CN105912839A - Method for optimizing reliability of structural noise based on dimension-by-dimension analysis strategy - Google Patents

Abstract

The invention discloses a method for optimizing reliability of structural noise based on a dimension-by-dimension analysis strategy. The method comprises following steps: firstly, establishing a structural noise optimizing model based on an interval reliability analysis model by using a quantitative structure for a model of interval value and environment and other uncertainties, determining orders and gauss integral points based on least-squares approximation responding to interval parameters by utilizing structural acoustics to response to nonlinearity degree related to interval parameters; secondly, utilizing gauss integral points and a quantitative model for interval numbers to sample parameter vectors of intervals, calculating response vector value at interval parameter sample points, establishing the best square approximation in order to determine the max-minimum point matrix such that response interval vectors can be calculated; and finally, utilizing response interval vectors and safety requirement to calculate reliability in intervals and finishing structural optimization under drive of the optimizing algorithm. The method for optimizing reliability of structural noise based on the dimension-by-dimension analysis strategy has following beneficial effects: by replacing classic safety factor with interval reliability, the method echoes the fine development trend of optimizing structural noise; conservativeness of a conventional optimizing method is effectively avoided; and the method has a broad application prospect.

Description

A kind of method based on the construct noise reliability optimization by dimension analysis strategy

Technical field

The present invention relates to equip the technical field of noise measurement, be specifically related to a kind of method of noise optimization in enclosed construction space, It is applicable to passenger plane, automobile and submarine etc. and there is the equipment inner space noise optimization of closed in spaces structure.

Background technology

Enclosed construction form (the passenger plane fuselage enclosed such as class cylinder thin-wall construction or submarine cabin, class cuboid thin-walled knot The inner space of vehicle that structure encloses) the most common in engineering field, elastic thin-walled structures vibrates under extraneous incentive action And to enclosed construction inner space radiated noise, high level noise not only has a strong impact on the physical and mental health of cabin interior occupant and comfortable Sexual experience, and the combination property of equipment or the function of key equipment assembly are had harm greatly, such as internal key equipment Acoustic fatigue destruction etc..Therefore, by structure design and optimization to reduce in its closed in spaces enclosed the sound at key position Properly functioning the having significantly of the acoustic pressure distribution key equipment internal to equipment overall performance and equipment arbitrarily downgraded or change in cabin Construction value.

During construct noise analysis, design and optimization, the factor such as such as " people, machine, material, method, ring, survey " is the most potential not With the uncertainty of degree, as the restriction of research worker Subjective level, the technique of machining device limit, structural material Quality and the difference of batch, theory analysis are assumed with simplifying of numerical analysis, the equipment fluctuation of Service Environment parameter, analysis of experiments Measurement error etc..Specifically, the fuselage pressure load change that aircraft cruising condition meets with unstable air-flow and causes, submarine The water pressure fluctuations caused because of briny environment change, the excitation difference that under motoring condition, uneven road surface causes, different temperatures bar The fluctuation of dielectric property (such as mass density and the velocity of sound) in closed in spaces under part.Traditional structural design and optimization are uncertain by these Utilize such as factor of safety etc. to limit it to structure design performance and the shadow of design function in the more rough mode of clean cut more Ring, fail effectively to combine current each subject and become more meticulous and high accuracy analysis development trend.Under this background, according to engineering demand And Sensitivity Analysis Method determines that the design variable of construct noise analysis and optimization problem (calls " design ginseng in the following text with uncertain parameter Number "), randomized optimization process based on design parameter large sample capacity test data are suggested and develop.But because of Subjective level With the restriction of objective test condition, for dividing into the extremely limited reality of sample size of meter parameter experiment data at all multi-states, Realize design parameter quantification with interval model, and then propose the structure acoustic radiation interval analysis side theoretical based on perturbation analysis Method.Not enough for existing methods in this field current, the present invention is with the interval model of design parameter for input, based on by dimension point Analysis strategy analyzes, with section reliability, a kind of method that construct noise reliability optimization invented by model.

