Case Study: Oncor's AMI

Sponsored by


Oncor, the largest electric transmission and distribution utility in Texas, accomplished a huge feat in 2012 when it completed the four-year deployment of advanced meters at more than 3.2 million homes and businesses across more than 100,000 square miles of service territory.

Case Study Oncors AMI

The successful deployment was the largest installation of the most advanced meters in the nation.

But along with the advanced metering infrastructure (AMI) deployment came the need for a cost-effective way to link results from infrastructure-heavy urban environments with that of less-dense areas of the company's footprint.

Case Study Oncors AMI

The problem

After the Oncor AMI project was complete and operational in many urban areas, the utility determined that the same system design used in more populated areas would not work in sparsely populated regions because projected costs per meter served would far exceed the cost per meter served in urban areas. Design methodology changes were needed to achieve a cost-effective deployment in rural areas where meter density is low.

The solution

Oncor developed a plan for rural areas that incorporated lessons learned in urban deployment using internal engineering resources and applying advanced radio frequency (RF) engineering design methods and tools.

Like most things in the RF world, design is an iterative process.

But with the many millions of potential RF links, some form of design automation software is critical.

An assessment was taken of engineering software tools that were capable of performing RF analysis of mesh networks and the optimal design methodologies for Oncor's geographically diverse service area.

The utility selected EDX SignalPro software, which combines geographic information system (GIS) mapping capabilities with propagation analysis and automated network layout tools that provide for efficient dimensioning and placement of infrastructure equipment.

The software incorporates a 3-D model of the geographical area of interest, which is built up using a digital elevation model, land-use data and building and structure data.

The 3-D model is critical so the physical issues that affect the performance of a complex AMI network can be accounted for in the design, particularly for rural areas with variant terrain.

After a model is created, the planning of the complex mesh architecture can be accomplished in a way that respects the unique physical issues of the service area and the capacity constraints of the vendor's equipment.

The product is designed to support large-scale AMI mesh networks and accurate modeling of morphologies from the dense, large pine trees of East Texas to the sparse vegetation and plateaus of West Texas.

AMI Mesh Architecture

As shown in Figure 1, the AMI mesh architecture implemented by Oncor consists of a data processing center, more than 400 collection radios called collectors, 11,000 repeaters or routers and electric meters on customer premises.

1 Oncor's AMI Mesh Network
Oncor's AMI Mesh Network

The collectors are connected to the data processing center by circuit or cellular modem.

Meters can report directly to a collector, router or another meter, adding complexity to the design process.

One or more routers can communicate directly to a collector.

The maximum number of router hops should not exceed 14.

The maximum reliable router hop count is limited to 10 in densely forested areas because the router-to-router RF connections often experience signal degradation.

Data Gathering

To perform an effective RF design, a utility must gather data from various sources.

The first set of data consists of spreadsheets that detail the meter number and location gathered from internal utility systems.

Another spreadsheet contains potential router-mounting asset data, such as pole information and location of all poles with transformers and streetlights in the design area, and was generated from distribution asset records.

Poles with streetlights and transformers are used as router locations because they have accessible power.

The last set of internal data is a list of substations and the location of each substation that can be used as a collector site.

AMI System Specifications

The specifications of the radios for the collectors, routers and meters are obtained from the manufacturer of the radio equipment.

The minimum data required includes the transmit power of all radios, antenna gain, polarization and receiver-required signal-to-noise ratio.

Network specifications provided by the manufacturer included hops for failure of a transmission to or from a meter, the number of meters that can report through a collector and the optimal spacing of routers around a collector.

RF Modeling, Analysis

After all site and equipment information is gathered, the information can be loaded into EDX SignalPro.

The next phase is to set the RF propagation tool parameters and databases for the design area.

The required information includes a database that contains the terrain elevations throughout the design area, street and road maps and a clutter database that models land usage.

Clutter data consists of desert, rangeland, farmland, forests and urban and suburban areas.

