Machine learning for steady state security assessment in power system
The objective of this paper is to investigate the reliability of the SSA in determining the security level of power system from serious interference during operation. Artificial Neural Network (ANN), Adaptive Network Based Fuzzy Inference System (ANFIS) and Decision Trees (DT) are implemented to cla...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Praise Worthy Prize S.r.l
2012
|
| Subjects: |
| _version_ | 1796859097692241920 |
|---|---|
| author | Saeh, Ibrahim S. Mustafa, Mohd Wazir |
| author_facet | Saeh, Ibrahim S. Mustafa, Mohd Wazir |
| author_sort | Saeh, Ibrahim S. |
| collection | ePrints |
| description | The objective of this paper is to investigate the reliability of the SSA in determining the security level of power system from serious interference during operation. Artificial Neural Network (ANN), Adaptive Network Based Fuzzy Inference System (ANFIS) and Decision Trees (DT) are implemented to classify the security status in the test power system, comparison are made in terms of computation time and accuracy of the networks. Impact of Attribute Selections on train and test set is proposed. The impact of attributes number and cross validation on performance of the train and test data set is proposed as well. Data obtained from Newton Raphson Load Flow (NRLF) analysis method are used for the training and testing purposes of the proposed AI techniques. The data are used also as a benchmark to validate the results from AI techniques to achieve high speed of execution and good classification accuracy. A new methodology of feature selection technique based on extracting variables has also been applied. The proposed techniques have been extended and tested on various IEEE test systems. Generally, the proposed AI techniques have successfully been applied to evaluate SSA for various IEEE test system. |
| first_indexed | 2024-03-05T19:22:09Z |
| format | Article |
| id | utm.eprints-47168 |
| institution | Universiti Teknologi Malaysia - ePrints |
| last_indexed | 2024-03-05T19:22:09Z |
| publishDate | 2012 |
| publisher | Praise Worthy Prize S.r.l |
| record_format | dspace |
| spelling | utm.eprints-471682019-03-05T02:03:38Z http://eprints.utm.my/47168/ Machine learning for steady state security assessment in power system Saeh, Ibrahim S. Mustafa, Mohd Wazir TK Electrical engineering. Electronics Nuclear engineering The objective of this paper is to investigate the reliability of the SSA in determining the security level of power system from serious interference during operation. Artificial Neural Network (ANN), Adaptive Network Based Fuzzy Inference System (ANFIS) and Decision Trees (DT) are implemented to classify the security status in the test power system, comparison are made in terms of computation time and accuracy of the networks. Impact of Attribute Selections on train and test set is proposed. The impact of attributes number and cross validation on performance of the train and test data set is proposed as well. Data obtained from Newton Raphson Load Flow (NRLF) analysis method are used for the training and testing purposes of the proposed AI techniques. The data are used also as a benchmark to validate the results from AI techniques to achieve high speed of execution and good classification accuracy. A new methodology of feature selection technique based on extracting variables has also been applied. The proposed techniques have been extended and tested on various IEEE test systems. Generally, the proposed AI techniques have successfully been applied to evaluate SSA for various IEEE test system. Praise Worthy Prize S.r.l 2012 Article PeerReviewed Saeh, Ibrahim S. and Mustafa, Mohd Wazir (2012) Machine learning for steady state security assessment in power system. International Review on Modelling and Simulations, 5 (5). pp. 2235-2242. ISSN 1974-9821 |
| spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Saeh, Ibrahim S. Mustafa, Mohd Wazir Machine learning for steady state security assessment in power system |
| title | Machine learning for steady state security assessment in power system |
| title_full | Machine learning for steady state security assessment in power system |
| title_fullStr | Machine learning for steady state security assessment in power system |
| title_full_unstemmed | Machine learning for steady state security assessment in power system |
| title_short | Machine learning for steady state security assessment in power system |
| title_sort | machine learning for steady state security assessment in power system |
| topic | TK Electrical engineering. Electronics Nuclear engineering |
| work_keys_str_mv | AT saehibrahims machinelearningforsteadystatesecurityassessmentinpowersystem AT mustafamohdwazir machinelearningforsteadystatesecurityassessmentinpowersystem |