Your Shopping Cart

It appears that your cart is currently empty!

CONTINUE SHOPPING

Big Data Analytics Platforms Architecture Requirements and Analysis Techniques

Abstract

•    Current - Architectural approach for Advanced Analytics - best practices and challenges.
•    Future - The rise of Infonomics – data monetization framework.

There are reportedly over 2,500 PMU devices in operation – and still counting - in the U.S. alone, and we’re being faced with how the PMU-generated data is reshaping the utility industry by “Big Data” analytics. The skyrocketing amount of PMU (synchrophasor) data creates demand for the new analytical techniques, but it is also raising questions about the acquiring, transferring and storing the high frequency data. Current IT infrastructure was not created to handle constantly streaming data from PMUs. Hence, the power system engineers and operators are calling for support from IT to meet the new needs. There are many solutions on the market, but not all of them address the requirements of modern WAMS and the prospective requirements imposed by WAPS/WACS transition.

Our implementation experience and expertise brought us to the following combination as the winning, most promising architecture and approach for the future to support the PMU data growth:
•    Micro-services approach to data intake, cleansing, processing and storage due to massive horizontal scalability of distributed architectures;
•    Hybrid cloud IaaS solutions with highly secure virtual cloud platforms, acting as a private network designed to work with sensitive data;
•    Integration of PMU Data processing and Decision-Making solutions with Data Science Workbenches, providing an in-house data scientist with a set of tools and infrastructure to build new models
•    Utilization of Advanced Analytics techniques like Machine Learning and Artificial Intelligence  


The existing analytical solutions have been presented by many well-developed synchrophasor applications, e.g. low-frequency oscillations analysis, angle stability monitoring, etc. However, the variety of new possibilities keeps increasing, and we are seeing a growing interest in the equipment condition monitoring, power plant and substation monitoring, real time control and system recovery, etc.
But, PMU based analytics expansion is facing several obstacles that require attention:  
•    Lack of a centralized repository for PMU data located in the cloud
•    Increased requirements for data quality – accuracy, completeness and timeliness
•    Protection against unauthorized access

A new and an ideal data platform to overcome the above obstacles would pave the way for adding other types of data (metering, weather, load data) to the platform and creating a foundation for monetizing utility data - Data-as-a-Service (DaaS) for energy and utility industry.
Monetization of data assets, or infonomics, was coined by Doug Laney, Data Strategy Analyst with Gartner, and it already became a driving force for the successful initiatives within telecommunication, healthcare and travel industries.  
Applying infonomics principals to energy and utilities would allow for:
•    Bringing additional revenue streams to utilities from power quality data, voltage profiles and customer data
•    Providing new business insights for the equipment manufacturing companies, service providers and financial organizations by utilizing predictive and prescriptive analytics
•    Addressing demand from R&D for the different levels of information granularity - raw data, aggregated data or specifically formatted data
The infonomics-based DaaS platform:
•    will be built on an integrated distributed system for data acquisition, transfer, storage and analysis;
•    will have an ability to expand the available storage to contain any type of data with the evolution of measurement infrastructure;
•    will support analysis techniques growth depending upon the market requests

 

Speaker

Viktor Litvinov

Vice President

GRT Corporation

Mr. Litvinov is an accomplished entrepreneur skilled in startup, business development, and operations taking vision to practical concept through successful implementation.

He led both national and international programs including Business Intelligence, Advanced Analytics and Information Security

Mr. Litvinov founded and led several successful startups including GRT Corporation that was included in Deloitte & Touche Technology 50 and 500 list and Inc.500 Fastest Growing Companies list.
Mr. Litvinov carries Ernst & Young Entrepreneur of the Year Award.

Prior to founding GRT, Mr. Litvinov had over fifteen years’ experience in senior software engineering and management positions.

He holds an MS degree in Electrical Engineering.