Latest Posts

Featured Posts

Filter By Categories:

Smart Grids: Challenges of Processing Heterogenous Data for Risk Assessment

By Pacevicius, Michael 30. October 2019

Lightning PerthRecent advances in IT-related fields are opening up a broad range of novel applications. This is especially true in the energy sector, where Smart Grid solutions are offering new opportunities for the monitoring of power transmission and distribution in electrical grids. However, optimal use of potentially accessible data sources is challenging, and most of the current Smart Grid projects continue to exhibit suboptimal utilization of heterogeneous information.

This situation is also faced when it comes to the assessment of risks associated to operation of electricity transmission and distribution networks. As a consequence, current management systems fail to provide accurate estimations of risk levels in real-world situations.

In the present article, we address this issue and contribute to the identification of possible solutions. We identify a number of heterogeneous data sources which could be relevant for risk assessment, but which are currently not fully exploited.

Smart Grid benefits and main challenges

Smart Grids have several benefits, such as higher demand response with minimized costs, reduction of the environmental impact and integration of renewable energy resources, and resilience to disturbances as well as electrical stability in the grid. Smart Grids may also enable system operators to reduce outage risks by getting access to previously unconsidered data, ranging from weather forecasts to social network data. Combining outage reports with weather reports could for example improve risk monitoring in regions with harsh climatic conditions.

This approach has been previously explored in some projects, but the full range of various data sources is far from being optimally exploited, especially when it comes to risk assessment. We demonstrate this challenge with three main arguments:

  1. The available frameworks and standards from industrial risk management (e.g., CSA Q850-97, ISO 31000:2009, NORSOK Z-013) are generally overlooked when it comes to the study of Smart Grids. This reduces the advantages that can be derived from actual digitalization and hinders the exploitation of the real-time feature offered by Smart Grids.

  2. Risk analysis is mainly based on collections of operating and maintenance data, without taking advantage of additional accessible data. In fact, enlarging the horizon of the observations, in association with new data sources, could allow controllers to detect, observe and potentially predict slow, long-term and non-trivial phenomena (e.g., mechanical fatigue, corrosion, dust accumulation) increasing the failure probabilities.

  3. The lack of cross-disciplines experts hinders decision makers from identifying relevant links between data sources, compromising the recognition of efficient combinations of data sources.

There is thus a need for new methods enabling continuous and effective integration of heterogeneous data for accurate risk assessment predictions. In the present article, we address this topic and focus on the first phases of Smart Grid dynamic risk management: collection and combination of relevant datasets.

Reduce the risk assessment gap in Smart Grids

Utilization of Smart Grid technologies affects the risk level in the context of power grid management, both positively and negatively. They represent for example a great opportunity for improved information management and optimization of energy management. They are however also synonym of new challenges, especially when it comes to their integration in existing installations and protocols, or due to the new threats they unlock (e.g. hacking).

Acknowledging this situation, one may wonder: How should we execute and optimize the integration and management of (potentially new) heterogeneous datasets in order to both, enhance the positive impacts of Smart Grid technologies in terms of risk reduction, and limit the negative consequences these can be responsible for?

To answer this question, we have identified and reported in our paper a list of data sources that can be used to better characterize risks. Moreover, we intended to highlight the links existing between the different sources in order to understand how the data should later be aggregated. A deepened review of storm and outage reports, as well as an intensive research among the existing literature and among online websites of power management stakeholders has allowed identifying main elements and factors involved in the emergence of risks, outages and accidents in power grids. This research has enabled to identify main categories of directly related data sources, as well as tools that are used to reduce the severity of such outcomes. 

Description of the results

With our work, we point out that various data sources - often originally unconsidered (e.g., collaborative platforms, simulated environment, open-access data sources, etc.), can support a more effective risk assessment in power grid management. More especially, we bring forward some potential benefits such sources can provide when combined with asset management data and daily monitoring data from the grid, such as:  

  • grid topology & asset information (age, location, failure rates, initial life-time models, etc.),
  • inspection and maintenance reports (preventive/corrective approaches),
  • outage reports (context dependent outage data, consequences, costs),
  • customer feedback and crew management decisions,
  • power-flow forecasts & real-time power flow measurements in the grid,
  • past, present and future design documents of the power grid.

The results highlight the plurality of data sources that can actually be relevant for risk assessment. Doing so, they enable to imagine the diversity of existing possibilities for the creation of proxies; which is a main advantage for the increase of resilience in a power grid. The sample of suggested scenarios described in our work also highlights the plurality of the applications a data source can be useful for.

Conclusion

With the work described in this article, we provide elements for the construction of a framework supporting better decision making in the management of risk in modern power grids. In order to provide actionable intelligence, there is nevertheless a need for better sharing of best practices and for better sharing of data. This will enable the principal stakeholders to get a more accurate overview of their infrastructures and so to better make decisions when it comes to risk management in their power grids.

Did you find this article interesting? Read the full paper here. 

Pacevicius, Michael's photo

By: Pacevicius, Michael

Michael Pacevicius is researcher at eSmart Systems and an industrial PhD candidate in the RAMS group at NTNU. He has an engineering degree in Operational safety, Risks and Environment from the Université de Technologie de Troyes (UTT) in France and a MSc. in Economics and Business Administration from the Technische Universität Braunschweig (TUBS) in Germany. He worked as a project coordinator and analyst in the Big Data business development department of SAP in Munich, Germany, before joining eSmart Systems in 2017.

  • Write a comment

A blog from eSmart Systems

 

eSmart Systems provides AI driven software solutions to the energy industry and service providers. Their cloud born platform is designed to handle and exploit IoT, Big Data and Analytics in real time. The company is based on more than 20 years of international experience in establishing and operating knowledge based, leading IT and energy related companies targeting global markets.

 

If you want to stay updated on artificial intellgence, the rapid development of the energy supply, the power grid and the energy consumption trends, Intelligence First is your water hole.

 

Visit esmartsystems.com