A variational Bayesian inference technique for model updating of structural systems with unknown noise statistics

Abstract

Dynamic models of structural and mechanical systems can be updated to match the measured data through a Bayesian inference process. However, the performance of classical (non-adaptive) Bayesian model updating approaches decreases significantly when the pre-assumed statistical characteristics of the model prediction error are violated. To overcome this issue, this paper presents an adaptive recursive variational Bayesian approach to estimate the statistical characteristics of the prediction error jointly with the unknown model parameters. This approach improves the accuracy and robustness of model updating by including the estimation of model prediction error. The performance of this approach is demonstrated using numerically simulated data obtained from a structural frame with material non-linearity under earthquake excitation. Results show that in the presence of non-stationary noise/error, the non-adaptive approach fails to estimate unknown model parameters, whereas the proposed approach can accurately estimate them.

1 Introduction

Bayesian model updating aims to estimate uncertain model parameters by minimizing the discrepancies between measured and predicted responses (Friswell and Mottershead, 2013). This technique has been extensively used for structural system identification (Behmanesh et al., 2015), parameter estimation (Ching et al., 2006; Astroza et al., 2014), damage identification (Doebling et al., 1996; Yang et al., 2006), and virtual sensing (Wenzel et al., 2007; Nabiyan et al., 2020). However, the performance of Bayesian model updating depends on the quality of prior knowledge about the prediction error, which includes the effects of modeling error and measurement noise (Beck and Yuen, 2004). In classical (non-adaptive) Bayesian model updating methods, the prediction error is assumed as a stationary, zero-mean Gaussian white noise process. However, this is not always the case in practice, and the prediction error can generally be a non-stationary, non-white, and non-Gaussian process due to the effect of modeling error (Sanayei et al., 2001; Law and Stuart, 2012; Nabiyan et al., 2022). The estimation accuracy of non-adaptive Bayesian methods can be adversely affected in practice when the prediction error deviates from stationary Gaussian assumption (Mehra, 1972; Xu et al., 2019).

To mitigate the need for prior knowledge about prediction error in model updating, several methods referred to as adaptive Bayesian model updating methods have been proposed (Akhlaghi et al., 2017; Amini Tehrani et al., 2020; Song et al., 2020b). Most of the adaptive Bayesian model updating methods in the literature consider a zero-mean Gaussian white noise with an unknown covariance matrix for modeling prediction error and estimate the error covariance matrix together with other model parameters or states. Zheng et al. (2018) developed a robust adaptive unscented Kalman filter (UKF) to improve the accuracy and robustness of state estimation of a non-linear system with uncertain noise covariance. In this method, first the states of the non-linear system are estimated using a standard UKF (Wu and Smyth, 2007), and then a covariance-matching method (Mehra, 1972) is utilized to estimate the covariance matrix of process noise and measurement noise. Astroza et al. (2019) used a similar approach to jointly estimate the unknown model parameters along with the diagonal entries of the covariance matrix of the prediction error. Huang et al. (2020) developed a hierarchical Bayesian model by combining sparse Bayesian learning (Tipping, 2001) with dual Kalman filters. Their hierarchical model employs two inference levels, state and parameter estimation and noise–parameter learning. They considered a zero-mean Gaussian distribution for the measurement noise in which the diagonal entries of its covariance matrix were learned solely from the measurement data up to the current time step. Yuen and Kuok (2016) proposed a Bayesian probabilistic algorithm to estimate the noise covariance matrix for the extended Kalman filter using the maximum a posteriori approach. Their method is also applicable for non-stationary noise with a time-variant covariance matrix. Song et al. (2020b) proposed two adaptive Kalman filters formulated based on covariance-matching techniques (Mehra, 1972) to jointly estimate the unknown model parameters along with the full covariance matrix of the prediction error. The validation studies show the superior performance of the presented adaptive filtering methods compared to standard UKF, in which the prediction errors have predefined distributions. The mentioned studies assumed a zero-mean Gaussian white noise for the prediction error. However, the modeling error may cause the prediction error to have non-zero mean (Sanayei et al., 2001). To address this issue, Kontoroupi and Smyth (2016) proposed a Bayesian method to estimate a biased (non-zero mean) prediction error. They assumed that the mean vector and covariance matrix of the prediction error are time-invariant and have Gaussian and inverse-Wishart distributions, respectively. In a previous work, Nabiyan et al., (2022) developed a two-step marginal maximum a posteriori (MAP) estimation approach to find a point estimation of the unknown model parameters and the prediction error statistics, where the mean vector and covariance matrix of the prediction error are considered to be time-variant.

