CN105205221A - Precision reliability analysis method for heavy numerical control machine tool - Google Patents

Abstract

The invention belongs to the field of reliabilities of numerical control machine tools, and particularly relates to a precision reliability analysis method for a heavy numerical control machine tool based on a multi-body system kinemics theory. The precision reliability analysis method comprises the following steps: constructing a relation model between the output precision error of the heavy numerical control machine tool and a factor affecting the precision of the machine tool, fitting a change relation of the factor affecting the precision of the machine tool along with the time by a random process, finally combining on-site data or data of the same type of sample, and introducing a Bayes evaluation method to update the precision error model of the heavy numerical control machine tool, thus realizing evaluation of the precision reliability of the machine tool. The precision reliability analysis method has the beneficial effects that the Bayes evaluation method is introduced to fit and correct the change tendency of respective factors causing a motion precision error of the machine tool, so that the difficulty that a sufficient statistical sample cannot be formed due to a 'small sample' is solved; and therefore, more scientific and reasonable evaluation is performed on the dynamic reliability of the precision error of the heavy numerical control machine tool.

Description

A kind of heavy digital control machine tool precision reliability analytical approach

Technical field

The invention belongs to Cnc ReliabilityintelligeNetwork Network field, be specifically related to a kind of heavy digital control machine tool precision reliability analytical approach based on Multibody Kinematics theory.

Background technology

Along with the development of modern industry, numerically-controlled machine experienced by and repeatedly updates, and its performance and technical merit are all developing rapidly.Particularly in the last few years, along with the development of the related disciplines such as infotech, electronic technology, mechanical technique, many performance index such as the speed of numerically-controlled machine, precision and the control number of axle all increase, but lathe integrity problem becomes one of subject matter of puzzlement numerical control equipment development.

The development level of heavy digital control machine tool often can embody a national Industry Development Level.Certain model heavy digital control machine tool of China's independent research is digital-control type side's ram movable landing milling-boring machine, and main shaft horizontal type is arranged, numerical control of machine tools coordinate is X, Y, Z, W, V, B, U axle, can realize four-axle linked arbitrarily.X-axis is that column moves, and Y-axis is that main spindle box moves, and Z axis is that boring axle moves, and W axle is that ram moves, and V, B are respectively rectilinear movement and the gyration of rotary table, and U axle is that horizontal rotary tool rest moves, and its geometric model figure as shown in Figure 2.Heavy digital control machine tool is the important manufacturing equipment of machinery manufacturing industry, there is batch few, long service life, the feature that maintenance cost is high, and the mechanism kinematic reliability of heavy digital control machine tool is mainly conducted oneself with dignity by key structure part, the scale error that parts manufacture, mechanism's power source drives error, the impact of deformation and wearing and tearing, once go wrong and will cause very serious loss in processing, in actual production, it is exactly the decline of machining precision that heavy digital control machine tool hydraulic performance decline the most directly shows, therefore, scientifically predict the precision reliability of heavy digital control machine tool, ensure that its machining precision tool is in-service of great significance.

The precision reliable of heavy digital control machine tool is analyzed and the precision reliability analysis of machine tool is distinguished very large, because heavy digital control machine tool has from great, the load that kinematic pair bears is large and process the features such as stroke length, heavy digital control machine tool precision in the course of the work is more easily subject to because the stand under load of machine tool component is out of shape, temperature distortion, the impact of the factors such as the manufacture rigging error of parts and wearing and tearing, the speed ratio machine tool finally causing the machining precision of heavy digital control machine tool to reduce is faster, therefore for the fail-safe analysis of heavy digital control machine tool, need on the basis of the precision reliability analytical approach of the more ripe machine tool of present stage development, in conjunction with the own structural characteristics of heavy digital control machine tool and the singularity of working environment, finally set up one more accurate, the trueness error model of rational heavy digital control machine tool.Closely during the last ten years, lot of domestic and international scholar does a lot of work in this field, achieves many creative achievements.Make a general survey of research both domestic and external, they have common feature, that is: usually using the small quantity of various error component as various correlated variables, utilize mechanism's operating analysis method, by suitable parametric variable process, derive lathe and export location attitude of the cutter error model.Such as, Multibody Kinematics theory is widely used when the machining precision error of analytical calculation lathe.

