Research On High Precision Magnetic Encoder Algorithm Based On Erroe Compensation Principle

In recent years,the industrial automation industry is developing rapidly at an unprecedented rate.As an indispensable information measurement component of industrial automation control systems,the demand for encoders is increasing year by year.The encoder is widely used in machine tools,servo motors,gimbals,industrial robots,elevators and many other industrial fields related to industrial control applications.Although the common photoelectric encoder has relatively high accuracy,it is not easy to integrate due to its complicated structure and easy to be interfered by external environment such as dust,oil,water vapor,mechanical vibration,etc.The actual accuracy performance is not ideal.And its cost is high,which brings a certain cost burden.The magnetic encoder has a simple structure and is easy to integrate.It can still guarantee stable working output even under harsh environmental conditions,and has great application prospects in the field of industrial control.This subject proposes a design scheme of high-precision magnetic encoder module based on error compensation algorithm,which can not only effectively overcome the harsh external environment,but also provide higher position feedback accuracy and higher stability than ordinary magnetic encoders.Through mathematical modeling analysis and theoretical derivation of the factors that affect the position feedback accuracy of magnetic encoders,this paper separately designs targeted error compensation algorithms to improve its accuracy.For phase errors and amplitude inconsistencies,phase error compensation and amplitude normalization algorithms are used to resolve them;for loss of accuracy caused by process noise,median filtering and mean filtering are used to co-process;For the dynamic angle error generated,a targeted dynamic angle error compensation algorithm is designed to reduce its dynamic error.The high-precision magnetic encoder designed in this paper mainly includes signal acquisition unit,analog-to-digital conversion unit and microcontroller unit(MCU)processing.The signal acquisition unit consists of four linear Hall sensors arranged perpendicular to each other.In order to suppress the temperature drift and the induction error caused by the fluctuation of the magnetic field,the four linear Hall output signals are differentiated in pairs and two-phase analog signals are output.The analog signals of the two-phase magnetic encoder are converted by a high-speed analog-to-digital converter and further output corresponding digital sampling signals.The two-phase digital sampling signal is sent to the MCU,which is processed in order by filtering,amplitude normalization,phase non-orthogonal compensation,and dynamic angle error compensation.Finally,the front-end signal is angle-decoded and calculated using the Coordinate Rotation Digital Computer(CORDIC)method.The final external output angle of the magnetic encoder is the sum of the result calculated by the angle solution algorithm and the result of angle error compensation.In this paper,the validity of the algorithm is verified on an industrial control development board equipped with an STM32 microprocessor chip.The results show that the magnetic encoder algorithm designed in this paper can finally make the magnetic encoder reach 10 ? 11 bit resolution,and the real-time accuracy error can be reduced to the range of ± 0.2 °.