CN110991101A

CN110991101A - Optimization design method for compression type piezoelectric accelerometer structure

Info

Publication number: CN110991101A
Application number: CN201911116341.4A
Authority: CN
Inventors: 黄志亮; 彭安南; 阳同光; 王中华; 孙文德
Original assignee: Hongfujin Precision Industry Shenzhen Co Ltd
Current assignee: Hongfujin Precision Industry Shenzhen Co Ltd
Priority date: 2019-11-15
Filing date: 2019-11-15
Publication date: 2020-04-10
Anticipated expiration: 2039-11-15
Also published as: CN110991101B

Abstract

The invention relates to the field of piezoelectric accelerometers, and discloses a structure optimization design method of a compression type piezoelectric accelerometer. The method comprises the steps of firstly, selecting design variables based on sensitivity analysis by taking the mass of the piezoelectric accelerometer as a design target and the inherent frequency and voltage sensitivity as constraints, so as to construct an accelerometer structure optimization model, and solving to obtain a preliminary solution. And secondly, considering uncertainty related to structure size and material parameters in engineering, measuring the comprehensive influence of the uncertainty on performance based on an interval analysis technology, improving an optimization model based on the comprehensive influence, and obtaining a reliable design result. The method effectively reduces the structure quality on the premise of ensuring the performance reliability of the accelerometer in the aspects of natural frequency, voltage sensitivity and the like, and provides a potential design tool for the optimization of the high-performance and high-reliability compression type piezoelectric accelerometer; in addition, the method has the advantages of simple principle, simple and convenient operation and good engineering practicability.

Description

Optimization design method for compression type piezoelectric accelerometer structure

Technical Field

The invention relates to the field of piezoelectric accelerometers, in particular to a structure optimization design method suitable for a compression type piezoelectric accelerometer.

Background

The compression type piezoelectric accelerometer has the characteristics of rapid response, wide frequency band, good linearity, high sensitivity and the like, so that the compression type piezoelectric accelerometer is widely applied to important links such as vibration test calibration, mechanical dynamic tests, modal analysis, fault diagnosis and the like of systems such as aerospace, energy and power, ship traffic and the like. Properties that are often important considerations in the design of accelerometers include: mass, natural frequency and sensitivity. The smaller the self weight of the accelerometer, the smaller the additional mass load introduced to the system under test, the less the influence on the measurement accuracy and the easier installation. 1/3, which is the first order natural frequency of the accelerometer, is usually selected as the maximum measurement range of the product in the design specification; in other words, a higher natural frequency means a larger span, and better engineering applicability. For most sensors, sensitivity is the core performance of the product. However, a great deal of engineering practice and theoretical research has shown that the mass, natural frequency and sensitivity of compressive piezoelectric accelerometers are in a balanced relationship with each other. Moreover, these critical properties are affected by the structural characteristics and material parameters of the internal components of the accelerometer. Therefore, the influence of the structural characteristics and material parameters of the element on key performances is analyzed, a simple and easy-to-use optimization design method for the structure of the compression type piezoelectric accelerometer is developed, and the method has important engineering significance for improving the comprehensive performance of the product.

As the requirements of engineering applications on the performance of accelerometers are increasing, the traditional design method based on trial and error is faced with many challenges. First, the structural dimensions can be designed and have a significant impact on performance, are highly dimensional and interrelated, and trial and error one by one and verified through experimentation will result in unacceptable design costs. Although numerical simulation techniques such as finite element analysis and the like have been introduced into the design process of the accelerometer, the design cost is greatly reduced compared with a real test; however, the design method based on trial calculation strategy still has difficulty in meeting the requirements of practical engineering. Secondly, in engineering practice, there are inevitable uncertainties in the structural dimensions and material parameters of the components in the accelerometer, and the combined effect of these uncertainties may cause the failure of the accelerometer in terms of key performance. Although some random optimization algorithms aiming at uncertain structures have been developed in the theoretical world, the theory is profound, and the process is complicated, so that the method has obstacles on engineering application.

Disclosure of Invention

The invention overcomes the defects of the prior art and adopts a structure optimization design method of the compression type piezoelectric accelerometer, and the method can comprehensively consider the key performances of the compression type piezoelectric accelerometer, such as quality, first-order inherent frequency, voltage sensitivity and the like, and introduce an uncertainty analysis technology into the method, thereby providing a high-efficiency design tool for the structural design of the compression type piezoelectric accelerometer with high reliability.

