From 1989 to 2004, I held a role as Manager of Electrical Distribution Engineering for IES Utilities (Alliant Energy). I had a staff of 15 Engineers and drafters reporting to me.
My role was multi-faceted. In parallel to planning for electric distribution power substation construction and decommission, my team was building a GIS (Geographic Information System) and engineering electric distribution lines.
During the late 1980s and 1990s – IES Utilities faced many business and financial pressures of maintaining electricity for over 1 million electric customers in Cedar Rapids and surrounding rural communities. IES had nearly 100 small (single 34.5KV distribution transformers) and aging distribution substations. Many of them were at high risk of mechanical failure, aging electrical insulators, potential containment failure and oil spillage, as well as a number of other possible mechanical failures. Many of the substations were non-standard and parts between substations were not inter-compatible. In addition to the aging infrastructure, Cedar Rapids and the Iowa City area was growing and the electrical service needs continued to stress an aging system. To further the stress of the situation, there was growing pressure to upgrade the electrical transmission system from 34.5KV to 69KV in the near future.
The traditional approach to facing these problems was to maintain status-quo. In short - continue to build small single transformer substations where they might be needed.
My individually developed concept for substation placement and an application of using multiple digital sources of information came 3 years prior to Gartner-Groups recognition of 3-D dimensional data analysis in its definition of “Big-Data”.
My approach to solving the growing problem included actions requiring the manipulation of large digital data sets. Keep in mind that in the late 1980’s and early 1990’s, compute power was not a particularly cheap or easy-to-use. This method of substation planning was completely a new idea using digital tools. There was no instruction book – it was all a creation of my-own based on the manipulation of digital data. Now known as Big-Data manipulation.
First I started with the individual distribution transformer data. This information contained both location and month-by-month kilowatt-hour usage of thousands of electric transformers spread over the city (app roximately 10 miles by 10 miles). From this information, I wrote a program in the programming language Pascal (citation). This program “smoothed” and scrubbed the raw data to create a grid 3-D surface plot that represented hills and valleys. In this case hills represented high load centers, and valleys represented low load centers. Similar to the graph below.
Secondly – I made an inquiry with the city of Cedar Rapids for their multi-year growth forecasts, road and planning data. Their information was clear and invaluable to the success of this project. I had a fair bit of manual work to do to convert their supplied information into a digital form that I could use to match up with the distribution information mentioned above.
Thirdly – I obtained and overlaid a digital version of all transmission lines in the Cedar Rapids area. Again, this data also required manual work to create a digital version that coincided with the two sets of information above.
Fourth – I digitally simulated new, planned or speculated large business load centers on my digital compilation of the materials mentioned above. This was both an art and a science as a majority of these potential business loads were speculative.
Fifth – Data scrubbing. This was a purely a manual effort to go over all existing load centers in the transformer data to ensure there were as few errors as possible. Due to the nature of the electro-mechanical devices of electric metering, some electric load meters report huge errors from month-to-month. This step was neither perfect and was very time consuming.
Sixth: Once again, all available real-estate information was gathered and digitally transformed into a format that could be used with the other sets of data mentioned above. The value of this was that it allowed quick and easy visual assessment of where substations could or could-not be placed due to the large number of real-estate restrictions.
Lastly – A multi-discipline engineering team worked together to design a more redundant, standardized, and aesthetically cleaner looking distribution substation. Since many of the older substations were many decades old, this meant compliance with a very large number of new regulations regarding electrical substation design and power distribution.
The end-result was a Big-Data analysis in digital graphical form that optimally placed a new and better electric substation in the place closest to where the electric load was to be served, closest to the existing transmission circuits, and built on real estate already owned by IES. Additionally, it gave the viewer a near instantaneous decision-making tool that would optimize future investments in distribution substations. And finally, it permitted a complete re-think of the approach to how distribution transformer stations would be built in the future.
With this aggregated Big-Data, the material was presented at public city meetings to help mitigate concerns about IES’s intentions on building substations, justification for new power-lines and the clear justification behind it all. With all the accumulated information from multiple sources – the resistance was easily managed with a few compromises from both sides, primarily related to aesthetics and landscaping.
Months after the conception of the idea and armed with the aggregated data, construction of the substation began in 1994 and was completed and put into service shortly thereafter.
This method of planning future substations was re-used over the past 33 years and as of today all of the previous small distribution substations have been replaced by the new design.