Research On Sensorless FOC Algorithm Of Brushless DC Motor Based On Single Chip Microcomputer

In recent years,as the core component of domestic industrial automation,intelligence,and miniaturization,the market demand for motors has been increasing year by year,and the demand for localization of motor modules has increased rapidly.Domestic research on micro-controller-based motor drive algorithms has become a hot spot in current industrial development.Brushless DC motor(BLDCM),as a kind of servo motor,is widely used in servo fields such as robotic arms,industrial control,medical,military,aerospace,and home appliances.Although the sensory drive algorithm has relatively high accuracy,its cost is high,and it is easily affected by temperature,humidity,mechanical installation and the sensor’s own factors.Therefore,the sensorless drive has received more attention in recent years.The current non-inductive drive algorithm direct torque control(DTC),although it can realize basic rotor position information estimation,can only be applied to some simple systems,and the estimation accuracy is not high.Therefore,how to estimate the rotor position information and confirm the rotor sector is the key to whether it can be driven normally.The selection and structural design of the position observer is particularly important for the non-inductive drive algorithm.This paper mainly studies and analyzes the sensorless FOC algorithm based on the traditional sliding mode observer(SMO),establishes the mathematical model,and designs and implements the software design and implementation of the LS2430single-chip microcomputer based on the ARM-M0 core,in addition of corresponding improved methods are proposed respectively for the shortcomings of the traditional SMO position observation algorithm.For the traditional SMO algorithm based on the sgn function,there is chattering caused by the discontinuity and unevenness of the signed function,and the sigmoid function is used to replace the switching function sgn to eliminate the chattering.For the low estimation accuracy of the traditional SMO algorithm,the position observation algorithm based on the extended Kalman filter(EKF)under non-linear conditions is used to improve the estimation accuracy of the rotor position information.For the position observation algorithm based on SMO,it is only applicable to the estimation of rotor position information at medium and high speeds.The position observation algorithm of high frequency voltage injection(HFP)realizes the estimation of the rotor position information at zero low speed,It solves the strong jitter caused by the unestimation of rotor position information at zero low speed.In this paper,the sensorless FOC algorithm is mainly composed of Clark transform,Park transform,inverse Park transform,PID algorithm,spatial voltage pulse-width modulation(SVPWM)algorithm and position observation algorithm.Among them,the Clark transformation transforms the three-phase sampling current into the current in the two-dimensional static coordinate system,the Park transformation transforms the current in the two-dimensional static coordinate system into the current in the rotating coordinate system,and the inverse Park transform transforms the AC-DC axis voltage in the rotating coordinate system into the AC-DC axis voltage in the static coordinate system,the PID algorithm performs speed loop modulation and current loop modulation on the given speed and estimated speed,the SVPWM algorithm realizes the confirmation of the rotor sector position,the position observation algorithm realizes estimation of rotor position information.Through dynamic debugging and building a system simulation model,the simulation results show that the improved algorithm not only eliminates the bucket vibration,but it improves the estimation accuracy of the traditional SMO algorithm for the rotor position information.The improved algorithm has a rotor speed estimation error of-310-210r/min is reduced to-50-50r/min,and the rotor position estimation error is reduced from 0-0.45 rad to-0.25-0.05 rad,and-0.25 rad only appears at the moment of startup.The overall estimation accuracy is somewhat different.Improve.It also realizes the estimation of the rotor position information of the motor at zero/low speed and reduces the jitter caused by the strong drag when the motor starts.