Power Systems Contingency Analysis using Artificial Neural Networks (original) (raw)
Related papers
Review Paper on Power System Contingency Analysis Using Artificial Neural Network
Contingency analysis is the important function of the power system security. The security assessment is an important task because it offers the information regarding the system state within the event of a contingency. Contingency analysis method is being widely used to predict the result of outages like failures of apparatus, conductor etc, and to require necessary actions to stay the facility system secure and reliable. The off line analysis to predict the impact of individual contingency could be a tedious task as a power system contains large number of parts. Various strategies of contingency analysis have been given in this review paper and additionally the utilization of artificial neural network for comntingency analysis has been cited.
Contingency Evaluation of Electrical Power System Using Artificial Neural Network
2010
This paper presents application of Artificial Neural Network (ANN) based contingency analysis of power system. The ANN has been chosen because of its high adaptation parallel information processing capability. Another feature that makes the ANN more suitable for this type of problems is its ability to augment new training data without the need for retraining. In this Multilayer Feed Forward network is used for contingency analysis in planning studies where the goal is to evaluate the ability of a power system to support a projected range of peak demand under all foreseeable contingencies. This work involves selection of network design, preparation of input patterns, training & testing. In order to generate the training patterns three system topologies were considered. Training data are obtained by load flow studies (NR Method) for different system topologies over a range of load levels using software simulation package (Mipower) and the results are compiled to form the training set....
Contingency Constrained Power System Security Assessment using Cascade Neural Network
A unified approach to power system security assessment and contingency analysis suitable for on-line applications is proposed. The severity of the contingency is measured by two scalar Performance Indices (PIs): Voltage-reactive power performance index, PIVQ and line MVA performance index, PIMVA. In this paper, a two stage cascade neural network is developed: Stage I employs Multi-Layer Perceptron (MLP) neural network trained by back propagation algorithm for estimating PIs and Stage II utilizes Kohonen's Self Organizing Feature Map (KSOFM) for contingency screening and ranking. The effectiveness of proposed methodology is tested on IEEE 39-bus New England system at different loading conditions corresponding to single line outage. The overall accuracy of the test results highlights the suitability of the approach for on-line applications to fast and accurate security assessment and contingency analysis.
POWER SYSTEM SECURITY ASSESSMENT AND CONTINGENCY ANALYSIS USING SUPERVISED LEARNING APPROACH
The most important requirement and need for proper operation of power system is maintenance of the system security. The security assessment analysis is done to determine until what period the power system remains in the safe operable mode. Contingency screening is done to identify critical contingencies in order to take preventive actions at the right time. The severity of a contingency is determined by two scalar performance indices: Voltage-reactive power performance index(í µí±í µí°¼í µí±£í µí±) and line MVA performance index(í µí±í µí°¼í µí±í µí±£í µí±). Performance indices are calculated based on the conventional method known as Newton Raphson load flow program. Contingency ranking is done based on the severity of the contingencies. In this proposed work, contingency analysis is done with IEEE 14 bus. Since the system parameters are dynamic in nature and keeps on changing, there is need of so ft computing technologies. Supervised learning approach that uses Feed-Forward Artificial Neural Network(FFNN) is employed using pattern recognition methodology for security assessment and contingency analysis. A feature selection technique based on the correlation coefficient has been employed to identify the inputs for thee FFNN. With these soft computing techniques, greater accuracy is achieved.
GSC Advanced Engineering and Technology, 2022
Power System Security and Contingency analysis is one of the most important tasks in power systems. In operation, contingency analysis assists engineers to operate the power system at a secure and safe operating point where equipment are loaded within their safe operating area (SOA). Power is dispatched to customers with acceptable quality standards. The results of off-line load flow calculations are used to estimate performance indices (PI flow, PI V). MATLAB toolbox was the proposed methodology used for the implementation. The proposed approach for contingency analysis was found to be appropriate for screening and ranking fast voltage and line flow contingencies.
