SynaptiQ, 3E’s portfolio performance monitoring platform, manages over 7000 parks and thus processes vast amounts of operational data, communicating with a community of specialized software services based on flexible and standardized interfaces.
Next to the management of operational data, 3E is committed to invest in the analysis of this data by developing tools designed to improve and automate the optimisation of performance. In this context, we developed the <PV Health Scan. This service offers data analytics based on artificial intelligence and machine learning.
3E now also applies this validated technology to its technical consultancy services, and in this case specifically to the verification of the performance behaviour of PV plants, one of the most important steps in the professional management of PV portfolios.
Investors and financiers are well-aware of this, so they impose sophisticated contractual conditions to the EPC and O&M contractors in order to assure that the guaranteed values of Performance Ratio (PR) and Availability are met. Otherwise, substantial Liquidated Damages (LD) are triggered, along with extensive discussions. Surprisingly enough, despite the well-structured testing protocols, complicated formulae and highly-technical definitions present in the contracts, the effective way performance indicators are calculated is far from sophisticated, mostly made via unmanageable Excel spreadsheets.
Make no mistake, the task is enormous. For instance, to calculate the Availability percentage based on fifteen-minute data input over 12 months, the tool created in Excel will have more than 35 000 rows. Add one column per data entry (strings, pyranometers, etc.) plus each contractual condition that needs to be modelled sequentially based on irradiation, pyranometer readings, grid outages, etc.
These spreadsheets are so heavy and complicated, that it is virtually impossible to detect data quality issues, wrong readings or inconsistencies in formulas across the different sheets. Hence, all the efforts dedicated to ensuring that the performance indicators are reached are diluted by an operational fact, creating frustration and ambiguity between the parties involved.
At 3E we think there should be a better method to perform this critical operation. That’s why our solar PV experts and big data researchers created a tool that will disrupt the way performance verification is calculated: the PV Automated Performance Verification or PV-APV.
By modelling the client’s PV plant in SynaptiQ, 3E’s advanced monitoring software, the PV-APV automated tool coded in Python will run through the data – robustly, regardless of the number of rows or columns – providing flawless results, not only PR and Availability values, but also valuable information such as data quality (e.g. detecting if there are gaps in the data and stating the percentage per month that, for instance, pyranometer readings had to be substituted by satellite values), which can serve as an independent sensor check.
The PV-AVP tool is another example on how 3E is playing a major role in the energy transition, bringing together the most advanced big data expertise with all the knowledge on PV and Wind gathered in almost 20 years in the market.