11.11.09Larsh Johnson, President and CTO, eMeter Corporation

Article Viewed 637 Times
4 Comments

The smart grid is a rapidly evolving combination of new intelligent networks, information systems and regulations. The promise of smart grid is to allow utilities and consumers to manage energy generation, distribution and usage in real-time, and to collaborate to drive energy efficiency. Arguably, this promise will be most acutely demonstrated in the "last mile" of the grid -- that portion encompassing the infrastructure and management processes where utility meets consumer and supply versus demand is optimized.

As the intelligent grid takes form, the organizational boundaries and business processes defining today's utility will be broken and reformed, with interconnected communities of collaborating suppliers and consumers. Utilities will migrate from an operational model designed to interact with consumers on a monthly basis to one that supports on-demand, two-way communications supported by information processed in real-time.

For the utility, the evolution to smart grid will involve a continuum of business process re-engineering (BPR), and information systems for grid management must change on a large scale to support this transition. In this environment, utilities and their IT organizations face material challenges:

• They must navigate this transition without impacting their existing business operations and customers.
• Too costly to replace, existing IT systems must be connected and extended to provide smart grid-useful information and automation for real-time business processes.
• Projects must demonstrate early return without becoming mired in complexity.
• Faced with continuous change in requirements that define "smart grid", utilities can ill-afford to make inflexible IT investments.

Despite these challenges, the transition to smart grid is underway and utilities cannot wait for the dust to settle before they begin deploying management systems that will drive operations in the immediate future. Implementing systems within a "smart grid ready" design framework can insulate utilities from complexity and risk in this fluid environment, and ensure utilities can adapt to meet the challenges and seize opportunities as the grid takes shape. This paper will outline the key components of the framework; discuss the requirements for, and respective benefits accrued, from each component.

Framework Definition

The Smart Grid Ready Framework is a strategic application blueprint that helps ensure grid management implementations deliver business value as expected. The framework is defined by five components:

• Support for real-time grid management
• Support for process interoperability and data exchange within and between enterprises
• Support for Web-based consumer engagement
• Inclusion of tools to manage deployment, operation and maintenance of AMI, HAN, and other smart grid systems
• Designed to Service-Oriented Architecture (SOA) standards

Utilities can use the framework as a top-down primer for evaluating, planning and deploying the IT systems that will power their transition to smart grid. Today these systems include Advanced Metering Infrastructure (AMI) networks and devices, Meter Data Management (MDM) systems, and Smart Grid Management applications. Together, these form the foundation for utility-specific applications such as Advanced Metering, Customer Service, Demand Response, Consumer Engagement or Distribution Grid Automation. Implementing systems within a smart grid ready framework can:

• Speed migration to AMI and smart grid management
• Ensure current and future smart grid technologies meet business expectations
• Insulate utilities from implementation complexity
• Minimize duplicate systems and application silos
• Enable systems to incrementally evolve without disruption
• Allow utilities/consumer collaboration to save energy

The following sections of this paper will provide an overview of the components:

1. Real-Time Grid Management

The need for a real-time infrastructure is a pervasive theme within the Energy Independence and Security Act of 2007 and U.S. Department of Energy specs for smart grid. This is driven by requirements for real-time monitoring and control of the distribution grid, time-based energy pricing and two-way consumer/supplier participation in the process. Effective adoption of smart grid and the processes that make it run must be based on a robust platform that can handle event-driven distribution of information and execution of business process in a way that is consistent, reliable and auditable.