In this paper, we introduce a completely different mathematical approach with better performance, in comparison to our previous work, (Nabiyan et al., 2022) for estimating both the unknown model parameters and statistical characteristics (mean vector and covariance matrix) of the prediction error, as well as approximating their joint posterior distribution. Exact calculation of this high-dimensional joint posterior distribution is intractable, so the process requires approximation (Šmídl and Quinn, 2006). Two approximation schemes can be used: stochastic or sampling methods such as Markov chain Monte Carlo (MCMC) (Bishop and Nasrabadi, 2006) and deterministic or variational frameworks such as variational Bayesian (VB) (Opper and Saad, 2001; Beal, 2003). In comparison to the sampling methods, the VB method is analytically tractable and is computationally less demanding (Beal, 2003). The VB method is used in this work as a tool to segregate the posterior distribution into separate components, which can help in solving the problem analytically. The VB method has been successfully applied for joint state and noise estimation in navigation, target tracking, and control-related applications (Huang et al., 2017; Zhang et al., 2018). In these applications, the adaptive VB Kalman filter method was used to jointly estimate the covariance matrix of a zero-mean prediction error and the state of linear (Sarkka and Nummenmaa, 2009; Sun et al., 2012; Huang et al., 2016; Huang et al., 2017) or non-linear (Sarkka and Hartikainen, 2013; Shi et al., 2018; Sun et al., 2018) state-space models. The VB method assumes that the approximate joint distribution is the product of some single- or multi-variable factors and uses the Kullback–Leibler (KL) divergence to minimize the difference between the approximation and the true posterior. In this paper, we introduce a new adaptive method for non-linear model updating based on the VB method to approximate the joint posterior distribution of the unknown model parameters and statistical characteristics of the prediction error at each time step.

The paper is structured as follows: Section 2 provides the model updating problem statement. Section 3 presents a detailed derivation of the proposed VB method for estimating the joint posterior distribution of unknown model parameters and statistical characteristics of the prediction error. The formulation of the VB method is then compared with that of the two-step marginal MAP estimation method (Nabiyan et al., 2022) in Section 4. In Section 5 and Section 6, the proposed method is verified by two model updating case studies: one with time-variant measurement noise and the other with modeling error. The results are compared to those from the two-step marginal MAP estimation method published in Nabiyan et al., (2022) and a non-adaptive Bayesian model updating method. Finally, the conclusions are presented in Section 7.

2 Model updating problem statement

We consider the measured response of a non-linear (or linear) dynamic system and its corresponding model [e.g., finite element (FE)] prediction , where is the vector of unknown model parameters. The parameter estimation problem at time can be formulated as (Haykin, 2004; Ebrahimian et al., 2015)where is the process noise and is the prediction error. In this study, the input forces are assumed to be known, so for notation brevity, the dependency of the model prediction response to the input forces is not shown explicitly in Eq. 2. The process noise is assumed to follow a zero-mean Gaussian white noise process with covariance matrix Q, i.e., . In non-adaptive Bayesian model updating methods, the prediction error is assumed to be a zero-mean Gaussian white noise process with a constant or time-invariant covariance matrix R, i.e., . For the parameter estimation problem defined in Eqs. 1, 2, the non-adaptive Bayesian methods can be used to find an estimate for the first two statistical moments of unknown model parameters (Astroza et al., 2015; Nabiyan et al., 2020). However, in the adaptive Bayesian model updating methods, the prediction error can be modeled as a non-stationary Gaussian random process with an unknown and time-variant mean vector and covariance matrix , i.e., , to be estimated recursively and jointly with the unknown vector of model parameters .

In this paper, we developed a new adaptive recursive Bayesian model updating method. Like other recursive Bayesian model updating algorithms, the proposed method has two steps at each time : “prediction” and “correction” (Astroza et al., 2017). In the “prediction” step, the new measurement at time is not given to the estimation process yet. Therefore, the prior estimates of , , and , denoted by minus superscript, are predicted through a dynamic model using their posterior estimates at the previous time step k-1. Eq. 1 can be used as the dynamic model for unknown model parameters of (Astroza et al., 2014; Nabiyan et al., 2022). For predicting the prior estimates of and , the dynamic models defined in Nabiyan et al., (2022) can be used, considering the forgetting factor parameters of and . These dynamic models result in and . In the “correction” step, the prior estimates are updated by the new measurement to obtain the posterior estimates, denoted by , , and . The updating process is further described as follows:

In our previous work (Nabiyan et al., 2022), we developed a two-step maximum a posteriori (MAP) estimation method to estimate , , and by maximizing the joint posterior distribution , i.e.,

To solve this MAP problem, we broke the problem into two iterative MAP estimation problems as and . In this paper, we aim to find the whole joint posterior distribution of unknown model parameters and noise, i.e., . Nevertheless, analytical working with this joint posterior distribution is not tractable because the number of variables is high, and the joint distribution is highly complex. To overcome this issue, we used the VB method to approximate this joint posterior distribution. The VB method and the derivation details are explained in the next section.

3 Variational Bayesian (VB) method

VB is a method to approximate a joint distribution by a joint distribution which can be factorized into single-variable or grouped-variable factors. The Kullback–Leibler (KL) divergence criterion is then used to make as close as possible to (Bishop and Nasrabadi, 2006). Using the VB method, we approximate the joint posterior distribution of the unknown model parameters and prediction error mean vector and covariance matrix by separating this joint posterior distribution into two factors as follows:where and are unknown distributions that can be obtained by minimizing the KL divergence between the right- and left-hand side of Eq. 4. The KL divergence is defined as

Using variational calculus to minimize the aforementioned KL divergence with respect to each of and while keeping the other one fixed will result in Eqs. 6, 7. The details of this derivation can be found in Weinstock, (1974); Tzikas et al., (2008).where denotes the expected value of with respect to x with the probability density function of , i.e., . The terms and denote the constants with respect to variables and , respectively. Since Eqs. 6, 7 are coupled through the term , analytical solutions are not available. Therefore, the fixed-point iteration algorithm can be employed to find approximate solutions for Eqs. 6, 7. To this end, we try to expand the right-hand sides of Eqs. 6, 7 from their innermost parentheses to the outer ones through the following steps.

