Dynamic compensation method and system for viscoelastic material force sensor

Through the first-order Volterra series gray box model and dynamic compensation technology of long memory truncation length, the creep problem of viscoelastic material force sensor under step excitation is solved, and the dynamic measurement accuracy and time domain follow-up performance of the sensor are improved, so as to achieve Effective compensation for high-order time-invariant viscoelastic systems.

CN114218792BActive Publication Date: 2024-11-15BEIJING UNIV OF TECH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202111534216.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-15
Publication Date
2024-11-15
Estimated Expiration
2041-12-15

AI Technical Summary

Technical Problem

Viscoelastic material force sensors have obvious creep under step excitation, which affects the stability and accuracy of the sensor output, and the dynamic response speed is slow. It is difficult for the existing technology to effectively improve dynamic measurement accuracy and time-domain following performance.

Method used

The first-order Volterra series gray box model is used to construct the viscoelastic material force sensor system model, and a dynamic compensation model is constructed based on the long memory cut-off length. By adjusting the cut-off length to meet the preset conditions, dynamic compensation for the sensor is achieved and dynamic compensation of the sensor is improved. Measurement accuracy and time domain following performance.

Benefits of technology

By accurately characterizing the viscoelasticity and high-order linear time-invariant characteristics of the sensor, series compensation for the high-order time-invariant viscoelastic sensing system is achieved, which improves the dynamic measurement accuracy and time-domain following performance of the sensor, and eliminates creep phenomenon. Improve response speed.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114218792B_ABST
    Figure CN114218792B_ABST
Patent Text Reader

Abstract

The present invention relates to a dynamic compensation method and system for a viscoelastic material force sensor. The dynamic compensation method for a viscoelastic material force sensor provided by the present invention can accurately characterize the viscoelasticity and high-order linear time-invariant characteristics of the sensor by establishing a sensor gray box model based on a first-order Volterra series, and construct a dynamic compensation model to achieve series compensation for a high-order time-invariant viscoelastic sensor system without the need to identify the gray box model, thereby improving the accuracy of dynamic measurement of the sensor and the time domain tracking performance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of sensor dynamic compensation, and in particular to a method and system for dynamic compensation of a viscoelastic material force sensor. Background Art

[0002] The sensitive materials of miniaturized force sensors - soft materials (silicone rubber, gel, polymer, etc.) are widely used because of their simple production process, low cost and high sensitivity. However, soft materials have high-order linear time-invariant characteristics, their viscoelasticity has strong historical dependence, and there is obvious creep phenomenon under step excitation, which seriously affects the stability and accuracy of the sensor output; in addition, the natural frequency and damping ratio of soft materials are usually small, which makes the dynamic response speed of the sensor slow.

[0003] The mechanical model of viscoelastic materials can be described by different combinations of springs and dampers. The Kelvin model, Maxwell model, generalized Kelvin model, etc. are commonly used to establish the strain model of viscoelastic materials under step excitation to study their creep phenomenon. Generally, the more springs and dampers there are in the model, the more accurate the description of the viscoelastic system is, but it also leads to an increase in the order of the system, which greatly increases the difficulty of sensor system identification and the computing power requirements of the sensor.

[0004] At the same time, some researchers use system identification and zero-pole placement methods to improve the dynamic characteristics of sensors. The system identification method treats the sensor as a black box and directly identifies the compensator based on calibration data and sensor output. It may ignore the influence of low-order components in high-order systems and limit the expansion of bandwidth. The compensation accuracy of the zero-pole placement method depends on the accuracy of the sensor dynamic model and is more suitable for low-order minimum phase systems.

