CN104035334A - Generalized resistance force based hydraulic dynamic leveling method - Google Patents

Abstract

The present invention is a hydraulic dynamic leveling method based on generalized retarding force, which belongs to the technical field of hydraulic synchronous control; the technical problem to be solved is to provide a hydraulic dynamic leveling method for solving the problems of poor synchronization and real-time performance; the adopted technical scheme In order to: use the surface leveling method with stronger real-time and synchronism to analyze the force of the moving multi-point hydraulic support platform, introduce generalized blocking force, establish a mechanical model, and establish the state equation and The observation equation introduces the extended state variable added with the generalized resistance, uses the strong tracking filter to estimate the time-varying model parameters and state variables at the same time, and corrects the model parameters online according to the input and output measurements of the next plane. The self-adaptive controller of the stagnation force tracks and controls the abrupt state of the traveling process, and quickly levels; the invention can be applied to a hydraulic synchronous control system to realize precise control of multiple actuators in the hydraulic system.

Description

Hydraulic pressure dynamic leveling method based on broad sense drag

Technical field

The present invention is based on the hydraulic pressure dynamic leveling method of broad sense drag, be particularly related to the dynamic leveling method of mobile radar special vehicle and hydraulic suspension heavy-load transport vehicle, belong to hydraulic synchronization control technology field, multiple hydraulic cylinders adopt electro-hydraulic proportional valve to control, and by the method based on the modeling of broad sense drag, mobile hydraulic platform are implemented to automatic leveling.

Background technology

Hydraulic leveling technology is widely used in modern national defense and civilian technology, the equipment such as such as mobile radar, Canon launching platform, large type drill, static pile press, heavy hoisting machine, heavy goods vehicle must carry out leveling to its carrying platform before work, make carrying platform fast, stable, accurately adjust to horizontal level.The static leveling technology that these all belong to hydraulic platform has a large amount of documents and patent to carry out relevant research to it before.

And dynamic leveling technology is had higher requirement to the real-time and the rapidity that regulate, for example in mobile radar special vehicle traveling process, in the tracking to aerial target and when interception, need to keep the level of platform, or heavy-load transport vehicle need to keep vehicle frame level and steadily of centre of gravity during by bend, ramp.Therefore hydraulic pressure dynamic leveling Technology Need adopts real-time responsiveness stronger, and the better control algolithm of control performance ensures that the Horizontal consistency of multi-hydraulic-cylinder is the horizontality of hydraulic platform.

Hydraulic leveling technology is static or dynamic leveling all belongs to hydraulic synchronization control research field, the factor of the machineries such as the characteristic that is synchronized with the movement of multi-hydraulic-cylinder is very complicated, the manufacturing and fixing error of the performance difference under different operating modes between leakage, the control element of the frictional resistance of the disturbance of load, actuator, system, the mixed volume of air and the each ingredient of system, electric and hydraulic pressure etc. all can have influence on the stroke of hydraulic cylinder; The rigidity also existing in plane when each hydraulic cylinder moves simultaneously simultaneously involves coupling, exists flow coupling between each hydraulic circuit.From the current study, hydraulic synchronization control research is mostly concentrated in the class same sex of two or more hydraulic cylinders, the similar synchronism that ensures hydraulic pressure of oil circuit, components and parts, topworks and control parameter, substantially ignored the otherness of each hydraulic cylinder, and the rigidity while not having to consider to be synchronized with the movement in plane involves coupling, all there is obvious defect in the description of its mathematical model and control model.Adopt based on data analysis because the shortcoming based on mechanism model has a lot of leveling algorithm researches, such as fuzzy control, ANN (Artificial Neural Network) Control etc., these algorithms need a large amount of data samples, and its real-time exists very large problem simultaneously.In sum, rigidity in the otherness of the kinetic characteristic of hydraulic cylinder, leveling process in variation, the plane of load variations, oil temperature involves coupling etc. has different impacts to hydraulic cylinder travel, and these impacts are difficult to the quantitative calculating of going, thereby therefore how to eliminate these impacts in hydraulic pressure dynamic leveling, to reach the problem of higher leveling precision just very crucial, and the present invention will address this problem exactly.