In order to achieve the purpose, the invention adopts the following technical scheme: a structure optimization design method for a compression type piezoelectric accelerometer comprises the following steps:

(1) according to an accelerometer to be optimized, an optimization target is predefined as mass M and a constraint condition is that a first-order natural frequency f is more than or equal to f₀Sum voltage sensitivity S is greater than or equal to S₀Selecting the mass M of the minimized accelerometer as an optimization target;

(2) selecting a designable structural dimension X based on the optimization target mass M according to engineering experience_iI 1,2, and N, and setting X_iValue range of [ X ]_i ^L,X_i ^R]And an initial value X_i ⁽⁰⁾(ii) a At the initial value X_i ⁽⁰⁾Calculate X one by one_iN sensitivity to the accelerometer mass M, selecting a variable as a design variable according to the desired sensitivity, and forming a design vector X ═ X₁,X₂,...,X_n]；

(3) Based on the design variable X_iEstablishing a mass function M (X) of the accelerometer;

(4) based on the design variable X_iEstablishing a function f (X) of the first-order natural frequency of the accelerometer;

(5) establishing a function S (X) of the voltage sensitivity of the accelerometer based on the design variables;

(6) constructing an accelerometer optimization model based on the function, solving the accelerometer optimization model, and outputting a preliminary solution;

(7) based on the preliminary solution and the uncertainty of the parameter, evaluating its effect on the performance of the accelerometer;

(8) calculating a safety factor for the constraint based on an effect of uncertainty on the accelerometer performance;

(9) updating the accelerometer optimization model based on the safety factor, solving the accelerometer optimization model and outputting an optimal solution;

in step (1), first-order natural frequency constraint f ≧ f₀22kHz, the voltage sensitivity constraint S is greater than or equal to S₀＝27mv/g。

Further, in the step (2), based on X_i ⁽⁰⁾Establishing a three-dimensional model of the accelerometer, inputting the material density of each component element of the accelerometer into three-dimensional analysis design software, and determining the mass M⁽⁰⁾Calculating the sensitivities S to M one by one at the initial value_i ^MCan be represented as S_i ^M＝|M(X⁽⁰⁾ ₁,...,1.1*X⁽⁰⁾ _i,...,X⁽⁰⁾ _n)-M⁽⁰⁾|/M⁽⁰⁾。

Further, the structural dimension X_iWherein the sensitivity S is calculated for M one by one with i being 1,2_i ^MWill S_i ^MArranged from large to small and co-ordinated with X_iOne-to-one correspondence, take the first 6X_iForm a design vector X ═ X₁,X₂,...,X₆]。

Further, in the step (4), 1/4 finite element models are established for the accelerometers, symmetric constraints are set on an X-direction symmetric plane and a Z-direction symmetric plane, modal analysis is carried out based on a finite element analysis software platform, and an initial design X is obtained⁽⁰⁾＝[X₁ ⁽⁰⁾,X₂ ⁽⁰⁾,...,X₆ ⁽⁰⁾]All natural frequencies of the accelerometer within the selected frequency range,inputting the value of X in the finite element analysis software interface to obtain the corresponding first-order natural frequency f (X), thereby constructing a function f (X) of the first-order natural frequency, wherein f (X) is an implicit function, and establishing a second-order approximate function without cross terms for the function through a response surface technology.

Further, in step (5), the voltage sensitivity S can be written as: s (X) 2000X₇d_YρX₄((X₁+X₂+X₃)²-4)/e_Y((X₁+X₂)²-X₁ ²) Where ρ is the density of the mass element; d_YAnd e_YRespectively representing the piezoelectric coefficient and the dielectric constant of the two piezoelectric block materials in the Y direction.

Further, in step (6), the accelerometer optimization model is:

solving the optimization model can obtain a preliminary solution: x⁽¹⁾＝[X₁ ⁽¹⁾,X₂ ⁽¹⁾,...,X_n ⁽¹⁾]。

Further, in step (6), the accelerometer optimization model is:

solving the optimization model can obtain a preliminary solution: x⁽¹⁾＝[X₁ ⁽¹⁾,X₂ ⁽¹⁾,...,X₆ ⁽¹⁾]。

The method for solving the optimization model to obtain the preliminary solution is a genetic algorithm.

