Prospect
Quick estimate of PV site's solar potential
Ground data verification
Ground-based solar and meteo measurements play a vital role in the solar industry. They are used for adapting models and evaluating the performance of solar power plants.
However, one of the key challenges of measured solar irradiance data is a high occurrence of anomalous values.
Before ground-based measurements are used for performance assessment or for adaptation of modeled time series, all suspicious values in the measured time series must be identified and flagged through quality control procedures.
Our Quality Control of Solar & Meteo Measurements service is based on Solargis experience handling measurements from thousands of locations globally. It helps you identify errors and prepare your datasets for the next steps of your project.
The magnitude of errors stemming from the cosine effect, temperature response, spectral sensitivity, stability, non-linearity, etc. depends on the quality of sensors and local conditions. So first, we review the technical specifications of sensors and their calibration certificates.
As a second step, we identify errors such as misalignment of sensors, shading by surrounding objects, etc., and flag the affected measurements.
In most cases, soiling of sensors (dust, snow, water droplets, frost, bird droppings, etc.) is difficult to prevent. However, we can identify these irregularities through thorough data analysis in Solargis Analyst.
In addition to flagging and removing errors, we substitute missing and erroneous data with inputs from our models. With that, we can create a harmonized, complete, gap-free time series dataset ready to use in performance evaluation.
In most cases, on-site data comes from several pyranometers. By harmonizing it, we replace erroneous values with modeled data and merge multiple datasets into one.
Unlike simple averaging, which can degrade the original quality of data, our statistical multicriteria approach preserves the natural features of measurements.
With our statistical approaches, we can adapt our model values to the same granularity as your measurements – for example disaggregating 15-min measured data to 1-min time series.
We issue a statement of uncertainty for the datasets that have been quality controlled. This enables you to use the data for bankable performance assessment and site adaptation of satellite-modeled historical time series.
