Energy-Storage.news Premium speaks with Michael Kirschner, managing director, US, for Habitat Energy, about optimising BESS projects in ERCOT.
Optimisation specialist Habitat Energy was founded in the UK in 2017 and has since expanded to Australia and the US with contracted assets globally.
The company is part of Quinbrook Infrastructure Partners. Habitat Co-founder and Director Ben Irons spoke with ESN about the challenges then facing optimisation firms in a 2023 interview.
Irons said of the lack of transparency, often cited as a challenge for optimisers: “With our model, you give us the keys and we’ll generate the revenue. Transparency isn’t as much of an issue as long as the revenue comes in. The project owners don’t necessarily need reporting on a day-by-day basis since they are not the ones pushing the buttons.”
Try Premium for just $1
- Full premium access for the first month at only $1
- Converts to an annual rate after 30 days unless cancelled
- Cancel anytime during the trial period
Premium Benefits
- Expert industry analysis and interviews
- Digital access to PV Tech Power journal
- Exclusive event discounts
Or get the full Premium subscription right away
Or continue reading this article for free
The company is experienced in the Electric Reliability Council of Texas (ERCOT) market, where the upcoming Real-Time Co-Optimisation Plus Batteries (RTC+B) market changes will soon take place.
RTC introduces a new process where energy and Ancillary Services (AS) are dispatched interchangeably in the Real-Time Market, according to ERCOT. Currently, ERCOT secures AS in the Day-Ahead Market and rarely reallocates them between resources in the Real-Time Market.
This change is highly significant for BESS, which in ERCOT primarily earns revenue from Ancillary Services. It will now be dispatched on a locational basis every five minutes, shifting from a system-wide approach. Furthermore, the Plus Batteries (+B) reform will treat BESS as single units rather than combining generation and load.
Energy-Storage.news: How does Habitat optimise BESS in ERCOT?
Michael Kirschner: I use the analogy of an F1 car. You build a fast car for a specific purpose, but it doesn’t translate to driving a Honda.
Storage as an asset involves new challenges, higher-frequency controls, sophisticated responsiveness, ancillary product stacking, and dynamic system operation. Managing multiple positions requires co-optimisation and shifting obligations based on system conditions like locational marginal pricing (LMP) or system-wide metrics.
Doing this manually takes 8,760 hours annually across all products, which is brutal, so our goal is automation. Think of it as a self-driving car: install the code and infrastructure, then let it handle most tasks, with human oversight for unexpected events.
Our integrated platform delivers this: instead of a generic software-as-a-service (SaaS), we offer a tailored package for traders, operated under your optimisation agreement. You can customise it per your risk preferences, manage collateral, and get actionable insights, allowing you to focus on your core strengths while we ensure seamless integration.
What are some of the biggest things you feel like have changed since Habitat’s been optimising BESS projects in ERCOT?
RTC+B has got to be number one. It’s effectively a full reset of the system. They’re updating all their code. All of history is effectively, I don’t want to say scrapped, but I’m old enough to remember when ERCOT went through the switch to LMPs back in 2008 or 2009, and COVID. Both of those events created a singularity in some sense, where on the other side of it, you can’t use history as a guide.
As we approach 5 December 2025, which is the cutover date for RTC+B implementation, and it looks like they’re on track.
We’ve had to effectively build our tech stack twice. We built it once for the original service offering, and now we’re rebuilding it for the RTC+B framework, making it ready for what we expect to be a more volatile period as everyone re-normalises to the rules.
What are the biggest changes through RTC+B?
Well, like all governmental or quasi-governmental rules, it’s fairly complex and obtuse. I don’t think anyone really understands it, but I’ll do my best. At a high level, it’s not that difficult to understand. Other ISOs, such as PJM, do it already.
The day-ahead transaction is now purely financial, unlike before, when it involved physical assets, which was especially significant for batteries in ancillary services. Traditionally, operators submitted forecasts for load, solar, and other variables before 10 am to determine clearing prices and asset dispatch.
Now, ancillary service obligations are adjusted in real time, rather than carried over from day-ahead commitments. This shift means units must adapt quickly to changes like outages or transmission issues, often encountering illiquid markets and high penalties, which made participants conservative.
The new system allows for more flexibility, enabling assets to be reassigned obligations in real time, likely increasing volatility while lowering average prices. This change aligns with a more efficient system, potentially benefiting battery storage, and effectively shifts the day-ahead process from physical to financial.
Will optimisers be well-positioned once RTC+B starts, or will it cause headaches initially before being figured out?
Batteries and optimisers focus on volatility, profiting from price spreads regardless of market direction.
The real challenges lie in load management, risk mitigation, and how operators, especially those with non-battery assets, like combined cycle gas turbines (CCGTs) and demand response, adapt. Different players will deploy at different times, based on economic factors, technical readiness, and confidence.
Some may exit markets under new rules, causing thinner trading and potential price volatility. High prices encourage participation, stabilising the market.
The optimiser’s role remains the same, act early, adapt as competition arises, and maximise value by shifting between markets. Overall, the approach may shift slightly, but priorities stay consistent.
How precise must these strategies be for optimisation?
At the end of the day, I’m an old options trader, and I think about all of this in the context of options trading. You’re always playing in a probabilistic world. If I knew exactly what was going to happen, I’d have retired already. The goal of modelling is not to predict the future. The goal of modelling is to predict a relationship between inputs that yields outputs, which manifest when they are present.
A good model doesn’t tell you the future. A good model tells you that when the weather is mild, prices are mild, and when weather is extreme, prices are extreme, and it gets them relatively matched in order of magnitude. Then the job of the modeller, the job of the optimiser, is to determine the probabilities of those things within the context. It’s like, ‘yeah, it’s a mild summer, but tomorrow could be really hot.’ There’s local information, and we’re trying to incorporate that and be very specific about it.
We’re not trying to have a crystal ball. Typically, if you know that much information, either you’re breaking a law or you don’t need to be in the business anymore. Our job is to identify opportunities, assess the probability weightings, and be strategic about how we dispatch them. We want to contextualise that with the risk preferences of our clients. It’s not often about just making an extra dollar. It’s being smart about the dollar that we’re making.
Sometimes, that dollar can work against you – if you experience an outage, unit trips, or a part of your asset fails, such as an inverter, or if you lose communication – you’ll have to buy back that position. If you fall down, and you were the reason the price was low because you sold too cheaply, when you buy it back, it’s going to cost you quite a bit to cover, and that’s going to hurt your performance.
We assess everything with a risk-adjusted approach, using probabilistic outcome distributions and context-specific methodology for local, regional, and ERCOT-wide situations. Automation ensures consistency, quick updates, and efficiency, reducing manual errors.
Our goal isn’t always to be number one, but to consistently stay in the top decile across uncertainties, accepting that mistakes happen, while aiming to play well.
Is there anything else you believe is important to note about the ERCOT market, either now or potentially in the future?
ERCOT is akin to a prototype space due to its energy-only focus and steady growth, making it an ideal environment for development.
Our current operations are ERCOT-only, but by 2026, we plan to expand to other ISOs, leveraging our platform to serve more clients across the US.
Real-time optimisation will evolve with the integration of virtual power plants (VPPs) and green technologies, like ammonia, hydrogen, and data centres, which can cycle up and down. Our digital twins, initially battery-based, can be adapted for these new assets to enable automation and portfolio optimisation.
ERCOT is uncertain about future changes but is exploring new technologies and services, like batteries for transmission support, which require new business models. ERCOT acts as an innovation incubator in the energy sector; it’s fun to watch the ride if nothing else.