The requirement of mechanism kinematic precision is a kind of performance index in essence, and when it can not meet the demands, mechanism is in a kind of failure state equally.As spindle drive and the tool feeding mechanism of lathe, due to the factor such as undue wear and gap, its static and dynamic errors exceeds permissible value, causes lathe cannot process the workpiece of the accuracy class specified in technical indicator.Now, though lathe can work, lose predetermined function, be in malfunction equally, need repair or scrap.Heavy digital control machine tool precision reliability declines, the product loss caused that falls short of specifications of processing is caused to be very huge, therefore, set up more reasonably heavy digital control machine tool trueness error model, scientifically predict that heavy digital control machine tool precision reliability operationally has very important engineering significance.

Summary of the invention

The object of the invention is to be different from the design feature of general NC lathe and special working environment according to heavy digital control machine tool, improve the weak point of existing analysis method for reliability, under the abrasion condition considering moving component, a kind of heavy digital control machine tool precision reliability analytical approach based on Multibody Kinematics theory is proposed, the innovation of the method is to use stochastic process to carry out matching to wear degradation process, establish random wear degradation model, then adopt bayes method, incorporate the wear degradation data of up-to-date acquisition each time, complete the real-time update of wear degradation Model Parameter, all data can be made to be fully utilized, also the model science more of foundation can be made, rationally.

To achieve these goals, technical scheme of the present invention is: propose a kind of heavy digital control machine tool precision reliability analytical approach based on Multibody Kinematics theory, main flow figure as shown in Figure 1, comprises the steps:

Step 1: according to the architectural feature of heavy digital control machine tool, when considering kinematic pair wearing and tearing on core component, adopting Multibody Kinematics theoretical, setting up the trueness error model of heavy digital control machine tool;

Step 2: carry out data acquisition stage by stage to the wear degradation data of the kinematic pair on core component, obtains the wear degradation data of operational phase different time lathe all parts;

Step 3: adopt the wear degradation process of rational stochastic process to each parts of lathe to carry out modeling, as Wiener process and Gamma process, use the wear degradation data that commitment gathers, or the wear degradation data of like product, Maximum-likelihood estimation is carried out to the parameter in wear degradation model, obtains the estimated value of wear degradation Model Parameter;

Step 4: the wear degradation data utilizing each parts of the heavy digital control machine tool of up-to-date collection, adopt bayes method, parameter value in the wear degradation model obtained in step 3 is upgraded, the maximum likelihood estimation of the wear degradation model parameter obtained in step 3 can as the average of prior distribution, and the determination of prior distribution type can be determined according to expertise or field data, by the renewal to wear degradation model parameter, achieve the real-time update to each component wear degenerative process of lathe;

Step 5: the trueness error model wear degradation process of each for the lathe of step 4 gained parts being substituted into the heavy digital control machine tool of step 1 gained carries out analytical calculation, obtaining can the machining precision error model of heavy digital control machine tool of real-time update;

Step 6: in conjunction with Practical Project requirement, determine heavy digital control machine tool machining precision failure threshold, and according to the heavy digital control machine tool trueness error model that can carry out real-time update that step 5 obtains, computational analysis obtains the fiduciary level of heavy digital control machine tool trueness error.

