Developing a smart application for automatic pre-analysis of disturbance recorder files for improved condition based monitoring of breaker assets
Background
Whenever a fault occurs in the power system and is detected by protection device, circuit breakers are involved in order to isolate the fault by breaking open the flow of current. High current are very difficult to break and very often electrical arcs occur during the current breaking period. This arcs can cause serious damage to the breaking mechanism, due to high heat and corrosion for example.
• Disturbance recorder files (comtrade) contain high resolution measurements that help power engineer understand what was the root cause of a power system fault.
• Comtrade files contain valuable information both for protection planning and asset health analytics, for example the current intensity at the instant of breaking.
• These files are made up of two different files, one configuration file containing information on how to understand the data, and one data file containing the data, in form of binary and analog channels.
Vattenfall has been able to show during a pilot project that it is possible to automatically forward the files to central level for analysis (for example condition based monitoring), but the files are stored with very little metadata which means that it is very difficult to sort them with greater accuracy.
Goal
The goal of this thesis is to build a smart application connected to a service oriented architecture that will automatically analyze disturbance recording files to produce relevant quick to access information for e.g. asset management users.
Scope of the thesis (30HP)
We would like the student to study the possibility to automatically import the files, analyze them and generate more metadata as well as produce calculated measurements from these files. This task would consist of two different parts:
• The first part related to communication and IT, in the form of creating an application capable of using generic web services in order to import and export information to/from the Service Oriented Architecture platform.
• The second part related to power system and fault analysis, in the form of creating an application capable of reading the Comtrade information and automatically identifying relevant patterns within the Comtrade data, be it with general understanding of fault analysis or by using machine learning algorithms and/or neural networks. The application should then be able to automatically generate extended metadata of the files and/or derived measurements for asset management and protection planning personal to use.
What else do I need to know?
• This master thesis is done with Vattenfall R&D in cooperation with Vattenfall Eldisitribution.
• The base location for this work will be at Vattenfall R&D in Solna near Stockholm, with probably at least one visit in Vattenfall’s office in Luleå, where some key IT personal reside.
Candidate
This work requires some knowledge in power system and especially fault analysis, but also some knowledge in IT integration (e.g. webservices, …) and machine learning. We see you as an interdisciplinary student or a student with a interests in both IT and the energy sector, as smart grids and grid analytics tend to bring together all these fields.
Process
CV, cover letter, and grades are to be submitted in the application. Application by email to the address below.
The interview process will be ongoing, as soon as relevant applicants are identified.
Date
Deadline for application: 2018-11-30.
Contact
Do not hesitate to contact us for any questions:
Vincent Gliniewicz
Vattenfall R&D
vincent.gliniewicz@vattenfall.com