Summary of the invention

The present invention overcomes the deficiency that prior art exists, technical matters to be solved is for providing a kind of for solving the problems such as the synchronism that exists in traditional leveling technology and real-time be poor, in hydraulic pressure dynamic leveling system, introduce broad sense drag, realize the hydraulic pressure dynamic leveling method of the multiloop real-time adaptive control of hydraulic leg.

In order to solve the problems of the technologies described above, the technical solution used in the present invention is: the hydraulic pressure dynamic leveling method based on broad sense drag, adopt that broad sense drag replaces that load variations in dynamic leveling process, oil temperature change, become when rigidity in each hydraulic cylinder plane involves coupling etc. and be difficult to the physical quantity of accurately surveying, and analyze the impact with output on the input of dynamic leveling system control model of these physical quantitys and broad sense drag, set up state equation and the input-output equation of the hydraulic pressure dynamic leveling system based on broad sense drag; The adaptive control algorithm that adopts strong tracking filter to time the nonlinear control system that becomes dynamic parameter, broad sense drag is estimated, obtains the precise control model of hydraulic pressure dynamic leveling system, according to the input quantity of system, eliminate and estimate residual error, control output quantity, complete leveling.

Hydraulic pressure dynamic leveling method based on broad sense drag, comprises the following steps:

A, introducing broad sense drag, analyze the impact of broad sense drag on hydraulic cylinder travel, sets up the broad sense drag θ that includes n hydraulic cylinder ₁, θ ₂θ _nstate equation and the input-output equation of hydraulic pressure dynamic leveling system, determine the sampling period T of system and regulate time t, according to the sampling period T of system, to its discretize, obtaining after discretize is proportioning valve valve position u with n hydraulic cylinder flow ₁, u ₂u _nfor input quantity, n hydraulic cylinder displacement Z ₁, Δ Z ₂Δ Z _nfor the system equation of output quantity, i.e. hydraulic pressure dynamic leveling system control model;

B, hydraulic platform are supported by n hydraulic leg, produce pitch angle, platform angle value α, the β that double-shaft tilt angle sensor is detected and the precision index α of platform when static ₀, β ₀change and import micro-control computing machine into by A/D;

C, write the adaptive control algorithm based on strong tracking filter, by this algorithm packing, programming, in implant controller, calls as a generalized subroutine, is used for according to inputoutput data to Parameter Generalized drag θ ₁, θ ₂θ _nlongitudinal stroke Δ Z with hydraulic leg ₁, Δ Z ₂Δ Z _ncarry out estimating and forecasting;

D, the α that initial testing is arrived, β value and precision index α ₀, β ₀after asking difference, carry out m decile, between point m regulatory region, plane is regulated, m is more than or equal to 2 natural number;

E, try to achieve for the first time the objective plane value α regulating ₁, β ₁, and calculate between first regulatory region i.e. α, β plane and α ₁, β ₁the longitudinal stroke Δ Z of an interplanar n hydraulic leg ₁₁, Δ Z ₁₂Δ Z _1n;

F, by the Δ Z in step e ₁₁, Δ Z ₁₂Δ Z _1nvalue is as setting value, in input control device, by the algorithm output controlled quentity controlled variable in step c, the valve position u of passing ratio valve ₁₁, u ₁₂u _1nregulation output, n hydraulic leg moves simultaneously and participates in platform adjusting, completes one-period and regulates;

G, the Plane Angle value α of actual measurement after release will be regulated ₁' β ₁' record, with the objective plane value α setting in step e ₁, β ₁compare and obtain estimating residual error, utilize the algorithm in step c to process this residual error, obtain n hydraulic cylinder broad sense drag θ ₁, θ ₂θ _none group be worth according to a preliminary estimate;

H, Repeated m-1 time step e, f, g all bring the broad sense drag θ that the last time obtains at every turn while repetition ₁, θ ₂θ _nvalue according to a preliminary estimate, what finally obtain is n hydraulic cylinder broad sense drag θ ₁, θ ₂θ _nestimated value, obtain broad sense drag θ ₁, θ ₂θ _nestimated value after, the algorithm in hydraulic pressure dynamic leveling system control model and step c in step a has just been determined;

