Abstract
This is the second part of a two-part paper on a new adaptive coordinated voltage control (ACVC) strategy. The overall basic scheme and the online control for prepared faults have been presented in Part I. In this paper, learning control is explored to accumulate knowledge, so that the online control performances can be improved for unknown emergencies. For unknown situations where no past experiences can be exploited, a learning scheme is used by which the control knowledge can be acquired gradually. The learnt knowledge goes into the database and is improved the next time a related situation happens. After full knowledge is acquired, the unknown fault becomes a prepared fault with prepared knowledge. With the learning scheme, control for any emergency can be realized in a rapid and effective response. The learning process is demonstrated by providing control for an unknown emergency in the New England 39-bus power system. The ACVC performance for a sequence of randomly generated fault and load event emergencies is presented at the end of this paper.
| Original language | English |
|---|---|
| Pages (from-to) | 1554-1561 |
| Number of pages | 8 |
| Journal | IEEE Transactions on Power Systems |
| Volume | 29 |
| Issue number | 4 |
| DOIs | |
| Publication status | Published - Jul 2014 |
| Externally published | Yes |
Keywords
- Adaptive control
- coordinated voltage control
- learning control
- model predictive control
- multi-objective optimization