Building intelligence layers for the energy transition

Intro

Cheaper energy enables us to make more technological progress — faster. This is for me the most exciting part of working on the energy transition. Albert Wenger highlights the importance of deploying massive energy capacity quickly in this post where he argues that we need to get out of the so-called “low energy trap”.

Strongly recommend to read Nathaniel Bullard’s superb presentation on Decarbonization

The exponential growth of renewable energy assets is leading to an explosion of new data at different levels. Data that spans from energy generation data to energy consumption data to granular machine-level consumption data. All this data needs to be transformed, interpreted, combined, forecasted, optimized — you name it.

One area where I believe big value creation will happen in the next few years, is in what I call the “intelligence layer”. In this short post, I want to explain the concept of this “intelligence layer” and share a few examples that fit into this thesis.

Value chain of the energy transition

When we think about renewable energy assets, we often refer to solar, heat pump, EV chargers and batteries today. While most of the deployment happens in this area right now, the deployment of other technologies such as thermal energy storage, carbon capture or hydrogen will follow. This includes technologies that researchers and scientists around the world are developing as I’m writing this post.

Now the way that I look at the value chain of the energy transition is three-fold, see the graphic below:

Let’s dive into the three segments.

1/Resources (Bottom)

This refers to the physical part, the different resources that are part of the energy transition, ie.

• renewable energy assets, eg. solar panels, heat pumps
• energy infrastructure, eg. microgrids, transmission lines
• material or equipment needed to install renewable assets, eg. vans, cables
• people who are actually doing the work, eg. electricians, plumbers (I don’t like to call people a “resource” but use the term here for simplification)

These resources are needed to rebuild our infrastructure and build a more sustainable world.

2/ Intelligence layer (Middle)

This refers to the “messy-middle” between the different resources and the user interface. The intelligence layer is particularly strong at the data-level. This often includes combining different data sources, live and historical data, and sometimes even data from own hardware devices.

The intelligence layer is needed to enhance the deployment and capabilities of renewable energy. Put differently, with intelligence layers, the deployment and the use of renewable energy will happen faster and more efficiently.

Data serves as the fundamental enabler for the intelligence layer. With the proliferation of sensors, meters, and smart grid technologies, massive amounts of data are generated from various sources such as renewable energy installations, grid infrastructure, load balancing, weather patterns, and energy consumption behaviors. Data availability is therefore key.

Intelligence layers enable use-cases such as the following:

• Analytics and engagement: with the rise of renewable energy solutions, passive participants such as consumers and companies can become active participants in the energy market. Data analytics provide insights into consumer behavior, preferences, and energy usage patterns, empowering them to make informed decisions about energy consumption — thus, strongly influence energy behavior and engagement in the energy market
• Forecasting and optimization: real-time monitoring of resources allow operators to anticipate potential issues and optimize resources for maximum efficiency
• Flexibility trading: balancing between renewable energy sources and demand peaks, ensuring stable and reliable energy supply
• Data-driven decision making: intelligence layers that optimize operations and make informed automated decisions about resource allocation, eg. for their field workers
• Policy and regulatory support: development of evidence-based policies to support renewable energy deployment, incentivize investment, and accelerate the transition to a low-carbon economy with the help of intelligence layers

3/ User interface (Top)

This refers to applications with a user interface for either b2c or b2b customers, ie.

• mobile App to monitor the energy production of solar panels
• a dashboard to detect gas leakages
• a collaboration workspace for the digital twin of a power plant
• routing software for mobile fleets

These user interfaces will become much more valuable if they are connected to an intelligence layer. Related to the graphic above, there are two scenarios:

1. the intelligence layer is disconnected from the user interface (see grahic above)
2. the intelligence layer is not separated from the user interface. The main value still lies in the intelligence layer but the user interface helps the users to interact with the intelligence layer. Then the graphic would more look like this:

In an ideal world for many use cases, the user interface will disappear. If the intelligence layer properly works, there should be no need for human intervention. The intelligence layer takes care of getting the job done itself.

Examples of intelligence layers

Below are some examples of intelligences layers that I find particularly exciting right now:

• KrakenFlex*: develops software to manage and optimize Distributed Energy Resources (DER) in real-time and at scale. This helps energy companies to control, monitor and optimize their assets real-time to match supply and demand
• Enode: often referred as a “Plaid for energy” that builds a SaaS API to connect to a universe of energy hardware systems. It enables energy companies to connect their apps to all kinds of hardware systems such as EVs, solar inverters or home batteries to control and optimize its users’ energy devices
• Descartes: offers data-driven parametric insurance against climate risk. Descartes uses data sources such as IoT, satellite imagery, stationary sensors, radar and third party data, and applies proprietary algorithms to unlock risk insights (eg. for floods, earthquakes)
• Electricity Maps**: centralizes, standardizes and forecasts global electricity data in real-time. This data helps eg. data centers to shift flexible loads. Tasks with no time constraint shift to times of the day that align with the availability of electricity with a lower carbon intensity

I came up with this simple abstraction since I was spending more time talking to different entrepreneurs in the field. So I’m looking forward to talking to more founders who are building these intelligence layers. If you ideate in that area, I’m also building a list of ideas that I find worth exploring. I’m sure this thesis will evolve and I’m excited to keep an open dialog about it with you, let’s talk.

*KrakenFlex is part of Kraken Technologies Ltd., the company I’m working for
**I’m an angel investor in the company

Also, thanks to Marco Holst, Jakob Banhardt and Henrik Grosse Hokamp for your feedback on this post.