Method of controlling photovoltaic power generation system

A method of controlling a photovoltaic power generation system including controlling an inverter such that an output voltage or an output current from a solar cell becomes generally equal to a photovoltaic output setting value, so as to deal with a variation in a photovoltaic voltage-power characteristic caused by a change of the solar irradiation, the method for controlling the photovoltaic output setting value to be an appropriate value that makes the output power of the solar cell generally equal to a maximum value, the method comprising:

• a genetic algorithm process including employing a GA controller based on a genetic algorithm in which the photovoltaic output setting value serves as a gene and the output power of the solar cell serves as an evaluation value of the gene;
• a first step including extracting a plurality of genes randomly or according to a predetermined condition out of a setting range of the photovoltaic output setting value thus to form an initial group of a first generation, sequentially activating the inverter based on each photovoltaic output setting value respectively corresponding to each gene of the initial group and storing the output power from the solar cell under operation as the evaluation value of each of the genes, and inputting the genes of the initial group in the GA controller, so that the GA controller selects the genes and causes the selected genes to crossover/mutate based on the evaluation value thereof, so as to output a predetermined number of genes thus to form a second generation group;
• a second step including sequentially activating the inverter based on the photovoltaic output setting values corresponding to the genes of the second generation group and storing the output power from the solar cell under operation as the evaluation value of each of the genes, and inputting the second generation genes in the GA controller, so that the GA controller selects the genes and causes the selected genes to crossover/mutate based on the evaluation value thereof, so as to output a predetermined number of genes thus to form a third generation group;
• and repeating the second step so as to converge the genes thus to update the generation of the gene group, until a difference between a maximum value and a minimum value among the photovoltaic output setting values corresponding to the genes falls within a predetermined range;
• and
• a hill climbing process including selecting a photovoltaic output setting value that has the greatest evaluation value among the converged genes as a reference photovoltaic output setting value;
• generating a value greater by a predetermined amount and lower by a predetermined amount than the reference photovoltaic output setting value;
• sequentially activating the inverter based on the three photovoltaic output setting values so as to set the photovoltaic output setting value that gives the greatest output power as an updated reference photovoltaic output setting value;
• and repeating such steps, so as to return to the genetic algorithm process when a predetermined switching condition for following up the variation in the photovoltaic voltage-power characteristic is satisfied.