Summary of the invention

The technical problem to be solved in the present invention is: overcome the large sample appearance to design variable experiment data in construct noise random optimization The restriction of amount demand, the restriction that in overcoming construct noise to analyze, fluctuated in design parameter minizone by Interval Perturbation analytic process, it is provided that Plant the construct noise analysis containing interval parameter and the method for reliability optimization.

It is said that in general, response involved during construct noise analysis and optimization can be divided into two classes, the sound of i.e. explicit analytical form Should be with the response by numerical analysis model (such as FEM (finite element) model or boundary element model) the non-analytical form of calculated implicit expression (such as sound Arbitrarily downgrade), with P, the present invention represents that the column vector of all implicit expression non-analytical form response composition is (such as space various location different frequency The sound pressure level column vector of lower sound pressure level composition).Because the calculation cost of the non-resolution response of implicit expression is far above display with calculating complicated process Resolution response, so present invention illustrates mainly for the non-resolution response of implicit expression.

The technical solution used in the present invention is: be primarily based on design parameter interval model propose construct noise Novel Interval Methods with The interval limit of computation structure noise response；The optimized algorithm analyzing model integrated maturation secondly based on section reliability realizes knot The reliability optimization of structure noise, implementation step is:

The first step: determine that construct noise optimizes the concrete variable that design variable vector x is comprised, become with thickness including length variable Amount, the design parameter comprised with design parameter vector h, including ambient temperature and density of material；According to design parameter vector h's Test data quantitatively turns to interval parameter vector h with interval model^I；

Second step: determine the natural frequency ω of the internal key equipment of enclosed construction₀With locus；Determine that key equipment is properly functioning Under the conditions of sound pressure level S scope S^IAnd critical reliability R_c；Determine that (x h), sets up construct noise reliability optimization object function f Corresponding section reliability Optimized model, selects the optimized algorithm intending using；

3rd step: the concrete variable that comprised according to design variable vector x in the first step and the initial value of given Optimized Iterative index K And design variable value x of k-th iteration step^(K)；

4th step: according to key equipment natural frequency ω in second step₀With locus, determine that response vector P is comprised concrete Response is the sound pressure level at different spatial different frequency point, assesses each response non-linear journey about design parameter vector h Degree, determines that response vector P is long-pending about exponent number N, Gauss integration point number s and the Gauss of each design parameter Least squares approach BranchUtilize interval parameter vector h in the first step^IWith Gauss integration pointDesign parameter vector h is sampled, by design parameter Vector sample point is stored in sample point matrix in block form M_hIn；

5th step: by design variable value x given in the 3rd step^(K)With the sample point matrix in block form M obtained in the 4th step_hLine by line Substitute in the numerical analysis model of response vector P, calculate response vector P response at each interval parameter vector sample point Value, and it is stored in response matrix in block form M_pIn；

6th step: according to the response matrix in block form M obtained in the 5th step_pAnd Least squares approach of based on Legnedre polynomial reason Opinion sets up the l component Least squares approach A about i-th interval parameter of response vector P^(l,i)(z)；

7th step: utilize the Least squares approach A obtained in the 6th step^(l,i)Z () calculates the l component of response vector P about the The maximum of points of i interval parameterAnd minimum pointTravel through all response component to obtain response vector P about i-th The maximum of points column vector of individual interval parameterWith minimum point column vectorTravel through all interval parameters, it is thus achieved that respond to The maximum of points matrix Z of amount P_maxWith minimum point matrix Z_min；

8th step: the maximum of points matrix Z of the response vector P that the 7th step is obtained_maxWith minimum point matrix Z_minMap to district Between in parameter vector h space, and calculate the interval limit vector P of response vector P^I；

9th step: utilize the response vector interval limit P that the 8th step obtains^IS is required with sound pressure level^IComputation interval reliability R, With critical reliability R given in second step_cRelatively, it is judged that the condition of convergence of object function；If being unsatisfactory for optimized algorithm rule, Then index K increases by 1, updates design variable and enters the 3rd step；If meeting optimized algorithm rule, export optimal case.

Described method achieves Analyzing Sound Radiation by Structure and the uncertain quantification in optimization with interval model.

Described method uses to calculate by dimension analysis strategy and asks Analyzing Sound Radiation by Structure with the uncertain parameter of interval model quantification The affecting laws of relevant response in topic.