[0005] Therefore, there is an urgent need in the art to provide a method or system for dynamic compensation of a viscoelastic material force sensor that can improve the accuracy of dynamic measurement and time domain tracking performance of the sensor. Summary of the invention

[0006] The object of the present invention is to provide a method and system for dynamic compensation of a viscoelastic material force sensor, which can improve the accuracy of dynamic measurement and time domain tracking performance of the sensor.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A dynamic compensation method for a viscoelastic material force sensor, comprising:

[0009] Constructing a first-order Volterra series gray box model; the first-order Volterra series gray box model is a viscoelastic material force sensor system model constructed based on the first-order Volterra series;

[0010] According to the first-order Volterra series gray box model and the long memory cutoff length, a dynamic compensation model is constructed with the original output of the viscoelastic material force sensor as input and the expected compensation value as output;

[0011] Determine whether the expected compensation value output by the dynamic compensation model meets the preset conditions. If so, use the dynamic compensation model to dynamically compensate the viscoelastic material force sensor. If not, adjust the long memory cutoff length until the expected compensation value meets the preset conditions.

[0012] Preferably, the constructing of a first-order Volterra series grey box model specifically includes:

[0013] Obtaining the system gain of the viscoelastic material force sensor system and the strain of the viscoelastic material;

[0014] The first-order Volterra series gray box model is constructed according to the system gain and the strain of the viscoelastic material; the first-order Volterra series gray box model is:

[0015]

[0016] Where m is the long memory cutoff length, ΔC(t) is the output, K is the system gain, ε(t) is the strain of the viscoelastic material, σ(ti) is the memory link, t is the sampling time, i is the sampling point number, A is the parameter of the first-order Volterra series, A = (a1,…,a i ,…,a m ).

[0017] Preferably, the dynamic compensation model is:

[0018]

[0019] Where y(t) is the ideal output of the viscoelastic material force sensor, ΔC(t) is the original output of the viscoelastic material force sensor, B is the parameter of the dynamic compensation model, B = (b1,…,b i ,…,b m ).

[0020] Preferably, a dynamic compensation model is constructed based on the first-order Volterra series gray box model and the long memory cutoff length, which takes the original output of the viscoelastic material force sensor as input and the expected compensation value as output, and then further includes:

[0021] The parameters of the dynamic compensation model are determined by fitting the ideal output of the viscoelastic material force sensor and the original output of the viscoelastic material force sensor.

[0022] Preferably, the preset condition is that the difference between the output of the viscoelastic material force sensor after dynamic compensation by the dynamic compensation model and the ideal output of the viscoelastic material force sensor is less than a preset value.

[0023] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0024] The dynamic compensation method for a viscoelastic material force sensor provided by the present invention can accurately characterize the viscoelasticity and high-order linear time-invariant characteristics of the sensor by establishing a sensor gray box model based on a first-order Volterra series. Moreover, without the need to identify the gray box model, a dynamic compensation model is constructed to achieve series compensation for a high-order time-invariant viscoelastic sensing system, thereby improving the accuracy of dynamic measurement and the time domain tracking performance of the sensor.

[0025] Corresponding to the above-mentioned viscoelastic material force sensor dynamic compensation method, the present invention also provides a viscoelastic material force sensor dynamic compensation system, comprising:

[0026] The first model building module is used to build a first-order Volterra series gray box model; the first-order Volterra series gray box model is a viscoelastic material force sensor system model built based on the first-order Volterra series;

[0027] A second model building module is used to build a dynamic compensation model with the original output of the viscoelastic material force sensor as input and the expected compensation value as output according to the first-order Volterra series gray box model and the long memory cutoff length;

[0028] The dynamic compensation module is used to determine whether the expected compensation value output by the dynamic compensation model meets the preset conditions. If so, the dynamic compensation model is used to dynamically compensate the viscoelastic material force sensor. If not, the long memory cutoff length is adjusted until the expected compensation value meets the preset conditions.

[0029] Preferably, the first model building module includes:

[0030] An acquisition unit, used for acquiring a system gain of a viscoelastic material force sensor system and a strain of the viscoelastic material;

[0031] A construction unit is used to construct the first-order Volterra series gray box model according to the system gain and the strain of the viscoelastic material; the first-order Volterra series gray box model is:

[0032]

[0033] Where m is the long memory cutoff length, ΔC(t) is the output, K is the system gain, ε(t) is the strain of the viscoelastic material, σ(ti) is the memory link, t is the sampling time, i is the sampling point number, A is the parameter of the first-order Volterra series, A = (a1,…,a i ,…,a m ).

