Historical Data Modeling Based on State Observation

Authors

Dong Ze, Hebei Engineering Research Center of Simulation & Optimized Control for Power Generation(North China Electric Power University), Baoding 071003, China;
Erxin Yin, Hebei Engineering Research Center of Simulation & Optimized Control for Power Generation(North China Electric Power University), Baoding 071003, China;

Abstract

Abstract: Aiming at the problem that the conventional historical data modeling method can not overcome the influence of disturbance on modeling accuracy, a historical data modeling method based on state observation is proposed. In this method, the data with the characteristic from dynamic to steady state are selected from the historical data as the modeling data, which is subjected to zero initialization processing based on the final steady state value. The processed data is then divided into two segments. The first segment is used to track the state of the prediction model, and the final observed state is used as the new initial state of the prediction model. The second segment is used to evaluate the accuracy of the estimated model parameters, and the particle swarm optimization is employed to find the optimal model parameters. The method not only has the advantages of the conventional historical data modeling, but also can effectively eliminate the influence of disturbance on the modeling process.

Recommended Citation

Ze, Dong and Yin, Erxin (2019) "Historical Data Modeling Based on State Observation," Journal of System Simulation: Vol. 30: Iss. 1, Article 3.
DOI: 10.16182/j.issn1004731x.joss.201801003
Available at: https://dc-china-simulation.researchcommons.org/journal/vol30/iss1/3

First Page

18

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.201801003

Last Page

27

CLC

TP391.9

Recommended Citation

Dong Ze, Yin Erxin. Historical Data Modeling Based on State Observation[J]. Journal of System Simulation, 2018, 30(1): 18-27.

DOI

10.16182/j.issn1004731x.joss.201801003