Abstract
In this paper identification and parameter estimation of a hydro generator thermal system using collected input and output data is presented. Third order state space model is constituted on the basis of a set of independent output variable (temperature) measurements derived from the system response to three thermal sources. Data were collected during a heat run and site acceptance test of a refurbished machine. Least squares (LS or RLS) methodology is used in parameter estimation procedure. Minimisation is performed on weighted measurement and model output distance squares. During the procedure system and input matrices were obtained. Thereafter, spectral properties of system matrix are analysed using calculated eigenvalues and eigenvectors. Eigenvalues assignment to the physical system objects is performed via L2 norm calculation and ranked to output variable sensitivities with respect to their time constants (negative reciprocal eigenvalues). Finally, simplified analogous lumped parameter scheme of the machine thermal system has been outlined.
Highlights
The main task of generator temperature monitoring system is to decrease temperature variations on critical locations regardless the actual electrical and magnetic load
Methodologies found in these sources relate to machine construction, thermal optimisation and temperature field distribution foremost by means of numerical approaches
Transient thermal phenomena are coupled with two main machine construction corpora: copper electric conductors and iron core as magnetic conductor
Summary
The main task of generator temperature monitoring system is to decrease temperature variations on critical locations regardless the actual electrical and magnetic load. Unfavourable temperature conditions, i.e. high temperature and its variations may lead to chemical and physical degradation of machine insulation and lifetime shortening Design of such a machine temperature monitoring system requires very detailed knowledge of dynamic parameters and properties of the generator thermal system. Referent state space identification and parameter estimation procedures of linear signals and systems (ARA, ARMAX) based on their detailed observations, especially of discrete systems can be found in [13] and [14] Based on such methodology the authors in [15] give lumped parameter estimation for fourth order thermal system of a semiconductor in order to predict the temperature dynamics in high current load conditions. Besides avoiding complex modelling of system behaviour, the main advantage of proposed method is possibility to assign gained eigenvalues to appropriate physical system objects via L2 norm ranking of system time response sensitivities with respect to parameters, i.e. time constants
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
