| Budget Amount *help |
¥1,600,000 (Direct Cost: ¥1,600,000)
Fiscal Year 1989: ¥500,000 (Direct Cost: ¥500,000)
Fiscal Year 1988: ¥500,000 (Direct Cost: ¥500,000)
Fiscal Year 1987: ¥600,000 (Direct Cost: ¥600,000)
|
|---|
| Research Abstract |
The main purposes of this research are to solve several types of problems, which arise along with implementation of state estimation in electric power systems, and to discuss synthetically practical use of dynamic estimators for tracking the system states. The results are summarized as follows:
1. Concerning optimal allocation of measurements, i.e., decision of redundant measurements, the effectiveness of the algorithms, in which variances of measurement noises are considered, was confirmed as well as standard values of redundancy. Furthermore, a graphical algorithm was established for the purpose of checking bad-data detectability and identifiability of measurement systems.
2. The performance of the dynamic state estimation was evaluated in detail by applying the estimator to a real-size model system. One of the concluding remarks is that the accuracy of estimation might be improved considerably by using a dynamic model because of the information at the previous time sample. The estimat
… More
or was also optimized for random variation of system states based on appropriate selection of a scalar parameter value included in the covariance matrix of system noises.
3. A novel method for hierarchical state estimation was developed by means of the goal coordination theory. The algorithm was applied to both of static and dynamic estimation, in which the final estimates are the same as the results obtained by the integrated state estimation basically. Based on the simulation results for model systems, it should be noted that the hierarchical dynamic estimators have excellent characteristics from the viewpoint of the computation time and the accuracy, although the coordination between an upper level and subsystems is approximated one.
4. Hierarchical estimation considering zero-power-injection busses as equality constraints was formulated and the algorithms were established. Especially this technique is effective for the so-called overlapping system decomposition, since the informations regarding tie-line-connected nodes can be used.
Finally, while the availability of the dynamic state estimators was clarified because of hierarchical structures, various functions, e.g., anomaly detection and identification, become easier for implementation. Less
|