I, hydraulic platform are at the middle generation pitch angle of advancing, and platform angle value α ', β ' that double-shaft tilt angle sensor is detected are changed and imported into micro-control computing machine by A/D, micro-control computing machine by detect angle value judge peak; Length travel between frontal plane and peak place surface level is calculated;

J, using the value in step I as setting value, in input control device, by the adaptive algorithm program in the system model in step a and step c, calculate output controlled quentity controlled variable, i.e. the valve position u of proportioning valve ₁, u ₂u _n, control in real time multiple hydraulic leg actions, adjust platform levelness.

After completing steps j, then repeat step I and j, complete the dynamic leveling of hydraulic platform.

The present invention compared with prior art has following beneficial effect.

The present invention can be applicable to only have double-shaft tilt angle sensor, does not need displacement, oil pressure, multi-point support hydraulic platform leveling in the advancing of the sensors such as flow, solve the problems such as strong coupling in hydraulic pressure dynamic leveling, varying duty, sudden change, realize real-time multiple spot and regulate, fast response.Changed in the past in leveling system, need multisensor measurement, the unidirectional adjusting of single-point, calculated off-line parameter, the problem such as response is poor while undergoing mutation.

Hydraulic pressure dynamic leveling method based on broad sense drag of the present invention can be applicable in hydraulic synchronous control system, realizes accurate control to the many topworkies of hydraulic system, is the development of hydraulic synchronization control theory and deeply, has more general application prospect.

Brief description of the drawings

Below in conjunction with accompanying drawing, the present invention will be further described in detail.

Fig. 1 is control structure figure of the present invention.

Embodiment

The present embodiment is described the dynamic leveling implementation procedure of a mobile radar test carriage its hydraulic platform in advancing.This mobile radar test carriage adopts 4 supports, and platform weighs 50 tons.

Leveling process is as follows:

1, the hydraulic system of this mobile radar test carriage in advancing carried out to force analysis, analyze the impact of broad sense drag on hydraulic cylinder travel, set up state equation and the input-output equation of the hydraulic leveling system of the broad sense drag based on 4 hydraulic legs, and by its discretize, obtain corresponding Input matrix in micro-control computing machine, obtain hydraulic pressure dynamic leveling system control model.

Obtain state-space expression, get quantity of state x ₁, x ₂, x ₃, be respectively:

$\left\{\begin{array}{c}{x}_1\left(t\right)=x_p\left(t\right)\\ x_2\left(t\right)=\stackrel{\·}{x_p}\\ x_3\left(t\right)=p_L\left(t\right)\end{array}\right.,$ X _pfor piston displacement, for piston speed, p _lfor equivalent piston pressure,

By 3 fundamental equations: flow lienarized equation, flow continuity equation, equilibrium equation form system of equations:

Wherein: x _v-main signal;

K _q-flow gain; K _c-flow pressure coefficient;

P _l-load pressure; A _p-piston effective area;

X _p-piston displacement; the flow that-piston movement is required;

V _t-hydraulic cylinder equivalence total measurement (volume); C _t-always reveal coefficient;

M _t-piston and being converted to the gross mass on piston by load;

B _pthe viscous friction coefficient of the movement parts such as-piston and load;

Environmental stiffness and oil when k-load movement;

F _l-act on other load force on piston;

Make valve position x _v(t)=u (t) is controlled quentity controlled variable, r (t)=F _l(t) load gravity is disturbance quantity,

Can obtain state-space expression:

Make x=[x ₁, x ₂, x ₃] ^t, have: x _v(t)=u (t), r (t)=F _l(t)

Known:

$C={[0,-\frac{F_L}{M_t},0]}^T={\left[\begin{array}{ccc}0& c& 0\end{array}\right]}^T$

(1) is carried out to discretize, can draw discrete state spatial expression:

