Research On Key Technologies Of Wireless Charging System Used In Substation Patrol Robots

Inspection work is the basis of ensuring safe and stable of the equipment in transformer substations.In recent years,the traditional manual inspection methods are facing severe challenges with the increase in the number and scale of substations.The method using substation patrol robots is intelligent,accurate and adaptability to the environment.It has gradually become a common means of testing substation equipment.However,there are many problems in contact charging method which is not conducive to the realization of unattended substation.Wireless charging technology of substation patrol robots has many advantages compared with contact charging method.There is no need for electrical connection and leakage interface.So wireless charging technology doesn’t have problems of contact sparking,wear aging and so on.The wireless charging system occupies less space because the transmitting coil of which is buried underground.What’s more,it can operate safely and steadily in harsh environment.It is found by actual test that the wireless charging system described in this paper needs to be studied in three key technologies:coil optimization,magnetic shielding and metal detection.So the system can run efficiently,stably and safely.The first one is that the transmission efficiency decreases greatly when the transmitting coil and the receiving coil are offset.The reason for this problem is that the transmission power as well as the size of transmitting coil and receiving coil are relatively small.The second one is that there is leakage magnetic field around substation patrol robots can interfere with parts of robots such as the encoder and communication module.It will influence normal work of robots while charging.Another question is that metal foreign bodies mixed into charging area will generate a lot of heat due to eddy current effect,which will reduce the transmission efficiency of the system and bring security risks.The basic theory of the magnetically coupled resonant wireless charging system is analyzed in this paper.Which includes the components,working principle,equivalent circuit,calculation formulas of power and efficiency,topological structure and control strategy of the system.On this basis,the optimization of coil structure and parameters is studied.We use beam-splitting coil structure with optimized parameters in order to optimize the conventional planar disc coil structure.It can improve the uniformity of magnetic field in the plane of charging area.And it will effectively improve the energy delivery efficiency when transmitting coil and receiving coil are offset.The parameter of beam-splitting coil are optimized with MATLAB.ANSYS Maxwell simulation software is used to model and simulate the new coil structure.And magnetic field distribution of the new coil structure and the conventional planar disc coil structure are compared and analyzed in this paper.Finally,the experimental platform is used to verify the result.The leakage flux of wireless charging system affects normal charging of the robot.In this paper,we use aluminum shell of robots combined with ferrite core as the main shielding method and resonant reactive coil as the auxiliary shielding method.The shielding principle of the main shielding method is analyzed.And we use the ANSYS Maxwell simulation software to establish simulation models and analysis shielding effect.The structure and parameters of resonant shielding coil are designed according to simulation results above.The shielding effect is simulated and analyzed and the experimental verification is carried out at last.A new method of metal detection using BP neural network is proposed in this paper aiming at the third problem.The sample data is collected through a large number of experiments using the wireless charging system.The feasibility of metal detection using BP neural network algorithm is discussed combining changing characteristics of input physical variables in different states during charging process.And the BP neural network model with the best performance is designed and simulated.The simulation result show that the BP neural network model established in this paper can meet the requirements of metal detection in wireless charging system and achieve high-precision metal detection function.