Key functional attributes are:

• Continuous validation and processing of meter interval, sensor and other data
• Real-time handling of alerts, exceptions, messages and network events
• On-demand, bi-directional exchange of current, consistent information between systems and users
• Support for just-in-time operations between dependent systems and processes
• Online provisioning of newly installed meters, HAN devices, distribution switches and other hardware

Smart Grid Real-time Applications

In a smart grid context, AMI collection and recording of meter interval data is necessary but insufficient. Providing data based on monthly, even daily, estimates is not useful. Interval data must be validated continuously and made immediately available for any use -- as it is received. In addition, exceptions, alerts or messages required to invoke action must be processed on-demand. In the new world, demand response requires the marriage of time-based rates with real-time consumer usage patterns so informed choices are made about how to optimize supply, cost and usage. Dispatchable time-of-use or critical peak pricing are event-driven processes requiring a real-time infrastructure. Soon, real-time control commands and receipt verification of home and commercial devices must be supported.

Today's Real-time Challenges

Future capabilities notwithstanding, timing is everything, especially with a utility's first smart grid step: meter-to-cash applications. These deployments link meter installation, provisioning, interval data validation, billing and customer service processes in a dependent fashion. Just-in-time events must be handled and current data must be available at the precise time required by each process. If billing is based on stale data, or today's service logistics are executed on yesterday's status, the process will be inefficient and invariably generate complaints from customers. With a real-time infrastructure, outage alarms can be received and power restoration checked immediately. Service can be enabled while a customer is on the phone. Meters (and other devices) can be provisioned while service crews are in the area to handle problems, and cut over when ready.

Scalability for Real-time Processing

Real-time smart grid management must be designed to handle extreme loads. Systems must be able to accommodate growth from thousands to millions of grid devices and data points -- and potentially billions of pieces of dynamic data. Advanced Metering networks, MDM solutions and their dependent applications should be designed to validate and process these data on a continuous basis, rather than with a batch approach. Finally, vendor systems should carry an independent benchmark to validate capability to handle multi-million meter deployments.

2. Multi-Process Interoperability

To effectively manage operations, improve customer interaction and increase energy efficiency in a smart grid context, utilities must extend their IT systems to support process interoperability and data exchange within and between enterprises.

Interoperability Within the Enterprise

Today, many of a utility's IT systems are monolithic applications designed to manage single discrete processes (e.g., billing, service logistics, call center, asset management, etc.). Process interoperability is driven by manual or periodic batch data exchange and synchronization. To manage events and exchange data across systems within the enterprise in an end-to-end flow, feeding each dependent process at the optimal time, utilities must re-engineer toward a more real-time, integrated view.

For example, when deploying initial AMI meter-to-cash applications, systems should be extensible beyond the stovepipe process of collecting interval data and feeding the billing system. Deployment should be easily extended to incorporate processes such as asset lifecycle management, service requests and incident resolution.

A smart grid ready infrastructure should also support availability of current data from multiple systems to complete real-time call center processes such as new accounts, turn on service or customer moves all while the customer is on the phone. The system should be extensible to migrate this process to a Web-based self service model.

Interoperability Between Enterprises

In an extended enterprise model in which several companies are involved in energy distribution, process execution must span company boundaries and access to data must be managed to each entity's authorized view. Beyond traditional utilities, participating entities can include aggregators, retailers, distributors, market operators, service providers and consumers. For example, a merged utility may need to support multiple operational models serving customers and regulatory bodies in multiple states, provinces or countries. A meter operator may have to deliver data from a common source to distribution companies, energy retailers and to a market settlement agency -- each with differing requirements.

These applications require interoperability between multiple legacy CIS, asset management, logistics systems and potentially multiple AMI/smart meter infrastructures, each with differing process flows and data models. With state-of the-art deployments, an MDM system is the core component to enable this integration. Integration should be enabled without disruption or changes to in-place application logic. A single platform should support multiple instances of "customized" process flow and provide the data management, audit trail and access control to deliver appropriate, authorized views of information created by the overall system. This includes automated aggregation of information from multiple operational systems to provide consistent datasets for planning, pricing, forecasting and other analytics.

Incremental Deployment

Finally, multi-process systems should be configurable for modular deployment: allowing a tailored implementation to meet the functional requirements of a pilot, but scaling in-place to add new capabilities. This will enable the utility to "think big" while starting small, enabling it to demonstrate early success and then scale fast.

