Chan-Nan Lu

Neural network based short term load forecasting

The artificial neural network (ANN) technique for short-term load forecasting (STLF) has been proposed previously. In order to evaluate ANNs as a viable technique for STLF, one has to evaluate the performance of ANN methodology for practical considerations of STLF problems. The authors make an attempt to address these issues. The results of a study to investigate whether the ANN model is system dependent, and/or case dependent, are presented. Data from two utilities are used in modeling and forecasting. In addition, the effectiveness of a next 24 h ANN model in predicting 24 h load profile at one time was compared with the traditional next 1 h ANN model. >

Distribution system state estimation

A three-phase distribution system state estimation algorithm is proposed in this paper. The normal equation method is used to compute the real-time states of distribution systems modeled by their actual a-b-c phases. A current based formulation is introduced and compared with other formulations. Observability analysis for the proposed distribution system state estimation is discussed. Test results indicate that the normal equation method is applicable to the distribution system state estimation and the current based rectangular form formulation is suitable for this application. >

Stochastic Analyses of Electric Vehicle Charging Impacts on Distribution Network

A stochastic modeling and simulation technique for analyzing impacts of electric vehicles charging demands on distribution network is proposed in this paper. Different from the previous deterministic approaches, the feeder daily load models, electric vehicle start charging time, and battery state of charge used in the impact study are derived from actual measurements and survey data. Distribution operation security risk information, such as over-current and under-voltage, is obtained from three-phase distribution load flow studies that use stochastic parameters drawn from Roulette wheel selection. Voltage and congestion impact indicators are defined and a comparison of the deterministic and stochastic analytical approaches in providing information required in distribution network reinforcement planning is presented. Numerical results illustrate the capability of the proposed stochastic models in reflecting system losses and security impacts due to electric vehicle integrations. The effectiveness of a controlled charging algorithm aimed at relieving the system operation problem is also presented.

Interconnected Network State Estimation Using Randomly Delayed Measurements

In an interconnected power system, the network security and power transfer capability analyses require a complete network base case solution. With the advent of communications among operations control center computers, real-time telemetered data can be exchanged for complete network modeling. Measurement time skew is a consideration when combining large area data received via a data communication network. In order to accommodate the effects of randomly varying arrival of measurement data, this paper presents an implementation of a stochastic extended Kalman filter (EKF) algorithm, which provides optimal estimates of interconnected network states for systems in which some or all measurements are delayed. The method relies on the delay statistics of exchanged data arrival at the control center and the delay is assumed to have binary statistics, i.e., either the measurements arrive in time or they are delayed by one time sample. Performance of the proposed method is compared with that of the standard weighted least square (WLS) technique.

Two-point estimate method for quantifying transfer capability uncertainty

A two-point estimate method is proposed in this paper to assess the power transfer capability uncertainty. This paper assumes that the uncertainty of the line parameters and bus injections involved in transfer capability calculations can be estimated or measured and shows how to estimate the corresponding uncertainty in the transfer capability. Instead of using a large number of simulations as required in the Monte Carlo approach, for a system with n uncertain parameters, the two-point estimate method uses 2n calculations of transfer capability to quantify the uncertainty. The proposed method uses a numerical method to calculate the moments of the transfer capability. The moments are then used in the probability distribution fitting. Using the obtained transfer capability uncertainty information and a desired level of reliability, an adequate transmission reliability margin can be determined for each transmission service. The proposed method can be used directly with a deterministic computer program and it does not require derivatives of the transfer capability. Test results of the proposed method are compared with those obtained from the Monte Carlo simulations and a truncated Taylor series expansion method.

Distribution Feeder Scheduling Considering Variable Load Profile and Outage Costs

In a deregulated power market, customers would have more choices for their power service and the improvement of service quality has become a challenge to power transmission and distribution companies. Distribution system reliability that was traditionally considered within the planning activities is now incorporated in the operational environment. This paper presents study results of a multiobjective feeder operation optimization problem that considers how to balance network efficiency, switching and reliability costs in a distribution network. The proposed method divides annual feeder load curve into multiperiods of load levels and optimizes the feeder configurations for different load levels in annual operation planning. Customer load profiles and seasonal varying data of feeder section failure rates and customer interruption costs are considered. Numerical simulations demonstrate the time-varying effects on the optimal distribution feeder configuration and operation costs. A binary particle swarm optimization (BPSO) search is adopted to determine the feeder switching schedule. Test results indicate that not considering time-varying effects and using only simplified fixed load and reliability parameters could underestimate the total loss to the utility and its customers.

Network constrained security control using an interior point algorithm

N. Karmarkars interior point method (1984), said to perform much faster than the simplex method in solving large scale linear programming problems, has attracted a great deal of attention. The authors present a preliminary implementation of the interior point algorithm and test results on the network constrained security control problem. Test results indicate that when solving various sizes of network constrained security control linear programming problems with increasing numbers of controls and constraints, both the total number of iterations and the overall execution time grow at a slower rate in the interior point method than in the simplex method. The number of iterations required by the interior point method is relatively insensitive to problem size and composition, while iteration counts for the simplex method tend to be much higher in the presence of large numbers of control variables and/or constraints. >

Non-Technical Loss Detection Using State Estimation and Analysis of Variance

Summary form only given. Illegal use of electric energy is a widespread practice in many parts of the world. Smart metering enables the improvement of customer load model, theft and stressed asset detections. In this paper, a state estimation based approach for distribution transformer load estimation is exploited to detect meter malfunction/tampering and provide quantitative evidences of non-technical loss (NTL). A measure of overall fit of the estimated values to the pseudo feeder bus injection measurements based on customer metering data aggregated at the distribution transformers is used to localize the electricity usage irregularity. Following the state estimation results, an analysis of variance (ANOVA) is used to create a suspect list of customers with metering problems and estimate the actual usage. Typical Taiwan Power Company (TPC) distribution feeder data are used in the tests. Results of NTL detection under meter defect and energy theft scenarios are presented. Experiences indicate that the proposed scheme can give a good trace of the actual usage at feeder buses and supplement the current meter data validation estimation and editing (VEE) process to improve meter data quality.

An external network modeling approach for online security analysis

An approach that combines the load flow and state estimation techniques is proposed to improve the numerical stability and provide an external network model for online security analysis. An initial load flow study that provides a preliminary solution for the external network is followed by a state estimation using pseudomeasurements and proper weighting factors. Methods for handling multiple observable islands in the external network modeling process are described, and simulation tests are carried out on a practical system with realistic complexity. >

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.