1998 Fiscal Year Final Research Report Summary
STUDY ON HIGH EFFICIENCY OPERATION OF SOLAR GENERATION PLANT USING FUZZY CONTROL WITH ADAPTIVE SCHEME.
| Project/Area Number |
08650342
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
電力工学・電気機器工学
|
|---|
| Research Institution | UNIVERSITY OF THE RYUKYUS |
|---|
Principal Investigator |
UEZATO Katsumi UNIVERSITY OF THE RYUKYUS,FACULTY ENGINEERING,PROFESSOR, 工学部, 教授 (70045029)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TANG Yue-Jin UNIVERSITY OF THE RYUKYUS,FACULTY ENGINEERING,ASSOCIATE PROFESSOR, 工学部, 助教授 (20284953)
SENJYU Tomonobu UNIVERSITY OF THE RYUKYUS,FACULTY ENGINEERING,ASSOCIATE PROFESSOR, 工学部, 助教授 (40206660)
|
|---|
| Project Period (FY) |
1996 – 1998
|
| Keywords | SOLAR GENERATION / MAXIMUM POWER POINT TRACKING CONTROL / FUZZY NEURAL NETWORK / DC / DC CHOPPER / INSOLATION / PARAMETER TUNING |
|---|
| Research Abstract |
In this research, fuzzy inference with adaptive scheme is applied to maximum power point tracking control to develop high efficiency solar generating plant, irrespective of operating condition changes and parameter variations of solar generating plant.
In 1996, solar energy distribution of Ryukyu Islands were investigated. The optimum setting angle and direction of solar panel to extract maximum power from photovoltaic array for one year were determined.
In 1997, laboratory shop-test plant for solar generation, in which control strategy for maximum power point tracking control is implemented with a personal computer, was constructed. The performance of the proposed control strategy was investigated on tracking performance for transient and steady state conditions. The good tracking performance was obtained for both natural and artificial insolation.
In 1998, a new maximum power point tracking control strategy with using fuzzy neural network was developed. The parameters of fuzzy control is tuned off-line by intuition and experience of experts. The tuning of fuzzy controller is tedious and difficult task for beginners. The fuzzy neural network can represent the fuzzy controller and tune fuzzy parameters to which a cost function is minimized. The fuzzy parameters of the fuzzy neural network are trained on-line, therefore, good tracking performance is obtained in condition changes, such as solar insolation change, temperature change, and partial shading of photovoltaic array.
The all of above mentioned researches was presented in domestic and international conference. Main research result has been accepted for international journal, and the paper will be published near future.
|