[0034] Preferably, it also includes:

[0035] The parameter determination module is used to determine the parameters of the dynamic compensation model by fitting the ideal output of the viscoelastic material force sensor and the original output of the viscoelastic material force sensor.

[0036] Since the technical effect achieved by the dynamic compensation system of the viscoelastic material force sensor provided by the present invention is the same as the technical effect achieved by the dynamic compensation system of the viscoelastic material force sensor provided above, it will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 A flow chart of a dynamic compensation method for a viscoelastic material force sensor provided by the present invention;

[0039] Figure 2 This is a diagram of the implementation structure of the dynamic compensation method of the viscoelastic material force sensor provided by the present invention;

[0040] Figure 3 A measurement principle diagram of a capacitive force sensor provided in an embodiment of the present invention;

[0041] Figure 4 A flow chart of sensor dynamic compensation provided by an embodiment of the present invention;

[0042] Figure 5 A schematic diagram of the compensation effect provided by an embodiment of the present invention;

[0043] Figure 6 This is a schematic structural diagram of the dynamic compensation system of the viscoelastic material force sensor provided by the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] The object of the present invention is to provide a method and system for dynamic compensation of a viscoelastic material force sensor, which can improve the accuracy of dynamic measurement and time domain tracking performance of the sensor.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] like Figure 1 As shown, the dynamic compensation method of the viscoelastic material force sensor provided by the present invention comprises:

[0048] Step 100: construct a first-order Volterra series gray box model. The first-order Volterra series gray box model is a viscoelastic material force sensor system model constructed based on the first-order Volterra series. The constructed first-order Volterra series gray box model is:

[0049]

[0050] Where m is the long memory cutoff length, ΔC(t) is the output, K is the system gain, ε(t) is the strain of the viscoelastic material, σ(ti) is the memory link, t is the sampling time, i is the sampling point number, A is the parameter of the first-order Volterra series, A = (a1,…,a i ,…,a m ).

[0051] Step 101: Based on the first-order Volterra series gray box model and the long memory cutoff length, a dynamic compensation model is constructed with the original output of the viscoelastic material force sensor as input and the expected compensation value as output. The constructed dynamic compensation model is:

[0052]

[0053] Where y(t) is the ideal output of the viscoelastic material force sensor, ΔC(t) is the original output of the viscoelastic material force sensor, B is the parameter of the dynamic compensation model, B = (b1,…,b i ,…,b m ).

[0054] Step 102: Determine whether the expected compensation value output by the dynamic compensation model meets the preset condition. If so, the dynamic compensation model is used to dynamically compensate the viscoelastic material force sensor. If not, the long memory cutoff length is adjusted until the expected compensation value meets the preset condition. The preset condition used in the present invention can be set according to actual needs. For example, the preset condition can be set to: the difference between the output of the viscoelastic material force sensor after dynamic compensation by the dynamic compensation model and the ideal output of the viscoelastic material force sensor is less than the preset value.

[0055] In order to further improve the accuracy of dynamic compensation, after constructing the dynamic compensation model, the present invention also uses the ideal output of the viscoelastic material force sensor and the original output of the viscoelastic material force sensor to fit and determine the parameters of the dynamic compensation model.

[0056] For force sensors made of viscoelastic materials such as soft materials, the analog electrical signal they output is usually linearly related to the strain of the soft sensitive material. Taking capacitive force sensors as an example, this type of sensor usually uses the plate variable spacing principle to measure normal pressure and the plate variable area principle to measure shear force, such as Figure 3 The following describes the specific implementation process and technical effects of the dynamic compensation method for the viscoelastic material force sensor provided by the present invention based on the measurement principle of the capacitive force sensor.