International Journal of Electrical Power Energy Systems, 2012
This paper presents the application of cascade neural network (CANN) based approach for integrated security (voltage and line flow security) assessment. The developed cascade neural network is a combination of one screening module and two ranking modules, which are Levenberg-Marquardt Algorithm based neural networks (LMANNs). All the single line outage contingency cases are applied to the screening module, which is 3-layered feed-forward ANN having two outputs. The screening module is trained to classify them either in critical contingency class or in non-critical contingency class from the viewpoint of voltage/line loading. The screened critical contingencies are passed to the corresponding ranking modules, which are developed simultaneously by using parallel computing. Parallel computing deals with the development of programs where multiple concurrent processes cooperate in the fulfillment of a common task. For contingency screening and ranking, two performance indices: one based on voltage security of power system (VPI) and other based on line flow (MWPI) are used. Effectiveness of the proposed cascade neural network based approach has been demonstrated by applying it for contingency selection and ranking at different loading conditions for IEEE 30-bus and a practical 75-bus Indian system. The results obtained clearly indicate the superiority of the proposed approach in terms of speedup in training time of neural networks as compared to the case when the two ranking neural networks were developed sequentially to estimate VPI and MWPI.
Enhancement of Neural Network Based Contingency Analysis by Power Network Reduction
IEEJ Transactions on Power and Energy, 1994
In this paper a new application of neural network for single line outage contingency (screening, evaluation) is proposed. At the contingency screening phase, neural network ANN 1 detects all critical lines whose single outages can cause at least one line overloaded. Then at the contingency evaluation phase, neural network ANN 2 evaluates the outage effect of each critical line on the power flow of other lines. However as the system topology expands and number of lines increases, depending on the existence of irrelevant and redundant data in the input/output patterns of ANN 2 it suffers from, poor convergence performance, long training time and dimensionality. In order to improve these problems, in this paper a line outage oriented power network reduction technique is presented by which the irrelevant and redundant data are eliminated from input/output patterns of ANN 2. Proposed neural networks in conjunction with power network reduction are applied for line outage contigency analysis of IEEE 30-bus (40 lines) system. Simulation results show a promising improvement in the convergence performance and size of neural network ANN 2, compare to the case without network reduction which has a very poor convergence.
ANN based integrated security assessment of power system using parallel computing
International Journal of Electrical Power & Energy Systems, 2012
This paper presents the application of cascade neural network (CANN) based approach for integrated security (voltage and line flow security) assessment. The developed cascade neural network is a combination of one screening module and two ranking modules, which are Levenberg-Marquardt Algorithm based neural networks (LMANNs). All the single line outage contingency cases are applied to the screening module, which is 3-layered feed-forward ANN having two outputs. The screening module is trained to classify them either in critical contingency class or in non-critical contingency class from the viewpoint of voltage/line loading. The screened critical contingencies are passed to the corresponding ranking modules, which are developed simultaneously by using parallel computing. Parallel computing deals with the development of programs where multiple concurrent processes cooperate in the fulfillment of a common task. For contingency screening and ranking, two performance indices: one based on voltage security of power system (VPI) and other based on line flow (MWPI) are used. Effectiveness of the proposed cascade neural network based approach has been demonstrated by applying it for contingency selection and ranking at different loading conditions for IEEE 30-bus and a practical 75-bus Indian system. The results obtained clearly indicate the superiority of the proposed approach in terms of speedup in training time of neural networks as compared to the case when the two ranking neural networks were developed sequentially to estimate VPI and MWPI.
Power system contingency classification using machine learning technique
Bulletin of Electrical Engineering and Informatics, 2022
One of the most effective ways for estimating the impact and severity of line failures on the static security of the power system is contingency analysis. The contingency categorization approach uses the overall performance index to measure the system's severity (OPI). The newton raphson (NR) load flow technique is used to extract network variables in a contingency situation for each transmission line failure. Static security is categorised into five categories in this paper: secure (S), critically secure (CS), insecure (IS), highly insecure (HIS), and most insecure (MIS). The K closest neighbor machine learning strategy is presented to categorize these patterns. The proposed machine learning classifiers are trained on the IEEE 30 bus system before being evaluated on the IEEE 14, IEEE 57, and IEEE 118 bus systems. The suggested k-nearest neighbor (KNN) classifier increases the accuracy of power system security assessments categorization. A fuzzy logic approach was also investigated and implemented for the IEEE 14 bus test system to forecast the aforementioned five classifications.