Each of these clutter types has a distinct effect on RF propagation.

The modeling tool accounts for each clutter type in a different way.

Figure 2 shows the terrain denoted by the bright green at the higher elevations and blue at the lower elevations.

Terrain Elevations
2 Terrain Elevations

Figure 3 shows clutter on top of the terrain. Dark green denotes forests, light green is open or rangeland and blue is lakes.

Terrain Clutter
3 Terrain Clutter

Once the parameters and terrain or geographic databases are defined and mapped, the RF design tool is used to automatically select router locations.

The RF modeling tool accomplishes this by:

  • Calculating the signal strength from the candidate router location to meter level along a radial from an omnidirectional antenna in 1-degree increments along the 360 radials around the antenna;
  • Using the terrain database to determine where terrain blockage of the signal occurs;
  • Using the clutter database to represent land coverage structures and types that will block or attenuate the RF signal;
  • Determining when the signal strength transmitted to or from a meter drops below an acceptable level;
  • Using point-to-point link calculations so a meter can hop through an adjacent meter to a router; and
  • Using automation techniques to analyze a large set of candidate routers and identifying the preferred router locations.

Once the router layer is defined, the design tool can be used to determine potential mesh routing from the routers back to the collectors.

The modeling tool uses similar modeling techniques described, but the hop count system parameter must be included in the router mesh layer.

Additional design considerations include:

  • Locating repeaters and routers on top of plateaus and other high spots to take advantage of the terrain;
  • Using company assets such as towers for collector locations; and
  • Placing intermediate routers on links that are encumbered by terrain or clutter.

Design Example

An example of the design analysis process is shown in Figures 4, 5 and 6. Figure 4 shows the location of all the premise meters that must be accessed through the AMI system.

Premise Meters That Must be Accessed Through the AMI System
4 Premise Meters That Must be Accessed Through the AMI System

Figure 5 shows the location of all the poles with transformers and street lights.

All Poles With Transformers, Streetlights
5 All Poles With Transformers, Streetlights

Figure 6 shows a completed design with all the collectors and routers placed to optimize meter coverage and minimize infrastructure costs.

 Completed Design With all Collectors, Routers
6 Completed Design With all Collectors, Routers

The lines are the mesh links between routers and routers and collectors.

A significant amount of link redundancy is evident.

Link redundancy is desirable to mitigate the effect that the failure of a single router can have.

Also, this design has two collectors that enable a ring topology.

Because a ring is formed, a collector failure can be overcome by routing around the loop.

The hop count will increase, but reliability is maintained.

Figure 7 is a detailed view of mesh linking.

Detailed View of Mesh Linking
7 Detailed View of Mesh Linking


The design process is complex, and a cost-optimized solution cannot be achieved by simple rule-of-thumb design.

Through its aggressive efforts and significant planning, Oncor implemented a new structure that provided a great cost savings in overall deployment and in rural areas.

The original infrastructure plan based upon the urban design model called for three collectors and 245 routers.

The final implementation, based on the method described here, used two collectors and 115 routers.

The savings in capital was some $340,000 for the 5,200 meters in the service area, or $65 per meter.

This design was successfully deployed, and all meters were served without further optimization.

Subsequently, this method was used for the remainder of the Oncor AMI infrastructure design in rural areas, which resulted in a cost savings of more than $4 million.

Oncor also is using the software and expertise developed during AMI deployment to optimize distribution automation RF mesh design and plans to use the tool for microwave path analysis and transmission dynamic line rating RF-based monitoring systems.

Richard Schertz is a senior network architect and professional engineer at Oncor. He recently implemented the RF design standards and processes for Oncor's AMI network and was instrumental in the RF engineering for and deployment of Oncor's smart grid networks. He has a bachelor's degree in electrical engineering from Texas Tech University and an MBA from the University of Texas at Dallas.