$\left\{\begin{array}{cc}\frac{x_1(\mathrm{\κ}+1)-x_1\left(\mathrm{\κ}\right)}{T}=x_2\left(\mathrm{\κ}\right)& (2-1)\\ \frac{x_2(\mathrm{\κ}+1)-x_2\left(\mathrm{\κ}\right)}{T}=a_{22}{x}_2\left(\mathrm{\κ}\right)+a_{23}{x}_3\left(\mathrm{\κ}\right)+\mathrm{cr}\left(\mathrm{\κ}\right)& (2-2)\\ \frac{x_3(\mathrm{\κ}+1)-x_3\left(\mathrm{\κ}\right)}{T}=a_{32}{x}_2\left(\mathrm{\κ}\right)+a_{33}{x}_3\left(\mathrm{\κ}\right)+\mathrm{bu}\left(\mathrm{\κ}\right)& (2-3)\end{array}\right.$

By (2-1), (2-2), (2-3) can solve x ₁(κ) about u (κ), the expression formula of r (κ), i.e. discrete differential equation, establishing the sampling period is T, z is difference operator,

Obtain (z-1) x by (2-1) ₁(κ)=Tx ₂(κ) (3-1)

Obtain (z-1-Ta by (2-2) ₂₂) x ₂(κ)=Ta ₂₃x ₃(κ)+Tcr (κ) (3-2)

Obtain (z-1-Ta by (2-3) ₃₃) x ₃(κ)=Ta ₃₂x ₂(κ) Tbu (κ) (3-3)

Simultaneous (3-1), (3-2), (3-3) formula can solve:

(z-1)[(z-1-Ta ₂₂)(z-1-Ta ₃₃)-T ²a ₂₃a ₃₂]x ₁(κ)＝Ta ₂₃bu(κ)+T ²C(z-1-Ta ₃₃)r(κ) (4)

Output y (κ)=x ₁(κ)

Separating (4) arranges and can obtain:

(z ³+a ₁z ²+a ₂z+a ₃)y(κ)＝bu(κ)+(C ₀+C ₁z)r(κ) (5)

Wherein a ₁=-Ta ₂₂-Ta ₃₃-3

a ₂＝Ta ₂₂+Ta ₃₃+(Ta ₂₂+1)(Ta ₃₃+1)-T ²a ₂₃a ₃₂+2 (6)

a ₃＝T ²a ₂₃a ₃₂-(Ta ₂₂+1)(Ta ₃₃+1)

Utilize difference operator to write out (5) formula:

y(κ+3)+a ₁y(κ+2)+a ₂y(κ+1)a ₃y(κ)＝b ₀u(κ)+c ₀r(κ)+c ₁r(κ+1)

Above formula can be rewritten as following form:

y(κ+1)＝-a ₁y(κ)-a ₂y(κ-1)-a ₃y(κ-2)+b ₀u(κ-2)+c ₀r(κ-2)+c ₁r(κ-1)

Make θ=[a ₁,-a ₂,-a ₃, b ₀, c _lc ₀] ^t,

(7)

2, write the adaptive control algorithm based on strong tracking filter, this algorithm is arranged, programmes, packed, in input micro-control computing machine, call as subroutine, be used for, according to inputoutput data, the longitudinal stroke Δ Z of Parameter Generalized drag θ and hydraulic leg is carried out to estimating and forecasting.

Consider the Discrete time Nonlinear Systems of the following form of a class:

$\left\{\begin{array}{c}x(k+1)=f_d(k,u\left(k\right),x\left(k\right))+\mathrm{\Γ}\left(k\right)v\left(k\right)\\ y(k+1)=h_d(k+1,x(k+1))+e(k+1)\end{array}\right.---\left(8\right)$

Wherein state x ∈ R ⁿ, input u ∈ R ^p, output y ∈ R ^m, nonlinear function f _d: R ^p× R ⁿ→ R ⁿand h _d: R ⁿ→ R ^mx is had to continuous partial derivative; Process noise v (k) ∈ R ^qbe zero-mean, variance is the white Gaussian noise of Q (k), measures noise e (k) ∈ R ^malso be zero-mean, variance is the white Gaussian noise of R (k), and Γ (k) is the matrix of known suitable dimension, and v (k) and e (k) add up independently.

