CN104806861A - Leveling method for multi-shaft support air-floating platform based on capacitive sensors - Google Patents

Abstract

The invention discloses a leveling method for a multi-shaft support air-floating platform based on capacitive sensors and belongs to the optical field. The method is used for solving the problems that the current method for adjusting a support air-floating platform is high in requirement on a measurement environment, is far limited during application, and is incapable of realizing high regulation precision and low complexity in regulation process at the same time. The specific procedures of the leveling method are as follows: three air feet are inflated with high pressure gas from the bottom by an air source, so as to form a gas film face; gas film heights of the three feet are collected respectively by three capacitive sensors; the collected gas film heights are output into a controller respectively by the three capacitive sensors; the controller calculates controlled quantity by adopting a neural network PID control algorithm; the controlled quantity controls pressure regulating valves of the three air feet after decoupling, so as to change the outlet pressure of the three air feet, regulate the air film heights and finish the leveling process. The leveling method is applied to satellite ground simulation.

Description

Based on the leveling method of the Multi-shaft support air floatation platform of capacitive transducer

Technical field

The present invention relates to a kind of leveling method of Multi-shaft support air floatation platform.

Background technique

Multiaxis air floating platform is the visual plant of satellite ground emulation, for microgravity, micro-damping environment of simulation space.The air film that air floating platform relies on pressurized air to be formed in pneumatic bearing supports, and realizes micro tribology and micro-damped motion.

In the middle of multiaxis air floating table emulation technology, be accurately vital core link to air floating table leveling.Produce static unbalance when there is drift for center of gravity relative to gyration center, and there is the problem of dynamic unbalance relative to turning axle in principal axis of inertia, first adopts manual leveling, then adopt self-leveling method, regulate the balance of air floating table.By the stage body levelness etc. of effective means of testing test air platform, the emulation for small satellite attitude control system provides effective, an accurate platform.In prior art, there is multiple leveling method, but there is many deficiencies.Such as: based on the laser centroid adjustment method of CCD, based on the barycenter intelligent adjusting method etc. of empirical modal method.The former governing speed is fast, but requires to measure have no occluder between measured article, has larger narrow limitation during application.The latter's degree of regulation is high, but operating process is quite complicated, and regulating time is long.

The domestic research to multiaxis air floating table is just carried out soon, and relevant every research is not deeply.Therefore the leveling method studying multiaxis air floating platform has vital effect for multiaxis air floating table emulation technology.

Summary of the invention

The present invention seeks to require high in order to solve the existing regulating method to air floating platform to measurement environment, during application, limitation is larger, the problem that degree of regulation is high and adjustment process complexity is low can not be reached simultaneously, provide a kind of leveling method of the Multi-shaft support air floatation platform based on capacitive transducer.

The leveling method of the Multi-shaft support air floatation platform based on capacitive transducer of the present invention, the detailed process of this leveling method is:

Step 1, source of the gas are filled with pressurized gas in the bottom of three gas foots, form air film face;

Step 2, employing three capacitive transducers gather the air film height of three gas foots respectively;

The air film height of collection exports in controller by step 3, three capacitive transducers respectively;

Step 4, controller adopt Neural Network PID to calculate controlled quentity controlled variable;

Control the pressure regulator valve of three gas foots after step 5, controlled quentity controlled variable decoupling zero, change the outlet pressure of three gas foots, adjustment air film height, completes leveling process.

The concrete grammar calculating controlled quentity controlled variable described in step 4 is:

$u\left(k\right)=u(k-1)+K\underset{i=1}{\overset{3}{\mathrm{\Σ}}}(w_i\left(k\right)/\underset{i=1}{\overset{3}{\mathrm{\Σ}}}|w_i\left(k\right)\left|\right)x_i\left(k\right);$