Described method establishes the section reliability of consolidation form and analyzes model, and will contain interval ginseng based on this reliability analysis model The construct noise Optimized model of number is converted and has been solved.

Present invention advantage compared with prior art is:

(1) present invention can process relevant parameter Analyzing Sound Radiation by Structure and optimization under test data sample size limited conditions Problem, compensate for the deficiency of randomized optimization process；

(2) instant invention overcomes Analyzing Sound Radiation by Structure based on perturbation theory and be only applicable to the fluctuation of uncertain parameter little scope Limit, there is the widely suitability；

(3) present invention is with section reliability model for tolerance, largely avoid traditional knot based on factor of safety The conservative of structure noise optimization method, has catered to each subject high accuracy analysis development trend.

Accompanying drawing explanation

Fig. 1 is ultimate principle figure based on the construct noise reliability optimization by dimension analysis strategy；

Fig. 2 is flow chart based on the construct noise reliability optimization by dimension analysis strategy；

Fig. 3 is the configuration picture that fuselage simplifies cylindrical structure；

Fig. 4 is comparison and the assessment figure of Different Optimization scheme；

Fig. 5 is comparison and the assessment figure of Different Optimization scheme in key equipment characteristic frequency neighborhood.

Detailed description of the invention

Below in conjunction with the accompanying drawings and detailed description of the invention further illustrates the present invention.

A kind of method based on the construct noise reliability optimization by dimension analysis strategy of the present invention, is beneficial to improve thin-wall construction and encloses Cabin interior key equipment shake the safety under load effect or the comfortableness of cabin interior occupant at sound.The method is with district Between model realization Analyzing Sound Radiation by Structure probabilistic quantification relevant to optimization problem, utilize by dimension analysis strategy calculate district Between the parameter affecting laws to structural acoustic response characteristic, incorporation engineering security requirement by the section reliability of consolidation form Analyze model and complete the reliability conversion of uncertain structure acoustic response, before utilizing optimized algorithm to realize meeting given security requirement The structure optimization put.First, with the uncertainty of interval number model quantification structure and environment etc., set up reliable based on interval Property analyze the construct noise Optimized model of model, determine response based on the response such as structural acoustic about the nonlinear degree of interval parameter The exponent number of Least squares approach and Gauss integration point about interval parameter.Next, with Gauss integration point with interval number model to district Between parameter vector be sampled, calculate at interval parameter sample point response vector value, set up Least squares approach to determine response Vector is about the value dot matrix of interval parameter, thus calculates response interval vector.Finally utilize response interval vector and safety Property require computation interval reliability, optimized algorithm drive under complete structure optimization.The present invention instead of classics with section reliability Factor of safety, has catered to the development trend that becomes more meticulous of structure noise optimization, effectively prevent the conserved property of traditional optimization, Application prospect is bright and clear.

As shown in Figure 1, 2, it is embodied as step and is:

The first step: according to engineering demand, research worker experience and sensitive analysis result, determine construct noise reliability optimization Design variable vector x and design parameter vector h.Test data according to design parameter vector h is with interval parameter vector h^IQuantitatively Changing, its lower bound vector is respectively h with upper bound vector^LAnd h^U, midrange vector h^cWith radius vectors h^rIt is calculated as the most respectively:

$h^c={\[h_1^c,h_2^c,...,h_m^c\]}^T=(h^L+h^U)/2---\left(1\right)$

$h^r={\[h_1^r,h_2^r,...,h_m^r\]}^T=(h^U-h^L)/2---\left(2\right)$

Second step: determine the natural frequency ω of the internal key equipment of enclosed construction according to engineering demand₀With locus, so that it is determined that Numerical analysis response vector P；Sound pressure level boundary S under given key equipment normal running (operation) conditions^IAnd critical reliability R_c；Determine Optimization aim f (x, h^I), then construct noise Optimized model can be expressed as:

$\left\{\begin{array}{c}\underset{x}{min}f(x,h^I)\\ s.t.g_l(x,h^I)\≥0,l=1,2,...,L\\ g(x,h^I)=P(x,h^I)-S^I\end{array}\right.---\left(3\right)$