The joint distribution in Eqs. 6, 7 can be factored aswhere is the likelihood function, is the prior distribution of , is the prior joint distribution of and , and is known because this is a recursive algorithm, meaning that at each time step, only the new measurement is used for updating parameters, so depends on the past measurements.

Here, we further expanded the terms on the right-hand side of Eq. 8. Based on Eq. 2, the likelihood function has a Gaussian distribution as

For the second and third terms on the right-hand side of Eq. 8, it is assumed, similar to (Nabiyan et al., 2022), that and have prior distributions of Gaussian and normal-inverse-Wishart (NIW), respectively. The NIW distribution is the product of a Gaussian (or normal) distribution and an inverse-Wishart (IW). The NIW is selected for the prior distribution because it is a conjugate prior for a Gaussian likelihood with an unknown mean vector and covariance matrix. The conjugacy guarantees the same functional form for the posterior and prior distributions (O'Hagan and Forster, 2004). Therefore, the second and third terms on the right-hand side of Eq. 8 can be written as follows:where and are the mean vector and covariance matrix, respectively, of the unknown model parameters given measurements but not . The minus superscripts represent the prior estimates. , , , and are the prior estimates of statistical parameters used in the NIW distribution. is the prior estimate for , and is the symmetric positive definite scale matrix. and are the confidence parameter and the degree of freedom parameter, respectively. They are scalar parameters and satisfy and , where is the number of measurement sensors.

By substituting Eqs. 9, 10, and 11 into Eq. 8, we obtain

The Gaussian (or normal) distribution and the inverse-Wishart (IW) distribution are proportional to the following expressions:where represents the determinant and denotes trace of a matrix. The sign “” representing “proportional to” is used in Eq. 13, as the normalizing terms are ignored.

Using Eq. 12 and the definitions of normal and IW distributions in Eq. 13, the term , which is used in Eqs 6, 7, can be expanded as follows:where denotes a constant with respect to all variables , and .

Now, having the expansion of in Eq. 14, we calculate the expectation terms in Eqs. 6, 7. By obtaining the expectation from each term of Eq. 14, the expectation term in Eq. 6, , can be expressed as follows:where is the summation of expectations of all terms in the right-hand side of Eq. 14, except the second and fourth terms, and is constant with respect to . It should be noted that the expectation of the fourth term of Eq. 14 is equal to itself because it does not depend on , and .

Using Supplementary Appendix Lemma S1 in the Appendix, the expectation term in Eq. 15 can be evaluated as

Now, we consider as a NIW distribution, i.e., , where the plus superscripts represent the posterior estimates. Therefore, by substituting the mean and covariance of , i.e., and in Eq. 16, and taking the expectation with respect to , we will have

It should be noted that in deriving Eq. 17, we use (Granström and Orguner, 2011), in which is the mean of IW distribution denoted as , and can be calculated as follows (O'Hagan and Forster, 2004).where (O'Hagan and Forster, 2004).

In Eq. 6, is proportional to the exponential of . So, by substituting Eq. 17 into Eq. 15 and taking the exponential of Eq. 15, it can be seen that

By linearizing in Eq. 19 by the first-order Taylor expansion about , i.e., , where is the sensitivity matrix of the FE model with respect to at , i.e., , Eq. 19 yields

The first exponential term in the right-hand side of Eq. 20 shows a Gaussian distribution for ignoring the normalization term. By using Supplementary Appendix Lemma S2 in the Appendix, the second exponential term in the right-hand side of Eq. 20 also represents a Gaussian distribution for . Therefore, the right-hand side of Eq. 20 is the product of two Gaussian distributions which, based on Supplementary Appendix Lemma S3 in the Appendix, results in a Gaussian distribution, i.e., . To obtain the parameters of this Gaussian distribution, we match the terms in the left-hand and the right-hand side of in Eq. 20, which results inwhere , , and . It is worth noting that Eq. (21-a) and Eq. (21-b) are similar to non-adaptive Bayesian model updating formulations, except for the term , which is added in Eq. (21-a) to consider non-zero mean prediction error.

In a similar way, we can evaluate the expectation term in Eq. 7 as follows. Getting the mathematical expectation of Eq. 14 with respect to leads towhere is sum of the expectations of the third, fourth, and last terms of the right-hand side of Eq. 14 and is constant with respect to and .

Now, considering and linearizing by the first-order Taylor expansion about , i.e., , where is the sensitivity matrix of the model with respect to at , the expectation term in the right-hand side of Eq. 22 can be obtained as follows using Lemma 1 in the Appendix.

