Episode 2: How we got here- the Old Grid vs. The New Grid

The grid was built to deliver electricity to the people to light, warm, and power their homes. 

In colder climates, the operation of the grid is critical to ensuring people have access to running, water, food, and most importantly warmth. Evaluating the grid based on its performance in the cold winter months is a fair benchmark to use. More recently the energy space has talked about peak summer demand during heat waves as the primary driver of requiring the grid to change. But this is the first time in a few years that so much of the US has been impacted by an extreme cold winter event. This last weekend’s winter storm Fern put most of the US under pressure with snow and freezing temperatures. 

Most of the discussion in our space about Winter Storm Fern has been about generation (renewables vs non-renewables) meeting high winter demand, but I think the most overlooked conversation has been around the reliable distribution, delivery, and recovery of power. The grid’s purpose is to deliver power to people over some distance. And most of this power is carried by cables stretched across poles. 

When lines break because of heavy ice or things falling, people will experience outages where whole neighborhoods can lose power or just a street can depending on where the break was. 

Operations to get people out to fix lines and get power up is monitored and managed using a combination of customer outage reports and automated alerting on major areas of the grid. When major interruptions occur at critical intersections of the grid, the utility can pinpoint a break in connection with equipment like smart switchers and re-closers making recovery easier. (distributed automation).

Source: LinkedIn Post

These technologies are making their way onto the distribution grid especially in target areas of the grid, but sometimes it is hard for utilities to locate faults or breaks because the exact mapping - like google maps mapping for what connects to what and where - may not exist in their system. If there's an outage in the areas where there are less smart devices or no mappings, operators have to figure out the locations of breaks in the system and the impacted size of each issue. 

I’m willing to bet that if we had full visibility into the grid, it would make operations during winter storms easier. 

Source: https://outagemap.dominionenergy.com/ 

We have to keep in mind that the old grid is also the original grid. And with the first iteration of anything there are usually a lot of assumptions and a list of todos that never get visited again. For example most utilities don’t have full mappings of sections of the grid. They have the mapping of the major intersections and main roads of the grid. But they don’t have street to street mapping within neighborhoods. They lack what is called secondary mapping from transformer to home. 

Source: https://en.wikipedia.org/wiki/Low-voltage_network#/media/File:Radial_Networks.svg

Secondary mapping was built out ad-hoc from the 50s to the 70s based on best trade practices and had little to no documentation.  So utilities know which set of homes are connected to a main road but not necessarily the specifics like what street comes first in the neighborhood or which house gets power first vs the last in the neighborhood. Only a handful of places actually maintain secondary mapping data like Eastern Australia, Hawaii, Manhattan where the grid has to either be super resilient or where there’s a lot of renewables connected to homes. 

The lack of secondary mapping data was something we experienced too during the preparation of the solar hosting capacity report and even the presentation. The utility expressed frustration at the lack of this data and the unfortunate impact it has had on making interconnection faster and easier. The lack of secondary mapping has not only become a bottleneck in approving interconnection requests for solar at homes but potentially in automating various operations. Lack of adequate mapping and therefore good models not only hinders interconnection for renewables and storage but also hinders the efficiency of operational recovery during things like winter storm outages. All lines have to be recovered if there’s widespread outages but knowing which breaks have the biggest impacts and being able to deploy resources in priority is definitely a plus. 

The new grid should have newer features, robust internal processes, and major bug fixes. And one thing to definitely do is addressing this lack of secondary mapping data. There’s whole companies trying to figure this out but these are still estimates. Utilities should determine how they want to get secondary mapping into their systems, what makes an estimate good enough, and then build good enough models that are useful in removing such bottlenecks. 

There’s some cool research to build out smart grid datasets, but I think the old grid guys at the utilities know their grid pretty well. The Aussies really know what they're doing too.  

So step 1 of old grid to new grid - getting the old grid mapped out - knowing what’s where and the specific values of lines, lengths, thickness, ratings, etc. Think about it like this - you’re planning an open concept kitchen in a home from the 50s and you’ll have to remove a wall, but you don’t even know if the wall is concrete or wood or both. The contractors doing the renovation would have to figure out what’s in the wall first- wires, pipes, if it’s load bearing - before removing it to plan the upgrade accordingly.

So the first step is getting the whole picture. Then if we were to upgrade anything to make our new grid dreams come true, I’d say we need to build redundancy into our grid. Redundancy means essentially having a backup component like backup lines between homes or main portions of the grid. Redundancy does exist between major sections of the grid like feeder to feeder and networked grids that need to be super resilient like Manhattan. But other than that there’s not much of it. 

To move the grid into the future, the grid needs resiliency, a strong foundation, to build upwards with trust. For my new grid, I would rebuild sections of the grid with redundancy to enable a street to potentially re-energize itself with stored energy during a winter time outage. Not a crazy amount of energy but enough for emergency services and heating for 6 to 12 hours. Getting enough batteries to power a street for 12 hours from homes and EVs is more feasible than it seems.

Redundancy enables energy recovery from energy storage during an emergency which is a newer grid concept taken from the microgrid world. We usually talk about this for datacenters or hurricane prone islands but we should consider this as an idea for upgrading the old grid to the new grid.