Further, in step (7), the process of evaluating the effect on the accelerometer performance is:

(7.1) according to engineering experience and existing sample data, adopting intervals to describe the uncertainty of the preliminary solution and the parameters, wherein the preliminary solution and the corresponding parameters can be collected in a vector P, and the jth component of P is uniformly written into P_j，P_jInterval I for uncertainty of_j＝[P_j ^a,P_j ^b]Description of the invention I_jIs written as P_j ^cAll the intervals have midpoints constituting a vector P^c＝(P₁ ^c,P₂ ^c,...,P_m ^c)；

(7.2) calculating the constraint function value of the lower boundary point of the interval, which can be written as:

f_j ^a＝f(P₁ ^c,P₂ ^c,...,P_j ^a,...,P_m ^c)，S_j ^a＝S(P₁ ^c,P₂ ^c,...,P_j ^a,...,P_m ^c)，j＝1,2,...,m；

(7.3) calculating the constraint function value of the boundary point on the interval, which can be written as:

f_j ^b＝f(P₁ ^c,P₂ ^c,...,P_j ^b,...,P_m ^c)，S_j ^b＝S(P₁ ^c,P₂ ^c,...,P_j ^b,...,P_m ^c)，j＝1,2,...,m。

further, in step (8), the process of calculating the safety factor of the constraint is as follows:

(8.1) for the first order natural frequency constraint, e.g. f_j ^a≤f_j ^bThen P is selected_j ^aOtherwise, P is selected_j ^b；

(8.2) adding the selected P_j ^aOr P_j ^bJ is collected in vector P1, 2_f ^wThe worst performance of calculating the first order natural frequency can be written as: f. of^w＝f(P_f ^w)；

(8.3) calculating a safety factor for the first order natural frequency constraint: c_f＝f(X⁽¹⁾)/f^w；

(8.4) for the voltage sensitivity constraint,such as S_j ^a≤S_j ^bThen P is selected_j ^aOtherwise, P is selected_j ^b；

(8.5) adding the selected P_j ^aOr P_j ^bJ is collected in vector P1, 2_S ^wThe worst performance of the voltage sensitivity is calculated and can be written as: s^w＝S(P_S ^w)；

(8.6) calculating a safety factor for the voltage sensitivity constraint: c_S＝f(X⁽¹⁾)/S^w。

Further, in step (9), the updated optimization model may be written as:

further, in step (9), the updated optimization model may be written as:

the method for solving the updated optimization model is a genetic algorithm.

Solving the optimization model to obtain an optimal solution X^*And in X^*Of the accelerometer mass M (X)^*) First order natural frequency f (X)^*) And voltage sensitivity S (X)^*)。

Compared with the prior art, the invention has the advantages that:

firstly, key performances such as quality, first-order inherent frequency, voltage sensitivity and the like of the compression type piezoelectric accelerometer are comprehensively considered, and a repeated trial calculation process based on engineering experience is avoided by constructing a standard optimization model, so that an effective design tool is provided for developing the compression type piezoelectric accelerometer with good comprehensive performance; secondly, in the optimization process, an uncertainty analysis technology is introduced, the influence of uncertainty factors on acceleration performance is measured quantitatively by adopting an interval method, an optimization model is effectively improved, and an optimization design result with high reliability is obtained. And thirdly, the invention has simple principle, simple and convenient operation and easy realization and popularization.

Drawings

FIG. 1 is a schematic flow diagram of the process of the present invention.

Fig. 2 is a schematic sectional view of an accelerometer in an embodiment of the invention.

FIG. 3 is a finite element analysis model of the first order natural frequency of an accelerometer in an exemplary embodiment of the invention.

Reference numerals: 20. an accelerometer; 21. a piezoelectric element; 211. an upper piezoelectric block; 212. a lower piezoelectric block; 213. an electrode sheet; 22. a mass element; 23. a tailstock; 24. a base; 25. an insulating pad; 26. a bolt; 30. a finite element model; 31. a Z-direction symmetry plane; 32. the X-direction symmetry plane.