[0057] In this embodiment, the specific implementation process of the dynamic compensation method of the viscoelastic material force sensor is as follows: Figure 2 shown.

[0058] In the measurement unit, there is the following linear relationship between strain and capacitance change:

[0059]

[0060] Where d0 is the thickness of the sensitive material layer, ε r is the relative dielectric constant of the sensitive material layer, ε0 is the vacuum dielectric constant, S is the area of ​​the positive pressure measurement unit electrode plane, L is the length of the electrode plane perpendicular to the direction of the detected shear force, ΔC n (t) and ΔC s (t) are respectively the tensile and compressive n (t) and shear force ε s (t) is the change in capacitance under the action of

[0061] Obviously, within the sensor range, the sensor input and the medium layer strain are in a linear relationship. Therefore, without loss of generality, the output of the sensor can be expressed as:

[0062] ΔC(t)=Kε(t)

[0063] If the material is an ideal elastic body, ε(t) in the above formula can be described as:

[0064]

[0065] Among them, E ideal and G ideal are the elastic modulus and shear modulus of an ideal elastic body, respectively.

[0066] However, soft materials are not ideal elastic bodies. The viscoelasticity of soft materials causes the strain ε(t) to change continuously under constant stress, affecting the stability and accuracy of the sensor. Therefore, accurately characterizing the continuous change of strain under stress (creep) is the key to achieving accurate sensor measurement.

[0067] Taking the generalized Kelvin model as an example, if the system input is σ(t), the sensor system can be described as:

[0068]

[0069] Where J(t) is the creep compliance in the generalized Kelvin model:

[0070]

[0071] Among them, K is the system gain, which represents the linear relationship between strain and capacitance output. T is the sampling time of the sensor, and i is the number of the sampling point. It can be seen from formulas (3) and (4) that the output of the force sensor based on soft materials is history-dependent, resulting in creep phenomenon, and its viscosity link will also cause the sensor to respond slowly.

[0072] The soft material viscoelastic system described by the generalized Kelvin model is an n+1-order linear time-invariant system. Generally, as the number of elastic and viscous units increases, the more accurate the model's description of the viscoelastic system, the higher the order of the system. However, the corresponding compensation method requires the derivation of the zero poles of the sensor in the frequency domain and the design of the corresponding compensator, which is difficult to handle the above high-order system.

[0073] However, in the time domain, if the historical dependency in equation (3) can be removed, the output of the compensated system is expected to be close to the ideal servo system, without the need for complex identification of the zeros and poles of the high-order system. The expected compensation result is expressed as:

[0074] y(t)=Ka0σ(t) (5)

[0075] where a0 represents the gain between stress and strain at time t.

[0076] In order to achieve the ideal compensation result shown in formula (5), it is necessary to eliminate the historical memory link σ(tT)~σ(t-nT) in formula (3). However, creep compliance and stress are both functions related to time t, and the convolution calculation (such as formula (3)) of the complex viscoelastic model (such as formula (4)) limits the design of the compensator.

[0077] The first-order Volterra series can describe linear time-invariant systems, and the description accuracy of the system increases with the increase of memory length. In practice, the first-order Volterra series with long memory links is used to represent the sensor viscoelasticity and high-order linear time-invariant characteristics as accurately as possible, while avoiding the complexity brought to the model by multi-component modeling.

[0078] At any time t, J(T)~J(iT) can be regarded as n fixed coefficients. Therefore, in order to effectively characterize the historical dependence of high-order linear time-invariant viscoelastic systems, a first-order Volterra series with long memory characteristics is introduced to describe the sensor system based on soft materials, namely:

[0079]

[0080] Where m is the long memory cutoff length of the model, A=(a1,…,a i ,…,a m ) is the parameter of the first-order Volterra series.

[0081] At the same time, as the long memory cutoff parameter m increases, the description accuracy of the sensor system increases synchronously.

