Self-adaptive inverse tracking control method for super-magnetostrictive tracking platform

Abstract

The invention discloses a self-adaptive inverse tracking control method for a super-magnetostrictive tracking platform. The self-adaptive inverse tracking control method comprises the following stepsthat firstly, a left inverse model of a super-magnetostrictive actuator is obtained through offline identification; secondly, based on a Filtered-epsilon LMS algorithm, two completely identical nonlinear filters are used for copying the left inverse model of the super-magnetostrictive actuator; finally, an error signal obtained by output subtraction of the two nonlinear filters is utilized to finda controller in a self-adaptive mode, the weight coefficients of the filters are adjusted online by adopting the LMS algorithm till the output of the super-magnetostrictive actuator is identical to the output of a reference model, wherein the input of one of the two completely identical nonlinear filters is an object output, and the input of the other nonlinear filter is the output of the reference model. By offsetting the delaying portion of the super-magnetostrictive actuator, accurate tracking control of the super-magnetostrictive actuator is achieved.

Description

technical field

[0001] The invention relates to an adaptive inverse tracking control method for a giant magnetostrictive tracking platform, belonging to the field of dynamic hysteresis nonlinear system modeling and control. Background technique

[0002] Smart materials commonly used in modern industry include piezoelectricity, giant magnetostriction and shape memory alloys. They exhibit coupling characteristics of electricity, heat, magnetism, and force fields, which can be used to design them as actuators or sensors. For positioning tracking of micron-scale displacement, giant magnetostrictive materials are usually used. However, this smart material has serious dynamic hysteresis nonlinear characteristics. The nonlinear characteristic of dynamic hysteresis not only reduces the control precision of the control system, but also reduces the stability of the closed-loop system and even causes the system to oscillate.

[0003] Adaptive inverse control uses the inverse of the ...

Examples

Embodiment 1

[0043] Embodiment 1 of the present invention: a kind of adaptive inverse tracking control method for giant magnetostrictive tracking platform, such as figure 1 As shown, the following steps are included: First, the left inverse model of the giant magnetostrictive actuator is obtained through offline identification; secondly, based on the Filtered-ε LMS algorithm, two identical nonlinear filters are used to replicate the giant magnetostrictive actuator The left inverse model of the device; finally, use the error signal obtained by subtracting the output of the two nonlinear filters to adaptively find the controller C, and use the LMS algorithm to adjust the weight coefficient of the filter on-line until the giant magnetostriction The output of the actuator is the same as the output of the reference model; wherein, the two identical nonlinear filters, wherein the input of one nonlinear filter is the object output; the input of the other nonlinear filter is the reference model ou...

Embodiment 2

[0056] Embodiment 2: a kind of adaptive inverse tracking control method for giant magnetostrictive tracking platform, such as figure 1 As shown, the following steps are included: first, the left inverse model of the giant magnetostrictive actuator is obtained through offline identification; secondly, based on the Filtered-εLMS algorithm, two identical nonlinear filters are used to replicate the giant magnetostrictive actuator The left inverse model of the left inverse model; finally, use the error signal obtained by subtracting the output of the two nonlinear filters to adaptively find the controller C, and use the LMS algorithm to adjust the weight coefficient of the filter online until the giant magnetostrictive action The output of the filter is the same as the output of the reference model; wherein, the two identical nonlinear filters, wherein the input of one nonlinear filter is the object output; the input of the other nonlinear filter is the output of the reference model...

Embodiment 3

[0058] Embodiment 3: a kind of adaptive inverse tracking control method for giant magnetostrictive tracking platform, such as figure 1 As shown, the following steps are included: first, the left inverse model of the giant magnetostrictive actuator is obtained through offline identification; secondly, based on the Filtered-εLMS algorithm, two identical nonlinear filters are used to replicate the giant magnetostrictive actuator The left inverse model of the left inverse model; finally, use the error signal obtained by subtracting the output of the two nonlinear filters to adaptively find the controller C, and use the LMS algorithm to adjust the weight coefficient of the filter online until the giant magnetostrictive action The output of the filter is the same as the output of the reference model; wherein, the two identical nonlinear filters, wherein the input of one nonlinear filter is the object output; the input of the other nonlinear filter is the output of the reference model...

technical field

[0001] The invention relates to an adaptive inverse tracking control method for a giant magnetostrictive tracking platform, belonging to the field of dynamic hysteresis nonlinear system modeling and control. Background technique

[0002] Smart materials commonly used in modern industry include piezoelectricity, giant magnetostriction and shape memory alloys. They exhibit coupling characteristics of electricity, heat, magnetism, and force fields, which can be used to design them as actuators or sensors. For positioning tracking of micron-scale displacement, giant magnetostrictive materials are usually used. However, this smart material has serious dynamic hysteresis nonlinear characteristics. The nonlinear characteristic of dynamic hysteresis not only reduces the control precision of the control system, but also reduces the stability of the closed-loop system and even causes the system to oscillate.

[0003] Adaptive inverse control uses the inverse of the ...

Examples

Embodiment 1

[0043] Embodiment 1 of the present invention: a kind of adaptive inverse tracking control method for giant magnetostrictive tracking platform, such as figure 1 As shown, the following steps are included: First, the left inverse model of the giant magnetostrictive actuator is obtained through offline identification; secondly, based on the Filtered-ε LMS algorithm, two identical nonlinear filters are used to replicate the giant magnetostrictive actuator The left inverse model of the device; finally, use the error signal obtained by subtracting the output of the two nonlinear filters to adaptively find the controller C, and use the LMS algorithm to adjust the weight coefficient of the filter on-line until the giant magnetostriction The output of the actuator is the same as the output of the reference model; wherein, the two identical nonlinear filters, wherein the input of one nonlinear filter is the object output; the input of the other nonlinear filter is the reference model ou...

Embodiment 2

[0056] Embodiment 2: a kind of adaptive inverse tracking control method for giant magnetostrictive tracking platform, such as figure 1 As shown, the following steps are included: first, the left inverse model of the giant magnetostrictive actuator is obtained through offline identification; secondly, based on the Filtered-εLMS algorithm, two identical nonlinear filters are used to replicate the giant magnetostrictive actuator The left inverse model of the left inverse model; finally, use the error signal obtained by subtracting the output of the two nonlinear filters to adaptively find the controller C, and use the LMS algorithm to adjust the weight coefficient of the filter online until the giant magnetostrictive action The output of the filter is the same as the output of the reference model; wherein, the two identical nonlinear filters, wherein the input of one nonlinear filter is the object output; the input of the other nonlinear filter is the output of the reference model...

Embodiment 3

[0058] Embodiment 3: a kind of adaptive inverse tracking control method for giant magnetostrictive tracking platform, such as figure 1 As shown, the following steps are included: first, the left inverse model of the giant magnetostrictive actuator is obtained through offline identification; secondly, based on the Filtered-εLMS algorithm, two identical nonlinear filters are used to replicate the giant magnetostrictive actuator The left inverse model of the left inverse model; finally, use the error signal obtained by subtracting the output of the two nonlinear filters to adaptively find the controller C, and use the LMS algorithm to adjust the weight coefficient of the filter online until the giant magnetostrictive action The output of the filter is the same as the output of the reference model; wherein, the two identical nonlinear filters, wherein the input of one nonlinear filter is the object output; the input of the other nonlinear filter is the output of the reference model...

Images