Total number of constraints during wherein L represents Optimized model.The present invention sets up the section reliability of following consolidation form and analyzes mould Type, has:

$R_{B^I\≥A^I}=\[(A^U-A^L)\·R_1+R_2+R_3/2\]/\[(A^U-A^L)(B^U-B^L)\]---\left(4\right)$

Wherein interval number A^IAnd B^IFor:

A^I=[A^L,A^U],B^I=[B^L,B^U] (5)

R₁=max (0, min (B^U-A^U,B^U-B^L)) (6)

R₂=min (B^U-B^L,max(A^U-B^L,0))·max(B^L-A^L,0) (7)

R₃=[max (0, min (A^U-B^L,A^U-A^L,B^U-A^L,B^U-B^L))]² (8)

Wherein R_i(i=1,2,3) section reliability corresponding to relative space position zones of different of two interval numbers is represented.

Utilize the section reliability model represented by formula (4), the construct noise Optimized model represented by formula (3) be converted into:

$\left\{\begin{array}{c}\underset{x}{min}\[\underset{h\∈h^I}{min}f(x,h),\underset{h\∈h^I}{\max }f(x,h)\]\\ s.t.R_{g_l^I\left(x\right)\≥0}\≥R_c\left(l\right)\\ g_l^I\left(x\right)=\[\underset{h\∈h^I}{\min }{g}_l(x,h),\underset{h\∈h^I}{\max }{g}_l(x,h)\],l=1,2,...,L\end{array}\right.---\left(9\right)$

WhereinConstraints g in expression (3)_l(x,h^IThe section reliability that) >=0 is set up, R_cL () represents the l constraints The critical section reliability set up, L represents total number of constraint function.

Conversion about object function can complete according to Practical Project demand, and the present invention is optimized for target with architecture quality, therefore by formula (9) It is further converted to:

$\left\{\begin{array}{c}\underset{x}{min}\underset{h\∈h^I}{min}f(x,h)\\ s.t.R_{g_l^I\left(x\right)\≥0}\≥R_c\left(l\right)\\ g_l^I\left(x\right)=\[\underset{h\∈h^I}{\min }{g}_l(x,h),\underset{h\∈h^I}{\max }{g}_l(x,h)\],l=1,2,...,L\end{array}\right.---\left(10\right)$

After setting up construct noise reliability optimization model (10), select appropriate optimized algorithm.

3rd step: set index K=1 and design variable vector value x^(K)。

4th step: determine that response vector P sets about each about the nonlinear degree of design parameter vector h according to response vector P The meter exponent number N of parameters optimal square approach, Gauss integration point number s and Gauss integration pointUtilize Gauss integration pointFormula (1) And formula (2), design parameter vector h is sampled, design parameter vector sample point is stored in sample point matrix in block form M_hIn, Have:

M_h=[S⁽¹⁾；S⁽²⁾；...；S⁽ⁿ⁾]^T (11)

Wherein about the interval parameter vector sample point matrix S of i-th interval parameter sampling⁽ⁱ⁾For:

$S^{\left(i\right)}(:,j)=h^c\left(j\right)+{\mathrm{\δ}}_{ij}\·h^r\left(j\right)\hat{z},j=1,2,...,m---\left(12\right)$

Symbol δ_ijFor Kronecker function, meet:

${\mathrm{\δ}}_{ij}=\left\{\begin{array}{cc}0,& i\≠j\\ 1,& i=j\end{array}\right.---\left(13\right)$

Wherein i, j represent the index value of interval parameter, and when the two is equal, Kronecker function value is 1, and otherwise value is 0.

5th step: by the sample point matrix in block form M represented by formula (11)_hSubstitute in the numerical analysis model of response vector P line by line, Calculate response vector P response value at each interval parameter vector sample point, and be stored in response matrix in block form M_pIn, have:

$M_p=\[M_p^{\left(1\right)};M_p^{\left(2\right)};...;M_p^{\left(m\right)}\]---\left(14\right)$

Wherein respond matrix in block formFor:

$M_p^{\left(i\right)}=\[P_1^{\left(i\right)},P_2^{\left(i\right)},...,P_s^{\left(i\right)}\],i=1,2,...,m---\left(15\right)$

The interval parameter vector sample point matrix in block form S of the i-th interval parameter sampling represented by formula (12)⁽ⁱ⁾The response vector composition at place Matrix, whereinRepresent the jth interval parameter vector sample point of corresponding i-th interval parameter.

