Unscented Kalman Filter in Gas pipeline leakage magnitude estimation and localization
| Autor: | Amirhossein Nikoofard, Hamid Khaloozadeh, Roya Doshmanziari |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0209 industrial biotechnology
Leak Discretization Computer science Multiphase flow 02 engineering and technology Kalman filter law.invention 020901 industrial engineering & automation Pressure measurement 020401 chemical engineering Method of characteristics Control theory law Boundary value problem Observability 0204 chemical engineering |
| Zdroj: | 2019 6th International Conference on Control, Instrumentation and Automation (ICCIA). |
| DOI: | 10.1109/iccia49288.2019.9030826 |
| Popis: | A model-based real-time transient modeling estimation process is proposed in a gas transmission line with the purpose of leak detection. The objective of this paper is aimed at determining the effectiveness of filtering techniques in leak detection and localization in a pipeline network through a synthetic pipe simulation. A model for a straight pipeline without any branch has been derived from Momentum and continuity equations and discretized based on the method of characteristics. In the next step by assuming proper boundary conditions, state-space representation of the system has been developed and stability and observability conditions have been explored. The unscented Kalman Filter has been selected as a nonlinear state observer. Simulations have been done in OLGA and PVTsim as multiphase flow simulators. The pressure measurements at the discrete points have been used as input to the state estimation algorithm. Finally, by defining different scenarios the effect of leak magnitude, location and time duration on the performance of the proposed method has been evaluated. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|