Greg Leon is director of business development at EDX Wireless and focuses on assisting smart grid vendors, consultants and utilities to better design and deploy their wireless networks. Leon has a bachelor's degree in biomedical engineering from the University of Iowa and a master's degree in telecommunications from the Interdisciplinary Telecommunications Program at the University of Colorado-Boulder, where he was a Digital Energy Fellow.

More PowerGrid International Issue Articles
PowerGrid International Articles Archives
View Power Generation Articles on

Sponsored by



FERC grants California ISO waiver of EIM pricing parameters

07/30/2015 FERC on July 21 issued an order conditionally accepting, subject to a further compliance filing, readiness provisions for ...

Ameren seeks eminent domain authority for Illinois Rivers Project


ATXI requested expedited approval of the petition, with a proposed deadline for decision on Sept. 10

Power demand records continue, ERCOT asks for conservation

07/30/2015 Electric demand this week has reached three consecutive July records, breaking 67,000 MW for the first time since 2011 

IPL to stop burning coal at Harding Street Station Unit No. 7


IPL also received approval to install and operate wastewater treatment technologies at Harding Street

Hopes rise for offshore wind as U.S. builds its first offshore wind farm

07/30/2015 U.S. Secretary of the Interior Sally Jewell said Monday that lenders, regulators and stakeholders can now see a path forwa...

SunPower buys 1.5 GW U.S. solar power plant pipeline


With the acquisition, SunPower has assumed ownership of projects in varying stages of development across 11 states

VIDEO: Riverside, California buys renewable energy from Iberdrola


Iberdrola Renewables will deliver about 101,000 MWh of firm renewable energy annually  

View more Job Listings >>



Solar + Storage: Capturing Opportunities and Overcoming Challenges

The convergence of solar and energy storage applications is under way. The combined application o...

Storage Enters the Global Stage: What’s the value proposition for your region?

For electricity storage, the events of the last two years have shown that the skepticism around t...

How to deploy mobile forms + devices to speed up field service processes & reporting

Overcoming the pain points associated with any process is much simpler and more cost effective th...

Beyond Integration: Three Dynamics Reshaping Renewables and the Grid

In a unique industry research initiative, DNV GL gathered views from over 1,600 energy sector par...

Engaging the Small and Mid-sized Business (SMB) Marketplace

This session will outline how Duke Energy leveraged Voice of Customer insights, previous experien...


Nuclear Horizontal Modular Storage (NUHOMS®)

AREVA TN’s NRC-licensed NUHOMS® system ensures the highest safety and lowest dose performance. With no tipping, tripping and sliding, you have the...

Decontamination & Decommissioning

Decommissioning strategy development, cost estimates, plant modifications; system and building decontamination; RV segmentation; large component re...

Cavitation Peening

AREVA developed its innovative solution for asset management, significantly extending the life of reactor components. Cavitation Peening is a perma...

ATRIUM™ 11 for BWR Fuel

ATRIUM™ 11 design fuel – in operation since 2012 – provides significantly lower fuel rod power levels for power uprate margins, operational flexib...


EtherMate® Industrial Ethernet - IO Modules

LioN-M ProfiNet device with 16 digital I/O channels, channels can be used universally as inputs or outputs, M12 socket, ro...

TRM® 2.0 - Test Results Manager PC Analysis and Reporting

Test Results Manager 2.0 (TRM® 2.0) is an all-in-one analysis, viewer and reporting tool designed for use with NOYES fiber...

OFL280 FlexTester - Handheld OTDR

AFL's OFL280 FlexTester offers an unmatched combination of fiber test functions, ease-of-use, portability, and value.

ADSS Mini-Span® Fiber Optic Cable

AFL Mini-Span® All-Dielectric Self-Supporting (ADSS) cable is designed for outside plant aerial and duct applications in l...

POWERGRID International

July 2015
Volume 20, Issue 7


June 2015
Volume 93, Issue 3