Data acquisition in a solar plant is critical for incident detection, performance analytics and data mining. This article details approaches to solar plant data acquisition based on Algo Engines experience.
In a solar plant we acquire data from inverters, combiner boxes, weather station, battery banks and transformers. The protocols used for data acquisition from each of the devices vary with modern devices using Modbus TCP, while older devices use Modbus RTU. A solar plant is usually broken up into zones or sections or units. Each unit is usually of 1 MW to 3 MW capacity and has a defined set of inverters, combiner boxes. The implementation of data acquisition devices vary based on the structure of each unit. Let us look at some of the standard structures that are present.
The data acquisition unit at the section or unit varies based on the combination used while all other components of the solution remain the same. Another call that we often discuss with clients is the needed for data redundancy. Based on the needed to maintain copies of data in case of failure of any component, we can plan data storage components at each of the unit or section level. The variation based on this is shown below.
The data acquired at site is displayed using the site monitoring tool and also replicated to the cloud to ensure remote monitoring. The replication of data on to the cloud will impact on the way the data can be analyzed. In many solutions the cloud is used only as a data backup rather than as the data analytics platform. This will impact the ability of customer to analyze across different sites, build machine learning models for maintenance based on multiple site data etc.