6th step: according to the response matrix in block form M represented by formula (14)_pAnd Least squares approach of based on Legnedre polynomial reason Opinion sets up the l component Least squares approach A about i-th (1≤i≤m) individual interval parameter of response vector P^(l,i)(z), it may be assumed that

$A^{(l,i)}\left(z\right)=\underset{k=0}{\overset{N}{\Σ}}{c}_k^{(l,i)}{L}_k\left(z\right)---\left(16\right)$

Wherein L_kZ () represents kth rank Legnedre polynomial, the coefficient of Least squares approachIt is calculated as:

$c_k^{(l,i)}=(2k+1)\underset{j=1}{\overset{s}{\Σ}}\frac{P_j^{\left(i\right)}\left(l\right)L_k\left(z_j\right)}{(1-z_j^2){\[L_s^{\′}\left(z_j\right)\]}^2},k=0,1,...,N---\left(17\right)$

Wherein z_jRepresent Gauss integration point vector in the 4th stepElement,Represent corresponding i-th interval parameter Jth interval parameter vector sample point, L '_sZ () represents the derived function of s rank Legnedre polynomial.

7th step: calculate Least squares approach A^(l,i)The zero point of the derived function of (z), i.e.

$\frac{{\mathrm{dA}}^{(l,i)}\left(z\right)}{dz}=0---\left(18\right)$

The solution of formula (18) is vectorial with the end points composition extreme point of standard interval [-1,1]Such that it is able to calculate l of response vector P Component is about the maximum of points of i-th interval parameterAnd minimum pointMeet:

$\begin{array}{c}{A}^{(l,i)}\left(z_{\min }^{(l,i)}\right)=\underset{z\∈\stackrel{\‾}{z}}{\min }{A}^{(l,i)}\left(z\right)\\ A^{(l,i)}\left(z_{\max }^{(l,i)}\right)=\underset{z\∈\stackrel{\‾}{z}}{\max }{A}^{(l,i)}\left(z\right)\end{array}---\left(19\right)$

Travel through all response component to obtain the response vector P maximum of points column vector about i-th interval parameterAnd minimum point Column vectorHave:

$\begin{array}{c}{z}_{\max }^{(:,i)}={\[z_{\max }^{(1,i)},z_{\max }^{(2,i)},...,z_{\max }^{(N,i)}\]}^T\\ z_{\min }^{(:,i)}={\[z_{\min }^{(1,i)},z_{\min }^{(2,i)},...,z_{\min }^{(N,i)}\]}^T\end{array}---\left(20\right)$

Travel through all interval parameters further, it is possible to obtain response vector P is about the maximum of points matrix Z of interval parameter vector h_maxWith Minimum point matrix Z_min, have:

$\begin{array}{c}{Z}_{\max }=\[z_{\max }^{\left(\mathrm{:,}1\right)},z_{\max }^{(:,2)},...,z_{\max }^{(:,m)}\]\\ z_{\min }=\[z_{\min }^{\left(\mathrm{:,}1\right)},z_{\min }^{(:,2)},...,z_{\min }^{(:,m)}\]\end{array}---\left(21\right)$

WhereinWithRepresent the response vector P maximum about i-th interval parameter respectively Point column vectors and minima value point column vectors, matrix Z_maxAnd Z_minIt is to be joined about interval by each component of response vector P respectively The matrix of the maximum of points vector sum minimum point vector composition of number, its l row represents that the l component of response vector P is the most respectively Big value point and minimum point.