Based on Eq. 7, is proportional to the exponential of . Substituting Eq. 23 into Eq. 22 and taking the exponential of Eq. 22, can be found as follows:

The right-hand side of Eq. 24 includes the product of two Gaussian distributions for and one IW distribution for . The product of these two Gaussian distributions leads to a scaled Gaussian distribution based on Lemma 3 in the Appendix. Therefore, the right-hand side of Eq. 24 results in a NIW distribution, which is a product of a normal distribution for and an IW distribution for . By substituting in the left-hand side of Eq. 24, and using Lemma 4 in the Appendix, the four parameters of the NIW distribution can be derived as follows by matching the terms in the left- and right-hand sides of Eq. 24.

Now having as an approximation for , we can evaluate the expectation of to represent point estimates of unknown model parameters and noise. Therefore, , , and represented in Eq. (21-a), Eq. (25-a), and (Eq. 18), respectively, can be considered the point estimates for , and . Eq. (21-a), Eq. (25-a), and (Eq. 18) are coupled equations that can be solved iteratively using a fixed-point iteration algorithm. In each iteration, the model-predicted responses and the sensitivity matrix need to be updated based on the updated unknown model parameters ; however, calculating the sensitivity matrix at each time step can be computationally demanding. To reduce the execution time, we can use the prior sensitivity matrix in Eq. (25-d). Since the convergence criteria are not changed, using instead of has no effect on the final estimation results. In this paper, a finite difference method is used to calculate the sensitivity matrix at each time step. The proposed algorithm of recursive VB for joint model and noise identification is presented in Figure 1. This framework can work with any FE modeling and simulation platform such as OpenSees, which is used in this study.

FIGURE 1

4 Comparison of the VB method with two-step marginal MAP estimation method

In this section, we compare the proposed variational Bayesian (VB) method with the two-step marginal MAP estimation method that was recently developed by Nabiyan et al. (2022) for joint estimation of unknown model parameters and the mean vector and covariance matrix of the prediction error. The formulations of the two methods are closely similar. There are two main differences between the two methods, as explained as follows. First, in the two-step marginal MAP estimation method, the mode of IW distribution is used as an point estimate, while in the VB method, the mean of IW distribution is assigned as (as shown in Table 1). The reason for this difference is that the MAP approach is used in the former method, while the expectation value is used in the latter. It is worth noting that the mode and mean of the IW distribution are not coincident (O'Hagan and Forster, 2004). Second, different equations are used to calculate the term as shown in Table 1. The equation used to calculate in the VB method has an additional term, i.e., .

In most applications, these two differences have small effects on the results because of the following reasons. First, the difference between the mode and mean of the IW distribution decreases through time as the value of increases in time in the denominator of the characterizing equations of . Second, as the covariance matrix of the unknown model parameters decreases through the Bayesian estimation (Song et al., 2020a), the effects of this additional term on the results also become negligible over time. However, the additional term in the VB method improves the estimation of at early iterations, which in turn results in better estimation of unknown model parameters, especially when the initial estimate of the prediction error covariance matrix is poorly selected. In the next section, the estimation results of these two methods are compared through two numerical case studies.

5 Case study 1: 3-Story 1-bay steel moment frame considering time-variant measurement noise

In this section, the performance of the proposed method is evaluated when applied to a numerical model of a 3-story 1-bay steel moment frame structure under earthquake excitation. The estimation results are compared with those of the two-step marginal MAP estimation method (Nabiyan et al., 2022) and a non-adaptive Bayesian model updating method (Ebrahimian et al., 2015). The story height and the bay width are 3.5 m and 6.0 m, respectively, as shown in Figure 2A. The frame’s geometric and material properties are similar to those used in our previous work (Nabiyan et al., 2022), and more details about the considered case study can be found there.

FIGURE 2

The numerical model of the frame is developed in OpenSees (McKenna, 2000). For this, force-based beam–column elements with seven integration points are used for columns and beams. A single fiber is used to represent each flange of beam and column cross-sections, while 10 fibers are used to discretize their webs. The uniaxial Giuffre–Menegotto–Pinto (GMP) material model (Filippou et al., 1983) with primary parametersis used to model the steel fibers and simulate the nominal/true dynamic response of the structure, where Young’s modulus, yield stress, and strain-hardening ratio. The first three parameters denoted by subscript “” are for columns, and the last three ones denoted by subscript “” are used for beams. A nodal mass , shown by the black circle in Figure 2A, is considered for each story at the beam–column nodes to represent dead and live mass. To model damping energy dissipation, Rayleigh damping with 2% damping ratio is considered for the first two vibration modes of the structure.

To simulate measurement data, the frame structure is excited by the Loma Prieta earthquake (0° component at Los Gatos station), as shown in Figure 2B. Then, the horizontal absolute acceleration response time histories of each floor (shown by black boxes in Figure 2A) are extracted and contaminated with artificial measurement noise to result in simulated measurement data. The measurement noise is considered a non-stationary Gaussian random process with time-variant mean vector and covariance matrix as follows: It is worth noting that non-stationary noise with time-variant mean and covariance is a common problem in realistic monitoring, and the assumed sinusoidal form for the mean and covariance is hypothetical and considered for the feasibility study.

