Photovoltaic pump maximum power point prediction system and method based on multistage neural network

• Main Authors: HU JUN, LI RUIDONG, KE SHENGNAN, XI JIARUI, YANG YANG
• Format: Patent
• Language: Chinese; English
• Published: 05.03.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROLLING; COUNTING; PHYSICS; REGULATING; SYSTEMS FOR REGULATING ELECTRIC OR MAGNETIC VARIABLES

The invention discloses a photovoltaic pump maximum power point prediction system and method based on a multistage neural network, and belongs to the technical field of photovoltaic pump control, and the system comprises a data acquisition module, a preprocessing module, a neural network model construction module and a prediction module. The method specifically comprises the following steps that S1, multiple sets of sample data are obtained through a photovoltaic solar water pump actual model; s2, inputting multiple groups of sample data into a multi-stage neural network; s3, determining structure parameters of the multi-stage neural network, and performing preprocessing; s4, constructing a multi-stage neural network model; and S5, predicting the maximum power of the photovoltaic solar water pump in the actual application process by using the trained multi-stage neural network model. According to the photovoltaic pump maximum power point prediction system and method based on the multilevel neural network, the