By: Beau Blumberg and Dave Sheehan, Infiswift Solutions
How new IoT concepts can be used to seamlessly aggregate data across technologically diverse assets
Over the last 10 years, the North American PV market has grown exponentially. One result of this explosive growth is that more and more installed plants are passing their 3, 5 or 10 year age mark. With these maturing plants comes a new set of challenges in managing them. Operations are passing from the installation firms that built the plants to the long term owner-operators; they’ll continue to own the performance of the plants through the rest of their contract lifespan. The management needs of the plant constructors (who typically manage the plants for the first years of operation) are vastly different to the needs of the long term owner-operators. EPCs are driven by the performance guarantee (production above and beyond their guarantee usually does not result in a higher payment to the EPC), while owner-operators have an economic interest in maximizing plant performance.
The different needs of these stakeholders can result in an owner-operator inheriting a disparate portfolio of data acquisition systems and management softwares. These data acquisition systems are almost always built to communicate with the equipment at one specific plant, making data aggregation and analysis across portfolios very difficult. API communication between systems is also difficult without a homegrown system, and in some cases, security regulations make data export to the cloud very complex. Without easy access to their data, long term plant managers can spend more time collecting and normalizing data than analyzing plant performance. This lack of interoperability can significantly reduce the efficiency of data analysis, resulting in more time required to discover actionable information (or missing it altogether).
There are, however, opportunities to use new technologies to overcome many of these issues with interoperability. The Internet of Things (IoT) uses low cost hardware, wireless networks and modern architectures (see figure 1 below) to connect a broad variety of data points in the cloud. By aggregating data that has historically been siloed with new data sources that were previously not economical to collect, this allows users to make better decisions based on more complete data. Organizations like SunSpec and companies like infiswift are working together to define standards and deploy solutions in the solar space to improve on interoperability issues faced by the industry.
In the past, interoperability was limited to moving data between different devices within the network. Now, application level standards allow this data to have meaning and usability beyond a single plant extended through an entire portfolio. However, there is still an issue of how to handle legacy plants that don’t have a feasible way to support these new standards without substantial upgrades. With open platforms that adopt micro-services, such as those shown in the table below, it can be easy to ingest and normalize data from nearly anywhere yet easily swap data sources and formats if the plant is updated. This means provisioning plants can occur over time, eliminating a massive one-off IT project or deployment.
|Data Contextualization||Technology Micro-Service||In order for data and insights to be meaningful for all parts of a distributed system, it is critical that everyone speaks the same language and dialect. Data contextualization allows components to know what data they’ve been given (e.g.: back of module temperature) and characteristics of how it may be used.|
|Edge Computing||System||Advanced grid services and regulations such as Rule 21 require more intelligence at the edge. Rather than building these as monolithic software components, better interoperability is achieved by breaking them down into smaller, well-defined sub-systems.|
|OTA Updates||System||As software becomes more important in PV management, long-term interoperability can be maintained by enabling commodity hardware to be updated with over the air updates through the entire life of the plant.|
|Aggregation Services||Energy Micro-Service||When data has context across an entire portfolio, revenue streams and ancillary services such as virtual power plants become possible.|
|Wireless Communication||Technology||Without adding cost to gateways, it becomes possible to reliably interconnect all types of telemetry. Wireless communication can be used in parallel with existing wired networks to enable temporary or permanent use of new data sources.|
One owner-operator of a portfolio of utility-scale PV systems has already begun to use IoT architectures to address issues with data access. For a multi-MW plant, infiswift was engaged to make data previously confined to the local SCADA historian available remotely. A dedicated Cloud Historian was set up, and with no hardware deployed, data was extracted and made available via API to both a Financial Asset Management Suite and a Work Order Management System. Furthermore, the owner-operator was able to perform analyses which highlighted inconsistencies between the data and the monthly performance reports from the O&M provider. As a result, a number of underperforming strings were identified and fixed, resulting in increased performance.
Future developments for this owner-operator would likely include deploying additional software at other plants. Aggregating data from a full portfolio of plants together with external systems such as Financial Asset Management, Work Order Management, Grid Data Feeds, and Asset Tracking will allow the owner to visualize and access data in a central location from multiple inverter vendors, streamline analysis and make significant operational improvements in managing the full portfolio. Without proper data interoperability, significant overhead is added to completing these critical operational tasks.
As the PV industry matures, new problems have surfaced that require modern solutions to overcome them. Interoperability is the gating issue that can hold all sorts of other innovation back if not handled properly. Once resolved, however, more advanced services can be implemented that use accessible and complete data to optimize production and change the industry.