Our goal is to estimate the unknown model parameters and compare them with their true values . The initial estimate of the unknown model parameters and its covariance matrix are selected as and , respectively. The initial mean vector and covariance matrix of the prediction error (measurement noise) are, respectively, assumed as ; , where is the identity matrix. Other initial parameters of the NIW distribution are selected as , , and , with . The process noise covariance matrix is selected as , and the forgetting factor parameters used for defining the dynamic model of the mean vector and covariance matrix of the prediction error are assumed as and , respectively Based on our study, the parameter estimation results are acceptable for and . However, choosing lower values for and may deteriorate the performance of the model updating process. The sensitivity of the estimation process to other filter tuning parameters (e.g., initial model parameter values, ,) has been the subject of another study presented in Astroza et al., (2019b).

Now, the proposed VB method is applied to jointly estimate the unknown model parameter vector and the mean vector and covariance matrix of prediction error. In this verification study, the results are compared with those of the two-step marginal MAP estimation method (Nabiyan et al., 2022) and the non-adaptive Bayesian model updating method (Ebrahimian et al., 2015) when using the same initial values. As mentioned before, in the non-adaptive Bayesian method, a zero-mean Gaussian white noise with a time-invariant diagonal covariance matrix is assumed for the prediction error , i.e., .

The estimated components of the mean vector and covariance matrix of the prediction error (measurement noise in this example) by the three methods (VB, two-step marginal MAP, and the non-adaptive) are compared with their true values in Figure 3 and Figure 4, respectively. The non-adaptive Bayesian model updating method, as mentioned before, does not estimate the mean vector and the covariance matrix of the prediction error, so they remain constant during the estimation process. However, both the VB and the two-step marginal MAP estimation methods can accurately track the trend of the true/nominal mean and covariance of error through time. As can be seen in Figure 4, in comparison to the two-step marginal MAP method, the VB method better estimates the covariance matrix of the prediction error at the early time steps because of the additional term discussed in the previous section.

FIGURE 3

FIGURE 4

Figure 5 shows the time histories of the unknown model parameters estimated by all three methods: the proposed VB method, the two-step marginal MAP estimation method, and the non-adaptive Bayesian method. It can be observed that the non-adaptive model updating method converges to incorrect unknown model parameters, or even diverges. However, adaptive methods can estimate unknown model parameters very well. The weakness of the non-adaptive method shows that estimation of the prediction error significantly affects the model updating results when the prediction error has a time-variant non-zero mean and, therefore, should not be ignored. Comparing the VB method with the two-step marginal MAP estimation method, the VB method better estimates the parameter (columns strain-hardening ratio) because of the better estimation of the covariance matrix of the prediction error at the early time steps.

FIGURE 5

6 Case study 2: 3-Story 3-bay steel moment frame considering modeling error

In this second study, we examine the proposed VB method in the presence of modeling error on a 3-story, 3-bay steel moment frame. The frame is taken from Song et al., (2020a), and the geometry, frame sections, and loads are shown in Figure 6. All beams and columns have wide flange profiles and are modeled with displacement-based beam–column elements. Rayleigh damping with 2% damping of the first and second modes is considered for structural damping. Distributed gravity loads are considered concentrated masses at nodes. The measured data are simulated using the steel constitutive model of Giuffre–Menegotto–Pinto (GMP) for beams and columns, in which their true properties are selected the same as in the previous example. The frame is excited by the Loma Prieta earthquake shown in Figure 2B, and the horizontal absolute acceleration responses at each floor (marked by black boxes in Figure 6) are recorded. Then, 1% RMS NSR Gaussian zero-mean white noise is added to these simulated acceleration responses to be considered measurement data.

FIGURE 6

While we use the GMP constitutive model Figure 7A to simulate the measurements, a bilinear model (Figure 7B) is used in the estimation process to add explicit modeling error to the model updating process.

FIGURE 7

For model updating in this example, the material properties of columns and beams are considered unknown model parameters, similar to the previous example. The initial estimate for the unknown model parameters and its covariance matrix are selected as and , respectively. The initial mean vector and covariance matrix of the prediction error are assumed as and , respectively. As mentioned before, the non-adaptive method does not update these two parameters and considers them constants during the estimation process. Other initial parameters of the NIW distribution required for both adaptive methods are selected as , , and , with . The process noise covariance matrix is selected as , and the forgetting factor parameters used for defining the dynamic model of the mean vector and covariance matrix of the prediction error are assumed as and .

Figure 8 shows the model updating results for the three methods: the proposed VB method, the two-step marginal MAP estimation method, and the non-adaptive Bayesian method. As can be seen, the non-adaptive method cannot estimate unknown model parameters correctly for all parameters, except and . The measured structural responses are less sensitive to parameters and , rather than , which results in estimation of and being more effected by prediction error (measurement noise + modeling error). The non-adaptive method limits the prediction error to a zero-mean and has fixed covariance matrix. However, adaptive methods release this assumption by updating the statistical parameters of the prediction error recursively at each time step, which results in far better estimations for unknown model parameters. In comparison with the proposed adaptive VB method with the two-step marginal MAP estimation method, the VB method improved the estimation values of column and beam stiffness hardening and because of better estimation of prediction error.