[0082] From equations (5) and (6), it can be seen that if the compensator can offset all parts of the historical memory link σ(t-1)~σ(tm) in the first-order Volterra series gray box model, efficient system compensation can be achieved to offset the high-order viscoelastic link of the sensor system, eliminate creep and improve the response speed. At this time, the expected output of the compensator is:

[0083]

[0084] Where y(t) is the ideal output of the sensor.

[0085] However, in the actual use of the measurement system, the compensator cannot directly obtain stress information. According to formula (6), the historical stress information should be obtained from the capacitance information. Therefore, based on the first-order Volterra series gray box model and its long memory truncation length m, a dynamic compensation model is established with the output as the expected compensation value and the input as the original output of the sensor. The form of the dynamic compensation model is:

[0086]

[0087] Where B=(b1,…,b i ,…,b m ) is the parameter of the compensator, ΔC(t)-ΔC(ti) can be measured in the sensor system, and equation (8) can be expanded into a form similar to equation (7) related to stress σ(t)~σ(tm). If there exists B that makes equations (7) and (8) valid, the compensator can achieve the expected compensation effect.

[0088] At this time, ΔC(t)-ΔC(tm) is expanded in the form of a first-order Volterra gray box model as follows:

[0089]

[0090] At this time, the ideal compensation output is expanded into:

[0091]

[0092] In order to obtain an ideal compensation effect, it is hoped that the coefficient of σ(t) in formula (6) is not zero, but the coefficients of the history-dependent links σ(t-1) to σ(tm) are all zero or close to zero, so that the compensated system is or is close to a follow-up system, achieving a good compensation effect. Therefore, let List the following equations:

[0093]

[0094] Formula (11) can be further expressed as:

[0095]

[0096] After sorting:

[0097]

[0098] Where A=(a1,…,a i ,…,a m ) is the long memory parameter describing the viscoelastic characteristics and high-order linear time-invariant characteristics of the sensor in the first-order Volterra series gray box model, B=(b1,…,b i ,…,b m ) are the compensator parameters to be fitted.

[0099] For different system parameters A, there is at least a least square solution B that makes the above equation valid, even if the actual output is close to the ideal follow-up output. Therefore, the calibrated ideal output y(t) and the original output ΔC(t) are used to fit the compensator.

[0100] Based on the above, when actually fitting parameter B, the ideal output y(t) is constructed as calibration data. According to formula (7), the closer the ideal output y(t) is to the output of the ideal sensor system, the closer y(t)-ΔC(t) is to the historical dependency part that needs to be eliminated in the system, and the better the effect of the compensator. Therefore, the step response method is used to dynamically calibrate the sensor, apply step excitation to each axis of the sensor, and record the original output ΔC(t) of the sensor without dynamic compensation for compensator parameter identification. The ideal output y(t) is constructed as:

[0101]

[0102] Among them, ΔC0 is the stable output of the sensor after a long time under ideal step excitation.

[0103] According to the step excitation time t0 and the truncation length m of the first-order Volterra gray box model, the ideal output y(t0)~y(t0+m-1) is obtained, and the corresponding capacitor output ΔC(t0-m)~ΔC(t0+m-1) is collected according to formula (8):

[0104]

[0105] Further, let y(t0)-ΔC(t0)=D(t0), F(t0,i)=ΔC(t0)-ΔC(t0-i), equation (15) can be expressed as:

[0106]

[0107] As shown in formula (16), where F mm is an m-order square matrix, B m1 is a matrix with m rows and 1 column, D m1 is a matrix with m rows and 1 column. The linear equations can be solved by LU decomposition, QR decomposition and other methods to obtain the parameter B. Therefore, by fitting the parameter B in the compensator (Equation 12) using the calibration data y(t) and the original output ΔC(t) of the sensor, an effective system compensator can be obtained to eliminate creep and improve the response time of the system. The fitting process is as follows: Figure 4 shown.

