The job of the distribution grid planner has become more challenging in recent years. Three interrelated forces are contributing to this situation.
1. The proliferation of applications to connect distributed energy resources (DERs). From storage (behind/before the meter), rooftop solar, electric vehicle (EV) chargers, to the sensors and communication technologies that enable smart cities, we face unprecedented demands on the distribution grid. The interaction effects can be quite complex, and the feasibility of any DER must be examined carefully to maintain grid reliability. However, current approaches to feasibility studies have not scaled at the same pace as DER applications, creating a backlog that frustrates solution providers, their customers, and policymakers who wish to enable a grid that is resilient, “green,” accessible and cost-effective. This is a serious impediment to grid modernization and the roll-out of the Internet of Things at the city and national level.
2. Data. The fundamental inputs to DER feasibility studies include diverse sets of data such as geolocation information, asset type, circuit loads, traffic patterns, solar insulation patterns, and socio-demographic data. Some of these data are located within the governing organization, however most of these datasets are often not integrated or interoperable, either because of legacy formats and technologies, or as mandated by policy as a security measure (or some combination). The result is a fractured information landscape that requires the grid planner to access multiple databases in series. Even with fairly routine feasibility studies, the time to assemble data sets needed to begin an assessment can be prohibitive – on the order of a day or more per project. Beyond this, there may be additional datasets that exist outside of the control of the grid planner that may be useful for current and future decision making. These data, mixed with distribution system operator (DSO) collected data, may offer valuable insights especially as interacting services become required, this comes with substantial risk and caveats.
3. Data security, data governance and privacy. Grid data is highly sensitive, and has been traditionally contained in specialized servers, most often on premises. Such data does not readily lend itself to comingling with outside sources, given security protocols and data policies. Furthermore, any data that could be used to infer personally identifiable information either directly or via derivative metadata, such as census data, maintenance personnel data, location or other kinds of behavioral information must be handled with appropriate privacy protocols. The general approach has been to have a bias against using additional datasets and related assessment algorithms given the high barrier to secure and safe data sharing. What this does is diminish the possibility to experiment with better assessment tools for current feasibility studies, or to look beyond such studies to develop data-driven predictions on how the grid should evolve. The issues of data governance are compounded by national and international privacy laws like the EU’s GDPR, which set severe penalties on enterprises that violate personal data rights. This is fundamentally a digital rights management problem.
In response to these challenges, German utility innogy SE (which owns Westnetz, the largest DSO in Germany), created DigiKoo, an automated grid information access platform that gives DSOs and other customers the ability to easily and securely access and share data within their organizations and beyond. At the heart of DigiKoo are a set trusted intermediary web services that securely store and manage disparate data sets in a fashion that complies with national privacy and data management laws, consumer protection frameworks and enterprise policies that the respective dataset owners specify. DigiKoo does this by using Intertrust Modulus, a data rights management platform built by Intertrust Technologies Corp, innogy’s close partner. DigiKoo provides a trusted analytics framework on top of Modulus curated datasets that allows for private and secure computation that gives planners the ability to quickly assess future scenarios that may affect the grid.
At its core the DigiKoo platform combines data virtualization and data governance with machine learning and physics-based power flow models. In addition, it provides software developers with a rich data and technology ecosystem that enables them to develop solutions and applications without requiring accessing DSOs directly. This enable efficient and quick application development and a route to market.
Data virtualization refers to a data management process that copes with the various data formats, software and operating systems, the result of which provides an error-free, and consistent view of all underlying data. This is important when multiple legacy systems within a company need to be accessed, which are not readily compatible, or when attempting to access data across companies where such challenges could be even more pronounced.
Data governance, or governance, is the process of ensuring the right entities gain access to the right data in the right context. This is done within DigiKoo by creating and enforcing fine-grained digital rules for managing access to data based on polices that DigiKoo and its partners define or program in the system. Governance takes place in a protected processing environment that ensures that the rights management occurs in a secure and predictable way. DigiKoo implements data governance using the Intertrust Modulus platform, which provides rights management capabilities that allow fine grain data control of data items, implementation of privacy and data policy and blind-analysis of data sets. Modulus not only allows for context-specific data management, but also for multi-party data analysis in a way that protects the data in a manner consistent with the data owner’s wishes.
The DigiKoo analytics framework provides several standardized applications that includes a product that identifies the cost optimal installation location of private electric vehicle charging stations, using a combination of powerflow data, socioeconomic data and geo-locational data of garages. The German government has allocated US$1 billion of support to upgrade garage infrastructure and this application can help to allocate these subsidies. A solar and storage application identifies the potential for rooftop solar at the circuit-level, while providing the likelihood of growth patterns and related grid impacts based on affinity measures of distributed generation and other sources. From a data quality perspective, the DigiKoo platform can visualize on a map where existing technical documentation may be incorrect and offer how to correct such mistakes.
DigiKoo continues to experiment and build its offerings through pilot projects. However, the current results are promising: The average time to complete a feasibility study for various DERs has been reduced from an average of 10 hours to 5 minutes. As utilities around the world adopt data-driven technologies to enhance planning in a world of smart cities and progressively more connected infrastructure, tools like DigiKoo will become as essential as volt meters in grid planning. This is especially true in counties with strict privacy regimes and in places where competing utilities must collaborate to provide planners with common data-driven interfaces to do their jobs.