In order to calculate the reliability of heavy digital control machine tool trueness error, needing the failure threshold determining machining precision error, according to different engineering demands, different failure threshold may be had, as failure threshold for definite value or may obey a distribution determined.Another innovation of the present invention is: trueness error often Normal Distribution actual in engineering, actual for heavy digital control machine tool machining precision error distribution distributed with the failure threshold of machining precision and be converted to generalized stress-Strength Interference Model, its fiduciary level can be expressed as:

$R\left(t\right)=\mathrm{\Φ}\left(\mathrm{\β}\right(t\left)\right)=\mathrm{\Φ}\left(\frac{{\mathrm{\μ}}_{{\mathrm{\σ}}_F}\left(t\right)-{\mathrm{\μ}}_{{\mathrm{\σ}}_{Fp}}}{\sqrt{s_{{\mathrm{\σ}}_F}^2\left(t\right)+s_{{\mathrm{\σ}}_{Fp}}^2}}\right)---\left(1\right)$

In formula, R (t) is fiduciary level, represents the fiduciary level of machine tool accuracy in example, and time correlation, and along with passage of time, machine finish declines, and the fiduciary level of machine tool accuracy also declines thereupon, and β (t) is reliability index, for generalized stress average, for generalized stress standard deviation, represent the actual machining precision error mean of lathe and standard deviation, equal and time correlation, for GENERALIZED STRENGTH average, for GENERALIZED STRENGTH standard deviation, represent average and the standard deviation of the failure threshold distribution of machining precision error.

Beneficial effect of the present invention is: because heavy digital control machine tool is the important foundation equipment of machinery manufacturing industry, have batch less, feature that long service life, maintenance cost are high, once go wrong and will cause very serious loss in processing.In actual production, it is exactly the decline of machining precision that heavy digital control machine tool hydraulic performance decline the most directly shows, and therefore, scientifically predicts the precision reliability of heavy digital control machine tool, ensures that its machining precision tool is in-service of great significance.But, due to the negligible amounts of heavy digital control machine tool, enough statistical samples cannot be formed, very difficult statistical method studies its reliability, at this, we introduce Bayesian Assessment Method to each cause the factor of machine tool motion trueness error over time trend carry out matching and correction, overcome " sample is little ", namely cannot form a difficult problem for enough statistical samples, more scientific and rational assessment has been made to the dynamic reliability of heavy Precision of NC Machine Tool error.

Accompanying drawing explanation

Fig. 1 main flow chart of the present invention.

Fig. 2 one embodiment of the invention institute is for heavy digital control machine tool structural representation.

Certain model heavy digital control machine tool generalized Topological structural representation of Fig. 3 step 1 of the present invention.

Certain model of Fig. 4 step 1 of the present invention heavy main spindle box of numerical control way rub schematic diagram.

Certain model of Fig. 5 step 1 of the present invention heavy numerical control machine tool ram wearing and tearing schematic diagram.

Certain model heavy digital control machine tool boring axle bearing of Fig. 6 step 1 of the present invention wearing and tearing schematic diagram.

The precision fiduciary level of certain model heavy digital control machine tool of Fig. 7 step 6 of the present invention is with the change of working time.

Embodiment

Now in conjunction with the embodiments, the invention will be further described for accompanying drawing: a kind of heavy digital control machine tool precision reliability analytical approach based on Multibody Kinematics theory, comprises the steps:

Step 1: according to the architectural feature of heavy digital control machine tool, when considering kinematic pair wearing and tearing on core component, adopting Multibody Kinematics theoretical, setting up the trueness error model of heavy digital control machine tool.

In the present embodiment, certain model heavy digital control machine tool physical model as shown in Figure 2, heavy digital control machine tool, primarily of ground 0, bed ways 1, slide 2, column 3, main spindle box 4, square ram 5, milling spindle 6, boring axle 7, cutter 8, table rest 9, universal stage 10, worktable 11 and workpiece 12, amounts to 12 basic building block compositions.Correlation parameter is as shown in table 1.

