Smart Grid: Energy Efficiency Requires Data Efficiency

A new approach to managing the electric grid has started to change completely the way that power is generated and used. The vast quantities of data and information generated by the emerging ‘smart grid’ will require major changes in the architecture and philosophy of traditional monitoring and control systems used by utilities; in order for them to be able to collect, analyze, secure, and visualize this data and information, while making it actionable. The first step to a ‘smart grid’ is being able to ‘see’ the grid and this involves collecting millions of data points in real time, analyzing this data and then displaying it in such a manner that it is actionable; not overwhelming.

The phrase 'smart grid' describes modernizing the transmission and distribution grids to facilitate greater competition between providers, to enable use of variable energy sources, especially renewable   and 'green,' to create the automation and monitoring capabilities needed for bulk transmission at transcontinental distances, and to enable market forces to drive energy conservation and compliance.  These disparate goals require integrating generation, transmission, distribution, and meter data management into a single system-one which must be built upon a secure enterprise infrastructure that can collect, store, analyze and visualize huge quantities of data.

The major driving forces to alter the current power grid can be divided into four, general categories.

1. Increase reliability, efficiency and safety of the power grid while increasing the use of renewable energy sources (prevent outages; lower CO2; reduce energy  bills)
2. Enable decentralized power generation so homes can be both energy client and supplier (provide consumers with interactive tools to manage energy usage)
3. Include flexibility for clients to choose power generation suppliers (this enables distributed generation, solar, wind, biomass)
4. Create new, clean energy jobs related to renewables, plug-in electric vehicles, etc.

While most of the focus around the smart grid seems to be on the consumer, the business driver most likely to initiate the move to the smart grid is the first one described, and most smart grid features such as smart meters serve the energy efficiency goal. When businesses and consumers see a direct economic benefit from energy efficiencies, they include energy cost of operation into decisions.  Long term, this will reduce the spinning reserve electric utilities have to keep on stand-by, as the load curve will level itself through market forces.

Any type of power infrastructure using intermittent renewable energy resources (wind and solar) must have the means to reduce electrical demand by shedding loads when natural phenomena needed to generate power do not occur. Smart grid customers have tools to use real-time electricity pricing or incentive-based and emergency load reduction signals. When they can see electricity prices rise at times when the required natural phenomena are not available, consumers will decrease consumption. 

In addition, it must be said a smart grid will improve the availability and use of renewable energy because it will improve the reliability of the energy sources through better predictability; for example a single wind tower/turbine may be unpredictable, but a wind farm with multiple units gets more predictable and reliable, and when the utility operators can see an aggregated group of wind farms the picture becomes even more predictable and reliable.  Add in weather effects and study them across a state or country, and the predictability, through forecasting techniques, goes up even further, as does reliability - the smart grid will look at all these resources and accomplish this.

Overall, the potential benefits to be realized from the smart grid are immense.  Robust two-way communications, advanced sensors, and distributed computing technology will improve the efficiency, reliability and safety of power delivery and use-saving billions of dollars in operating costs over the next two decades; while advancing technologies for other uses.  In essence, the smart grid constitutes a digital upgrade of transmission and distribution facilities to optimize operation and open new markets for alternative energy production. The increased efficiency and reliability of the smart grid is expected to save consumers money and help reduce CO2 emissions

But these benefits require a data infrastructure with a scale and speed unique in industrial applications.

The infrastructure for the smart grid must be able to operate in real-time with absolute reliability over large distances, be able to handle disparate types of inputs, and be able to interact intelligently with the edge user at the meter (and possibly beyond).

A secure, real-time enterprise infrastructure for smart grid operations
The OSIsoft PI System provides utilities a secure, real time enterprise infrastructure for smart grid operations.  It spans generation, transmission, distribution, and meter data management. Approaching three  decades of proven performance across millions of utility data streams has fortified PI to meet the massive data and information demands of smart grid applications-requirements including demand response, distributed generation integration, renewable resource management and advanced distribution monitoring and control.