FIGURE 8

To investigate the capability of the updated model with material modeling error in predicting the responses, absolute acceleration responses at each floor and moment–curvature response at the base of the first story’s inner column are predicted from the updated model using all three methods and compared with their true counterparts in Figure 9 and Figure 10, respectively. Although, as can be seen in Figure 9, the discrepancies between measured and predicted acceleration responses are minimized, and the estimated parameters are biased for the case of non-adaptive, and to a lesser extent for the two-step MAP method. As can be observed in Figure 10, the response predictions are considerably improved by both adaptive methods, and their predictions agree with the true moment–curvature response. Comparing the proposed VB method with the two-step MAP method, the VB method better predicts the response because of better estimation of unknown model parameters.

FIGURE 9

FIGURE 10

7 Conclusion

In this paper, we exploited the variational Bayesian (VB) approach and proposed an adaptive variational Bayesian model updating method for joint model and noise identification. A detailed mathematical derivation is provided in the paper. The performance of the proposed method is demonstrated through two numerical case studies. Two non-linear steel moment frames subjected to earthquake excitation were used, in which six parameters characterizing the constitutive models of the steel beams and columns were considered unknown. In the first case study, absolute acceleration responses at each floor contaminated by Gaussian noise with a time-variant mean vector and covariance matrix were considered measurement data. For considering modeling error in the second case study, a steel constitutive model of Giuffre–Menegotto–Pinto (GMP) is used for data simulation, and a bilinear constitutive model is used in the estimation process. The estimation results of both case studies showed that the proposed VB-based method performs well in the presence of time-variant prediction error (measurement noise and modeling error). The proposed VB-based method was also compared to a recently developed two-step marginal MAP estimation method, and the non-adaptive Bayesian model updating method. The results showed that both adaptive methods have comparable performance, while the non-adaptive method resulted in significantly biased estimations due to the adverse effects of non-stationary prediction error. The future scope of this work is to extend the algorithm to estimate the dynamic inputs, which will result in a joint input–parameter–noise estimation, and to validate the algorithm in real-world applications where the modeling errors can result in divergence or significant bias in regular model updating algorithms.

References

• 1

  AkhlaghiS.ZhouN.HuangZ. (July 2017). Adaptive adjustment of noise covariance in Kalman filter for dynamic state estimation, Proceedings of the 2017 IEEE power and energy society general meeting (IEEE). Chicago, IL, USA. 10.1109/PESGM.2017.8273755

  • CrossRef
  • Google Scholar

• 2

  Amini TehraniH.BakhshiA.YangT. T. (2020). Online probabilistic model class selection and joint estimation of structures for post-disaster monitoring. J. Vib. Control27, 1860–1878. 10.1177/1077546320949115

  • CrossRef
  • Google Scholar

• 3

  AstrozaR.AlessandriA.ConteJ. P. (2019). A dual adaptive filtering approach for nonlinear finite element model updating accounting for modeling uncertainty. Mech. Syst. Signal Process115, 782–800. 10.1016/j.ymssp.2018.06.014

  • CrossRef
  • Google Scholar

• 4

  AstrozaR.EbrahimianH.ConteJ. P. (2014). Material parameter identification in distributed plasticity FE models of frame-type structures using nonlinear stochastic filtering. J. Eng. Mech.141 (5), 4014149. 10.1061/(ASCE)EM.1943-7889.0000851

  • CrossRef
  • Google Scholar

• 5

  AstrozaR.EbrahimianH.ConteJ. P. (2017). “Batch and recursive bayesian estimation methods for nonlinear structural system identification,” in Risk and reliability analysis: Theory and applications. Springer series in reliability engineering. Editor GardoniP. (Berlin, Germany: Springer), 341–364. 10.1007/978-3-319-52425-2_15

  • CrossRef
  • Google Scholar

• 6

  AstrozaR.EbrahimianH.ConteJ. P. (2019b). Performance comparison of Kalman−based filters for nonlinear structural finite element model updating. J. Sound. Vib.438, 520–542. 10.1016/j.jsv.2018.09.023

  • CrossRef
  • Google Scholar

• 7

  BealM. J. (2003). Variational algorithms for approximate Bayesian inference: Dissertation. University of London, University, London UK.

  • Google Scholar

• 8

  BeckJ. L.YuenK.-V. (2004). Model selection using response measurements: Bayesian probabilistic approach. J. Eng. Mech.130 (2), 192–203. 10.1061/(asce)0733-9399(2004)130:2(192)

  • CrossRef
  • Google Scholar

• 9

  BehmaneshI.MoaveniB.LombaertG.PapadimitriouC. (2015). Hierarchical Bayesian model updating for structural identification. Mech. Syst. Signal Process64, 360–376. 10.1016/j.ymssp.2015.03.026

  • CrossRef
  • Google Scholar

• 10

  BishopC. M.NasrabadiN. M. (2006). Pattern recognition and machine learning. New York, NY, USA: Springer.