[0108] In addition, the longer the long memory truncation length m, the higher the accuracy of the first-order Volterra series in describing the sensor, and the better the effect of the compensator (Formula 12). However, considering the computing power of the sensor embedded system, the truncation length of the compensator will be limited. In general, the ratio of 2 times the original output ΔC(t) rise time to the sampling time T is selected as the long memory truncation length m to fit the compensator. Then verify the compensation effect and adjust the truncation length of the compensator (such as Figure 4 ), especially the step response settling time and overshoot of the compensated sensor.

[0109] Finally, the solved parameter B is written into the program of the sensor embedded system, and the compensation effect is as follows: Figure 5 As shown, Figure 5 Where ΔC(t) is the original data of the sensor, and Y(t) is the response of the sensor to step excitation after compensation.

[0110] Based on the above description, the dynamic compensation method of the viscoelastic material force sensor provided in this embodiment, first, establishes a sensor gray box model based on the first-order Volterra series to characterize the viscoelastic mechanical characteristics and high-order linear time-invariant characteristics in the sensor, and improves the accuracy of the model through the Volterra long memory cutoff length. Secondly, without the need to identify the gray box model, a compensator is constructed, that is, a sensor system characterized by a first-order Volterra series gray box model, and with the follow-up system as the target, an online compensator is derived to achieve series compensation for a high-order time-invariant viscoelastic sensor system. Then, a step response experiment is performed on the sensor to obtain dynamic calibration experimental data for compensator parameter identification, and then the parameter fitting of the online compensator is realized based on the calibration data. Finally, the dynamic compensator based on the first-order Volterra series gray box model after fitting is used to dynamically compensate the measured output of the sensor, thereby eliminating creep and effectively improving the dynamic performance of the sensor in the time-frequency domain.

[0111] Corresponding to the above-mentioned viscoelastic material force sensor dynamic compensation method, the present invention also provides a viscoelastic material force sensor dynamic compensation system, such as Figure 6 As shown, the system includes:

[0112] The first model building module 600 is used to build a first-order Volterra series grey box model. The first-order Volterra series grey box model is a viscoelastic material force sensor system model built based on the first-order Volterra series.

[0113] The second model building module 601 is used to build a dynamic compensation model with the original output of the viscoelastic material force sensor as input and the expected compensation value as output according to the first-order Volterra series gray box model and the long memory cutoff length.

[0114] The dynamic compensation module 602 is used to determine whether the expected compensation value output by the dynamic compensation model meets the preset conditions. If so, the dynamic compensation model is used to dynamically compensate the viscoelastic material force sensor. If not, the long memory cutoff length is adjusted until the expected compensation value meets the preset conditions.

[0115] The first model building module 600 includes: an acquisition unit and a building unit.

[0116] The acquisition unit is used to acquire the system gain of the viscoelastic material force sensor system and the strain of the viscoelastic material.

[0117] The building block is used to construct a first-order Volterra series grey box model based on the system gain and the strain of the viscoelastic material. The first-order Volterra series grey box model is:

[0118]

[0119] Where m is the long memory cutoff length, ΔC(t) is the output, K is the system gain, ε(t) is the strain of the viscoelastic material, σ(ti) is the memory link, t is the sampling time, i is the sampling point number, A is the parameter of the first-order Volterra series, A = (a1,…,a i ,…,a m ).

[0120] To further improve compensation accuracy, the system provided by the present invention also preferably includes a parameter determination module for determining parameters of the dynamic compensation model by fitting the ideal output of the viscoelastic material force sensor and the original output of the viscoelastic material force sensor.

