Optimal Design Of RBWO Repetitive Pulsed Guiding Magnetic Field System Based On Genetic Algorithm

Relativistic backward wave oscillator?RBWO?is one of the most competitive and promising high power microwave devices with characteristics like high power,high efficiency and the best adaptability for repetitive pulse operating conditions.The repetitive pulse operating condition of RBWO requires a strong guiding magnetic field,which is the key factor that sets limit to the miniaturization of the system and the operating frequency.It also directly affects the microwave output power and efficiency.Thorough and deep research in the guiding magnetic field system is of great significance to promote repetitive pulse high power microwave technology to develop in the direction of practical applications.The repetitive operation of RBWO can effectively improve the average microwave output power,but it is limited by the thermal effect of the magnet,the power of the power supply system and other factors.In this paper,optimization has been done to the design of RBWO's repetitive pulse guiding magnetic field system according to its practical requirements.Because the system demands a high frequency for its repetitive operation,the author designed a discharge topology with energy feedback to improve the energy utilization of the power supply and reduce the need to supplement energy for each discharge,so as to improve the repetitive operation frequency of the system.When the system is working,there are coupling relations among its electromagnetic field,thermal field,mechanical field and other physical fields,for example,the temperature rise of the magnet will result in changes of the coil's resistance.When the system works continuously,the parameters of the magnet change over time,so we have to do iterative calculations on all the parameters of the system with its multiple physical fields coupled,which puts a high demand on the design of the system.In this paper,the basic characteristics of the power system and magnet system are modeled respectively,and an iterative calculation model of the repetitive operation process was established in MATLAB.Taking the magnet's height?h?,diameter?a₂?and the capacity of the capacitor?C?as variables to be designed,the magnet's stress and power supply as constraints and the maximum repetitive operating frequency and the magnet's temperature as optimization goals,the author did a joint optimization to the design of the power system and the magnet system based on a Post-decision Pareto frontier multi-objective genetic algorithm in MATLAB.It provides an optimization design method for the RBWO's repetitive pulse guiding magnetic field system.While optimizing the temperature rise of the magnet,it also provides RBWO with a higher repetitive operating frequency and reduces the interval between two discharges.On the basis of all optimizations above,this paper also deals with the hardware design for the charging circuits and software design for the control system.To verify the effects of the optimization,the author ran several simulations in MATLAB and COMSOL according to the parameters derived from the optimization design,made the magnet which is supposed to generate the guiding magnetic field and did a series of experiments using the specifically designed circuits.The author finally achieved a 3T magnetic field with its repetitive frequency at 20Hz and verified the validity of the optimization.