Blog

Navigating Uncertainty: How HIRO brings real-time Insight to Hydropower Operations Decision Making

5
minute read

As we set our sights on a future powered by the storage potential of hydropower, hydro asset managers are exploring ways to maximize and leverage existing flexibility, on day-ahead and intra-day markets.

Time is of the essence and faster water management models are just starting to replace traditional ones. Chances are you are not using heuristic algorithms in your production planning yet, as they are harder to build and fine-tune to the intricacies of hydropower planning, but undoubtedly faster. HIRO – Heuristic Integrated Robust Optimization is the result of 7 years of our team’s passion and expertise, combining state-of-the-art heuristics with robust optimization techniques.

HIRO and Black Swans  – Ready for the Unknown Unknowns

Our model empowers operators to store and dispatch energy following grid requirements in real-time, while accounting for all expected and unexpected limitations. It is in this realm of uncertainty that heuristic models like HIRO truly shine. These powerful decision-making frameworks not only navigate around events which can be expected, but also prepare us for the unknown unknowns – the notorious “black swans”. In the current global climate, it is no longer a question of whether a black swan is heading our way, but rather a question of when it will strike. This is how to use HIRO to navigate such events:

Play the long game

Singular events, although impactful, should not dictate the success or failure of our hydropower production. Instead, it is the ability to optimize and plan for the long-term that sets you apart. In our experience, asset managers who prioritize long-term optimization over short-term gains are better equipped to weather singular events like black swans without compromising their overarching goals. Heuristics models can be designed to prioritize long-term goals over short-term fluctuations, which aligns well with the idea of withstanding black swan events without sacrificing overall objectives.

Stay on your toes

HIRO Heuristics models are often more adaptable and flexible, enabling them to adjust quickly when confronted with unusual events. They can modify their decision rules based on the current situation, making them more robust in the face of unforeseen circumstances.

Don’t rely on the past

Traditional models heavily rely on probabilities and historical data, by assuming that future events will follow past patterns. Black swan events, by definition, defy historical patterns, making such models less effective in dealing with such extreme events. By contrast, heuristic models offer a non-probabilistic approach, and make decisions based on the current available information and expert judgment.

Work around missing links

In real-world scenarios, missing data is not uncommon, making traditional models more susceptible to performance issues. Heuristic models, on the other hand, can leverage the available data effectively, making informed decisions even in the absence of complete information. From fluctuating water levels to changing energy demands, it is crucial to have a strategy that can adapt and thrive in the face of constant change.

Heuristic models have the power to optimize hydropower production strategies, considering a multitude of variables and uncertainties.

The specifics of hydropower production involve some invariable data, such as turbine efficiency points, reservoir size, some operational constraints, and certain environmental constraints. These set the scene for the available flexibility of your hydropower asset.

The two main sources of variable data impacting production planning are market prices and expected inflow, followed by inflow dependent environmental and operational constraints. Each asset manager will have to prioritize different metrics, depending on their topology, mid- and long-term planning.

Good news is, heuristic algorithms can be crafted to fit any needs precisely, while traditional off-the-shelf solutions have limited flexibility.

HIRO – Heuristic Integrated Robust Optimization

As the name implies, HIRO is more than a hydropower adapted heuristic model.

HIRO uses an integrated approach to hydropower operations, by collecting and using all data you have available in its computations: data on weather, on water levels, on power market prices, turbine efficiency, environmental restrictions, and regulations, and much more. The better the data is, the more accurate HIRO’s work will be. One example is our inflow forecast. Here we use machine learning to create the most accurate estimate possible, given weather data. Our forecast is better than a naïve forecast by a median of 60%, of crucial importance for RoR powerplants.

With this data, HIRO computes a real-time water value calculation that can drive operations through real insight. This makes HYDROGRID Insight faster and easier to use, but it also yields more reliable results for both water and energy systems.

We’ve put a lot of thought and effort into making HIRO robust by minimizing computation time and complexity to achieve true real-time reaction. In doing that, we ensure that we optimize between time and accuracy, because you don’t need the perfect estimate one day too late; you need a good enough estimate within minutes.

By running all calculations under the same algorithm, we enable you to optimize all parameters against each other, and prioritize the metrics that matter to you. Using HYDROGRID Insight, you can now automatically achieve the best scenario for your priorities, which remain fully customizable and editable at any time.

HIRO ensures we make the most of every drop, every time. It’s time to tap into the immense potential of hydropower with HIRO and turn ripples into renewable power waves.

Make every drop of water count with HYDROGRID INSIGHT!

Get in touch with our hydro consultants to learn how!

Get Started
Ready to take your hydropower
operations to the next level?

Get in touch with our hydro consultants to learn how!

Get Started
Whitepapper
The impact of digitalization and data-driven decisions in hydropower
Hydropower has immense potential to help accelerate the global energy transition. While wind and solar are powerful forms of renewable energy, water remains a ubiquitous and reliable contender.
Download PDF
Share:
author
Ciro Taranto
Optimisation & Machine Learning Lead, HYDROGRID
Linkedin

Ciro Taranto is the Optimization & Machine Learning Lead at HYDROGRID, known for his passion for numbers and translating real-world problems into mathematical models and code. With expertise in machine learning and AI, he brings analytical precision and a pragmatic approach to solutions. Ciro is a dedicated Python developer, scientific communicator, and team player, driven by curiosity and determination.