• Proven, scalable, secure, mission critical enterprise technology
• Rapid and sustained creation of value
• Secure and reliable smart grid infrastructure
• Unmatched partner eco system
• Sustainable and flexible innovation platform

A proven enterprise technology
The PI System is an infrastructure of interoperable products which are highly configurable, easily deployed and user friendly.  With over twenty-five years in business, 14,000 installs of the PI System technology, and core resources focused on technological enhancements, OSIsoft has developed a highly scalable, mission-critical system infrastructure that enables an enterprise to make business decisions in real time using operational data-putting the power of data in the hands of the decision makers at all levels.  It is an industry standard in the utilities industry:

• 17 of the top 20 owners/operators of wind generation assets in the world leverage OSIsoft's PI System
• Approximately 60% of the average daily power generated in the United States is being monitored by OSIsoft PI Systems
• 100% of the ISOs and 50% of the RTOs in the United States leverage OSIsoft's PI System.

Secure and reliable smart grid infrastructure
The PI System incorporates the key attributes for smart grid initiatives.  Currently it provides data access, mobile data capture, analytics, web-based visualization, and data management in a highly flexible and open environment in a number of areas in the utility.   In addition, OSIsoft has significant experience in renewable integration.

Rapid and sustained creation of value
The PI System manages the underlying details to provide a robust, event and real-time environment for a real-time and historical perspective of the process being managed-enabling  better business decisions from the time of installation.   The PI System manages a myriad of details like high-speed data retrieval and storage, cross time-zone information management, security, and seamless integration of multiple data sources, event propagation throughout an IT infrastructure, all through  desktop and Internet presentation of information.  Hosting all of the data in one central, mission critical repository provides users with a complete operational overview and eliminates the effort spent moving data between systems for reporting and prevents data fragmentation.  The PI System allows logical presentation and visualization of data and information.  With the PI System users are presented with "one version of the truth" because the same accurate and secure information is always available to all users.

Sustainable and flexible innovation platform
OSIsoft operates on the premise that system architecture must integrate the different aspects of the smart grid.  A data integration strategy for the smart grid needs to include:

• Distributed intelligence and analytic capabilities from corporate and operations down to the substation level
• Multiple data management zones (DMZs) based on trust levels and sources of data - generation, transmission, operations, substations, automation, metering (C&I versus Residential); and potentially different DMZ's depending upon the originating control/set point signals. The sources of control-paths and the elements of security will define these DMZs.
• Data collection and reconciliation of disparate systems from generation, transmission, operations, substation automation, metering, corporate business centers, and other market and economic data-including continuous updates to all parts of the distributed intelligence system comprising the smart grid.
• Continuous/Complex Event Processing (CEP)-an ability to extract and present intelligent information to users and to other business and analysis systems, which includes, at any single moment, available Distributed Energy Resources (DERs) and Demand Response (DRs); Aggregated KPI's for system conditions (dynamic ratings), available green energy and other economic detractors-plus rapid outage response, and notifications for predicted asset failures. All such business rules must be easily and rapidly modified based on best-system-practices/observations and statistical calculations, without forcing a redesign.
• The ability to point & click for an informative, at-a-glance view of the smart grid KPI's
• Incorporation of advanced technologies- such as Remedial Action Schemes, WAMS, and C&I Curtailment programs
• A push for all automation technologies to comply in real-time or near real-time, such as the timely management, reconciliation, and analysis of advanced metering interval and event data
• Health monitoring of the IT infrastructure reliability and performance (including CPU's, communication paths, DMZ's, and interfaces for security)
• Utilization of industry standards, when appropriate

These are some of the smart grid initiatives OSIsoft users are currently engaged in:

Fleet Performance Analysis
Fuel Optimization
NOX Reduction
CBM/RCM
WAMS
Remedial Action Schemes
Load Curtailment Monitoring
Renewable Integration
System Loss Evaluation
Substation Monitoring
Distribution Transformer Load Monitoring
Identify Equipment Failures
Customer Energy Management
Monitor IT and Communication Health
Smart City
Substation Integration
Critical Facilities Management
Data Center Management

Unmatched partner eco system
From its inception the PI System was designed for interoperability, connectivity and rapid integration by utilizing the resources already available in the user environment, such as the Microsoft platform,  and by interfacing with all standard data sources.  The enriched partner ecosystem in which PI operates enables it to provide unique advantages in moving toward the smart grid.

