Neural-networks for predicting the operation of an under-frequency load shedding system

Abstract

Dynamic security assessment is of special importance to island power systems. The CPU time required in order to apply conventional methods for those calculations do not allow real time application. The fast calculation time is, therefore, an important advantage of artificial neural-networks compared to other methods. This paper presents two neuralnetwork models that were designed to calculate the minimal frequency and the load shedding system operation during a forced outage of a generating unit. The minimal frequency and the extent of the load sheddiiig are strong indications of the severity of the fault. Hence, it is a significant part of the dynamic security assessment procedure.

Load Sheddingtechniques For System With Cogeneration: A Review

“…Shows the corresponding load shedding result of the schemes [2]. However they have some drawbacks [16] as described below.…”

“…These time delays as a sum result into a slow response [5]. • Usually setting of each frequency relay is based on the most severe disturbance conditions, so incorrect loadshedding may take place [16]. • Frequency relay settings require simulation of hundreds of transient stability studies which are not actually dependent on the real time operating conditions [16].…”

“…Possible contingencies are evaluated, and required load shedding is assembled [42]. However, this table is independent of any changes that may take place in the system [16,42]. PLC systems are faster and accurate than underfrequency relays [16].…”

“…However, this table is independent of any changes that may take place in the system [16,42]. PLC systems are faster and accurate than underfrequency relays [16]. However, PLC systems, uses underfrequency relays as a backup [16] due to reasons such as:…”

“…PLC systems are faster and accurate than underfrequency relays [16]. However, PLC systems, uses underfrequency relays as a backup [16] due to reasons such as:…”