Research On BP Neural Network Control Method For Harmonic Filter

In the production of enterprises,the power grid harmonic pollution is serious,and has a great impact on production efficiency and production safety.Harmonic governance,to enhance the production efficiency,increase the economic efficiency of enterprises,has a strong practical significance.In this paper,a BP neural network control method for harmonic filtering device is studied for the dynamic harmonic governance,supported by 2015 Wuhan Science and Technology Support Project(2015010101010013).The harmonic filter device can carry out harmonic filtering on the grid,realize intelligent dynamic harmonic control of power grid.The main work and achievements of this paper are as follows:(1)The research on BP neural network control algorithm of harmonic filter is completed.Based on the research of topology and mechanism of harmonic filtering system,the control principle of harmonic filter device is analyzed and studied.Through the research on the control principle of the filter device,the BP neural network algorithm is adopted as the control strategy of the filter device,which can meet the real-time dynamic control requirement of the harmonic filter device.Based on the theoretical analysis,the BP neural network structure of the harmonic filter device is designed.And,the BP neural network is achieved on the MATLAB platform.The BP control strategy of the harmonic filter device is analyzed and analyzed.(2)The simulation study of filtering effect of harmonic filter device is completed.Firstly,based on the topology structure of the harmonic filtering system,the system simulation model is constructed on MATLAB / Simulink,and the simulation scheme is put forward.Then,the main simulation modules of the system are designed.Finally,the parameters of the harmonic filter are analyzed and set up.The simulation results verify that the harmonic filtering effect of the BP neural network control method for harmonic filtering device is significant and the scheme is feasible.The simulation study provides technical support for the design of the harmonic filter using BP neural network control.(3)Completed the software design and hardware design of the harmonic filter device.The software design mainly includes the design of the overall structure of the software,the main program of the harmonic filter device,the power quality data processing program and the harmonic control program,and the human-computer interaction interface of the harmonic filtering device.The hardware design mainly includes the design of the hardware overall structure,as well as the selection and design of the main control module,the power signal acquisition module,the thyristor trigger module,the switch quantity signal acquisition and output module,the host computer communication module.Under the condition of the laboratory,the prototype load experiment is carried out to verify the filtering effect of the harmonic filter device using BP neural network control method.The scheme has the industrial significance.The research results of this paper can be applied to the production of enterprises,to meet the demand for dynamic harmonic control of power grid,reduce the cost of enterprises,and have practical application value.The device is expected to be popularized and applied in enterprises.