  • Google Scholar

• 11

  Cesmd (2019). Center for engineering strong motion data, cesmd‐ A cooperative effort. https://strongmotioncenter.org.

  • Google Scholar

• 12

  ChingJ.BeckJ. L.PorterK. A. (2006). Bayesian state and parameter estimation of uncertain dynamical systems. Probabilistic Eng. Mech.21 (1), 81–96. 10.1016/j.probengmech.2005.08.003

  • CrossRef
  • Google Scholar

• 13

  DeisenrothM. P.FaisalA. A.OngC. S. (2020). Mathematics for machine learning. Cambridge University Press, Cambridge, UK.

  • Google Scholar

• 14

  DoeblingS. W.FarrarC. R.PrimeM. B.ShevitzD. W. (1996). Damage identification and health monitoring of structural and mechanical systems from changes in their vibration characteristics: A literature review. NM, USA: Los Alamos National Lab. 10.2172/249299

  • CrossRef
  • Google Scholar

• 15

  EbrahimianH.AstrozaR.ConteJ. P. (2015). Extended Kalman filter for material parameter estimation in nonlinear structural finite element models using direct differentiation method. Earthq. Eng. Struct. Dyn.44 (10), 1495–1522. 10.1002/eqe.2532

  • CrossRef
  • Google Scholar

• 16

  FilippouF. C.PopovE. P.BerteroV. V. (1983). Effects of bond deterioration on hysteretic behavior of reinforced concrete joints. EERC 83–19. Earthquake Engineering Research Center. Berkeley, CA, USA: Univ. of California.

  • Google Scholar

• 17

  FriswellM.MottersheadJ. E. (2013). Finite element model updating in structural dynamics. Springer Science and Business Media, Berlin, Germany.

  • Google Scholar

• 18

  GranströmK.OrgunerU. (2011). Properties and approximations of some matrix variate probability density functions. Linköping University Electronic Press, Linköping, Sweden.

  • Google Scholar

• 19

  HaykinS. (2004). Kalman filtering and neural networks. John Wiley and Sons, New York, NY, USA.

  • Google Scholar

• 20

  HuangY.YuJ.BeckJ. L.ZhuH.LiH. (2020). Novel sparseness-inducing dual Kalman filter and its application to tracking time-varying spatially-sparse structural stiffness changes and inputs. Comput. Methods Appl. Mech. Eng.372, 113411. 10.1016/j.cma.2020.113411

  • CrossRef
  • Google Scholar

• 21

  HuangY.ZhangY.LiN.ZhaoL. (2016). Design of sigma-point Kalman filter with recursive updated measurement. Int. J. Circuits, Syst. Signal Process.35 (5), 1767–1782. 10.1007/s00034-015-0137-y

  • CrossRef
  • Google Scholar

• 22

  HuangY.ZhangY.WuZ.LiN.ChambersJ. (2017). A novel adaptive Kalman filter with inaccurate process and measurement noise covariance matrices. IEEE T Autom. Contr63 (2), 594–601. 10.1109/TAC.2017.2730480

  • CrossRef
  • Google Scholar

• 23

  KolloT.Von RosenD. (2006). Advanced multivariate statistics with matrices. Springer Science and Business Media, Berlin, Germany.

  • Google Scholar

• 24

  KontoroupiT.SmythA. W. (2016). Online noise identification for joint state and parameter estimation of nonlinear systems. ASCE-ASME J. Risk Uncertain. Eng. Syst. A Civ. Eng.2 (3), B4015006. 10.1061/AJRUA6.0000839

  • CrossRef
  • Google Scholar

• 25

  LawK. J.StuartA. M. (2012). Evaluating data assimilation algorithms. Mon. Weather Rev.140 (11), 3757–3782. 10.1175/mwr-d-11-00257.1

  • CrossRef
  • Google Scholar

• 26

  McKennaF.FenvesG. L.ScottM. H. (2000). Open system for earthquake engineering simulation. Berkeley, CA, USA: University of California.

  • Google Scholar

• 27

  MehraR. (1972). Approaches to adaptive filtering. IEEE T Autom. Contr17 (5), 693–698. 10.1109/tac.1972.1100100

  • CrossRef
  • Google Scholar

• 28

  NabiyanM. S.EbrahimianH.MoaveniB.PapadimitriouC. (2022). Adaptive bayesian inference framework for joint model and noise identification. J. Eng. Mech.148 (3), 4021165. 10.1061/(ASCE)EM.1943-7889.0002084

  • CrossRef
  • Google Scholar

• 29

  NabiyanM. S.KhoshnoudianF.MoaveniB.EbrahimianH. (2020). Mechanics‐based model updating for identification and virtual sensing of an offshore wind turbine using sparse measurements. Struct. Contr. Health Monit.28, e2647. 10.1002/stc.2647

  • CrossRef
  • Google Scholar

• 30

  O'HaganA.ForsterJ. J. (2004). Kendall's advanced theory of statistics, 2. 2, Arnold, London, UK.

  • Google Scholar

• 31

  OpperM.SaadD. (2001). Advanced mean field methods: Theory and practice. MIT Press, Cambridge, MA, USA.

