New strategies and analysis for high impedance fault detection in distribution power systems
High impedance faults affect the reliability and safe operation of the distribution system. The previous represents a challenge to the Distribution System Operator (DSO) on the identification and isolation of low current faults not detected by conventional overcurrent functions such as ANSI 50/51. C...
- Autores:
-
Holguín Cárdenas, Juan Pablo
- Tipo de recurso:
- Fecha de publicación:
- 2022
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/55025
- Acceso en línea:
- http://hdl.handle.net/1992/55025
- Palabra clave:
- Falla de alta impedancia
Detección de condiciones anormales
Sistema eléctrico
Confiabilidad
Sistemas de distribución
Ingeniería
- Rights
- openAccess
- License
- https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
| Summary: | High impedance faults affect the reliability and safe operation of the distribution system. The previous represents a challenge to the Distribution System Operator (DSO) on the identification and isolation of low current faults not detected by conventional overcurrent functions such as ANSI 50/51. Currently, multiple models seek to represent a high impedance fault (HIF). However, very few HIF representations make it in a realistic approach according to principal characteristics such as randomness, non-linearity, intermittency, asymmetry, build and shoulder phenomena, plus those models do not incorporate low and high frequencies in the waves of current and voltage. Likewise, there are detection methodologies based on simple techniques such as harmonic detection that end up being few selective and complex algorithms based on machine learning which execute very slowly and where not all types of HIF are evaluated. This article presents an alternative in the HIF modeling and detection that meets all the characteristics, by developing an algorithm based on the detection of abnormal conditions in the electrical system using the Clarke transformation and the detection of zero crossings of the current waveform. Subsequently, it evaluates whether those conditions are due to a high impedance fault using criteria known as the wavelet transform, Fourier transform, asymmetry evaluation, and zero sequence current. Computational simulations used a simplification of the IEEE 34 node system on ATPDraw for generating the test waveforms, and the algorithm is implemented in Matlab. The results obtained reflect the accuracy of detection of 98.48% with times between one and three cycles. |
|---|