A Review of Bayesian Networks Applications for Electrical Systems

Abstract

The development in the field of electrical energy has been growing increasingly due to the need of this energy in daily life. The reliability and safety of electrical power systems and equipment represent complex problems that are difficult to solve by conventional methods such as Fuzzy Logic and Artificial Neural Networks. Bayesian network is recently used to overcome some limitations in the conventional methods. This paper represents a bibliographic review about the use of Bayesian networks in the field of electric systems. This paper seeks to answer the following questions: (i) What are the areas of interest? (ii) What are the most active countries in this field?? (iii) Who are the most participating authors in this field? (iv) which year witnesses the largest number of publications? (v) What is the most widespread field related to this research? (vi) What is the most used system in terms of application? This field witnesses a slight increase in the number of publications in the last two decades (1999–2021), with a note of sharp increase in publishing in the last two years. It is observed that reliability assessment and fault diagnosis are the most common fields. Furthermore, it is found that China and USA are the highest active countries in this topic. Electric Power and Energy Systems Journal and IEEE Transactions on Power Systems Journal are the lead source documents, and most of the documents used electric power systems as an application. This paper will help researchers to know the versability features of BN and to identify the gaps in the use of BN in electric domains.

Research on Distributed Photovoltaic Grid-Connected Fault Diagnosis Methods: A Review

“…The main problems in applying neural networks are; the need for a large number of typical samples for training and long training time, and for new systems, re-training [24]; the ability to give only numerical results between 0 and 1, and lack of the ability to interpret the results; the inability to provide information to help the operator to diagnose the components of the malfunction; and the difficulty of modeling for the complex power transmission network.…”