The core component correlation parameter of certain model heavy digital control machine tool of table 1

Project	Parameter
		Boring spindle diameter	160mm
Milling spindle end diameter	280mm
		Ram cross section (wide × high)	440×480mm
Spindle taper hole	ISO 50(7:24)
		Column stroke (X)	X＝6000mm
Main spindle box stroke (Y)	Y＝3000mm
		Boring spindle travel (Z)	Z＝1000mm
Ram stroke (W)	W＝900mm

Theoretical according to Multibody Kinematics, the generalized Topological structure between 12 components as shown in Figure 3.The wearing and tearing of this example main consideration main spindle box guiding rail, ram are worn and torn and the wearing and tearing of boring axle bearing cause heavy digital control machine tool machining precision error, and error mould is:

$\begin{array}{c}E=\left(\begin{array}{cccc}1& 0& 0& x\\ 0& 1& 0& 500\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 3000\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& -{\mathrm{\ΔY}}_{zc}& 0\\ 0& {\mathrm{\ΔY}}_{zc}& 1& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 300\\ 0& 0& 1& 150\\ 0& 0& 0& 1\end{array}\right)\\ \·\left(\begin{array}{cccc}1& -{\mathrm{\Δ\γ}}_z& 0& {\mathrm{\Δx}}_z\\ {\mathrm{\Δ\γ}}_z& 1& -{\mathrm{\Δ\α}}_z& {\mathrm{\Δy}}_z\\ 0& {\mathrm{\Δ\α}}_z& 1& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 600\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& -{\mathrm{\ΔY}}_{yc}& 0& 0\\ {\mathrm{\ΔY}}_{yc}& 1& -{\mathrm{\Δ\α}}_{xc}& 0\\ 0& {\mathrm{\Δ\α}}_{xc}& 1& 0\\ 0& 0& 0& 1\end{array}\right)\\ \·\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& 260\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{c}0\\ -650\\ 245\\ 1\end{array}\right)-\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& 1500\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 2500\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{c}0\\ 650\\ -245\\ 1\end{array}\right)\end{array}---\left(2\right)$

In formula, $\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& -{\mathrm{\ΔY}}_{zc}& 0\\ 0& {\mathrm{\ΔY}}_{zc}& 1& 0\\ 0& 0& 0& 1\end{array}\right),\left(\begin{array}{cccc}1& -{\mathrm{\Δ\γ}}_z& 0& {\mathrm{\Δx}}_z\\ {\mathrm{\Δ\γ}}_z& 1& -{\mathrm{\Δ\α}}_z& {\mathrm{\Δy}}_z\\ 0& {\mathrm{\Δ\α}}_z& 1& 0\\ 0& 0& 0& 1\end{array}\right),\left(\begin{array}{cccc}1& -{\mathrm{\ΔY}}_{yc}& 0& 0\\ {\mathrm{\ΔY}}_{yc}& 1& -{\mathrm{\Δ\α}}_{xc}& 0\\ 0& {\mathrm{\Δ\α}}_{xc}& 1& 0\\ 0& 0& 0& 1\end{array}\right)$ Represent that main spindle box the wear and tear trueness error transition matrix that causes and the boring axle bearing of the trueness error transition matrix, the ram that cause that wear and tear weares and teares the trueness error transition matrix caused respectively, and have:

${\mathrm{tan\ΔY}}_{zc}=\frac{u_{4z}-{u^{\′}}_{4z}}{1000}---\left(3\right)$

${\mathrm{tan\Δ\α}}_z=\frac{u_{3x}}{1000}---\left(4\right)$

${\mathrm{tan\Δ\γ}}_z=\frac{u_{3y}}{1000}---\left(5\right)$

${\mathrm{\Δx}}_z=(1000+600)\frac{u_{3x}}{1000}---\left(6\right)$

${\mathrm{\Δy}}_z=(1000+600)\frac{u_{3y}}{1000}---\left(7\right)$

${\mathrm{tan\Δ\α}}_{xc}={\mathrm{tan\ΔY}}_{yc}=\frac{u_{1y}+u_{2y}}{1000}---\left(8\right)$

Wherein, u _4z, u' _4zbe respectively main spindle box guiding rail wear-thickness, as shown in Figure 4;

U _3x, u _3yfor ram wear-thickness, as shown in Figure 5;

U _1y, u _2y, u _1x, u _2xfor boring axle bearing wear-thickness, as shown in Figure 6;

Δ Y _zcrepresent the angle that main spindle box rotates around the X-axis of the coordinate system be cemented on main spindle box;

Δ γ _z, Δ α _zrepresent the Z axis of ram around the coordinate system be cemented on ram and the angle of X-axis rotation respectively;

Δ x _z, Δ y _zrepresent the distance that the initial point of the coordinate system be cemented on ram offsets along X-axis and Y direction respectively;

Δ α _xc, Δ Y _ycrepresent the X-axis of boring axle around the coordinate system be cemented on boring axle and the angle of Z axis rotation respectively.