Strong tracking filter adaptive control algorithm design is as follows:

$\begin{array}{c}\hat{x}(k+1|k+1)=\hat{x}(k+1|k)+K(k+1)\mathrm{\γ}(k+1)\\ \hat{x}(k+1|k)=f_d(k,u\left(k\right),\hat{x}\left(k\right|k))\end{array}$

$\begin{array}{c}P(k+1|k)=\mathrm{LMD}(k+1)F(k,u\left(k\right),\hat{x}\left(k\right|k))P\left(k\right|k)\\ F^T(k,u\left(k\right),\hat{x}\left(k\right|k))+\mathrm{\Γ}\left(k\right)Q\left(k\right){\mathrm{\Γ}}^T\left(k\right)\end{array}$

$P(k+1|k+1)=[I-K(k+1)H(k+1,\hat{x}(k+1|k))]P(k+1|k)$

$\mathrm{\γ}(k+1)=y(k+1)-h_d(k+1,\hat{x}(k+1|k))$

$F(k,u\left(k\right),\hat{x}\left(k\right|k))=\frac{\∂f_d(k,u\left(k\right),x\left(k\right))}{\∂x}{|}_{x=\hat{x}\left(k\right|k)}$

$H(k+1,\hat{x}(k+1|k))=\frac{\∂h_d(k+1,x(k+1))}{\∂x}{|}_{x=\hat{x}(k+1|k)}$

3, automatic horizontal control system is launched, hydraulic leg lands, and enters duty.Radar test car produces pitch angle when static, and double-shaft tilt angle sensor detects angle value α, β, changes by A/D, is passed to micro-control computing machine.

4 micro-control computing machines by detect angle value judge peak; According to leveling time and sampling period, will between this frontal plane and peak place surface level, divide 10 decile planes into; Ask the objective plane value regulating for the first time:

α ₁=0.9 (α-1), β _i=0.9 (β-1), by (α, β) plane and (α ₁, β ₁) plane can obtain the setting regulated quantity Δ Z of each hydraulic leg ₁Δ Z ₄; Due to α, β is less, can be similar to think cos α=1, and cos β=1, can try to achieve:

ΔZ ₁＝0ΔZ ₂＝asinα-asinα ₁ΔZ ₃＝asinα+bsinβ-asinα ₁-bsinβ ₁ΔZ ₄＝bsinβ-bsinβ ₁。

5, using the value in step 4 as setting value, in input control device, by the algorithm output controlled quentity controlled variable of step 2, the valve position u of passing ratio valve ₁₁, u ₁₂u _1nregulation output, controls 4 hydraulic leg actions, adjusts hydraulic platform levelness.Double-shaft tilt angle sensor detects the angle [alpha] of next sampling instant ₁' β ₁', change by A/D, be passed to micro-control computing machine, in controller with set plane (α ₁, β ₁) compare, produce estimation residual error, utilize the algorithm of step 2 to process this residual error, a group of broad sense drag that obtains 4 hydraulic legs is worth according to a preliminary estimate.Detailed process is shown in accompanying drawing 1.

6, repeat 9 steps 4,5, repeat all to bring into the value according to a preliminary estimate of the broad sense drag that the last time obtains, the estimated value that is 4 hydraulic leg broad sense drag finally obtaining at every turn.

7, when radar test garage enters, produce pitch angle, double-shaft tilt angle sensor detects angle value α ', β ', changes by A/D, is passed to micro-control computing machine; Micro-control computing machine by detect angle value judge peak, will calculate when the length travel between frontal plane and peak place surface level.

8, using the value in step 7 as setting value, in input control device, by the algorithm output controlled quentity controlled variable in system model and step 2 in step 1, control 4 hydraulic leg actions, adjust platform levelness.

9, repeating step 7,8, completes the dynamic leveling of hydraulic platform.

Claims (3)

1. the hydraulic pressure dynamic leveling method based on broad sense drag, it is characterized in that: adopt that broad sense drag replaces that load variations in dynamic leveling process, oil temperature change, become when rigidity in each hydraulic cylinder plane involves coupling etc. and be difficult to the physical quantity of accurately surveying, and analyze the impact with output on the input of dynamic leveling system control model of these physical quantitys and broad sense drag, set up state equation and the input-output equation of the hydraulic pressure dynamic leveling system based on broad sense drag; The adaptive control algorithm that adopts strong tracking filter to time the nonlinear control system that becomes dynamic parameter, broad sense drag is estimated, obtains the precise control model of hydraulic pressure dynamic leveling system, according to the input quantity of system, eliminate and estimate residual error, control output quantity, complete leveling.