8th step: by the maximum of points matrix Z represented by formula (21)_maxWith minimum point matrix Z_minConvert to interval parameter vector Maximum of points matrix H is formed in space_maxWith minimum point matrix H_min, have:

Wherein symbol ο represents that two vectorial corresponding elements are multiplied.Matrix H_maxAnd H_minRow k represent response vector P's respectively Kth component is in interval parameter vector space h^IInterior maximum of points and minimum point, substitute into formula (22) in numerical analysis model, May determine that the lower bound vector P of response vector P^L, upper bound vector P^UWith response interval vector P^I。

9th step: according to formula (4), utilizes response interval vector P^IS is required with sound pressure level in second step^IComputation interval reliability R, With critical reliability R given in second step_cRelatively, it is judged that the condition of convergence of object function.If being unsatisfactory for optimized algorithm rule, Then index K value increase by 1, and update design variable x^(K)Rear entrance the 3rd step；If meeting optimized algorithm rule, then enter next Step.

Tenth step: output optimal case x^(K), and assess feasibility.

The cylindrical structure simplified with the airframe structure shown in Fig. 3 and come is as object, and foundation characteristic frequency is certain key of 80Hz Equipment is arranged on the specific location in the cabin that cylindrical shell encloses, when in its locus, place foundation characteristic frequency neighborhood The sound pressure level of (75Hz～85Hz) is in normal operating conditions less than this equipment under the conditions of critical sound pressure level (85dB).Set with key Standby normal operation is constraint, with different section inner cylindrical housing thickness T_i(i=1,2 ..., 5) it is variable, total with cylindrical shell Quality is target, it is considered to the impact on cylinder acoustic response characteristic of the uncertainty of elasticity modulus of materials, mass density and the velocity of sound, Realize the loss of weight optimization of cylindrical shell.Cylindrical shell and operatic tunes medium with finite element discretization, boundary condition be operatic tunes rigid plane with The node of cylindrical shell intersection location is clamped.Utilize deterministic optimization based on factor of safety (factor of safety is 1.005 and 1.1) and The present invention (section reliability index is 0.95) realizes cylindrical structure noise reliability optimization.In view of the cylinder operatic tunes and load Axial symmetry characteristic, arrange observation station 1 (0,0,0.4) and observation station 2 (0,0,0.5), with 1Hz as step in frequency range 75Hz～85Hz Long, calculate each observation station maximum sound pressure level at all Frequency points or sound pressure level interval limit, and be translated into correspondence about Bundle condition.The optimal case of Different Optimization method is shown in Table 1, is estimated prioritization scheme, and result is as shown in Figure 3 and Figure 4.

Table 1

Prioritization scheme is estimated, has:

(1) initial scheme is unsatisfactory for the sound pressure level requirement that key equipment is properly functioning, deterministic optimization scheme and section reliability Prioritization scheme all achieves in equipment characteristic frequency section with the cost of sacrifice cylindrical shell quality and the spatial distribution of change quality The reduction of sound pressure level.

(2) deterministic optimization based on factor of safety is it needs to be determined that factor of safety value, but the value of factor of safety not only with treat Analysis particular problem is relevant, also closely related with priori.Meanwhile, prioritization scheme based on given factor of safety value is in district Between parameter effect under response characteristic also can be caused to fluctuate, the effect of factor of safety is in the way of rough clean cut to limit sound Characteristics fluctuation is answered to be worth (this problem be maximum) given clear marginal value in (this problem be 85dB) most, and exceed marginal value i.e. Think inefficacy.For this angle, deterministic optimization scheme 1 causes equipment fault because factor of safety is too small, and definitiveness is excellent Change scheme 2 the most strictly limits sound pressure level in clear marginal value.

(3) actual for response fluctuation value is exceeded clear marginal value and is considered as failure condition by deterministic optimization based on factor of safety, Therefore factor of safety is caused to select to guard for avoiding losing efficacy in engineering field.Conversely, the conservative source of factor of safety It is the subjective mind of inefficacy in response reality fluctuation value is exceeded clear marginal value.Section reliability optimization thinks that equipment is super Normal operating conditions can also be kept in crossing the certain limit of clear marginal value, and scope is the biggest more easily breaks down, and utilize district Between reliability limit and respond actual fluctuation range in the degree of critical interval range.For this angle, deterministic optimization scheme 1 and the section reliability of scheme 2 be up to 0.9992 and 1.0.And in theory, section reliability optimization can be by given critical Interval range and section reliability require to realize maximum objective optimization.

