Performance assessment of primary frequency control responses for thermal power generation units using system identification techniques

Extending SFR model to incorporate the influence of thermal states on primary frequency response

“…Although in the practical operation, the initial thermal state of a boiler may deviate from the rated point and changes dynamically, the aforementioned two aspects could be neglected with massive thermal units and abundant spinning reserves. However, considering the integration of VREs and the substitution of thermal power units, the impacts of boiler thermal states on PFR capability are becoming more and more prominent [8]. For instance, in the ‘9·19’ Jinping‐Sunan DC bipolar locking fault of East China, the frequency deviation velocity and magnitude exceed initial expectations, whose post‐mortem analysis indicates that neglecting boiler thermal states is a key factor [9].…”

“…On the other hand, the PFR capability can be measured via the output power of generating units [8, 22, 23] and the frequency response of the power system [24–26]. More specifically, the quantification of output power is mainly used to regulate the remunerations or penalisations of PFR provision.…”

“…More specifically, the quantification of output power is mainly used to regulate the remunerations or penalisations of PFR provision. A system identification method has been proposed to estimate the response time, settling time, and speed droop ratio of generators [8]. Moreover, according to the acceptable maximum frequency deviation, a constraint of PFR capability with different delivery speeds has been developed [22].…”

“…The assumption may lead to larger errors as the thermal power units are gradually replaced by VREs. Moreover, different indices and methods have been adopted to analyse PFR capability with different aims, but the impacts of thermal states have drawn little attention [8, 22–26]. Hence, the extension of a boiler sub‐model that incorporates the thermal dynamics and initial thermal states in the SFR model and the analysis of corresponding influence are underlying problems to be solved.…”

Linear Time-Varying Data Model-Based Iterative Learning Recursive Least Squares Identifications for Repetitive Systems

“…System identification is of great importance in many engineering fields such as in chemical process [1], [2], power system [3], [4], biomedical system [5], [6], and so on. Many identification methods [7]- [9] have been proposed over the last several decades to identify time-invariant parameters of the system.…”