Detailed Description

The invention will now be further described with reference to the following examples, which are not to be construed as limiting the invention in any way, and any limited number of modifications which can be made within the scope of the claims of the invention are still within the scope of the claims of the invention.

As shown in fig. 1-3, the present invention provides a method for optimally designing a structure of a compression type piezoelectric accelerometer, which comprises the following processing steps:

step 1: optimization objectives and constraints are predefined according to the accelerometer to be optimized. As shown in fig. 2, the compressive piezoelectric accelerometer 20 to be optimized is composed of a piezoelectric element 21, a mass element 22, a tailstock 23, a pedestal 24, an insulating pad 25 and a bolt 26; the piezoelectric element 21 includes two annular upper and lower piezoelectric blocks (211, 212) and an electrode piece 213; the mass element 26, the piezoelectric element 21, and the tailstock 23 are fixed together by a bolt 26 and are integrally placed on the base 24, and an insulating pad 25 is provided between the base 24 and the tailstock 23. When acceleration exists in the axial direction, the piezoelectric element 21 generates electric charge under the action of the inertial force of the mass element 22, and the electric charge is captured and output through the electrode sheet 213, thereby achieving measurement of the acceleration. Selecting the mass M of the minimized accelerometer 20 as an optimization target; two constraints are set: the first order natural frequency constraint f ≧ f of accelerometer 20₀22kHz, the voltage sensitivity constraint S is greater than or equal to S₀＝27mv/g。

Step 2: and extracting design variables with higher sensitivity based on the optimization target. As shown in fig. 2, the structural dimension X_i1,2, 13 are programmable, setting X according to engineering experience_iValue range of [ X ]_i ^L,X_i ^R]Initial value X_i ⁽⁰⁾(ii) a Based on X_i ⁽⁰⁾A three-dimensional model of accelerometer 20 is created and the material densities of the various components listed in Table 1 are input, and mass M of accelerometer 20 is analytically obtained based on three-dimensional design software⁽⁰⁾64.30 g; calculating sensitivities S to M one by one at an initial value_i ^MIt can be expressed as:

S_i ^M＝|M(X⁽⁰⁾ ₁,...,1.1*X⁽⁰⁾ _i,...,X⁽⁰⁾ ₁₃)-M⁽⁰⁾|/M⁽⁰⁾；X_ithe information of (a) and the resulting sensitivity are listed in table 2. Selecting X with higher sensitivity₁-X₆As a design variable, a design vector X ═ X can be written₁,X₂,...,X₆]。

TABLE 1

TABLE 2

And step 3: based on the selected design variables, a mass function of accelerometer 20 is established. The quality function can be expressed as:

M(X)＝0.0562X₄((X₁+X₂+X₃)²-4)+0.0244X₅((X₁+X₂+6.4)²-4)-1.676X₆+24.93。

and 4, step 4: based on the selected design variables, a first order natural frequency function of accelerometer 20 is established. As shown in fig. 3, for accelerationThe meter 20 establishes 1/4 finite element model 30, sets symmetry constraints on X-direction symmetry plane 32 and Z-direction symmetry plane 31, and performs modal analysis based on finite element analysis software platform to obtain initial design X⁽⁰⁾＝[X₁ ⁽⁰⁾,X₂ ⁽⁰⁾,...,X₆ ⁽⁰⁾]The time accelerometer 21 has all natural frequencies within 80KHz, of which the first order natural frequency f (X)⁽⁰⁾) 22.67 KHz; by inputting the value of X in the finite element analysis software interface, the corresponding first order natural frequency f (X) can be obtained, and the function f (X) of the first order natural frequency is constructed. (x) is an implicit function, and a second-order approximation function without cross terms is established for the implicit function through a response surface technology, and can be written as:

f(X)＝24.627-0.309X₁+1.006X₂+6.826X₃,-1.604X₄+0.139X₅+0.241X₆-0.264X₁ ²-0.362X₂ ²-1.267X₃ ²+0.062X₄ ²-0.050X₅ ²-0.039X₆ ²

and 5: based on the selected design variables, a functional function of the voltage sensitivity of accelerometer 20 is established. The accelerometer 20 voltage sensitivity S can be written as: s (X) 2000X₇d_YρX₄((X₁+X₂+X₃)²-4)/e_Y((X₁+X₂)²-X₁ ²). Wherein, as mentioned above, the structural dimension X₇2.65mm, the density ρ of the mass element 22 is 0.0179g/mm³；d_YAnd e_YRespectively representing the piezoelectric coefficient and dielectric constant of the materials of the upper and lower piezoelectric blocks (211, 212) in the Y direction as shown in FIG. 3, d_Y＝3.92e-10m/V，e_Y1.95 e-8F/m. At X⁽⁰⁾S (X)⁽⁰⁾)＝27.20mV/g。

