Abstract: The present invention are a system and a method for controlling solar photovoltaic power generation on the basis of machine learning, the system comprising: solar photovoltaic modules; node control units for switching off a connected solar photovoltaic module when measured current, voltage and power data do not satisfy control data; a gateway unit for storing measured data; a real-time control module for classifying, comparing and analyzing data and storing same, and transmitting a control command to the gateway unit; and machine learning for monitoring a device and data, learning on the basis of machine learning, and extracting functional data required for controlling solar photovoltaic power generation so as to provide control service data according to the result of performed modeling.

Abstract: The present invention relates to a system and a method for controlling photovoltaic balancing, the system comprising: photovoltaic modules wherein the photovoltaic modules are connected in series with each other; node balancing control units for blocking and switching the connected photovoltaic modules when measured data fall below current, voltage, and power control data set to control; a gateway unit for storing the measured data; a real-time control module for classifying, comparing, and analyzing the measured data, storing same in a database, and transmitting a control command to the gateway unit; and an integrated information server for monitoring photovoltaic component devices and measured values, analyzing and processing profile information of the component devices, and providing same to the real-time control module.

Abstract: The present invention relates to a method of extracting a roof edge image for solar panel installation by using machine learning, the method comprising: a training step for passing original rooftop image data through a second generation unit of an image extraction system to output an image similar to a target image, and passing image data, from which a rooftop edge has been detected, through a first generation unit of the system to identify the image data from an original image; a step for segmenting an obstruction hiding a roof edge, and receiving, by a second discriminator unit, an image in which the roof edge has been detected; a step for optimizing the weight of a parameter, and training the second generation unit and the second discriminator unit again; and a step for automatically connecting edge portions after extracting edges, and generating a complete roof edge image.