Showing papers by "Mayank Dave published in 1997"

Kohonen neural network classifier for voltage collapse margin estimation

Abstract: Owing to frequent black-outs world wide, voltage collapse has received much attention in the electric utilities industry. This paper presents an artificial neural network based method for on-line voltage collapse margin estimation. Homotopy Continuation based Newton-Raphson method is used to drive system operating point to knee of nose curve. The distance of operating point from critical point, measured in terms of system loading may be regarded as margin to voltage collapse. This paper utilizes Kohonen classifier to estimate margin so that computational efforts are reduced compared to conventional methods. Also there is ample of saving in training time compared to error back propagation for parameter estimation. Kohonen neural network classifier transforms input patterns into neurons on the 2-dimensional grid. Power system conditions are assigned to neurons on the grid based on self-organized feature mapping. Finally these neurons are allocated voltage collapse margins corresponding to their sys...

Input-features based comparative study of intelligent transient stability assessment

Abstract: With the growing stress on today's power system, it is operated much closer to its stability limit. Under such circumstances it is highly desirable that one must be able to assess the security and stability of the electric power system when exposed to disturbances/faults. In the post-fault transient analysis of interconnected systems, the transient energy margin which is a complex function of prefault system conditions, structure of fault (type and location) and network topology at the specified fault clearing time gives a quantitative idea about the stability of the system. High adaptation capabilities of artificial neural networks make them capable of synthesizing the complex mapping that transform the input features in to a single-valued space of energy margin. Appropriate input feature selection has a direct bearing on the consistency and accuracy of mapping. This issue has been addressed in the present paper by comparing the prediction results based on approaches (Sobajic and Pao, 1989), (So...

Ann-based voltage collapse proximity prediction

Abstract: In the present work, an artificial neural network technique is applied to determine the distance from system operating point to voltage collapse. The distance is measured in terms of the existing loading/generating scenario. The critical/bifurcation point is approached in steps while moving along the nose curve. Generation resheduling is done at each step to ensure economical operation of the system. The homotopy continuation-based Newton Raphson load flow method takes care of numerical instabilities associated with the singularity of Jacobian while approaching critical point. Proximity to critical loading of present operating point is a complex function of operating point attributes and loading/generating pattern followed to approach voltage collapse. The high adaptation capabilities of artifidal neural networks make it feasible to synthesize the function that maps system state attributes (bus power injections and tap settings of transformers) to distance from voltage collapse for uniform dispatch strategy. A three-layer (one hidden layer) feed-forward artificial neural network (ANN) is trained to predict the newness of current operating point to voltage collapse in terms of existing loading condition. It has been demonstrated here that the predicted values are well in tune with their actual ones. The technique is tested on a Ward-Hale 6-bus system and an IEEE 14-bus system.