With the expansion of modern power systems, the stability issues become more and more prominent. Transient stability emergency control is usually designed in open-loop schemes and applies proper actions to avoid system collapse when transient stability cannot be guaranteed in serious contingencies. Taking transient stability and economic efficiency of power system into consideration, the emergency control problem can be modeled as an optimal control problem, which is computational expensive. In this paper, an optimal control method with constraint aggregation is proposed to reduce computational complexity. The yield nonlinear problem is a fairly small-scale optimization problem which can be efficiently solved by predictor–corrector interior point method. The adjoint sensitivity analysis (ASA) is employed to evaluate the first-order derivative while Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm is used to obtain the second-order derivative. Besides, very dishonest Newton (VDHN) method and reusage of LU factorization results are explored to accelerate the forward and backward integration phase of ASA, respectively. The proposed approach is tested on four cases with different scales, and shows its potential in computational efficiency.

An Efficient Optimal Control Method for Open-Loop Transient Stability Emergency Control

Quantitative methods for both transient stability assessment and control decision making are urgently required for coordinating the economy and security of the power market. These methods should be able to identify all potentially unstable modes, to show their stability margins, to recognize dangerous reloading directions, to calculate their limits of generation and interface flow, to estimate the stability domain in the injection space and to optimize preventive/emergency actions. Besides, it should be fast enough for tracking system changes. FASTEST integrates EEAC, a unique and rigorously quantitative method, with numerical integration software such as ETMSP, a state-of-the-art time domain simulation program, is just such a tool. This paper presents its principle, development and applications.

Fast analysis of stability using EEAC and simulation technologies

A new tool for dynamic security assessment of power systems

Practically Negative Effects of Emergency Controls

The Ertan Hydroelectric Power Development Project with a total capacity of 3300 MW and the associated 500 kV power transmission project is the largest of its kind in China before the Three Gorges Power Transmission Project. An on-line pre-decision system (OPS-1) has been built and tested in the laboratory for its emergency control. Based on on-line stability analysis and pre-decision-making, OPS-1 follows the measured operating condition and refreshes the look-up decision table periodically. Therefore, it is adaptive to operating condition change and structure evolution of the power system, and is a fundamental innovation on the present emergency control system. OPS-1 integrates the unique EEAC technique for quantitative analysis of stability, an integer programming for nonconvex nonlinear optimization of transient stability control, an object-oriented distributed processing platform for contingency-oriented parallel computation into distributed devices for emergency control. The control system structure, decision searching algorithm and laboratory test results are presented in this paper.

An on-line pre-decision based transient stability control system for the Ertan power system

Achievements in on-line transient stability assessment of the Northeast China Power System are reported in this paper. The improved dynamic extended equal area criterion (DEEAC) is interfaced to a state estimator with DBMS in an advanced EMS environment. Thousands of DEEAC results, recorded from snapshots of on-line estimations have been thoroughly checked with an off-line numerical integration method. For first swing stability assessment of both symmetrical and asymmetrical faults, there are only less than 1% of cases whose errors exceed 10%. Moreover, all large errors occur with mild contingencies only. Each critical clearing time (CCT) calculation takes about 10 s CPU time of a VAX-11/785 computer. In an HABITAT environment (developed by ESCA Corp.), the on-line EEAC package is very user-friendly. >

Y. Xue

Papers

On-line transient stability assessment in operation-DEEAC in Northeast China Power System