[0121] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0122] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for dynamic compensation of a viscoelastic material force sensor, characterized in that: include: Construct a first-order Volterra series grey box model; The first-order Volterra series grey box model is a viscoelastic material force sensor system model constructed based on the first-order Volterra series; According to the first-order Volterra series gray box model and the long memory cutoff length, a dynamic compensation model is constructed with the original output of the viscoelastic material force sensor as input and the expected compensation value as output; Determine whether the expected compensation value output by the dynamic compensation model meets a preset condition, if so, use the dynamic compensation model to dynamically compensate the viscoelastic material force sensor, if not, adjust the long memory cutoff length until the expected compensation value meets the preset condition; The construction of the first-order Volterra series grey box model specifically includes: Obtaining the system gain of the viscoelastic material force sensor system and the strain of the viscoelastic material; The first-order Volterra series gray box model is constructed according to the system gain and the strain of the viscoelastic material; the first-order Volterra series gray box model is: Where m is the long memory cutoff length, K is the system gain, ε(t) is the strain of the viscoelastic material, σ(ti) is the memory link, t is the sampling time, i is the sampling point number, A is the parameter of the first-order Volterra series, A = (a1,…,a i ,…,a m ); The dynamic compensation model is: Where y(t) is the ideal output of the viscoelastic material force sensor, ΔC(t) is the original output of the viscoelastic material force sensor, B is the parameter of the dynamic compensation model, B = (b1,…,b i ,…,b m ).

2. The method for dynamic compensation of a viscoelastic material force sensor according to claim 1, characterized in that: According to the first-order Volterra series gray box model and the long memory cutoff length, a dynamic compensation model is constructed with the original output of the viscoelastic material force sensor as input and the expected compensation value as output, and then it also includes: The parameters of the dynamic compensation model are determined by fitting the ideal output of the viscoelastic material force sensor and the original output of the viscoelastic material force sensor.

3. The method for dynamic compensation of a viscoelastic material force sensor according to claim 1, characterized in that: The preset condition is that the difference between the output of the viscoelastic material force sensor after dynamic compensation by the dynamic compensation model and the ideal output of the viscoelastic material force sensor is less than a preset value.

4. A viscoelastic material force sensor dynamic compensation system, characterized in that: include: The first model building module is used to build a first-order Volterra series grey box model; The first-order Volterra series grey box model is a viscoelastic material force sensor system model constructed based on the first-order Volterra series; A second model building module is used to build a dynamic compensation model with the original output of the viscoelastic material force sensor as input and the expected compensation value as output according to the first-order Volterra series gray box model and the long memory cutoff length; A dynamic compensation module, used to determine whether the expected compensation value output by the dynamic compensation model meets a preset condition, and if so, dynamically compensate the viscoelastic material force sensor using the dynamic compensation model; if not, adjust the long memory cutoff length until the expected compensation value meets the preset condition; The first model building module includes: An acquisition unit, used for acquiring a system gain of a viscoelastic material force sensor system and a strain of the viscoelastic material; A construction unit is used to construct the first-order Volterra series gray box model according to the system gain and the strain of the viscoelastic material; the first-order Volterra series gray box model is: Where m is the long memory cutoff length, K is the system gain, ε(t) is the strain of the viscoelastic material, σ(ti) is the memory link, t is the sampling time, i is the sampling point number, A is the parameter of the first-order Volterra series, A = (a1,…,a i ,…,a m ); The dynamic compensation model is: Where y(t) is the ideal output of the viscoelastic material force sensor, ΔC(t) is the original output of the viscoelastic material force sensor, B is the parameter of the dynamic compensation model, B = (b1,…,b i ,…,b m ).

5. The viscoelastic material force sensor dynamic compensation system according to claim 4, characterized in that: Also includes: The parameter determination module is used to determine the parameters of the dynamic compensation model by fitting the ideal output of the viscoelastic material force sensor and the original output of the viscoelastic material force sensor.

Citation Information

Patent Citations

  • Temperature sensor temperature response lag compensation method

    CN106500857A

  • Vibration, creeping and lagging cascade compensation method used for piezoelectric cantilever beam

    CN108490795A

  • Method and system for online identification and compensation of hysteresis nonlinearity of piezoelectric ceramic actuator by using Maxwell model

    CN108710301A

  • Sensor system

    EP2392898A1