Kapernikov enables Umicore to increase plant capacity
Kapernikov helped Umicore to make its precious metal recycling processes more efficient. This enable ...
In order to be able to anticipate the impact of selective shedding and power outages, Infrabel needs to have precise information about the configuration of its electrical network. Generally, local supervisors use a variety of methods to collect electrical cable data. Kapernikov consolidates this data and collects all useful information in the field. As a result, the heads of the electrical division have a consistent and up-to-date view of the network.
Kapernikov assists Infrabel’s internal electrical grid operator in building an accurate and consistent overview of its cable assets and a topology of its networks. The end result is a high-quality electrical model of the infrastructure.
In 2019, Kapernikov planned, implemented and managed a campaign to collect the necessary data and transfer it into the company’s asset database in a consistent way.
The goal of the campaign was twofold:
- Centralize the data in order to get an overview of the cable network.
- Build an accurate representation of the network and the associated topology in order to better understand the energy flows.
This topology in turn allows Infrabel to model the impact of power disruptions in a better way.
In order to achieve these two goals, Kapernikov needed access to as much data as possible. We divided the project in phases, with each phase preparing for the next and improving the quality and reliability of collected data.
Phase 1: consolidate historical structured data
In a first phase, we queried the existing database. A similar effort had been done in the years before our data collection campaign. Since the data was sitting in multiple databases, we needed to go through different steps of consolidation and processing before we could select the right data according to the source system.
Phase 2: gather local non-structured data
A lot of data had already been compiled locally in one form or another (spreadsheets, AutoCAD drawings…). In a second phase, we asked all engineers and local supervisors to send us local versions of the data. This would allow us to compile the data and enter it in a temporary database for further validation.
Next, we carried out an intensive search in the library of digital plans. This enabled us to extract the information we needed and to convert it into usable data. This part was useful to prepare the third phase, as a significant part of the network was already laid out in the plans.
Phase 3: data validation
The third and last phase started with a data management team meeting with local engineers to consult them on the actual layout of the network. Based on this meeting, we entered data into a QGIS-based tool created to map the energy grid.
Finally, after this data collection campaign, a custom network visualizer allowed us to investigate the network and detect any instance of faulty data according to a set of rules (based on the allowed tension or the ideal topology of the network).
Since the two goals were closely linked, an adequate data model and data input workflow had to be designed. This was done using a combination of two open-source tools: QGIS for data input/mapping and PostgreSQL/PostGIS for data storage.
Selective scheduling calculations
After the data validation and network correction were finalized, we could use the representation of the network to map its energy flows. The representation also enabled us to understand how the network would react to potential disruptions due to power outages or selective shedding. Next, we loaded the cable data into SAP. Today, the local engineers who use this SAP system have a much better overview of their electrical network.
It seems like you're really digging this article.
Subscribe to our newsletter and stay up to date.
