Principle And Characteristic Studies Of Remotely Measuring Angular-velocity With A Pair Of Photovoltic Cell Arrays

Rotating machinery is a common mechanical structure form,such as ship propellers,aiming and tracking apparatus of a weapon's actuators,transmitted chains of various processing machine tools,etc.Rotation parameters,such as angular-velocity,angular displacement and angular acceleration,are important parameters to determine the working status and the movement characteristics of these rotating devices or rotary systems.The accurate measurement of transient angular-velocity can help us to understand the operation states and performance,and to decide the kinetic characteristics for improving the accuracy of movement process controls of rotating systems,and also to acquire fault information in machine diagnoses.Therefore,the exact measurement of angular-velocity is vitally important and it has been widely applied to mechanical testing.Considering the common deficiency that an extra tags or a precise circular-division-disk installed coaxially are needed for the angular-velocity sensor at present,this dissertation has presented a new approach to remotely measuring angular-velocity with a pair of photovoltaic cell arrays,based on spatial filtering technique?SFT?.Here,the principle of sensing angular-velocity is studied,and the characteristic of this sensing method are analyzed.After introducing the existing research actuality of rotation velocity measurement in the world,the development history and technical features and applications of the SFT are described,and the principle of sensing rotation velocity with SFT is presented in detail.Two linear photovoltaic-arrays are constructed by using several photocells,then a spatial filtering angular-velocity sensor?SFAS?is made up by arranging the two photovoltaic linear arrays with a certain interval with each other.The SFAS has both functions of spatial filtering and photovoltaic conversion,and it can be used to measure the rotation velocity and rotational direction of a target body.This dissertation has suggested two methods of identifing the rotational center.One is based on the relative change ratio of central frequencies,and another is on the relative variation of spectrum peaks,in remotely sensing angular-velocity with SFT.In the relative-change-ratio based method,a coefficient j is calculated by using two central frequencies f₁ and f₂,and their variations,which are obtained from the output signals of the SFAS.Then depending on the sign of the coefficient j,together with the relation of f₁ and f₂,the location zone of the imaged rotational center is distinguished.In the method on the relative variation of spectrum peaks,a mixed signal is obtained at first by adding two output signals of SFAS together,then the fast Fourier transform?FFT?is applied to obtaining two spectral peaks of the mixed signal,which are corresponding to two central frequencies of the mixed signal.Then the SFAS is moved with a certain distance in the radial direction,and two spectral peaks of the mixed signal are got again by using the FFT.At last,the location zone of the imaged rotation center is determined by the change tendency of the two spectral peaks before and after moving the SFAS.These suggested methods have solved the problem of identifing rotational center in the SFT.Two identification methods of the rotational direction in remotely measuring angular-velocity with SFT were proposed,which include power-spectrum-based method and differential phase method.In the power-spectrum-based method,the identification coefficients d₁ and d₂ of two linear-velocity directions are calculated according to the feature that the power spectra of two output signals from one differential spatial filter?DSF?with duty ration of 0.5 are conjugated.The rotational direction and center-location of measured body are identified according to the signs of the coefficients d₁ and d₂,together with the relationship between f₁ and f₂.The power-spectrum method has been improved to be the differential phase method which is realized by reconstructing the SFAS to be a complex-differential-spatial-filter-based SFAS?CDSFAS?whose DSFs are with duty ratio of 0.25.One phase difference is derived from the two orthogonal output signals of a complex differential spatial filter?CDSF?in the CDSFAS,and its sign shows the movement direction of the image on the CDSF.Two phase differences D₁ and D₂ can be obtained from the four output signals of the CDSFAS.Then the transient rotation direction and center location zone of the target object are determined by the signs of D₁ and D₂,together with the relation between f₁ and f₂.Both the power-spectrum-based method and the differential phase method can solve the difficult problem of identifying rotational direction in the SFTInfluences of the transmission function and the amounts of the spatial period that acting on the performance of the spatial filter,and influence of the duty ratio that acting on the spectrums of the spatial filtering signals were analyzed,respectively.The results have indicated that there are different amplitude and phase of the outputs of the spatial filter when the amounts of the spatial period and the duty ratio are different.Additionly,the influences of the fluctuation of the magnification of the optical system,the installation eccentricity of the sensor,and the axial endplay and the radial waggle of the measured body that acting on the measurement of the rational velocity were also presented.And the results have shown that the rotational velocity sensing method is without any circular indexing plate,and is insensitive to the axial endplay,radial fluctuation,and parallel movement of the target object.Confirmation experiments of two rotation-center identification methods were took,and their results have illustrated that the identification methods can effectively distinguish the location zone of the rotation center in case of the rotation velocity is stable.Additionally,the correctness of the proposed rotational direction identification methods is confirmed.And the results have shown that the power-spectrum-based method can effectively identify the rotational direction in the remote angular-velocity measurement with SFT,and that it has the features as less computational amount and convenient application.The differential phase method can exactly distinguish the instantaneous rotation-direction of the measured body,which has the features as being single and easy for applications,and having strong adaptability,excellent instantaneity and high accuracy.The angular-velocity remote-sensing method based on a pair of linear photovoltaic arrays with self-scale-division and non-contact,can be used to detect the rotation velocity and the rotational direction of a target object.This sensing method has overcome the difficult problem that a precise circular indexing plate must be coaxially installed on the measured object,and its sensing system is simple and its environmental suitability is nice.What's more,the suggested sensing method has solved a difficult problem of identifying the rotational direction of the measured body in real-time.As a result,the sensing method with strong universality and good application prospect is established for remotely measuring rotation velocity of solid body or vortex flow velocity of liquid.