Step 2: carry out data acquisition stage by stage to the wear degradation data of the kinematic pair on heavy digital control machine tool core component, obtains the wear degradation data of not lathe all parts in the same time.

In this step, the acquisition of the wear degradation data of the kinematic pair on lathe core component is the wear degradation information monitored in product testing process or actual moving process.Specify that in the model of step 1 and need the wear degradation data obtained to have main spindle box guiding rail wear-thickness, ram wear-thickness and boring axle bearing wear-thickness in this example, and in order to simplify length, we only provide boring axle bearing wear-thickness data u _1xas the analysis of step 3, one is divided into the wear data that double teacher acquires five lathes respectively, wherein the first four months is early stage image data, for the undetermined parameter of degradation estimation stochastic process in step 3, the data that May gathers are latest data, for upgrading the parameter value of degeneration stochastic process in step 4, revise degradation model, concrete wear data is as shown in table 2, and unit is um.

Table 2 boring axle bearing wear-thickness data

Step 3: adopt the wear degradation process of rational stochastic process to each parts of lathe to carry out modeling, as Wiener process and Gamma process, use the wear degradation data that commitment gathers, or the wear degradation data of like product, Maximum-likelihood estimation is carried out to the parameter in wear degradation model, obtains the estimated value of wear degradation Model Parameter.

It is random and the stochastic process of dullness that Gamma process is very suitable for describing degenerative process, because its increment has independence and nonnegativity, Gamma process has α, and β two parameters, have:

u(t)～Gamma(αt,β)(9)

In formula, u (t) is amount of degradation.

This example adopts Gamma process to carry out the wear degradation process of matching machine tool component, provides the wear degradation data of the first four months, estimate α with Maximum Likelihood Estimation Method according to step 3, the value of β, can obtain α, the point estimate of β is respectively 6.152534 and 0.3291, therefore has:

u _1x～Gamma(6.152534t,0.3291)(10)

Step 4: the wear degradation data utilizing each parts of the heavy digital control machine tool of up-to-date collection, adopts bayes method, upgrades, realize the real-time update of each component wear degenerative process of lathe to the parameter value of the wear degradation model obtained in step 3;

In order to upgrade the degradation model parameter in step 3 in this step, have employed bayes method, its main thought thinks that the parameter in probability distribution is unknown stochastic variable, by certain probability distribution, the prior distribution of parameter can be determined before the test according to certain prior imformation, with α in step 3 in this example, the maximum likelihood estimation of β is for referencial use, the prior distribution of α is interval for [3.15, being uniformly distributed 9.15], the prior distribution of β is Gamma distribution, form parameter and scale parameter are respectively 3.291 and 10, the data that recycling test sample is degenerated or lost efficacy determine the Posterior distrbutionp of parameter, this method is relevant to sample, it takes full advantage of prior imformation, heavy digital control machine tool sample can be solved preferably little, the problem of data deficiencies, bayes method is the method for carrying out statistical inference according to Bayes' theorem, core formula is as shown in Equation 11:

$\mathrm{\π}\left(\mathrm{\θ}\right|x)=\frac{P\left(x\right|\mathrm{\θ})\mathrm{\π}\left(\mathrm{\θ}\right)}{\underset{\mathrm{\Θ}}{\∫}P\left(x\right|\mathrm{\θ})\mathrm{\π}\left(\mathrm{\θ}\right)d\mathrm{\θ}}---\left(11\right)$

Wherein, π (θ) is known parameter θ prior distribution, can determine according to field data or expertise; π (θ | x) be parameter θ Posterior distrbutionp.