  • Google Scholar

• 32

  PaolellaM. S. (2018). Linear models and time-series analysis: Regression, ANOVA, ARMA and GARCH. John Wiley and Sons, New York, NY, USA.

  • Google Scholar

• 33

  SongM.BehmaneshI.MoaveniB.PapadimitriouC. (2020). Accounting for modeling errors and inherent structural variability through a hierarchical bayesian model updating approach: An overview. Sensors20 (14), 3874. 10.3390/s20143874

  • CrossRef
  • Google Scholar

• 34

  SanayeiM.AryaB.SantiniE. M.Wadia‐FascettiS. (2001). Significance of modeling error in structural parameter estimation. COMPUT-AIDED Civ. Inf.16 (1), 12–27. 10.1111/0885-9507.00210

  • CrossRef
  • Google Scholar

• 35

  SarkkaS.HartikainenJ. (September 2013). Non-linear noise adaptive Kalman filtering via variational Bayes, Proceedings of the 2013 IEEE international workshop on machine learning for signal processing (MLSP), (IEEE), Southampton, UK.

  • Google Scholar

• 36

  SarkkaS.NummenmaaA. (2009). Recursive noise adaptive Kalman filtering by variational Bayesian approximations. IEEE T Autom. Contr54 (3), 596–600. 10.1109/TAC.2008.2008348

  • CrossRef
  • Google Scholar

• 37

  ShiY.TangX.FengX.BianD.ZhouX. (2018). Hybrid adaptive cubature Kalman filter with unknown variance of measurement noise. Sensors18 (12), 4335. 10.3390/s18124335

  • CrossRef
  • Google Scholar

• 38

  ŠmídlV.QuinnA. (2006). The variational Bayes method in signal processing. Springer Science and Business Media, Berlin, Germany.

  • Google Scholar

• 39

  SongM.AstrozaR.EbrahimianH.MoaveniB.PapadimitriouC. (2020). Adaptive Kalman filters for nonlinear finite element model updating. Mech. Syst. Signal Process143, 106837. 10.1016/j.ymssp.2020.106837

  • CrossRef
  • Google Scholar

• 40

  SunC.ZhangY.WangG.GaoW. (2018). A new variational Bayesian adaptive extended Kalman filter for cooperative navigation. Sensors18 (8), 2538. 10.3390/s18082538

  • CrossRef
  • Google Scholar

• 41

  SunJ.ZhouJ.GuX. (2012). Variational Bayesian two-stage Kalman filter for systems with unknown inputs. Procedia Eng.29, 2265–2273. 10.1016/j.proeng.2012.01.299

  • CrossRef
  • Google Scholar

• 42

  TzikasD. G.LikasA. C.GalatsanosN. P. (2008). The variational approximation for Bayesian inference. IEEE Signal Process. Mag.25 (6), 131–146. 10.1109/MSP.2008.929620

  • CrossRef
  • Google Scholar

• 43

  WeinstockR. (1974). Calculus of variations: With applications to physics and engineering. Dover Publications, New York.

  • Google Scholar

• 44

  WenzelT.BurnhamK.BlundellM.WilliamsR. (2007). Kalman filter as a virtual sensor: Applied to automotive stability systems. Trans. Inst. Meas. Control29 (2), 95–115. 10.1177/0142331207072990

  • CrossRef
  • Google Scholar

• 45

  WuJ. (2005). Some properties of the normal distribution. https://cs.nju.edu.cn/_upload/tpl/00/ed/237/template237/paper/Gaussian.pdf.

  • Google Scholar

• 46

  WuM.SmythA. W. (2007). Application of the unscented Kalman filter for real‐time nonlinear structural system identification. Struct. Contr. Health Monit.14 (7), 971–990. 10.1002/stc.186

  • CrossRef
  • Google Scholar

• 47

  XuD.WuZ.HuangY. (2019). A new adaptive Kalman filter with inaccurate noise statistics. Int. J. Circuits, Syst. Signal Process.38 (9), 4380–4404. 10.1007/s00034-019-01053-w

  • CrossRef
  • Google Scholar

• 48

  YangJ. N.LinS.HuangH.ZhouL. (2006). An adaptive extended Kalman filter for structural damage identification. Struct. Contr. Health Monit.13 (4), 849–867. 10.1002/stc.84

  • CrossRef
  • Google Scholar

• 49

  YuenK. V.KuokS. C. (2016). Online updating and uncertainty quantification using nonstationary output-only measurement. Mech. Syst. Signal Process66, 62–77. 10.1016/j.ymssp.2015.05.019

  • CrossRef
  • Google Scholar

• 50

  ZhangQ.YangY.XiangQ.HeQ.ZhouZ.YaoY. (2018). Noise adaptive Kalman filter for joint polarization tracking and channel equalization using cascaded covariance matching. IEEE Photonics J.10 (1), 1–11. 10.1109/JPHOT.2018.2797050

  • CrossRef
  • Google Scholar

• 51

  ZhengB.FuP.LiB.YuanX. (2018). A robust adaptive unscented Kalman filter for nonlinear estimation with uncertain noise covariance. Sensors18 (3), 808. 10.3390/s18030808

  • CrossRef
  • Google Scholar