The content not being described in detail in description of the invention belongs to prior art known to professional and technical personnel in the field.

The above is only the preferred embodiment of the present invention, it is noted that for those skilled in the art, Under the premise without departing from the principles of the invention, it is also possible to make some improvements and modifications, these improvements and modifications also should be regarded as this Bright protection domain.

Claims (4)

1. a method based on the construct noise reliability optimization by dimension analysis strategy, it is characterised in that comprise the following steps:

The first step: determine that construct noise optimizes the concrete variable that design variable vector x is comprised, become with thickness including length variable Amount, the design parameter comprised with design parameter vector h includes ambient temperature and density of material；According to design parameter vector h's Test data quantitatively turns to interval parameter vector h with interval model^I；

Second step: determine the natural frequency ω of the internal key equipment of enclosed construction₀With locus；Determine that key equipment is properly functioning Under the conditions of sound pressure level S scope S^IAnd critical reliability R_c；Determine that (x h), sets up construct noise reliability optimization object function f Corresponding section reliability Optimized model, selects the optimized algorithm intending using；

3rd step: the concrete variable that comprised according to design variable vector x in the first step and the initial value of given Optimized Iterative index K And design variable value x of k-th iteration step^(K)；

4th step: according to key equipment natural frequency ω in second step₀With locus, determine that response vector P is comprised concrete Response is the sound pressure level at different spatial different frequency point, assesses each response non-linear journey about design parameter vector h Degree, determines that response vector P is long-pending about exponent number N, Gauss integration point number s and the Gauss of each design parameter Least squares approach BranchUtilize interval parameter vector h in the first step^IWith Gauss integration pointDesign parameter vector h is sampled, by design parameter Vector sample point is stored in sample point matrix in block form M_hIn；

5th step: by design variable value x given in the 3rd step^(K)With the sample point matrix in block form M obtained in the 4th step_hLine by line Substitute in the numerical analysis model of response vector P, calculate response vector P response at each interval parameter vector sample point Value, and it is stored in response matrix in block form M_pIn；

6th step: according to the response matrix in block form M obtained in the 5th step_pAnd Least squares approach of based on Legnedre polynomial reason Opinion sets up the l component Least squares approach A about i-th interval parameter of response vector P^(l,i)(z)；

7th step: utilize the Least squares approach A obtained in the 6th step^(l,i)Z () calculates the l component of response vector P about the The maximum of points of i interval parameterAnd minimum pointTravel through all response component to obtain response vector P about i-th The maximum of points column vector of individual interval parameterWith minimum point column vectorTravel through all interval parameters, it is thus achieved that respond to The maximum of points matrix Z of amount P_maxWith minimum point matrix Z_min；

8th step: the maximum of points matrix Z of the response vector P that the 7th step is obtained_maxWith minimum point matrix Z_minMap to district Between in parameter vector h space, and calculate the interval limit vector P of response vector P^I；

9th step: utilize the response vector interval limit P that the 8th step obtains^IS is required with sound pressure level^IComputation interval reliability R, With critical reliability R given in second step_cRelatively, it is judged that the condition of convergence of object function；If being unsatisfactory for optimized algorithm rule, Then index K increases by 1, updates design variable and enters the 3rd step；If meeting optimized algorithm rule, export optimal case.

Method based on the construct noise reliability optimization by dimension analysis strategy the most according to claim 1, its feature exists In, described method achieves Analyzing Sound Radiation by Structure and the uncertain quantification in optimization with interval model.

Method based on the construct noise reliability optimization by dimension analysis strategy the most according to claim 1, its feature exists In, described method uses and calculates the uncertain parameter with interval model quantification to Analyzing Sound Radiation by Structure problem by dimension analysis strategy The affecting laws of middle relevant response.

Method based on the construct noise reliability optimization by dimension analysis strategy the most according to claim 1, its feature exists In, described method establishes the section reliability of consolidation form and analyzes model, and will contain interval ginseng based on this reliability analysis model The construct noise Optimized model of number is converted and has been solved.