Wherein: k represents sampling instant, k>=0, controlled quentity controlled variable when u (k) represents that the moment is k, controlled quentity controlled variable when u (k-1) represents that the moment is k-1, K represents neuronic scaling factor, K>0, w _ithe controlled quentity controlled variable of i-th gas foot when () represents that the moment is k k, i=1,2,3, x _ik () represents the difference of i-th gas foot k moment and the input of k-1 moment error;

w ₁(k)＝w ₁(k-1)+η _PZ(k)u(k)x ₁(k)，

w ₂(k)＝w ₂(k-1)+η _IZ(k)u(k)x ₂(k)，

w ₃(k)＝w ₃(k-1)+η _DZ(k)u(k)x ₃(k)；

W ₁(k-1) controlled quentity controlled variable of the 1st gas foot when representing that the moment is k-1, w ₂(k-1) controlled quentity controlled variable of the 2nd gas foot when representing that the moment is k-1, w ₃the controlled quentity controlled variable of the 3rd gas foot when () represents that the moment is k-1 k; x ₁k () represents the difference of the 1st gas foot k moment and the input of k-1 moment error, x ₂k () represents the difference of the 2nd gas foot k moment and the input of k-1 moment error, x ₃k () represents the difference of the 3rd gas foot k moment and the input of k-1 moment error; η _pZthe learning rate of expression ratio, η _iZrepresent the learning rate of integration, η _dZrepresent the learning rate of integration;

x ₁(k)＝e(k)-e(k-1)，x ₂(k)＝e(k)，x ₃(k)＝e(k)-2e(k-1)+e(k-2)，z(k)＝e(k)；

E (k) represents that error inputs.

Advantage of the present invention: in the present invention, adopts capacitive transducer to gather the altitude signal of air floating platform, after highly carrying out accurate Calculation, regulates the level of air floating platform.The resolution of capacitive transducer is very high, can reach nanometer.The leveling method antijamming capability that the present invention simultaneously proposes is strong, and Environmental Conditions is not limited to, and applies efficient and convenient.

Accompanying drawing explanation

Fig. 1 is the FB(flow block) of the leveling method of the Multi-shaft support air floatation platform based on capacitive transducer of the present invention;

Fig. 2 is the structural representation of Multi-shaft support air floatation platform of the present invention.

Embodiment

Embodiment one: present embodiment is described below in conjunction with Fig. 1, based on the leveling method of the Multi-shaft support air floatation platform of capacitive transducer described in present embodiment, the detailed process of this leveling method is:

Step 1, source of the gas are filled with pressurized gas in the bottom of three gas foots, form air film face;

Step 2, employing three capacitive transducers gather the air film height of three gas foots respectively;

The air film height of collection exports in controller by step 3, three capacitive transducers respectively;

Step 4, controller adopt Neural Network PID to calculate controlled quentity controlled variable;

Control the pressure regulator valve of three gas foots after step 5, controlled quentity controlled variable decoupling zero, change the outlet pressure of three gas foots, adjustment air film height, completes leveling process.

Embodiment two: present embodiment is described further mode of execution one, the concrete grammar calculating controlled quentity controlled variable described in step 4 is:

$u\left(k\right)=u(k-1)+K\underset{i=1}{\overset{3}{\mathrm{\Σ}}}(w_i\left(k\right)/\underset{i=1}{\overset{3}{\mathrm{\Σ}}}|w_i\left(k\right)\left|\right)x_i\left(k\right);$

Wherein: k represents sampling instant, k>=0, controlled quentity controlled variable when u (k) represents that the moment is k, controlled quentity controlled variable when u (k-1) represents that the moment is k-1, K represents neuronic scaling factor, K>0, w _ithe controlled quentity controlled variable of i-th gas foot when () represents that the moment is k k, i=1,2,3, x _ik () represents the difference of i-th gas foot k moment and the input of k-1 moment error;