2. the hydraulic pressure dynamic leveling method based on broad sense drag according to claim 1, is characterized in that comprising the following steps:

A, introducing broad sense drag, analyze the impact of broad sense drag on hydraulic cylinder travel, sets up the broad sense drag θ that includes n hydraulic cylinder ₁, θ ₂θ _nstate equation and the input-output equation of hydraulic pressure dynamic leveling system, determine the sampling period T of system and regulate time t, according to the sampling period T of system, to its discretize, obtaining after discretize is proportioning valve valve position u with n hydraulic cylinder flow ₁, u ₂u _nfor input quantity, n hydraulic cylinder displacement Z ₁, Δ Z ₂Δ Z _nfor the system equation of output quantity, i.e. hydraulic pressure dynamic leveling system control model;

B, hydraulic platform are supported by n hydraulic leg, produce pitch angle, platform angle value α, the β that double-shaft tilt angle sensor is detected and the precision index α of platform when static ₀, β ₀change and import micro-control computing machine into by A/D;

C, write the adaptive control algorithm based on strong tracking filter, by this algorithm packing, programming, in implant controller, calls as a generalized subroutine, is used for according to inputoutput data to Parameter Generalized drag θ ₁, θ ₂θ _nlongitudinal stroke Δ Z with hydraulic leg ₁, Δ Z ₂Δ Z _ncarry out estimating and forecasting;

D, the α that initial testing is arrived, β value and precision index α ₀, β ₀after asking difference, carry out m decile, between point m regulatory region, plane is regulated, m is more than or equal to 2 natural number;

E, try to achieve for the first time the objective plane value α regulating ₁, β ₁, and calculate between first regulatory region i.e. α, β plane and α ₁, β ₁the longitudinal stroke Δ Z of an interplanar n hydraulic leg ₁₁, Δ Z ₁₂Δ Z _1n;

F, by the Δ Z in step e ₁₁, Δ Z ₁₂Δ Z _1nvalue is as setting value, in input control device, by the algorithm output controlled quentity controlled variable in step c, the valve position u of passing ratio valve ₁₁, u ₁₂u _1nregulation output, n hydraulic leg moves simultaneously and participates in platform adjusting, completes one-period and regulates;

G, the Plane Angle value α of actual measurement after release will be regulated ₁' β ₁' record, with the objective plane value α setting in step e ₁, β ₁compare and obtain estimating residual error, utilize the algorithm in step c to process this residual error, obtain n hydraulic cylinder broad sense drag θ ₁, θ ₂θ _none group be worth according to a preliminary estimate;

H, Repeated m-1 time step e, f, g all bring the broad sense drag θ that the last time obtains at every turn while repetition ₁, θ ₂θ _nvalue according to a preliminary estimate, what finally obtain is n hydraulic cylinder broad sense drag θ ₁, θ ₂θ _nestimated value, obtain broad sense drag θ ₁, θ ₂θ _nestimated value after, the algorithm in hydraulic pressure dynamic leveling system control model and step c in step a has just been determined;

I, hydraulic platform are at the middle generation pitch angle of advancing, and platform angle value α ', β ' that double-shaft tilt angle sensor is detected are changed and imported into micro-control computing machine by A/D, micro-control computing machine by detect angle value judge peak; Length travel between frontal plane and peak place surface level is calculated;

J, using the value in step I as setting value, in input control device, by the adaptive algorithm program in the system model in step a and step c, calculate output controlled quentity controlled variable, i.e. the valve position u of proportioning valve ₁, u ₂u _n, control in real time multiple hydraulic leg actions, adjust platform levelness.

3. the hydraulic pressure dynamic leveling method based on broad sense drag according to claim 2, is characterized in that: after completing steps j, then repeat step I and j, complete the dynamic leveling of hydraulic platform.