Market Power Assessment Using Hybrid Fuzzy Neural Network

Market power assessment is an important aspect of electric market analysis and operation. Market power problems are more complicated in an electric market than those in other markets due to the specific properties of electricity. A comprehensive and dynamic market power assessment has been proposed in this paper to protect and improve the open electricity market. This paper proposes a multi output fuzzy neural network (FNN) for market power assessment and for finding the on line market power ranking status of GENCOS in a competitive power system using a fuzzy composite market index (FCMI). This index is formulated by combining (i) Lerner Index, (ii) Relative market power and (iii) Nodal Cost. In the proposed FNN a trained multioutput neural network is being used as a fuzzy inference engine. The input of FNN consists of real loads and a bipolar code to represent a trading interval while the output consists of the fuzzy values of FCMI. To train the FNN a number of training patterns, covering the full operating range of the power system, are generated using the system data such as offer prices and operating constraints. OPF results are used to compute the above three market power indices and the corresponding FCMI. Once the network is trained it is capable of predicting the FCMI values in five fuzzy classes (GENCO ranking) for any given operating scenario, on line, instantaneously, without bothering about the computational burden of OPF. The computational effort is required only for training the network which is an off line process. Since the training of ANN is extremely fast and test results are accurate, they can be directly floated to OASIS (open access same time information system) and any other web site. An Independent system operator(ISO) and customers can access this information instantly. The performance of the proposed method has been tested on an IEEE 14 bus system.

[1]  M. K. Soni,et al.  Artificial neural network based peak load forecasting using Levenberg-Marquardt and quasi-Newton methods , 2002 .

[2]  Manjaree Pandit,et al.  Hybrid fuzzy-neural network-based composite contingency ranking employing fuzzy curves for feature selection , 2009, Neurocomputing.

[3]  Ettore Francesco Bompard,et al.  An IPV auction model for strategic bidding analysis under incomplete and asymmetric information , 2003 .

[4]  Yinghua Lin,et al.  A new approach to fuzzy-neural system modeling , 1995, IEEE Trans. Fuzzy Syst..

[5]  Manjaree Pandit,et al.  Voltage contingency ranking using fuzzified multilayer perceptron , 2001 .

[6]  Fushuan Wen,et al.  Development of bidding strategies in electricity markets using possibility theory , 2002, Proceedings. International Conference on Power System Technology.

[7]  A. J. Svoboda,et al.  Revenue adequate bidding strategies in competitive electricity markets , 1999 .

[8]  J Nahman,et al.  Fuzzy logic and probability based real time contingency ranking , 2000 .

[9]  George Gross,et al.  OPF tools for optimal pricing and congestion management in a two sided auction market structure , 2001, 2001 IEEE Porto Power Tech Proceedings (Cat. No.01EX502).

[10]  S. M. Shahidehpour,et al.  AI approach to optimal VAr control with fuzzy reactive loads , 1995 .

[11]  D. J. Wu,et al.  Strategic gaming in electric power markets , 2001, Eur. J. Oper. Res..

[12]  Fernando L. Alvarado,et al.  Contingency ranking for voltage collapse via sensitivities from a single nose curve , 1999 .

[13]  A. David,et al.  Optimal bidding strategies and modeling of imperfect information among competitive generators , 2001 .

[14]  Peng Wang,et al.  Nodal market power assessment in electricity markets , 2004 .

[15]  G. A. Hamoud,et al.  Assessment of transmission congestion cost and locational marginal pricing in a competitive electricity market , 2004 .

[16]  Enrico Carpaneto,et al.  The role of load demand elasticity in congestion management and pricing , 2000, 2000 Power Engineering Society Summer Meeting (Cat. No.00CH37134).

[17]  Manjaree Pandit,et al.  Fast voltage contingency selection using fuzzy parallel self-organizing hierarchical neural network , 2003 .

[18]  Chen-Sung Chang,et al.  Neuro-fuzzy networks for voltage security monitoring based on synchronized phasor measurements , 1998 .

[19]  Edward J. Anderson,et al.  Optimal Offer Construction in Electricity Markets , 2002, Math. Oper. Res..

[20]  W.-H.E. Liu,et al.  An integrated application for voltage sag analysis , 1997 .

[21]  J. W. Lamont,et al.  Strategic bidding in an energy brokerage , 1997 .

[22]  Sushil Chauhan,et al.  Fast real power contingency ranking using counter propagation network: feature selection by neuro-fuzzy model , 2005 .

[23]  Donald V. Bourcier,et al.  Locational Market Power Screening and Congestion Management: Experience and Suggestions , 2002, IEEE Power Engineering Review.

[24]  Enrico Carpaneto,et al.  Reactive load modeling impacts on nodal prices in pool model electricity markets , 2000, 2000 Power Engineering Society Summer Meeting (Cat. No.00CH37134).

[25]  Leigh Tesfatsion,et al.  Market power and efficiency in a computational electricity market with discriminatory double-auction pricing , 2001, IEEE Trans. Evol. Comput..

[26]  M. Ilić,et al.  Dynamic games-based modeling of electricity markets , 1999, IEEE Power Engineering Society. 1999 Winter Meeting (Cat. No.99CH36233).

[27]  Manjaree Pandit,et al.  Coherency-based fast voltage contingency ranking employing counterpropagation neural network , 2007, Eng. Appl. Artif. Intell..

[28]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[29]  Peter B. Luh,et al.  Optimization based bidding strategies in the deregulated market , 1999 .

[30]  Hong Chen,et al.  Transaction impact analysis and its application in electricity markets , 2003, Large Engineering Systems Conference on Power Engineering, 2003.