S. Surender Reddy

Realistic and Transparent Optimum Scheduling Strategy for Hybrid Power System

This paper addresses the transparent and realistic optimum day-ahead (DA) scheduling for a hybrid power system by explicitly considering the uncertainties. The basic components of the hybrid power system include conventional thermal generators, wind farm, and solar photovoltaic (PV) modules. A set of batteries is available for energy storage and/or discharge. The most critical problem in operating a wind farm or solar PV module is that these renewable energy resources cannot be dispatched in the same manner as conventional plants, because they involve climatic factors such as wind velocity and solar irradiation. This paper proposes the optimal scheduling strategy taking into account the impact of uncertainties in wind, solar PV, and load forecasts, and provides the best-fit DA schedule by minimizing both DA and real-time adjustment costs including the revenue from renewable energy certificates. This strategy consists of a genetic algorithm (GA)-based scheduling and a two-point estimate-based probabilistic real-time optimal power flow. The simulation for the IEEE 30-bus system with the GA and two-point estimate method, and the GA and Monte Carlo simulation have been obtained to test the effectiveness of the proposed scheduling strategy.

Real-Time Economic Dispatch Considering Renewable Power Generation Variability and Uncertainty Over Scheduling Period

Real-time economic dispatch (RTED) is performed every 5-15 min with the static snapshot forecast data. During the period between two consecutive schedules, generators participate in managing power imbalance, based on participation factors (PFs) from previous economic dispatch (ED). In modern power systems with considerable renewable energy sources that have high variability, this conventional approach may not adequately accommodate the economic implication of the said variability. This paper proposes the evaluation of “best-fit” PFs by taking into account the minute-to-minute variability of solar, wind, and load demand, for a scheduling period. Since “best-fit” PFs are evaluated only once, i.e., at the start of scheduling interval, the dimensionality of optimization problem remains the same as that of conventional approach. The proposed approach is suggested for sequential and dynamic variants. Results for two test systems have been obtained to verify the benefit of the proposed approach.

Short term electrical load forecasting using back propagation neural networks

This paper presents a new approach for short term electrical load forecasting (STLF) using artificial neural networks (ANN), and examines the feasibility of various mathematical models for STLF. To make these mathematical models to yield satisfactory and acceptable results, various system models are formulated considering various combination of parameters like base load component, day of the week, load inertia, short term trends, autocorrelation, length of the past data, etc. Various modifications of Back Propagation Algorithm (BPA) have been proposed, to explore the ideal combination that suit the forecasting need of large utilities like regional electricity grids. Further, the load dynamics are extensively studied to identify the parameters for system modeling.

Differential evolution-based efficient multi-objective optimal power flow

This paper proposes a novel-efficient evolutionary-based multi-objective optimization (MOO) approaches for solving the optimal power flow (OPF) problem using the concept of incremental load flow model based on sensitivities and some heuristics. This paper is useful in robust decision-making for the system operator. The main disadvantage of meta-heuristic-based MOO approach is computationally burdensome. The motivation of this paper is to overcome this drawback. By using the proposed efficient MOO approach, the number of load flows to be performed is reduced substantially, resulting to the solution speed up. Here, three objective functions, i.e., generator fuel cost minimization, loss minimization, and L index minimization are considered. The proposed approach can effectively handle the complex non-linearities, discontinuities, discrete variables, and multiple objectives. The potential and suitability of the proposed efficient MOO approach is tested on the IEEE 30 bus system. The results obtained with the proposed efficient MOO approach are also compared with the meta-heuristic-based non-dominated sorting genetic algorithm-2 (NSGA-II) technique. In this paper, the proposed efficient MOO approach is implemented using the differential evolutionary (DE) algorithm. However, it is a generic one and can be implemented with any type of evolutionary algorithm.

Application of swarm intelligent techniques with mixed variables to solve optimal power flow problems

This paper proposes a new swarm based evolutionary algorithm called LBEST PSO with dynamically varying sub-swarms (LPSO DVS). Swarm based algorithms are meta-heuristic search methods whose mechanics are inspired by the collaborative behaviour of biological populations. The performance of four swarm based algorithms, i.e., particle swarm optimisation (PSO), fuzzy adaptive particle swarm optimisation (FAPSO), fitness distance ratio particle swarm optimisation (FDRPSO) and LPSO DVS are also compared with genetic algorithm (GA) and improved GA when applied to the power system optimal power flow (OPF) problem. OPF optimises the power system operating objective function, while satisfying the set of system operating constraints. The objective functions considered in this OPF problem are fuel cost (FC) minimisation, voltage stability enhancement index (VSEI) minimisation, transmission loss minimisation (LM) and voltage deviation (VD) minimisation. Simulation results for the IEEE 30 bus system are presented and the comparison is made among the numerical results obtained using the different evolutionary algorithms.