w ₁(k)＝w ₁(k-1)+η _PZ(k)u(k)x ₁(k)，

w ₂(k)＝w ₂(k-1)+η _IZ(k)u(k)x ₂(k)，

w ₃(k)＝w ₃(k-1)+η _DZ(k)u(k)x ₃(k)；

W ₁(k-1) controlled quentity controlled variable of the 1st gas foot when representing that the moment is k-1, w ₂(k-1) controlled quentity controlled variable of the 2nd gas foot when representing that the moment is k-1, w ₃the controlled quentity controlled variable of the 3rd gas foot when () represents that the moment is k-1 k; x ₁k () represents the difference of the 1st gas foot k moment and the input of k-1 moment error, x ₂k () represents the difference of the 2nd gas foot k moment and the input of k-1 moment error, x ₃k () represents the difference of the 3rd gas foot k moment and the input of k-1 moment error; η _pZthe learning rate of expression ratio, η _iZrepresent the learning rate of integration, η _dZrepresent the learning rate of integration;

x ₁(k)＝e(k)-e(k-1)，x ₂(k)＝e(k)，x ₃(k)＝e(k)-2e(k-1)+e(k-2)，z(k)＝e(k)；

E (k) represents that error inputs.

Fig. 2 is the structural representation of Multi-shaft support air floatation platform of the present invention, this Multi-shaft support air floatation platform comprises pedestal (1), supporting leg (2), gas foot (3) and platform (4), source of the gas is filled with pressurized gas in the bottom of gas foot (3) and forms air film face, and platform (4) is moved at the upper Micro-friction that produces of pedestal (1).Capacitive transducer is arranged between pedestal (1) gentle foot (3).

Claims (2)

1. based on the leveling method of the Multi-shaft support air floatation platform of capacitive transducer, it is characterized in that, the detailed process of this leveling method is:

Step 1, source of the gas are filled with pressurized gas in the bottom of three gas foots, form air film face;

Step 2, employing three capacitive transducers gather the air film height of three gas foots respectively;

The air film height of collection exports in controller by step 3, three capacitive transducers respectively;

Step 4, controller adopt Neural Network PID to calculate controlled quentity controlled variable;

Control the pressure regulator valve of three gas foots after step 5, controlled quentity controlled variable decoupling zero, change the outlet pressure of three gas foots, adjustment air film height, completes leveling process.

2. according to claim 1 based on the leveling method of the Multi-shaft support air floatation platform of capacitive transducer, it is characterized in that, the concrete grammar calculating controlled quentity controlled variable described in step 4 is:

$u\left(k\right)=u(k-1)+K\underset{i=1}{\overset{3}{\mathrm{\Σ}}}(w_i\left(k\right)/\underset{i=1}{\overset{3}{\mathrm{\Σ}}}|w_i\left(k\right)\left|\right)x_i\left(k\right);$

Wherein: k represents sampling instant, k>=0, controlled quentity controlled variable when u (k) represents that the moment is k, controlled quentity controlled variable when u (k-1) represents that the moment is k-1, K represents neuronic scaling factor, K>0, w _ithe controlled quentity controlled variable of i-th gas foot when () represents that the moment is k k, i=1,2,3, x _ik () represents the difference of i-th gas foot k moment and the input of k-1 moment error;

w ₁(k)＝w ₁(k-1)+η _PZ(k)u(k)x ₁(k)，

w ₂(k)＝w ₂(k-1)+η _IZ(k)u(k)x ₂(k)，

w ₃(k)＝w ₃(k-1)+η _DZ(k)u(k)x ₃(k)；

W ₁(k-1) controlled quentity controlled variable of the 1st gas foot when representing that the moment is k-1, w ₂(k-1) controlled quentity controlled variable of the 2nd gas foot when representing that the moment is k-1, w ₃the controlled quentity controlled variable of the 3rd gas foot when () represents that the moment is k-1 k; x ₁k () represents the difference of the 1st gas foot k moment and the input of k-1 moment error, x ₂k () represents the difference of the 2nd gas foot k moment and the input of k-1 moment error, x ₃k () represents the difference of the 3rd gas foot k moment and the input of k-1 moment error; η _pZthe learning rate of expression ratio, η _iZrepresent the learning rate of integration, η _dZrepresent the learning rate of integration;

x ₁(k)＝e(k)-e(k-1)，x ₂(k)＝e(k)，x ₃(k)＝e(k)-2e(k-1)+e(k-2)，z(k)＝e(k)；

E (k) represents that error inputs.