He Renmu

Composite load modeling via measurement approach

The accuracy of the load model has great effects on power system stability analysis and control. Based on our practice in China on modeling load from field measurements, this paper systematically develops a measurement-based composite load model. Principles guiding the load modeling practice are discussed based on detailed analysis on stochastic characteristics of the modeling procedure. The structure of the measurement-based composite load model is presented. A multicurve identification technique is described to derive parameters. The generalization capability of this built load model is also investigated in this paper. Two cases are studied to illustrate the accuracy of the developed load model on describing the load dynamic characteristics in the actual power system.

Load modeling by finding support vectors of load data from field measurements

The representation of load dynamic characteristics remains an area of great uncertainty and has become a limiting factor for power system analysis and control. The random nature of the load makes load modeling a very difficult problem, which becomes even more challenging when the field measurements increase and the recorded dataset becomes large. This paper proposes a novel concept of modeling load based on support vectors (SVs) of load data. A three-stage procedure to find SVs of the recorded load dataset is presented. Then the load model is built on the SVs. Although the model is derived from only a small subset of the original dataset, it has a strong generalization capability to describe dynamics of the whole dataset. However, the computational burden on the modeling process is much relieved since only a small subset of data is involved. The proposed method also answers the question on how data should be grouped and how many load models should be built as data are accumulated. This paper infers that, although the data space where the load varies seems indefinite and big, its characteristic can be captured and modeled in a much smaller subspace. The presented method is shown to be effective by the case study on Hushitai substation.

Input dimension reduction for load forecasting based on support vector machines

The traditional methods for load forecasting can not supply the required accuracy for the engineering application because we only get limited history data sets and the factors that affect the load forecasting are complex. This paper presents a new framework for the power system short-term load forecasting: firstly, this paper establishes the feature selection model and uses floating search method to find the feature subset; then this paper makes use of the support vector machines to forecast the load and takes full advantage of the SVM to solve the problem with small sample and nonlinear. Hence the accuracy of the estimation result is improved and a better generalization ability is guaranteed. The EUNITE network is employed to demonstrate the validity of the proposed approach.

Measurement-based load modeling-model structure

Referenced to the results of component-based load modeling approach, the paper presents an improved composite load model structure that can be applied into measurement-based load modeling area. The improved composite load model can keep its parameters concentrated when the load changes. Besides that, comparing with the typical dynamic load model, the improved composite load model can describe the reactive power of the real loads more accurately. Finally, a great lot of data measured from site were dealt with and the effectiveness of both measurement-base load modeling and load model structure was proved.

Impacts of DFIG-based wind farm on load modeling

It is expected that increasing amounts of new generation technologies will be connected to electrical power systems in the near future. Most of these technologies are of considerably smaller scale than conventional synchronous generators and are therefore connected to distribution grids, as the DFIG-based wind farm. When connected in small amounts, the impact of wind farm on power system transient stability can be negligible. However, if its penetration level becomes higher, wind farm may start to influence the dynamic behavior of the power system as a whole. In this paper, the impact of DFIG-based wind farm on the load modeling is investigated. It is found if the influence of wind farm is not negligible, the composite induction motor load model fails to describe the actual load characteristic effectively. A new load model-asynchronous generator +ZIP is used in this paper. The validity of the load model is also verified via various simulations. Then, composite induction-motor and asynchronous generator +ZIP load models are unified in one program.

Iterative Prony Method Based Power System Low Frequency Oscillation Mode Analysis and PSS Design

This paper presents an identification method based on the iterative Prony to identify transfer function and configure PSS parameters using residue way. The iterative Prony is a very useful method to acquire power system oscillation modes, by which the system eigenvalue, oscillation frequency, amplitude, damping and relative phase can be directly estimated using the response of given input signals. Considering the influence of input signal on output signal, the iterative Prony can acquire the transfer function of power system, then, according to the corresponding residues we conducted coordinated setting of power system stabilizer (PSS) by residue method. The simulation results for two testing system examples prove that this proposal method is effective in the power system oscillation mode analysis and parameters determination of PSS, can improve eigenvalues of system and can enhance system damping after PSSs whose parameters are optimized by above method are built into power system

Measurement-based Load Modeling Validation by Artificial Three-phase Short Circuit Tests in North East Power Grid

The researches on measurement-based load modeling have been continued for more than twenty years in China. However, since lacking of system measured data of major disturbance, it is doubtful for load model generalization capability, and more attentions are also caused for the issue that whether dynamic simulations adopting the measurement-based load model accord with the real system. The measurement-based load modeling technology was put into practice in North East Power Grid in 2002, and load models based on field measurement were relatively constructed. There were four artificial three-phase short circuit tests in 2004 and 2005 respectively in North East Power Grid, and the measured dynamic data from PMU and load data from load characteristics recorders validated the constructed load models and checked the generalization capability of the load model. The well generalization capability of the measurement-based load models verifies the validation of measurement-based load modeling technology.

Applications of data mining technique for power system transient stability prediction

This paper presents a data mining framework for the historical data of measurement and simulation units. Taking example for transient stability prediction, this paper establishes a data mining flow. The data market of transient stability is built up by all kinds of data sources. The data market is convenient for online analytical processing. At the same time, many model of data mining can be constructed based on the data market. We can acquire more knowledge of the power system transient stability. The IEEE 39-Bus test system is employed to demonstrate the validity of the proposed approach.

A reduced order method for swing mode eigenvalue calculating based on fuzzy coherency recognition

This paper presented a reduced-order method for swing mode eigenvalue calculating based on fuzzy coherency recognition. First, we recognize the coherent generator groups using the fuzzy clustering method. Then we aggregated the generators in a coherent group into a single equivalent generator that the dimension of the state equation reduced evidently. Using QR algorithm to the reduced-order state equation we calculated the eigenvalues of the inter-area mode. The eigenvalues of local mode calculated by using QR algorithm to the sub-state matrices corresponding to the coherent groups separately. Thus, all eigenvalues of swing mode can be calculated. We have given detailed results of both the coherent generator groups recognition and the eigenvalues calculating of the 10-machine New England power system. The results shows that the method for eigenvalue calculation is simple and practical.

Genetic algorithm based design of power system stabilizers

This paper presents a GA based optimization scheme for simultaneous coordination of multiple machines power system damping controllers. The proposed algorithm is applied to tuning of single and multiple power system stabilizers. Controller design is tested on the small and mid-sized power systems to prove its effectiveness. PSSs are designed using MATLAB control system toolbox and optimized using GAOT toolbox. All models and simulations are carried out using MATLAB and SSAT developed by Powertech Labs Inc, Canada.