MADA Analytics: We are processing energy the way we process data

Log in

Remember me

Forgot your password?
Free trial

MADA Analytics: We are processing energy the way we process data

Date: September 8^th 2021

Author: Alenka Lena Klopčič

Category: En.vision

Topic: Renewables , New technologies , Economy , En.vision

Check how Israeli-based but run by an American, startup MADA Analytics reduces the uncertainty of renewable energy project cash flows caused by weather and other factors!

How exactly do they reduce the renewables project uncertainty and how MEPS™, a software solution designed to optimize the way users store, use, invest, and mitigate risk in renewable energy, was ‘born’ is explained by Daniel Gimpel, who a co-founded MADA Analytics (MADA means science or knowledge in Hebrew) after spending four years immersed in the field of energy storage project development and financing renewable energy project development, as well as working as a venture capitalist (VC) for some time.

In a third video interview which is a part of the ‘Cleantech entrepreneurs for a sustainable future’ series conducted by Energetika.NET for its energy sector audience and supported by WorldChicago, the Chicago-based organization which among other things supports young entrepreneurs in cleantech industries, Danny Gimpel told us just how renewable project uncertainty could be reduced.

In order to get the project financed, investors have to come forward with a planned schedule and optimized energy storage or insurance warranties - and that is where MADA Analytics comes in. Their approach is based on an understanding of the uncertainty in forecasting data that's accurately measured on the data side, the engineering side, and the financing side. And then, when the simulations are run in the different configurations, MADA’s system determines different price sensitivities, different prices for insurance warranties and storage systems operating within their specification constraints. “And that's how we try to deliver that data to customers at the end of the day”, Gimpel explains.

“In the way the software is utilized and the way it's delivered, deliverables to customers consist of three main things: an insurance pricing of a warranty to a sized system, a time series schedule - that's how the system is to be operated and integrated with solar or wind - and a financial model, how it impacts financing. So we get those three deliverables from the software”.

What matters is not the quantity, but quality of the data!

Gimpel begins by explaining how the system works: “In the machine learning simulation, you are always looking for different cost sensitivities and ranges. Uncertainty is driven in renewable energy from the source of the energy, which is weather, and wind patterns and solar patterns are just not sufficiently certain to always be consistent. So that's one driver of uncertainty, and then the other is actually in the market, in the financing of projects. And that comes in the form of pricing of energy, and it also comes in the form of financing projects from banks. How to de-risk that uncertainty? There's a saying that big data is really not that impressive, that it is not the quantity of data, but rather the quality of data. And so that's where it really begins: with the accuracy of the data. And the second is really what you do with the data. You have to really know the financial implications. So you have to understand project finance and you have to understand the engineering control constraints of wind, solar and storage systems,” Gimpel explains, adding that finally “you have to know the market and the market pricing”.

Next, we ask ourselves what the best solution or best cost benefit is to guarantee that the energy will be delivered or revenue ensured to meet debt financing payments? “If that's an energy storage system, how will you be able to capture energy or charge the system and be able to deliver the energy when you have to, and in order to be certain that the storage system will perform as expected – priced warranties can be priced and issued to cover the system risk That's one scenario. But the second scenario is if there is a vacuum, with no storage system and there is no generation, how do you ensure cash generation, delivery of revenues and the ability to pay debt and bank financing? And that's really a matter of performance insurance, so that regardless of patterns, or over a probabilistic period, you'll have the certainty that even if there's a low day or a low week or a low month on average, over the course of the year, you will make up the pattern if it's a volume risk and a fixed price. capturing the energy and delivering it at a different time. If it is a fluctuating energy price risk, there are other ways to mitigate the risk with swaps and price hedges.”

“So, you have a backup of insurance that can actually make banks more comfortable even though there's low performance in this month, it will be made up, the insurance company will earn the revenue further on, and it will cover the difference based on this kind of performance. The main performance insurance that we focus on is actually within storage systems to understand how, if you're going to maximize the system all the way to its fullest lifecycle, do you make sure that you don't run the system out too soon, or how the cycling will impact degradation,” Gimpel explains energetically.

Also, he continues, there are different ways you can run the system. “You can oversize the system, or you can add within the process of the system, all of those types of simulations and scenarios are run through MADA’s software. And that way we know many, many, many different configurations and scenarios to identify the best way to plan and design a schedule going forward, an operational schedule. So when the project is running, you're adjusting - adjusting the schedule and optimizing operations as you're running forward”.

Technically, it doesn’t really matter where in the world MADA’s user is based, Gimpel confirms.

When one starts working with a visionary…

At the beginning of his energy carrier, Gimpel got involved in renewables as a partner with late Arnold Goldman. “He was really quite a visionary and quite a brilliant genius. And he was well ahead of his time in seeing that you need to integrate multiple perspectives of renewable energy”, he says of his partner and ‘energy mentor’. “So, you need to understand the engineering side and the finance side. And when you look at renewable energy, when you deal with it from the performance side, the operational engineering side, you have to deliver electricity”, Gimpel notes, relating it to the energy side.

“Electricity can come from wind or solar, but it can also come from gas or coal, which are not clean. But if it’s the performance that you must meet, then you might be able to charge storage from the grid or be able to deliver electricity from a gas turbine if it's within a system. We were approaching utilities in the United States, in the North-west, in Washington state, Oregon and California. And we were looking at projects that entailed multiple inputs, which is a type of software simulation engine, i.e., MIMO - multi input, multi output we developed.

“We have looked at processing energy the way you process data. You can process data like a stream, a fast stream of data, like a video. Or you could store data in deep memory, like it's stored in fossil fuel, like coal and gas. Or you can store it in a battery, kind of like RAM memory, like a computer. There's text, there's audio, there's video. All these different types of data are streamed at different speeds. And that's how we looked at energy. If you could manage energy and model energy by looking at it from random wind, to batteries with a time limitation, to deep energy storage, which is in deeply-embedded fossil fuels”. That is when their MEPS™, a software solution designed to optimize the way users store, use, invest, and mitigate risk in renewable energy, was ‘born’.

Sustainability goes far beyond energy

This summer, MADA Analytics presented itself at the pitch day of the Techhouse Accelerator in Graz, Austria, where the main focus was on climate change and a sustainable future. As a startup entrepreneur, how does he personally see the future, and how could all of us, including your potential clients, contribute to it? Answering this question, Danny Gimpel thinks beyond energy.

“Sustainability in the future really is something of a consciousness, because everyone who wants a sustainable future has to think about it really, as a kind of lifestyle or value. It impacts your daily decisions. What kind of foods do you eat? Are they organic or do they contain pesticides? Do you deal with plastics or do you buy permanent? What kind of care do you drive - a car with gas or a hybrid or electric car? And where do you get your energy from? Do you prefer clean energy over fossil fuel energy, which might be cheaper for sustainability? People have to begin making conscious decisions, they have to be aware. But it also has to be either driven by a desire for a sustainable world, more environmentally beneficial things for our future, or for financial benefit, so that they say, ‘wow, it's actually economically better for me to use clean energy rather than dirty fossil fuels’”.

So, sustainability has to begin with awareness and consciousness, stresses the CEO and co-founder of a startup that has ambitious plans and already a few promising projects with their European partners, including one with EUREKA and their possible selection (in the competition of 300 startups) to the Austrian Verbund’s accelerator. (At the time of publication it was still not known if they made it to the very final selection.)

Watch the whole video interview in which Daniel Gimpel also shares more about how and why his switch from the VC to the startup side happened.

If you want to learn more about MADA Analytics, visit their webpage: www.madaanalytics.com.