The second part of this two-part article will cover the final three components: consumer engagement, tools for deployment of smart grid devices, and adaptive, Service-Oriented Architecture (SOA).

We know you have something to say!

There is an immediate need for articles on the hot topics in the Power Industry! EnergyPulse, like no other publication, also provides a means for our readers to immediately interact with experts like you.

Contribute Today!

Please view our Author Guidelines and send submissions to the editor.

Reader's Comments

Date	Comment
Bob Amorosi 11.11.09	Nice article describing the complexities in managing the smart grid transformation of a utility company. I hope the future Part 2 of the article will entertain the subject of how a utility company will bear the huge costs of deploying smart grid, especially with equipping residential consumers with the in-home technology tools they need to engage them much more in a smart grid.
German Ibanez 11.18.09	Thank you for your article. In terms of data storage, how long is real time data kept? Compared to current technology grids, what is the additional cost in data storage to implement a smart grid system?
Len Gould 11.19.09	German: I have some preliminary answers for that. Let's assume the following table structure for this data: CREATE TABLE METER_DATA (METER_ID NUMBER(10,0) NOT NULL, READ_DT TIMESTAMP(0) NOT NULL, PREDICTED_WATTS NUMBER(10,0) NULL, CONSUMED_WATTS NUMBER(10,0) NULL, CUSTOMER_ID NUMBER(7,0) NOT NULL, SUPPLIER_ID NUMBER(10,0) NOT NULL PRIMARY KEY (METER_ID, READ_DT) USING INDEX PCTFREE 10 INITRANS 2 MAXTRANS 255 TABLESPACE avail_index STORAGE ( INITIAL 163840 NEXT 163840 PCTINCREASE 0 MINEXTENTS 1 MAXEXTENTS 2147483645 ) ) We can presume that the database system can store numeric data in internal binary numeric form, so any positive integer up to 2 billion will require only 4 bytes, a Date+Time will require 8 bytes, each meter has an associated scaling factor to handle a broad range of possible customer sizes, and the entire data structure is stored on the primary key index. This results in a row storage requirement of 30 bytes per row. The outcome is that 1 day's 15 minute interval readings for 1 million customers generate 2.9 gigabytes. 31 days = 90 gig. 12 months = 1,100 gig. Assuming a 250 gig disk drive costs $200 (my desktop at home has two of those), and all data is stored in a RAID1 array which duplicates all data, requiring 2,200 gig per year, the total year's disk storage costs 2,200 / 250 * $200 = about $2,000. That's about $0.002 per customer per year. Of course there will be additional data structures required, so let's quintuple that amount, to $0.01 per customer per year.
john Marsh 5.26.10	urrent smart meter technology is still modeled after the old conventional electromechanical service meters where each has a electronic display on its web templates face that shows cumulative energy consumption at that customer's site. To credit the customer for a portion of the displayed total home based business each time another customer's vehicle is charged would mean the meter must also adjust downwards its display to reflect only the energy billed to the property owner. Similarly the meter at the vehicle owner's site would be appropriately adjusted upwards to reflect his billing. If these meter displays are not adjusted, it opens the door to widespread billing disputes between customers and the utility, which is a nightmare the utilities will fiercely avoid at any cost.Moreover, here in Canada we have the Measurement Canada agency of our federal government that tests and approves all utility meter designs before authorizing them for commercial use on homes by utilities. One of their criteria for certification of smart meters is they will only mobile homes for sale permit meters that are tamper proof, where customers AND the utilities are both prevented from altering a meter's internal programming for any reason, especially anything that would change the total energy consumption displayed. As such a utility cannot even upgrade a meter's software for any reason without replacing the cheap travel insurance meter, even when the meters are designed to be field upgradeable using simple software reprogramming.

Do you agree or disagree with this article? Send in your own article.

Add your comments:

Please log in to leave a comment!

back to top