System and method for controlling solar photovoltaic power generation on basis of machine learning

A machine-learning-based photovoltaic power generation control system comprising:

• a plurality of photovoltaic modules configured to convert photovoltaic energy into electrical energy in order to perform photovoltaic power generation, the photovoltaic modules being connected to each other in series so as to be arranged in at least two rows and at least two columns;
• node controllers configured to measure voltage, current, and electric power data produced by the photovoltaic modules and, when the measured data are less than set current, voltage, and electric power control data, to switch off the photovoltaic modules connected thereto;
• a gateway unit configured to receive the voltage, current, and electric power data measured by the photovoltaic modules from the node controllers, to parse the measured data, and to store the measured data;
• a real-time control module configured to receive the parsed data from the gateway unit, to sort, compare, and analyze the received data, to store the data in a database, and to transmit a control command for setting control data necessary to control the node controllers to the gateway unit;
• and
• a machine learning server connected to the real-time control module over an Internet or a communication network, the machine learning server being configured to monitor a photovoltaic power generation construction device comprising the photovoltaic modules and data, to learn the data transmitted from the real-time control module based on machine learning, to extract functional data necessary to control photovoltaic power generation to perform modeling, and to provide control service data based on a result of modeling to the real-time control module,
• wherein each of the node controllers comprises:
• a measurement unit configured to measure voltage, current, and electric power data produced by a corresponding one of the photovoltaic modules;
• an analysis and control unit configured to compare the measured data with the control data transmitted from the real-time control module, the control data being defined to control current and voltage, and to determine whether the measured data are less than the defined control data to switch an operation of a corresponding one of the photovoltaic modules;
• a data transmission and management unit configured to transmit data stored in the node controller to an outside, to receive the control data from the real-time control module, and to transmit the received control data to the node controller;
• and
• a switching unit configured to switch off or on the photovoltaic module connected to the node controller based on a result of determination of the analysis and control unit.