Carry out needing in renewal process to face complicated integral operation to wear degradation model parameter using bayes method, we adopt Markov chain Monte Carlo (MCMC) method, its basic thought is the sample that the Markov chain of π (θ) obtains π (θ) by setting up a stationary distribution, do various statistical inference based on these samples again, detailed step repeats no more.By WinBUGS Software tool, upgrade the wear degradation model parameter of step 3 gained, result is as shown in table 3.

Contrast before and after table 3 wear degradation model parameter upgrades

Step 5: the trueness error model wear degradation process of each for the lathe of step 4 gained parts being substituted into the heavy digital control machine tool of step 1 gained carries out analytical calculation, obtaining can the machining precision error model of heavy digital control machine tool of real-time update;

One by one Gamma process model building is being used to the wear degradation data of each machine tool component, re-use after bayes method upgrades degradation model parameter, the wear degradation model set up is substituted into the trueness error model of the heavy digital control machine tool of step 1 gained, finally complete the foundation of whole model.

Step 6: in conjunction with Practical Project requirement, determine heavy digital control machine tool machining precision failure threshold, and according to the heavy digital control machine tool trueness error model that can carry out real-time update that step 5 obtains, computational analysis obtains the fiduciary level of heavy digital control machine tool trueness error.

In this example, failure threshold is defined as: the average permissible error of all directions is 0.04mm, and variance is 0.01mm ², and Normal Distribution.According to put forward the methods of the present invention, after simulation calculation is carried out to completed model, the precision fiduciary level of heavy digital control machine tool with actual run time change as shown in Figure 7, achieve the analytical calculation of the DYNAMIC RELIABILITY of type Precision of NC Machine Tool error.

This example specifically describes the implementation procedure of put forward the methods of the present invention, first the generalized Topological structure between heavy digital control machine tool all parts and coupled relation is analyzed, determine the kinematic pair that lathe easily occurs to wear and tear in the course of the work, finally adopt Multibody Kinematics theory to carry out modeling to the machining precision error model of lathe, in model, contain main spindle box guiding rail wear-thickness u _4z, u' _4z, ram wear-thickness u _3x, u _3ywith boring axle bearing wear-thickness u _1y, u _2y, u _1x, u _2xeight time dependent stochastic variables, namely in fact they are mutual independently eight stochastic processes, as which kind of stochastic process of employing can determine the degree of fitting of concrete wear data according to stochastic process to be selected.With u in this example _1xfor example, we use Gamma process to u _1xdata carried out matching, because wear data collection is carried out stage by stage, in this example, use the wear data of four months previously obtained, adopt Maximum Likelihood Estimation Method to carry out parameter estimation to the Gamma process of obeying of degenerating, obtain α, the estimated value of β.If we obtain again new wear data present stage, be the data obtained for five month in this example, just bayes method can be adopted, new data for upgrading the parameter alpha in the Gamma process of wearing and tearing obedience, β, making when constantly obtaining new data, can constantly upgrade the parameter in the degenerative process of wearing and tearing obedience, make model more accurately, rationally.The analytic process of remaining seven wear variable is all identical.After wear degradation model is set up, the machining precision error model being substituted into heavy machine tool just can obtain machining precision error relation over time, after determining machining precision failure threshold and obeying the normal distribution that average and variance all determine, use simulation means just can obtain heavy digital control machine tool machining precision fiduciary level relation over time.

Those of ordinary skill in the art will appreciate that, embodiment described here is to help reader understanding's principle of the present invention, should be understood to that protection scope of the present invention is not limited to so special statement and embodiment.Those of ordinary skill in the art can make various other various concrete distortion and combination of not departing from essence of the present invention according to these technology enlightenment disclosed by the invention, and these distortion and combination are still in protection scope of the present invention.

