Estimation Method of Freguency Bias Using Normal Operating Data

Abstract

Accurate estimation of the freguency bias of electric power systems is becoming more important in order to maintain system frequency within operating limits and to make the best use of generating and transmitting facilities.
The freguency bias was supposed to use the data of the governor-cut test, but it was not the same as the data of normal operating conditions.
This paper proposes that using frequency fluctuation data, total power generation data and tie-line power flow data, calculations be made from the input/output spectrum.
It is not clear whether calculation outputs are right or not, so we investigated evaluation methods to confirm whether calculation outputs using observational data are correct.
Some 400 calculation cases enabled us to evaluate calculation outputs.
The estimation of the freguency bias (K) is expressed as the function of the standard deviation of frequency fluctuation (σ).
Although this esimation is larger than the value of load frequency characteristics mentioned so far, it corresponds well with the value of generator-frequency characteristics calculated using generators' speed regulation.