Stochastic Voltage/Var control with load variation

Renewable Energy sources are the requirement of the present power system, not only as replenish source but also as distributed generation source. This paper develops a stochastic optimization for Voltage/Var for Optimal Power Flow (OPF) involving load variation. The Renewable Energy Resources (RERs) also invite some imputes along with it such as stochastic behavior. The planning and operation of certain system is also a major challenge for the industry. So, OPF for such system is intriguing problem. One of the important challenge i.e., Voltage/ VAR control is a prime source of complexity and reliability. Therefore, it is fundamental requirement of all the utility companies. There is a need for robust and efficient Voltage VAR optimization technique to meet the peak demand and reduction of the losses. The voltages beyond limit may damage costly sub-station devices and equipment at consumer end as well. Especially, RERs introduces more disturbance and some of the RERs are not capable enough to meet the VAR demand. Hence, there is a strong need for Voltage/ VAR control in RER environment. This paper aims at development of OPF for Voltage VAR control involving RERs using the best available technology. The developed optimization scheme with RERs is tested on 24 bus system with Load Variation into the system.

Review of optimization techniques for Renewable Energy Resources

This paper presents a review of optimization techniques for Renewable Energy Resources (RERs) optimization framework. Review of probabilistic power flow, voltage/VAR control, modeling of voltage VAR problem has been presented. Various objective functions and formulations for Voltage/VAR problem has been presented in this paper. Summary of different OPF methods and their formulations for voltage / VAR problem has been presented. Various examples of stochastic programming using Recourse model, Simple Average Approximations, Chance Constrained Programming and chance constraint are presented in this paper.

Optimal power flow using clustered adaptive teaching learning-based optimisation

In this paper, optimal power flow (OPF) with non-convex and non-smooth generator cost characteristics is presented using clustered adaptive teaching learning-based optimisation (CATLBO) algorithm. The proposed OPF formulation includes active and reactive power constraints; prohibited zones, and valve point loading (VPL) effects of generators. In the problem formulation, transformer tap settings and reactive power compensating devices settings are also considered as control variables. OPF is a complicated optimisation problem, hence there is a need to solve this problem with an accurate algorithm. In the proposed CATLBO algorithm, the class is divided into different sections, and allot different teacher to every section depending on the performance of that particular section. This sectioning of the class makes the proposed technique more robust and less prone to trapping in local optima. The OPF solution is obtained by considering generator fuel cost, transmission loss and voltage stability index as objective functions. The effectiveness of proposed CATLBO algorithm is validated on IEEE 30 bus test system, and the simulation results obtained with CATLBO algorithm are compared with other optimisation techniques presented in the literature.

Combined Economic and Emission Dispatch using Radial Basis Function

This paper proposes an Artificial Neural Network approach for solving the Combined Economic and Emission Dispatch (CEED) problem using Radial Basic Function (RBF) based neural network. The goal of CEED is to minimize both the operating fuel cost and emission level simultaneously, while satisfying load demand and operational constraints. This multi-objective CEED problem is converted into a single objective function using a modified price penalty factor approach. In this paper, centres which are chosen randomly from input space are chosen such that they are fairly far apart from each other, and covering the whole of input space. Here, centres and weights memorization is done which resulted in improving RBF Network convergence and computation time. The proposed approach is implemented on two test systems and the obtained results are compared with conventional technique.

Optimal location of FACTS for ATC enhancement

The objective of this paper is to enhance the Available Transfer Capability (ATC) from source area to sink area in deregulated power system using Continuous Power Flow (CPF) method during normal and contingency cases with optimal location and control parameter of Flexible AC Transmission System (FACTS) devices such as Thyristor Controlled Series Capacitor (TCSC) or Static VAR Compensator (SVC) on IEEE 24 bus reliability test systems. Real parameter Genetic Algorithm (RGA) is used to determine optimal location and control parameter of TCSC or SVC. ATC is dependent on many factors, such as the base case of system operation, system operation limits, network configuration, specification of contingencies, etc. FACTS technology has a severe impact to the transmission system utilization with regards to those constraints on ATC. Therefore, maximum use of existing transmission assets will be more profitable for Transmission System Operators (TSO), and customers will receive better services with reduced prices.