Standards – OSIsoft's long-standing reputation for advancing and embracing standards, and years of experience in operations is reflected in the PI System today. 

In addition, a partnership with SISCO enables OSIsoft to support electric utility industry standards like CIM, ICCP-TASE.2 and other open international standards. OSIsoft and SISCO have extensive experience working together to deliver solutions to electric utilities.  SISCO also supports  IEC61850, IEC61970, IEC61968 and IEC61970 standards with their PI adaptor.

Interfaces – OSIsoft has extensive experience in developing interfaces.  There are currently over 480 standard OSIsoft Interfaces and COM Connectors that provide connectivity to most energy management systems, SCADA, PLC, DCS, lab, and other business information systems.  OSIsoft offers standard interfaces to all major automation vendors' systems.  OSIsoft also has interfaces to support numerous standards including: OPC, ODBC, XML, and ASCII.  OSIsoft Interfaces provide high-speed links to external data sources, sending real-time, fault-tolerant data permanently to PI.  COM connectors are used to make data from foreign data sources presentable through the PI System without storing the information in PI.  Many OSIsoft interfaces support collection of sub-second data. Next generation interfaces automatically discover and set up the devices/meters in PI and begin collecting data without human intervention

Scalability and Accessibility – OSIsoft has proven the ability to perform at large scale systems. A single  PI System is able to sustain  inputs at 250,000 values/second and 8 million values/second out, and it is lab tested to 21 million data streams (real-time) on 64-bit Windows server.  The data is accessible to users as fast as collection systems can provide it. The design gives users and analysis systems rapid access to years of data for ad-hoc analysis or complex algorithms.

Flexibility – OSIsoft 's core competency  and focus on developing interfaces and robust infrastructure provide users the ability and flexibility to manage new devices, data frequencies, applications, and new business rules.   Two example attributes are 1) the ability to make mass changes to device attributes and configurations and 2) manage multiple network views such as an asset, static and dynamic system views.

Continuous Event Processing – Asset models and business rules are easily configurable and real-time calculations and equations are constantly being processed for events and notifications to the appropriate people/applications. Market modeling and seasonal dynamics will be highly dependent on a subset of calculated data-especially with regard to self-healing grids and dynamic response to predicted and real-time system constraints or outages. OSIsoft cooperated in the development of and is incorporating advanced low latency Complex Event Processing (CEP) technology from Microsoft  into the PI System;  this CEP platform will capture data from system-level, application-level and external events and correlate them into patterns to provide users with better insight.

Security – OSIsoft strategy is to be a good corporate citizen in the cyber security community.

• The PI system is an interface focal point for connectivity with critical infrastructure systems. PI is deployed in a topology that creates one or more cyber barriers designed to prevent access into critical systems. Data archived in PI is then used to service users and applications. This mechanism is a key security aspect; essentially using PI to reduce the critical infrastructure attack surface. Fewer access points means less to monitor and manage... resulting in less risk.
• The PI system is a market leader; as such the system is subject to expert security assessments. Major versions include PI version 3.3 by Electric Power Research Institute and PI version 3.4 by Idaho National Lab. The assessments are collaborative and help improve the overall security posture.
• Critical infrastructure systems are also able to access the PI system. This feature is often used to support optimization and supervisory control logic. PI monitors the control logic to assure integrity. For example if a smart grid customer changes a temperature setting, PI is able to confirm the new setpoint and monitor response to the change.
• The PI system includes several technologies for data replication. PI to PI is often used between geographically isolated PI servers. OSIsoft strategy leverages infrastructure security services including VPN or IPSEC technologies to protect data in transit.

NERC CIP includes both prescriptive and non-prescriptive, user defined security controls.  The PI system meets the prescriptive requirements and can be used to support requirements that involve site specific implementation.  Specifically, the IT Monitor components facilitate convergence of required cyber and physical security data.

John O'Shea
Senior Vice President Sales and Business Development, OSIsoft, Inc.