Semiphysical modelling architecture for dynamic assessment of power components loading capability

Dynamic loading of power components in a deregulated electricity market requires reliable models that are able to predict the thermal behaviour when the load exceeds a particular value. The thermal stress of the components is known to be the most critical factor to the assessment of network load capability. Predicting the evolution of the thermal stress during overload conditions is essential to estimate the loss of insulation life and to evaluate the consequent risks of both technical and economical nature. The paper discusses an innovative grey-box architecture for integrating physical knowledge modelling (also know as white-box) with machine learning techniques (also known as black-box) in dynamic load capability assessment of power components. To evaluate the effectiveness of the proposed solution, a specific case study concerning a system of medium-voltage power cables is presented.

[1]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[2]  Pasquale Daponte,et al.  A neural diagnostic system for the monitoring of transformer heating , 1996 .

[3]  Mauro Birattari,et al.  Lazy Learning Meets the Recursive Least Squares Algorithm , 1998, NIPS.

[4]  G. Nokes.  Optimising power transmission and distribution networks using optical fibre distributed temperature sensing systems , 1999 .

[5]  Mauro Birattari,et al.  A Model Selection Approach for Local Learning , 2000, AI Commun..

[6]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[7]  Huan Zhao,et al.  Underground power cable environment on-line monitoring and analysis , 2000, 2000 Power Engineering Society Summer Meeting (Cat. No.00CH37134).

[8]  Ghavameddin Nourbakhsh,et al.  Probabilistic cable rating based on cable thermal environment studying , 2000, PowerCon 2000. 2000 International Conference on Power System Technology. Proceedings (Cat. No.00EX409).

[9]  A. Carotenuto,et al.  Effect of radiative and convective heat transfer on thermal transients in power cables , 1995 .

[10]  Bernhard Sick,et al.  Recurrent and non-recurrent dynamic network paradigms: a case study , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[11]  Mauro Birattari,et al.  The Lazy Learning Toolbox. For Use with Matlab , 1999 .

[12]  Mauro Birattari,et al.  Lazy Learners at work: the Lazy Learning Toolbox , 1999 .

[13]  Alfredo Vaccaro,et al.  Application of Local Learning Techniques to power transformer thermal overload protection , 2001 .

[14]  Mauro Birattari,et al.  Data-driven techniques for direct adaptive control: the lazy and the fuzzy approaches , 2002, Fuzzy Sets Syst..

[15]  Andrew W. Moore,et al.  Locally Weighted Learning , 1997, Artificial Intelligence Review.

[16]  Alfredo Vaccaro,et al.  Neural diagnostic system for transformer thermal overload protection , 2000 .

[17]  Martine Baelmans,et al.  Detailed Modelling of Thermal Transients in Power Transformers : New trends in hot spot temerature rise modelling in power transformers , 1999 .

[18]  Alfredo Vaccaro,et al.  Application of local memory-based techniques for power transformer thermal overload protection , 2001 .

[19]  Mauro Birattari,et al.  Local Learning for Iterated Time-Series Prediction , 1999, ICML.

[20]  Lennart Ljung,et al.  Tools for semiphysical modelling , 1995 .

[21]  A. Y. Chikhani,et al.  Thermal analysis of power cable systems in a trench in multi-layered soil , 1998 .

[22]  Gianluca Bontempi,et al.  Local learning techniques for modeling, prediction and control , 2000 .