Rung-Fang Chang

Security-constrained optimal power flow by mixed-integer genetic algorithm with arithmetic operators

This paper presents an efficient real-coded mixed-integer genetic algorithm (MIGA) for solving non-convex optimal power flow (OPF) problems with considering transmission security and bus voltage constraints for practical application. In the MIGA method, the individual is the real-coded representation that contains a mixture of continuous and discrete control variables, and two arithmetic crossover and mutation schemes are proposed to deal with continuous/discrete control variables, respectively. The objective of OPF is defined that not only to minimize total generation cost but also to enhance transmission security, to reduce transmission loss, to improve the bus voltage profile under normal or contingent states. Moreover, the valve-point loading effect of thermal units should be taken into consideration. The effectiveness of the proposed method is demonstrated for a 26-bus and the IEEE 57-bus systems, and it is compared with the evolutionary programming (EP) in terms of solution quality and evolutionary computing efficiency. The experimental results show that the MIGA-based OPF method is superior to the EP

Transmission System Loadability Enhancement Study by Ordinal Optimization Method

Due to the growth of electricity demands and transactions in power markets, existing power networks need to be enhanced in order to increase their loadability. The problem of determining the best locations for network reinforcement can be formulated as a mixed discrete-continuous nonlinear optimization problem (MDCP). The complexity of the problem makes extensive simulations necessary and the computational requirement is high. An ordinal optimization (OO) technique is proposed in this paper to solve the MDCP involving two types of flexible ac transmission systems (FACTS) devices, namely static var compensator (SVC) and thyristor controlled series compensator (TCSC), for system loadability enhancement. In this approach, crude models are proposed to cope with the complexity of the problem and speed up the simulations with high alignment confidence. Test results based on a practical power system confirm that the proposed models permit the use of OO-based approach for finding good enough solutions with less computational efforts.

Prediction of frequency response after generator outage using regression tree

The ability of a regression tree method to properly interpolate among recorded data to give an estimate of the frequency decline following a generator outage is examined in this letter. The proposed method is a nonparametric technique that can select those system characteristics and their interactions that are most important in determining the relation between the generation/load imbalance and the frequency decline. The information obtained from the proposed method can be used online for scheduling fast-acting reserve or load shedding for severe generator outage incidents.

Feeder reconfiguration for accommodating distributed generations interconnection

Previous studies have suggested that, with proper planning, distributed generation (DG) could provide system operation and planning benefits such as reduced losses and enhanced reliability. However, the allowable DG interconnection capacity at each site of the existing network is often restricted by fault current level, voltage variation and power flow constraints. An improved feeder reconfiguration technique is proposed in this paper to maximize DG penetration in the distribution network. Tie switch locations in the network are determined based on crude load flow model and minimum voltage deviation criterion in the proposed particle swarm optimization (PSO) solution procedure to improve computational efficiency. A simplified distribution network is used to test the performance of the proposed method and compare with a basic PSO based method. Numerical results indicate that the proposed method can efficiently provide suitable feeder structure for loss reduction and accommodating higher DG integrations.

Distribution feeder bus load estimations by measurement matching

The effectiveness of the entire distribution management system (DMS) relies on the data accuracy of load modeling/load estimation (LM/LE) in the distribution network. In this paper, feeder bus load estimations results obtained from a load scaling (LC) and backward/forward sweep (BFS) procedure are presented. The proposed technique makes use of the actual measurements (AM) data available from distribution automation (DA) and Advanced Metering Infrastructure (AMI) to adjust priori bus load data considered as pseudo measurements (PM). Two measurement matching procedures used to match the calculated data with the different types of measurement data at different feeder locations are described. Test results demonstrate that the proposed technique could lead to a load flow solution that matches AM and provides final load scaling values for PM.

Modified particle swarm optimization for solving distribution feeder reconfiguration problem with distributed generation

An ordinal optimization (OO) technique in conjunction with a particle swarm optimization (PSO) method is presented in this paper to solve a feeder reconfiguration problem that aims to maximize distributed generation (DG) penetration in an existing distribution network. A set of good enough solutions obtained from crude models used in OO approach are taken as the initial population of particles for PSO solution in order to speed up solution efficiency in subsequent iterations. A simplified Taipower distribution network is used in this study to test the performance of the proposed method. Test results indicate that the proposed method can be used to obtain a good solution for maximizing DG integration in the distribution network.

Optimal PID controller for high-speed rail pantograph system with notch filter

The pantograph device must exert enough upward force to maintain sliding contact with the catenary at all times to avoid loss of contact due to excessive transient motion. This paper presented a global-oriented particle swarm optimization (GPSO) technique with Cauchy mutation for tuning the optimal control gains of PID controller in a high-speed rail pantograph device with notch filter. The proposed method used the global-oriented mechanism to enhance the computation efficiency of traditional PSO algorithm and used a Cauchy operator for escaping local minima. To verify the advantage of the proposed GPSO-PID controller, a refined integrated time weighted squared error (RITSE) performance criterion was employed and a pantograph system with notch filter was tested. The results showed that the proposed method was indeed adaptive and robust in quick search of the PID controller parameters. The GPSO-PID controller can decrease rapidly the contact force variation between the pantograph and the catenary.

Modified DC Load Flow Based Optimal DG Installations with Fast Feeder Reconfiguration

With proper installation of distributed generation (DG), the performances of distribution grid, such as system loss, power quality and reliability, can be improved effectively. In this paper, the particle swarm optimization (PSO) based solution procedure, including the simplified load flow and the proposed modified DC load flow models, is used to solve for the optimal DG installation problem with the proposed fast feeder reconfiguration procedure. The computing efficiency is increased with the load flow fast estimated by the simplified load flow model and the tie switches to open to make a radial feeder structure selected based on the modified DC load flow model. Numerical results indicate that the proposed method can efficiently make a suitable feeder structure with a small system loss and a big DG penetration level.