement is performed in the first stage, while balancing power or curtailments and load losses are performed in the second stage.

Wind power plant capacity allocation in areas with limited transmission capacity

This paper proposes a simulation framework for assessing the possibility of wind power plant integration into the existing transmission network in an optimal manner that takes into account the wind potential at each location scheduled for WPP construction. Wind power plant capacity allocation is analyzed from the static aspect considering maximum line loading under normal system operating conditions with WPP production modeled as stochastic variable. Primal-dual interior point method with a preprocessing phase which eliminates redundant constraints is employed to solve the optimization problem.