Step 6: based on the established function, an optimization model of the accelerometer 20 is constructed, and a preliminary solution is output after solving the optimization model. The optimization model is as follows:

solving the optimization model by adopting a classical genetic algorithm to obtain a preliminary solution: x⁽¹⁾＝[2.19,2.02,2.42,4.75,4.02,3.80](ii) a At X⁽¹⁾F (X)⁽¹⁾)＝26.66kHz，S(X⁽¹⁾) 27.03mV/g, which meets the constraint requirement; at this time, the mass M (X) of the accelerometer 21⁽¹⁾)＝41.50g。

And 7: based on the preliminary solution and the uncertainty of the parameters, the effect of the uncertainty on the performance of the accelerometer 20 is evaluated. And describing the uncertainty of the initial solution and the parameters by using intervals according to engineering experience and the existing sample data. The parameters include d_YAnd e_YAnd elastic modulus E of the insulating pad 25 and the upper and lower piezoelectric block materials (211, 212)₂₅，E₂₁(ii) a The preliminary solution and the above parameters may be collected in a vector P, i.e., P ═ X₁ ⁽¹⁾,X₂ ⁽¹⁾,...,X₆ ⁽¹⁾,d_Y,d_Y,E₂₅,E₂₁]. For convenience of presentation, the jth component of P is hereinafter collectively written as P_j(e.g., P)₁＝X₁ ⁽¹⁾)。P_jInterval I for uncertainty of_j＝[P_j ^a,P_j ^b]Description of the invention I_jIs written as P_j ^cAll the intervals have midpoints constituting a vector P^c＝(P₁ ^c,P₂ ^c,...,P₁₀ ^c). The constraint function value of the lower boundary point of the interval is calculated and can be written as: f. of_j ^a＝f(P₁ ^c,P₂ ^c,...,P_j ^a,...,P₁₀ ^c)，S_j ^a＝S(P₁ ^c,P₂ ^c,...,P_j ^a,...,P₁₀ ^c) (ii) a The same reasoning can be obtained for the upper boundary point. P, I_jAnd f obtained_j ^a，f_j ^bThe results are shown in Table 3.

TABLE 3

And 8: a safety factor for the constraints is calculated based on the effect of uncertainty on the performance of accelerometer 20. For natural frequency constraints of the accelerometer, e.g. f_j ^a≤f_j ^bThen P is selected_j ^aOtherwise, P is selected_j ^b(ii) a The selected boundary points are collected in a vector:

P_f ^w＝[2.55,2.09,3.43,3.51,4.07,3.75,3.72e-10,1.83e-8,8.1,44.6]。

based on P_f ^wThe worst performance of calculating the first order natural frequency of the accelerometer can be written as: f. of^w＝f(P_f ^w) 25.30 kHz; and further calculating a safety factor of the natural frequency: c_f＝f(X⁽¹⁾)/f^w1.054. Voltage sensitivity constraints for accelerometers, e.g. S_j ^a≤S_j ^bThen P is selected_j ^aOtherwise, P is selected_j ^b(ii) a The selected boundary points are collected in a vector:

P_S ^w＝[2.55,2.09,3.33,3.41,3.97,3.65,3.72e-10,2.07e-8,8.1,44.6]。

based on P_S ^wThe worst performance of the accelerometer voltage sensitivity is calculated and can be written as: s^w＝S(P_S ^w) 23.06 mV/g; and further calculating a safety factor of the natural frequency: c_S＝S(X⁽¹⁾)/S^w＝1.172。

