Weerakorn Ongsakul

Optimal Power Flow by Improved Evolutionary Programming

This article proposes an improved evolutionary programming (IEP) for solving optimal power flow (OPF) with nonsmooth and nonconvex generator fuel cost curves. Initially, the whole population is divided into multiple subpopulations, which are used to perform the parallel search in divided solution space. IEP includes Gaussian and Cauchy mutation operators in different subpopulations to enhance the search diversity, selection operators with probabilistic updating strategy to avoid entrapping in local optimum, and reassignment operator for every subpopulation to exchange search information. The proposed IEP was tested on the IEEE 30 bus system with three different types of generator fuel cost curves. It is shown that IEP total generator fuel cost is less expensive than those of evolutionary programming, tabu search, hybrid tabu search and simulated annealing, and improved tabu search, leading to substantial generator fuel cost savings. Moreover, IEP can easily facilitate parallel implementation to reduce the computing time without sacrificing the quality of solution.

Optimal power flow with multi-type of FACTS devices by hybrid TS/SA approach

In this paper, a hybrid tabu search and simulated annealing (TS/SA) approach is proposed to minimize the generator fuel cost in optimal power flow (OPF) control with multi-type of flexible AC transmission systems (FACTS) devices. The problem is decomposed into the optimal setting of FACTS parameters subproblem that is searched by the hybrid TS/SA approach and the OPF with fixed FACTS parameters subproblem that is solved by the quadratic programming (QP). Four types of FACTS devices are used: thyristor-controlled series capacitor (TCSC), thyristor-controlled phase shifter (TCPS), unified power flow controller (UPFC), and static VAr compensator (SVC). Test results on the modified IEEE 30 bus system with multi-type of FACTS devices indicate that the proposed hybrid TS/SA approach can obtain better solutions and require less CPU times than genetic algorithm, SA, or TS alone. Using multi-type of FACTS devices results in less total generator fuel cost than using individual FACTS device.

Parallel micro genetic algorithm for constrained economic dispatch

This paper proposes a parallel micro genetic algorithm (PMGA) for solving ramp rate constrained economic dispatch (ED) problems for generating units with nonmonotonically and monotonically increasing incremental cost (IC) functions. The developed PMGA algorithm is implemented on the 32-processor Beowulf cluster with Ethernet switches network on the systems with the number of generating units ranging from 10 to 80 over the entire dispatch periods. The PMGA algorithm carefully schedules its processors, computational loads, and synchronization overhead for the best performance. The speedup upper bounds and the synchronization overheads on the Beowulf cluster are shown on different system sizes and different migration frequencies. The proposed PMGA is shown to be viable to the online implementation of the constrained ED due to substantial generator fuel cost savings and high speedup upper bounds.

A multi-agent based power system restoration approach in distributed smart grid

The objective of this paper is to design, develop and implement a multi-agent system (MAS) that provides intelligent and enables real-time management to a smart grid located at a distribution level (so called distributed smart grid). The MAS application development is discussed concerning suitable agent development framework, agent specification, agent architecture, and implementation of MAS. The paper illustrates MAS application in power systems. As faults and outages are inevitable and likely to occur in distribution systems, an efficient and fast switching operation scheme is required to detect the fault location, isolate the fault, and restore power to de-energized areas. The system under study consists of both physical (microgrid) and cyber elements (MAS). Finally, the simulation result indicates that the developed MAS for power system restoration applications can provide an effective and timely solution to manage microgrid given the existence of fault in the system.

Artificial Intelligence in Power System Optimization

With the considerable increase of AI applications, AI is being increasingly used to solve optimization problems in engineering. In the past two decades, the applications of artificial intelligence in power systems have attracted much research. This book covers the current level of applications of artificial intelligence to the optimization problems in power systems. This book serves as a textbook for graduate students in electric power system management and is also be useful for those who are interested in using artificial intelligence in power system optimization.

An efficient load-balancing processor scheduling algorithm for parallelization of Gauss-Seidel type algorithms

Abstract The paper extends our earlier results on the parallelization of Gauss-Seidel (G-S) algorithms for power flow analysis. In the earlier paper, we formulate the parallelizing process as a basic coloring problem, which satisfies the constraint that no directly connected vertices have the same color, without worrying about the constraint on the number of available processors. In this paper, this extra constraint is considered. A heuristic approach is developed to maximize the processor efficiency under the number of processor constraint. The idea is to fully utilize the processor resource, to balance the computational load, and to maximize the use of newly computed data for faster convergence. Some examples and test results are described in this paper.

An adaptive SOR algorithm and its parallel implementation for power system applications

In our earlier papers,we investigated the parallelization and implementation of Gauss-Seidel (G-S) and Successive Overrelaxation (SOR) power flow analysis on shared memory, (SM) and distributed (DM) machines. For the SOR case, constant acceleration factors obtained from experiments are used to speedup convergence. In this paper, we introduce a new adaptive nonlinear SOR (ANSOR) algorithm which uses an approximated optimal acceleration factor obtained during the iteration process. The algorithm is shown to be faster due to the significant reduction in the number of iterations, and to converge robustly under heavily-loaded conditions on large power systems. We also implement parallel and sequential versions of our ANSOR algorithm on the nCUBE2 machine, and show that our algorithm is competitive with the fast decoupled load flow (FDLF) algorithm. Moreover, the portability of the parallel ANSOR code is demonstrated by porting the code to the Intel iPSC/860 hypercube and the Paragon mesh MIMD machines. However, our new algorithm is not a panacea for all problems, as we demonstrate with an example from transient stability analysis.<<ETX>>

An efficient task allocation algorithm and its use to parallelize irregular Gauss-Seidel type algorithms

The parallelization and implementation of Gauss-Seidel power flow analysis have been investigated. The desired properties to maximize the speedup, such as minimum communication overhead and balanced computational load, have been described. In this paper, we investigate a two-stage parallelization scheme to achieve the desired properties for distributed memory machines. In the first stage, we introduce a new efficient heuristic clustering algorithm which reduces the communication time and balances the computational load. In the second stage, we devise a coloring algorithm whose purpose is to minimize the synchronization overhead and coordinate the information exchange among processors. It is shown that the parallelization scheme effectively increases the speedup and the associated upper bound of the Gauss-Seidel algorithm on the nCUBE2 machine.<<ETX>>

Managing the bottlenecks of a parallel Gauss-Seidel algorithm for power flow analysis

The parallelization and implementations of Gauss-Seidel (G-S) algorithms for power flow analysis have been investigated on a Sequent Balance shared memory (SM) machine. In this paper, the authors generalize the idea to more general computer architectures and demonstrate how to effectively increase the speedup upper bounds of G-S algorithms by properly managing the bottlenecks.<<ETX>>

A new relaxation algorithm for power flow analysis

The possibility of revitalizing the relaxation algorithm and improving its reliability and speed at the same time is investigated. Some comparison issues such as how to compare parallel algorithms using sequential computers are discussed. The argument is illustrated by analyzing a new algorithm based on polar coordinates for power flow analysis. This relaxation nonlinear iterative scheme is described in detail and tested on a few scenario systems such as the IEEE and EPRI test systems. It is compared with classical Gauss and Gauss-Seidel methods.<<ETX>>