Claims (3)

1. the invention belongs to Cnc ReliabilityintelligeNetwork Network field, be specifically related to a kind of heavy digital control machine tool precision reliability analytical approach based on Multibody Kinematics theory, comprise the steps:

Step 1: according to the architectural feature of heavy digital control machine tool, when considering kinematic pair wearing and tearing on core component, adopting Multibody Kinematics theoretical, setting up the trueness error model of heavy digital control machine tool;

Step 2: carry out data acquisition stage by stage to the wear degradation data of the kinematic pair on core component, obtains the wear degradation data of operational phase different time lathe all parts;

Step 3: adopt the wear degradation process of rational stochastic process to each parts of lathe to carry out modeling, as Wiener process and Gamma process, use the wear degradation data that commitment gathers, or the wear degradation data of like product, Maximum-likelihood estimation is carried out to the parameter in wear degradation model, obtains the estimated value of wear degradation Model Parameter;

Step 4: the wear degradation data utilizing each parts of the heavy digital control machine tool of up-to-date collection, adopt bayes method, parameter value in the wear degradation model obtained in step 3 is upgraded, the maximum likelihood estimation of the wear degradation model parameter obtained in step 3 can as the average of prior distribution, and the determination of prior distribution type can be determined according to expertise or field data, by the renewal to wear degradation model parameter, achieve the real-time update to each component wear degenerative process of lathe;

Step 5: the trueness error model wear degradation process of each for the lathe of step 4 gained parts being substituted into the heavy digital control machine tool of step 1 gained carries out analytical calculation, obtaining can the machining precision error model of heavy digital control machine tool of real-time update;

Step 6: in conjunction with Practical Project requirement, determine heavy digital control machine tool machining precision failure threshold, and according to the heavy digital control machine tool trueness error model that can carry out real-time update that step 5 obtains, computational analysis obtains the fiduciary level of heavy digital control machine tool trueness error.

2. a kind of heavy digital control machine tool precision reliability analytical approach based on Multibody Kinematics theory according to claim 1, it is characterized in that, the wear degradation process of the rational stochastic process of employing in described step 3 and 4 to each parts of lathe carries out the process of modeling, and concrete steps are:

Step 11: choose suitable stochastic process and carry out matching wear degradation process;

Step 12: gather the wear data that each parts of lathe are up-to-date;

Step 13: adopt bayes method, the wear data incorporating up-to-date collection upgrades the parameter in stochastic process;

Step 14: wear degradation model parameter is upgraded, realizes the real-time update of each component wear degenerative process of lathe.

3. a kind of heavy digital control machine tool precision reliability analytical approach based on Multibody Kinematics theory according to claim 1, it is characterized in that in described step 6 to calculate the reliability of heavy digital control machine tool trueness error, need the failure threshold determining machining precision error, according to different engineering demands, different failure threshold may be had, as failure threshold for definite value or may obey a distribution determined.Another innovation of the present invention is: trueness error often Normal Distribution actual in engineering, actual for heavy digital control machine tool machining precision error distribution distributed with the failure threshold of machining precision and be converted to generalized stress-Strength Interference Model, its fiduciary level can be expressed as:

$R\left(t\right)=\mathrm{\Φ}\left(\mathrm{\β}\right(t\left)\right)=\mathrm{\Φ}\left(\frac{{\mathrm{\μ}}_{{\mathrm{\σ}}_F}\left(t\right)-{\mathrm{\μ}}_{{\mathrm{\σ}}_{Fp}}}{\sqrt{s_{{\mathrm{\σ}}_F}^2\left(t\right)+s_{{\mathrm{\σ}}_{Fp}}^2}}\right)---\left(1\right)$

In formula, R (t) is fiduciary level, represents the fiduciary level of machine tool accuracy in example, and time correlation, and along with passage of time, machine finish declines, and the fiduciary level of machine tool accuracy also declines thereupon, and β (t) is reliability index, for generalized stress average, for generalized stress standard deviation, represent the actual machining precision error mean of lathe and standard deviation, equal and time correlation, for GENERALIZED STRENGTH average, for GENERALIZED STRENGTH standard deviation, represent average and the standard deviation of the failure threshold distribution of machining precision error.