And step 9: based on the obtained safety factor, the optimization model of the accelerometer 20 is updated, and an optimal solution is output after the optimization model is solved. The updated optimization model can be written as:

solving the optimization model by adopting a classical genetic algorithm to obtain an optimal solutionX^*＝[X₁ ^*,X₂ ^*,...,X₆ ^*]And in X^*Of the accelerometer mass M (X)^*) First order natural frequency f (X)^*) And voltage sensitivity S (X)^*)。

To demonstrate the beneficial effects of the proposed optimal design method, the calculation at X is as described in step 8^*And X⁽⁰⁾The worst performance values for first order natural frequency and voltage sensitivity are shown in table 4, along with other results before and after optimization. From the results, firstly, the mass of the accelerometer is greatly reduced compared with that before optimization, and the reduction amplitude reaches 1/3; therefore, the additional load of the accelerometer on the tested system due to self weight is greatly reduced. Secondly, optimizing the worst performance values of first-order natural frequency and voltage sensitivity at the solution by considering the uncertainty of the structure size and the material parameters to meet the design requirement; while the initial solution may be at risk of failure in the presence of uncertainty, neither the worst performance values (21.20kHz, 23.49mV/g) of its first order natural frequency nor voltage sensitivity meet the design requirements (22kHz, 27 mV/g). In summary, in the embodiment, by the proposed optimization design method, the quality of the accelerometer is greatly reduced on the premise of ensuring the reliability of the accelerometer performance under the uncertain condition, so that the comprehensive performance and reliability of the accelerometer are effectively improved.

TABLE 4

Claims

1. A structure optimization design method of a compression type piezoelectric accelerometer is characterized by comprising the following steps:

(1) according to an accelerometer to be optimized, an optimization target is predefined as mass M and a constraint condition is that a first-order natural frequency f is more than or equal to f₀Sum voltage sensitivity S is greater than or equal to S₀Wherein f is₀Nominal natural frequency, S, given by the design requirements₀Indicating the nominal sensitivity given by the design requirements.

(2) Based on theOptimizing the target mass M, determining the structural dimension X according to engineering experience_iHas a value range of [ X ]^L _i,X^R _i]1,2, N, wherein X is^L _i、X^R _iRespectively representing the lower and upper bounds of the value range, with an initial value of X_i ⁽⁰⁾At said initial value X_i ⁽⁰⁾Calculate X one by one_iN sensitivity to the accelerometer mass M, selecting a variable as a design variable according to the desired sensitivity, and forming a design vector X ═ X₁,X₂,...,X_n]；

2. the optimized design method for the structure of the compression-type piezoelectric accelerometer according to claim 1, wherein in the step (2), the design method is based on X_i ⁽⁰⁾Establishing a three-dimensional model of the accelerometer, inputting the material density of each component element of the accelerometer into three-dimensional analysis design software, and determining the mass M⁽⁰⁾Calculating the sensitivities S to M one by one at the initial value_i ^MCan be represented as S_i ^M＝|M(X⁽⁰⁾ ₁,...,1.1*X⁽⁰⁾ _i,...,X⁽⁰⁾ _n)-M⁽⁰⁾|/M⁽⁰⁾。

3. The method of claim 2, wherein the dimension X is a structural dimension_iWherein the sensitivity S is calculated for M one by one with i being 1,2_i ^MWill S_i ^MArranged from large to small and co-ordinated with X_iOne-to-one correspondence, take the first 6X_iForm a design vector X ═ X₁,X₂,...,X₆]。

4. The optimized design method for the structure of the compression-type piezoelectric accelerometer of claim 3, wherein in the step (4), 1/4 finite element models are established for the accelerometer, symmetry constraints are set on the X-direction symmetry plane and the Z-direction symmetry plane, and modal analysis is performed based on a finite element analysis software platform to obtain an initial design X⁽⁰⁾＝[X₁ ⁽⁰⁾,X₂ ⁽⁰⁾,...,X₆ ⁽⁰⁾]And (2) inputting the value of X in a finite element analysis software interface to obtain corresponding first-order natural frequency f (X) of the accelerometer in all natural frequencies in a selected frequency range, thereby constructing a function f (X) of the first-order natural frequency, wherein f (X) is an implicit function, and establishing a second-order approximate function without cross terms for the function through a response surface technology.

