Towards a theory of voltage collapse in electric power systems

Several voltage collapses have had a period of slowly decreasing voltage followed by an accelerating collapse in voltage. The authors clarify the use of static and dynamic models to explain this type of voltage collapse, where the static model is used before a saddle-node bifurcation and the dynamic model is used after the bifurcation. Before the bifurcation, a static model can be used to explain the slow voltage decrease. The closeness of the system to bifurcation can be interpreted physically in terms of the ability of transmission systems to transmit reactive power to load buses. Simulation results show how this ability varies with system parameters. It is suggested that voltage collapse could be avoided by manipulating system parameters so that the bifurcation point is outside the normal operating region. After the bifurcation, the system dynamics are modeled by the center manifold voltage collapse model. The essence of this model is that the system dynamics after bifurcation are captured by the center manifold trajectory. The behavior predicted by the model is found simply by numerically integrating the system differential equations to obtain this trajectory. >

On voltage collapse in electric power systems

A multilayer feedforward neural network is proposed for short-term load forecasting. To speed up the training process, a learning algorithm for the adaptive training of neural networks is presented. The effectiveness of the neural network with the proposed adaptive learning algorithm is demonstrated by short-term load forecasting of the Taiwan power system. It is found that, once trained by the proposed learning algorithm, the neural network can yield the desired hourly load forecast efficiently and accurately. The proposed adaptive learning algorithm converges much faster than the conventional backpropagation-momentum learning method. >

Short term load forecasting using a multilayer neural network with an adaptive learning algorithm

A new method is presented for calculating the nose curves and critical loading conditions of power systems. The nose curve (PV curve), which donates the relationship between total load and system voltages, is calculated by a new approach based on the homotopy continuation method. The critical loading condition, which might be called the bifurcation point, is also calculated precisely as the final point of the nose curve. This method does not require an exhausting cut-and-try process or a rough-approximation process. It is based on the conventional Newton-Raphson load flow calculation, but it overcomes the numerical difficulties associated with the singularity of the Jacobian matrix. The results of applying the proposed method to the IEEE 118-bus system and to other large practical systems (e.g., a 496-bus system) verify its robustness and feasibility. >

Calculation of critical loading condition with nose curve using homotopy continuation method

This paper provides a comprehensive list of books, reports, workshops and technical papers related to voltage stability and security

Bibliography on voltage stability

Power system security assessment

The authors provide an approach that enables a power system operator or planner to evaluate his or her alternatives in selecting a certain voltage collapse criterion. Specifically, the selection of a criterion must guarantee that the risk of making a wrong decision is minimal for any operating point and disturbance. This is done by recasting the various voltage collapse criteria in terms of a decision framework. This framework is based on statistical decision theory and provides a method for evaluating the risk of making the wrong decision in terms of the probability of a missed voltage collapse or a false alarm. Some examples are given to illustrate the effectiveness of the proposed approach. >

On the evaluation of voltage collapse criteria

Brief paper: Optimal control of a solar collector loop using a distributed-lumped model

The authors demonstrate the feasibility of shape and position control of large flexible structures with collocated sensors and actuators using direct position-plus-velocity feedback. This result is important for large space structures where the number of modes is very large and eventually unknown. When the number of inputs equals the number of outputs, the stabilizing feedback gain stands for any positive definite matrix, including diagonal matrices. This result may explain the success of decentralized adaptive controllers when a reasonable number of sensors is used to satisfy the observability assumption. >

Direct position plus velocity feedback control of large flexible space structures

The utility of trained neural networks in calculating the network state and classifying its security status under different load and contingency conditions is demonstrated. In particular, a two-layer multiperceptron is used to screen contingent branch overloads. The performance of this approach is evaluated using a six-bus example. The results indicate that the proposed tasks can be performed reliably by back-propagation-trained multiperceptrons. >

Robert Fischl

Papers

Power system security assessment

On the evaluation of voltage collapse criteria

Brief paper: Optimal control of a solar collector loop using a distributed-lumped model

Direct position plus velocity feedback control of large flexible space structures

Screening power system contingencies using a back-propagation trained multiperceptron