Solar power forecasts for single PV plants and entire portfolios
alitiq forecasts the output of your photovoltaic plants from 5 minutes to 10 days ahead. Every plant gets its own model, optimised with its measurements. And because you can retrieve forecasts and measurements at any time, you can check the quality yourself.
The alitiq solar forecast at a glance
- 5 min – 10 days
- forecast horizon, resolution of your choice (default: 15 minutes)
- 5.7%
- nRMSE of the day-ahead forecast in our case study: 15 MWp PV plant in Switzerland
- Minutes
- until the first baseline forecast after a plant is created
- €18
- per MWp and month, all-inclusive, first 30 days free
How the alitiq solar forecast works
Three steps, and you only do the first two once.
Create your plants
Location, installed module and inverter capacity, orientation and tilt: in the web portal or via API. We model plants with several orientations, trackers (single-axis with backtracking or dual-axis) and self-consumption sites accurately.
Send measurements
Send measurements from up to the last 90 days and then current values continuously, plus curtailments and unavailability. That's how the model learns how exactly your plant behaves.
Retrieve forecasts
For single plants or the whole portfolio, in your time zone, unit and resolution: via REST API, Python package or as files. The baseline forecast is ready minutes after a plant is created, the optimised forecast with measurements from the next day.
What our forecast quality is built on
A good solar forecast doesn't come from a single weather model, but from combining meteorology, plant knowledge and measurements.
One model per plant
Our optimised model combines the strengths of several weather sources with your plant's past behaviour. The forecast reflects how your plant actually produces, not just what the datasheet says.
Several weather models, not one
We use ICON-D2 and ICON-EU from the German Weather Service, HARMONIE-AROME from DMI, ARPEGE from Météo-France, the Met Office Unified Model and GFS, among others. Each model has its own strengths; the optimised model weights them based on the results at your site.
Curtailments kept separate
When your plant is curtailed, for example by the grid operator, for maintenance or for market reasons, you report it via API. The model doesn't learn from distorted measurements, and the forecast stays realistic.
Continuously updated
Continuous updates every 5 minutes capture rapidly changing weather, so new information reaches your day-ahead and intraday decisions in time.
Case study: day-ahead forecast vs. measurement
A PV plant in Switzerland with 15 MWp installed capacity, 13 days in September 2026: the day-ahead forecast of our optimised model, exactly as it was calculated the day before, next to the measured output.
- nRMSE
- 5.7 %
- MAE
- 2.7 %
- Bias
- +0.2 %
- Measured
- Day-ahead forecast
- Absolute error
Accuracy you can check yourself
You shouldn't have to take a forecast on trust. That's why the API gives you not only the latest forecast but every earlier calculation too, so you can compare it with the measurements you've sent us.
The most meaningful test runs for several weeks on your own plants, which is exactly what the free 30 days are for. How to evaluate forecasts fairly, and why a single metric is rarely enough, is explained in our guide.
- Earlier forecasts retrievable by calculation time
- Submitted measurements viewable at any time
- Direct line to our team: support@alitiq.com
Solar forecast API, Python or files
The forecast arrives where you need it, without a big IT project.
REST API
Retrieve forecasts for single plants or the whole portfolio, send measurements and curtailments, using your own plant IDs. Authentication by API key.
Python package
With pip install alitiq and the alitiq-py package, you get forecasts directly as pandas-compatible data.
File delivery
Forecasts as files via sFTP, email or S3, with no development effort on your side.
Web portal
Create and manage plants, see your portfolio in MWp, manage API keys and optionally secure your account with two-factor authentication.
Example: retrieve a portfolio forecast with Python
# pip install alitiq
from alitiq import alitiqSolarAPI
solar_api = alitiqSolarAPI(api_key="your-api-key")
forecast = solar_api.get_forecast_portfolio(
power_measure="MW",
timezone="Europe/Berlin",
interval_in_minutes=15,
portfolio_sum_column=True,
)Who the alitiq solar forecast is for
Direct marketers
Forecasts for your entire portfolio in one request, with an optional sum column, for schedule nomination and intraday trading.
Municipal utilities and energy suppliers
Your own PV plants and your customers' generation for procurement and balancing group management, combinable with the alitiq load forecast.
Plant operators
A reliable forecast for your plant, whether a rooftop system with several orientations or a ground-mounted plant with trackers, and the comparison with your measurements.
Keep balancing costs under control
Every deviation from your nominated schedule costs balancing energy. See how more accurate forecasts shrink your imbalances and how much you could save:
Reduce balancing costs and calculate your savingsSimple, transparent pricing
Start with a free trial, then pay only for the installed capacity you forecast.
All-inclusive forecast service
18 € /month per MWp
One price, based on your installed capacity. No hidden fees, no different plans.
- All forecast services included: single plants, portfolio, all weather models
- Optimised forecasts with your measurements
- Volume discounts for larger portfolios
- Try it free for 30 days, on your own plants
Frequently asked questions about solar forecasting
What is a solar power forecast?
A solar power forecast predicts how much electricity a photovoltaic plant will generate over the coming minutes to days. It combines weather forecasts, especially for solar irradiance and cloud cover, with the plant's technical data. Direct marketers, utilities and operators use it for schedule nomination, energy trading and to avoid balancing costs.
What data does alitiq need for a solar forecast?
To get started, the location, installed module and inverter capacity, and orientation (azimuth) and tilt are enough, per subsystem if the plant has several orientations. Trackers need a few geometric details. Measurements are optional but are the basis for the optimised forecast.
How quickly is the first forecast available?
A baseline forecast based on the weather models is ready a few minutes after the plant is created. Once measurements are available, you get the optimised forecast from the next day.
What horizon and resolution does the forecast have?
The forecast covers 5 minutes to 10 days ahead. You choose the resolution when retrieving it; the default is 15-minute values, matching schedules and quarter-hour products.
Can I retrieve forecasts for a whole portfolio?
Yes. One request returns the forecasts for all plants in your portfolio, with an optional sum column. You use your own plant IDs, so the data flows straight into your systems.
Are curtailments taken into account?
Yes. You send curtailments and unavailability, for example by the grid operator, for maintenance or for market reasons, via API. The model doesn't learn from distorted measurements, and the forecast stays realistic.
How much does the alitiq solar forecast cost?
€18 per MWp of installed capacity per month, with all forecast services included and no different plans. Larger portfolios get volume discounts, and the first 30 days are free.
How can I check the forecast quality?
The API lets you retrieve earlier forecasts as well as the latest one and compare them with your measurements. The most meaningful test runs for several weeks on your own plants, which is exactly what the free trial period is for.
Test the solar forecast with your own plants
Free for 30 days: you create your plants, we deliver the forecasts, and you compare them with your measurements. Then you decide.