With the recent push to integrate renewable energy into the existing energy infrastructures, it is becoming clear that there is a need to adjust its operation mode. This need is apparent because most renewable energy sources depend on the weather and are not easy to predict or plan with. Moreover, the power generated from such sources as wind energy and solar energy is highly stochastic. This situation calls for the application of advanced technologies for renewable energy forecasting and scheduling.
Renewable energy forecasting helps foresee what changes are expected in the amount of energy that will be generated in the future. This prior knowledge is informative for energy suppliers to plan the input they put into generating systems. Renewable energy scheduling also works side-by-side with forecasting because it is mainly determined by the predictions made by the energy forecasting models.
How renewable energy forecasting and scheduling work
Recent advancements in artificial intelligence have improved the job of weather forecasting (done by meteorologists) through machine learning. As a result, grid operators can leverage machine learning techniques to determine the amount of renewable energy that will be used and purchased by consumers at a particular time.
Machine learning (which is used for renewable energy forecasting) works because a software system learns patterns from recent data and develops an improved analysis for the future. In order to achieve this, a forecasting model is designed to fit a particular situation over several days. In addition, the data collected must be valid, accurate, reliable, consistent, and complete to be effective.
What You Should Know About Renewable Energy Forecasting And Scheduling
Here are five important things about renewable energy forecasting and scheduling you should know;
1. Renewable energy forecasting is built around short-term forecasting
Forecasting can be done with different horizons: short, medium, and long-term. Short-term forecasting involves forecasting from a few minutes to a few days ahead. It is used for day-to-day activities, and this time frame applies to renewable energy prediction.
The lead times in short-term forecasting are such that the changes in weather over a short period can be analyzed and used to predict the data to be used the next time. Renewable energy forecasting and scheduling require updated and recent data as frequently as possible, and short-term forecasting achieves that. The amazing part of it all is that there would be little or no human interruption with the presence of technology. Such way, errors would be significantly minimized.
2. Decentralized computing plays a prominent role in renewable energy scheduling.
Decentralized computing involves the allocation of both software and hardware to various points of duty. It is not like centralized computing, where all activities stem from a particular place. This form of computing (decentralized computing) is necessary for renewable energy scheduling because of the nature of locations in renewable energy generation and consumption.
For example, in an energy community, several houses may produce power at a time, and some utilities need power. The allotment of power to different places can be done effectively with decentralized computing technologies like the blockchain. It is effective because control has to happen independently from various locations when forecasting predictions require an adjustment.
3. Smart grids allow for renewable energy forecasting and scheduling.
Smart grids are electrical distribution points that are not like the conventional grid. The difference is that they contain many operation and control systems, advanced metering systems, intelligent circuit breakers and boards, and most importantly, renewable energy sources fit in well. Their operations are more efficient and can be readily evaluated because of the availability of needed information at the click of the finger. It is in such a system that renewable energy forecasting and scheduling can thrive.
Advanced forecasting models can be introduced and used to plan how the plants would run, whether solar photovoltaics or wind turbines. The ease in integration occurs because the smart grid already has smart IoT devices for thermal sensing, smart meters, phasor management networks, and the likes.
4. Grids with renewable energy attain stability easier with forecasting models in place.
Grid stability is of utmost importance for the sake of the life span of grids. Grid operators cannot consistently have variations in input and output in the grid happen repeatedly. With accurate renewable energy forecasting models, proper preparation and scheduling would be done, and there would be less frequent stability problems. Renewable energy forecasting and scheduling cut back most excesses when it comes to grid management.
5. The weather is a significant factor in renewable energy forecasting and scheduling.
Renewable energy forecasts are usually a combination of accurate weather predictions and the availability of plants and systems. The weather is a great factor, as the weather changes cause significant changes in the renewable power generated. For example, the variable speed of the wind is proportional to the amount of power generated by wind turbines. In the same way, the intensity of sun rays and the positioning of clouds play a big role in the fluctuations when it comes to solar power.
What makes renewable energy forecasting and scheduling interesting is that it studies the highly influencing factors of effective power generation. This kind of study is immediately applied, rather than just being carried out for nothing. It turns out that analyzing the weather, as Meteomatics does, has a vital role in the forecasting done by Hive Power’s Forecaster.
Renewable Energy Forecasting and Scheduling Solution – Hive Power
Hive Power’s Forecaster is one of our Flexibility Operator’s modules that performs short-term forecasting in a very accurate manner. It simply considers various factors involved in renewable energy-based power generation and uses them in forecasting. Its machine learning models make predictions on the amount of energy that would be used and generated in the future, based on previous data. This data is real-time data which is very helpful because it is used as soon as it is delivered.
Renewable energy forecasting and scheduling are essential for the effectiveness of renewable energy systems. With more observations in the needs of a renewable energy system, new technologies keep springing up, and it fosters development. Therefore, it is crucial to embrace these technologies as they come, especially when they are practical and efficient like this (in renewable energy forecasting and scheduling).
The Forecaster: using machine learning and weather forecasts to more accurately predict energy consumption and generation.
“It’s hard to make predictions – especially about the future.”- Robert Storm Petersen
Predicting the future might well be hard, but it’s often necessary to run the business. Energy forecasting is of primary importance in day-to-day market operations. Short-term forecasting generally involves forecast horizons that range from a few minutes to a few days ahead. Energy Suppliers and Distribution System Operators benefit from accurate predictions of power demand and generation because they can optimally orchestrate their flexible assets to achieve their business goals.
Hive Platform’s Forecaster module computes short-term stochastic forecasts of aggregated energy consumption and PV generation, using the most advanced machine learning methods available. Stochastic means that we not only predict the “shape” of the load curve over time but also calculate confidence intervals around this curve.
Prediction intervals are essential when it comes to risk management applications, such as predicting the daily or monthly peak of consumption and efficiently activating peak shaving mechanisms.
How does it work?
There is no crystal ball (unfortunately). Instead, the Hive Platform’s Forecaster predicts future consumption and generation by learning what happened in the past and using information available from the future. The Forecaster consists of an ensemble of machine learning models trained on and continuously re-informed by a rich dataset.
The data fed into these models includes the power signal itself, numerical weather predictions (temperature, solar radiation, wind speed and direction, humidity, and heaps of other parameters), seasonality information (the time of the day, the day of the week, the month of the year, etc.), public holidays, school holidays, and custom events (lockdowns, white nights, major events, religious events, etc.).
Weather is king
Weather is by far the most important external factor affecting energy consumption and generation. For this reason, Hive Power decided to conduct an internal research study to review and select the most performant numerical weather prediction provider.
One important criterion we considered was the availability of historical weather forecasts. Many weather providers do not archive and store their predictions. However, historical weather forecasts are of crucial importance to train an energy prediction model. To be robust and reliable, a model should be trained on the same type of data that will be used at inference time.
To understand this point, think of the following. If we trained the model on actual weather observations, the model would learn to trust the weather signals to a certain level. When using this model to predict the future, we would need to replace weather observations with weather predictions, which will not be 100% accurate by definition of prediction. Hence the model will give too much weight to the weather predictions and fool itself into error.
Conversely, if we train the model on weather predictions, the model will learn to trust less the weather parameters and commit less error at inference time. Still sceptical? Try out yourself. Some of our most wary scientists did so and found exactly what was expected.
Weather forecast benchmark
We looked at about a dozen different providers, filtered out those that did not tick our boxes, and ended up with five finalists. We then asked them for a sample of their data to build a benchmark. We requested a year’s worth of hourly predictions of ground-level temperature and solar radiation generated for a single location at around midnight and covering from 24 to 48 hours ahead, which is the typical forecast horizon of our energy prediction models. We compared these predictions with actual local observations and were stunned to discover Meteomatics’ superior performance.
In the figures below, we show some results. In Figure 1, we plotted the distribution of the discrepancy between observed and predicted temperature. In Figure 2, we overlay the five distributions for a more convenient comparison. A similar situation was found for the solar radiation parameters. It was then clear to us that Meteomatics numerical weather predictions were the most accurate and well-calibrated.
Figure 1 – Distribution of the discrepancy between observed and predicted ground temperature for five different weather providers. The vertical dashed lines indicate the mean of each distribution (only Meteomatics’ mean error is centred on zero). The Mean Absolute Error (MAE) is reported on each chart (the lower, the better).
Figure 2 – The same error distributions of Figure 1 overlaid (after estimating their kernel density). Meteomatics’ curve is the narrowest and the only one that is zero-centred.
The beginning of a strategic partnership
“We were looking for a one-stop-shop that could provide us with up-to-date and detailed weather data covering Europe, which needed to be conveniently available via a RESTful API. We took quite some time to evaluate several numerical weather prediction providers. It was not an easy choice because of our demanding list of criteria, but Meteomatics checked all the boxes and exceeded our expectations. We were after a provider that covered the whole world, with a focus on Europe and especially the Alps, which are a challenging region when it comes to high-resolution weather forecasting. It was clear from our own forecast benchmarking exercise that Meteomatics was the most accurate weather forecaster, particularly over the short term. We were pleased to discover the rich plethora of standard and advanced weather parameters that Meteomatics’ API offers. We are thrilled to have switched to Meteomatics as our weather data provider of choice, and we foresee a long and fruitful partnership with them.”, said Gianluca Corbellini, Managing Director at Hive Power.
You can read more about this in Meteomatics’ latest press release: https://www.meteomatics.com/en/meteomatics-and-hive-power-agree-strategic-partnership/ .
Stay tuned for an upcoming webinar (26th of October 2021) where Hive Power and Meteomatics will guide you through an in-depth look at how numerical weather predictions inform energy prediction models.
Meteomatics and Hive Power Agree Strategic Partnership: Bringing machine learning to more accurately predict energy consumption and generation.
Meteomatics AG is a private weather business, making a rich database of weather information (>7 Petabytes) and weather insights available to users across the globe. Meteomatics provides incredible detail and accuracy with weather forecasts downscaled to 90 meters and up to 5 minutes temporal resolution, globally. All through an easy-to-use RESTful API endpoint.
Providing downscaled forecasts at 90-meter resolution and up to 5-minute temporal resolution for anywhere on the planet, were very important reasons why Hive Power chose to switch to Meteomatics.
Hive Power is a smart grid analytics company with a strong focus on creating innovative solutions to improve grid operations for energy suppliers and grid operators through data-driven artificial intelligence. One of their innovations is the Hive Power Forecaster which computes stochastic energy forecasts to simplify how energy retailers and grid operators manage the aggregated energy production and consumption.
“We were looking for a one-stop-shop that could provide us with up to date and detailed weather data covering Europe, which needed to be conveniently available via a RESTful API. Hence why we were very excited about partnering with Meteomatics, and the commercial possibilities Meteomatics’ API could generate for Hive Power. We took quite some time to evaluate several numerical weather prediction providers. It was not an easy choice because of our demanding list of criteria, but Meteomatics checked all the boxes and exceeded our expectations. We were after a provider that covered the whole world, with a focus on Europe and especially the Alps, which are a challenging region when it comes to high-resolution weather forecasting. It was clear from our own forecast benchmarking exercise that Meteomatics was the most accurate weather forecaster, particularly over the short term. We were pleased to discover the rich plethora of standard and advanced weather parameters that Meteomatics’ API offers. Moreover, Meteomatics was one of the few providers that easily allowed the retrieval of archived historical weather forecasts, which are of critical importance to correctly train our machine learning models. We are thrilled to have switched to Meteomatics as our weather data provider of choice, and we foresee a long and fruitful partnership with them.’’, said Gianluca Corbellini Managing Director at Hive Power.
Meteomatics Weather API allows Hive Power to inform and enrich their proprietary energy consumption and production forecasting models. Powered by the latest and most advanced machine learning techniques, Hive Power’s Forecaster computes short-term probabilistic predictions to simplify how energy retailers and grid operators manage the aggregated energy production and consumption. Accurate energy predictions are crucial for effective peak-shaving, performant energy trading, robust grid stability and avoidance of congestions.
Meteomatics unique approach to forecast downscaling allows Meteomatics’ API to resolve the challenges local topography can bring to weather, enabling Meteomatics to achieve a very high degree of forecast accuracy. Plus, the breadth of Meteomatics’ weather database covering all forecast timescales: historical data (from 1979), nowcasts, forecasts, probabilistic and seasonal, allows agricultural businesses to improve operational efficiencies and forecast yields.