Dynamic security constrained rescheduling using stability sensitivities by neural network as a preventive tool

Abstract

In this paper, a novel approach for generation rescheduling as a preventive control for enhancing dynamic security using neural network is presented.. Critical clearing time (CCT) associated with each fault including the effect of system controllers and limitation, is adopted as dynamic security criteria. A Dynamic Security Analyzer Neural Network (DSANN) is trained to estimate CCTs associated with different system faults. For each given operating point, DSANN evaluate system CCTs by using steady state pre fault operating condition as input pattern. The most interesting feature of the proposed neural network application is evaluation of sensitivity of CCT with respect to generation pattern. These sensitivities are derived from the information stored in the weighting factors of trained DSANN. The sensitivity of CCT is used as a guideline for selecting the most effective pair of generators to reschedule their MW generation in the process of generation rescheduling aimed security enhancement. The proposed method has been demonstrated on the IEEE-39 bus system with promising results for enhancing dynamic security by generation rescheduling using sensitivity characteristic of neural network.

A novel approach for improving voltage stability margin by sensitivity analysis of Neural Network

“…The memory requirement and processing time can be reduced either by reducing the dimension of the input data or by reducing the number of training patterns. In this paper, the dimension of input pattern space is reduced by extracting its dominant features in a lower dimension space by using principle component analysis (PCA) [20,21]. Principle component analysis is one of the well-known feature extraction techniques and a standard technique commonly used for data reduction in statistical pattern recognition and signal processing.…”

“…For this purpose, the sensitivity analysis of VSM with respect to bus voltages is performed to find the most effective buses for compensation. The sensitivity of VSM with respect to each bus voltage magnitude can be calculated using information stored in the weighting factors of VSANN and input data by (5) [21]: In order to find the most effective bus for injecting capacitive reactive power and consequently for enhancing VSM, it is necessary to evaluate the sensitivity of VSM with respect to reactive power compensation. For this purpose, the network Jacobain matrix as shown in (6) is used.…”