5. The optimized design method for the structure of the compression-type piezoelectric accelerometer according to claim 4, wherein in the step (5), the voltage sensitivity S can be written as: s (X) 2000X₇d_YρX₄((X₁+X₂+X₃)²-4)/e_Y((X₁+X₂)²-X₁ ²) Where ρ is the density of the mass element; d_YAnd e_YRespectively representing the piezoelectric coefficient and the dielectric constant of the two piezoelectric block materials in the Y direction.

6. The optimization design method of the structure of the compression type piezoelectric accelerometer according to claim 1 or 2, wherein in the step (6), the accelerometer optimization model is as follows:

7. The optimization design method of the compression type piezoelectric accelerometer according to any one of claims 3-5, wherein in the step (6), the accelerometer optimization model is as follows:

solving the optimization model using genetic algorithms can yield a preliminary solution: x⁽¹⁾＝[X₁ ⁽¹⁾,X₂ ⁽¹⁾,...,X₆ ⁽¹⁾]。

8. The optimized design method for the structure of the compression piezoelectric accelerometer according to claim 6, wherein in the step (7), the process of evaluating the influence of the optimized design method on the performance of the accelerometer is as follows:

(7.1) according to engineering experience and existing sample data, adopting intervals to describe the uncertainty of the preliminary solution and the parameters, wherein the preliminary solution and the corresponding parameters can be collected in a vector P, and the jth component of P is uniformly written into P_j，P_jInterval I for uncertainty of_j＝[P_j ^a,P_j ^b]Is described, wherein P_j ^a、P_j ^bRespectively represent P_jLower and upper bounds of uncertainty interval, I_jIs written as P_j ^cAll the intervals have midpoints constituting a vector P^c＝(P₁ ^c,P₂ ^c,...,P_m ^c)；

f_j ^a＝f(P₁ ^c,P₂ ^c,...,P_j ^a,...,P_m ^c)，S_j ^a＝S(P₁ ^c,P₂ ^c,...,P_j ^a,...,P_m ^c) J is 1, 2.. times.m, wherein f_j ^a、S_j ^aIs represented by the formula (P)₁ ^c,P₂ ^c,...,P_j ^a,...,P_m ^c) A first order intrinsic frequency value and a voltage sensitivity value of the accelerometer;

f_j ^b＝f(P₁ ^c,P₂ ^c,...,P_j ^b,...,P_m ^c)，S_j ^b＝S(P₁ ^c,P₂ ^c,...,P_j ^b,...,P_m ^c) J is 1, 2.. times.m, wherein f_j ^b、S_j ^bIs represented by the formula (P)₁ ^c,P₂ ^c,...,P_j ^b,...,P_m ^c) A first order intrinsic frequency value and a voltage sensitivity value of the accelerometer.

9. The structural optimization design method of the compressive piezoelectric accelerometer according to claim 6, wherein in the step (8), the process of calculating the safety factor of the constraint is as follows:

(8.2) adding the selected P_j ^aOr P_j ^bJ 1,2, m is collected in a vector

The worst performance of calculating the first order natural frequency can be written as:

wherein f is^wTo represent

A value of a first order natural frequency of the accelerometer;

(8.4) constraint on the voltage sensitivity, e.g. S_j ^a≤S_j ^bThen P is selected_j ^aOtherwise, P is selected_j ^b；

(8.5) adding the selected P_j ^aOr P_j ^bJ is collected in vector P1, 2_S ^wThe worst performance of the voltage sensitivity is calculated and can be written as: s^w＝S(P_S ^w) In which S is^wTo represent

A value of the accelerometer voltage sensitivity;

(11.6) calculating a safety factor for the voltage sensitivity constraint: c_S＝f(X⁽¹⁾)/S^w。

10. The structural optimization design method of the compressive piezoelectric accelerometer according to claim 7, wherein in the step (8), the process of calculating the safety factor of the constraint is as follows:

(8.2) adding the selected P_j ^aOr P_j ^bJ 1,2, m is collected in a vector

11. The optimized design method for the structure of the compression-type piezoelectric accelerometer according to claim 9, wherein in the step (9), the updated optimization model can be written as:

12. the optimized design method for the structure of the compression-type piezoelectric accelerometer according to claim 10, wherein in the step (9), the updated optimization model can be written as: