Research On Harmonic Detection Of Metro Power Supply System Based On Adaptive LMS Algorithm

Metro rail transportation has the characteristics of saving land resources,low per capita energy consumption,large passenger traffic,low pollution emissions,safety and reliability,and has become the backbone of public transportation in large and medium cities.However,the power electronics in the subway power supply system generate a lot of harmonics,which will cause resonance,shorten equipment life,increase power loss,interfere with the normal operation of communication functions and relay protection,and threaten the safe and stable operation of the power system.In order to create a better power supply environment to ensure the orderly operation of subway rail transit,it is urgent to control the harmonics generated in the subway power supply system.Harmonic detection is a prerequisite for harmonic control,and a faster and more accurate the harmonic detection algorithm is of great significance.Aiming at the characteristics of the harmonics in the subway power supply system,the adaptive least mean square(LMS)algorithm commonly used in harmonic detection of subway power supply systems is introduced.Based on the adaptive noise cancellation technology,the work of the adaptive LMS algorithm the principle and convergence performance are analyzed in detail.Aiming at the contradiction between convergence speed and steady-state error in the traditional fixed-step adaptive LMS algorithm,the idea of variable-step adaptive LMS algorithm was introduced.The variable step size adaptive LMS algorithm is typically a variable step size algorithm based on the hyperbolic tangent function,but this algorithm does not significantly solve the contradiction between convergence speed and steady-state error.Based on this,this paper proposes an LMS harmonic current detection algorithm based on an improved hyperbolic tangent function.This algorithm uses the ratio of the current time error to the historical error to dynamically control the step size of the algorithm,and introduces the automatic error signal system.Correlation time average estimation is used to control the influence of past data on the current state.At the same time,two iterations of the weight coefficient are performed,and the results of the two iterations are used as the new weight coefficient.The results of simulation experiments prove that the algorithm has both fast convergence speed and high detection accuracy,and it can effectively solve the contradiction between convergence speed and steady-state error.Finally,a Matlab/simulink platform is used to build a simulation model of the subway power supply system.The above three harmonic detection algorithms are applied to the simulation model for harmonic current detection and analysis.The simulation results show that the LMS harmonic current detection algorithm based on the improved hyperbolic tangent function converges significantly faster than the other two algorithms,and the detection accuracy is significantly higher than the other two algorithms,which can be faster and more accurate.The harmonics in the subway power supply system were detected by ground to verify the effectiveness and feasibility of the proposed algorithm.