Adaptive Single Neuron Based Control for Superconducting Magnetic Energy Storage Unit

LI Yan,CHENG Shi-jie,PAN Yuan
DOI: https://doi.org/10.3321/j.issn:1000-3673.2005.20.011
2005-01-01
Abstract:The adaptive single neuron unit based control is such a control approach: the single neuron is connected to environmental system and by use of the iterative algorithm, in which the features of Hebbian study and supervision study are combined, the connection weight value is modified in on-line mode to realize the control objective. When above-mentioned control approach is applied to the design of controller for superconducting magnetic energy storage (SMES) unit, the modeling of the system is not required, the structure of the obtained controller is simple and its algorithm is convenient, so the obtained controller is suitable for engineering application. Under the environment of MATLAB, the authors build up a more precise high order nonlinear model of power system including SMES unit and examine the performance of the adaptive single neuron based controller by means of time domain simulation. The results show that the controller possesses satisfying control effect and better robustness.
What problem does this paper attempt to address?