IoT Blog
IoT Blog

Why the Energy Sector Needs a More Intelligent Distributed IoT Solution

by Olivier Amiot, Director of Marketing, Energy
The IoT is transforming the energy industry by eliminating tradeoffs between operation, SCADA systems, maintenance and new services for assets deployed in industrial and power facilities, buildings and across the grid.

When it comes to building the best IoT system for your business application, it’s vital to keep your use case and business requirements at the forefront of your technical design strategy. In the energy industry, accessing and collecting data at the edge from disparate, heterogenous, multi-site, fixed topologies and transferring that data efficiently to the cloud to perform analytics and action business decisions is still the greatest challenge. Mission-critical data collected from the edge is integral to energy facility operations and cannot be excluded or corrupted. 

Key challenges for the energy industry to maximize the transformative value of the IoT include: 

Scaling infrastructure
Building the backbone infrastructure to connect smart grid sensors, transformers and RTUs in substation and industrial facilities, or solar and wind farms remains complex; scaling and provisioning these systems is difficult, in part because of the many diverging connectivity requirements. The infrastructure must be robust and include urban and rural coverage, bandwidth to transmit data to potentially millions of assets and resilient and secure connectivity that can evolve over time with business and environmental changes. 

banner-blog-coverage analysis LTE-M

Management of incoming data
Data collected from the edge is mission critical for the grid, power monitoring at the utility, municipalities or any industrial energy facility operation and is therefore intolerant to missed, corrupted, or erronious data. The number of connected endpoints and sensors used in smart metering and grid monitoring at utilities facilities is growing. Over time, this results in a massive tsunami  of incoming data, requiring a cost-prohibitive increase in processing and analytics capabilities on premises or in the cloud to deal with the information collected from the edge. In addition, more bandwidth is required to transmit this data, impacting the performance of the critical backbone infrastructure when, in the end, much of the raw data turns out to be unnecessary or, in the worst case, hardly usable to action decisions. Huge amounts of data from smart metering is stored by utilities , but is currently not leveraged.

To better handle incoming data, companies are employing big data systems designed by SAP, Oracle, ABB, General Electric and others that use predictive maintenance and demand forecasting to add value. Other forms of future analytics, such as machine learning and artificial intelligence (AI) to give added context to incoming data, are not yet available today; having contextualization ready at the edge will be required to unlock the value of the data for AI and machine learning tasks.

Optimization of development resources
The focus on operation enablement and expert developer resources is shifting from a complex embedded world to an abstacted cloud environment that requires a web-based tool chain, efficient protocols, and API connectors to manage change in real-time.  

Today, technology gaps prevent IoT application developers from overcoming these limitations and harnessing the full power of edge-to-cloud data and analytics. The solution must add value to the data by contextualing it to reduce bandwidth and optimize resources to improve infrastructure performance, and address the limitations and deficiencies mentioned above. This includes:

Closing the Technology Gaps
There are technology gaps today that need to be filled to overcome these challenges. Key solutions to harmonize the uncoordinated, data-burdened bandwidth, network resources, data collection processes and analytics requires:
Intelligent edge processing
Building an edge device shadow on the energy cloud, allowing an additional level of abstraction that provides very resource-constrained LPWA IoT devices with virtual machine capabilities filled with rules and action to dynamically perform certain processing functions at the edge.

Intelligent data orchestration

Edge devices must recognize and transfer only valuable, select data from smart meters and grids, RTUs and inverters. By filtering, storing and transferring the data that decision-makers need, when they need it, edge devices can save resources and optimize their power use. 

Linking data flows for contextualization
By linking different flows of data (weather conditions, device tampering, downtime) and aggregating this information on the cloud, we can create new event streams, translating to new actions at the edge and more efficient operations.

Adapting to changing conditions

IoT solutions for the energy industry must be intelligent and dynamic enough to adapt the rules to changing conditions, including abnormal power surges outages, in real-time, and centralized or decentralized operation in runtime. A cloud developer needs to be able to do all of this securely with the same set of web-based toolchains scaling from a small pilot to thousands of endpoints. 

The ideal IoT solution for the energy industry will use distributed intelligence processing between the edge and the cloud to support smart metering and grid infrastructure for power monitoring, electricity, gas and water management. Unlocking the value of data streams and virtualizing the edge will help the energy industry overcome current limitations in edge-to-cloud analytics and deliver on cost and power consumption benefits of LPWA LTE-M technology. Start with Sierra to find out how to leverage LPWA and cutting-edge data processing in your energy deployments.