• Smart Structures and Systems
• /
• 제34권2호
• /
• pp.97-116
• /
• 2024

The standard l₁-norm regularization is recently introduced for impact force identification, but generally underestimates the peak force. Compared to l₁-norm regularization, l_p-norm (0 ≤ p < 1) regularization, with a nonconvex penalty function, has some promising properties such as enforcing sparsity. In the framework of sparse regularization, if the desired solution is sparse in the time domain or other domains, the under-determined problem with fewer measurements than candidate excitations may obtain the unique solution, i.e., the sparsest solution. Considering the joint sparse structure of impact force in temporal and spatial domains, we propose a general l_p-norm (0 ≤ p < 1) regularization methodology for simultaneous identification of the impact location and force time-history from highly incomplete measurements. Firstly, a nonconvex optimization model based on l_p-norm penalty is developed for regularizing the highly under-determined problem of impact force identification. Secondly, an iteratively reweighed l₁-norm algorithm is introduced to solve such an under-determined and unconditioned regularization model through transforming it into a series of l₁-norm regularization problems. Finally, numerical simulation and experimental validation including single-source and two-source cases of impact force identification are conducted on plate structures to evaluate the performance of l_p-norm (0 ≤ p < 1) regularization. Both numerical and experimental results demonstrate that the proposed l_p-norm regularization method, merely using a single accelerometer, can locate the actual impacts from nine fixed candidate sources and simultaneously reconstruct the impact force time-history; compared to the state-of-the-art l₁-norm regularization, l_p-norm (0 ≤ p < 1) regularization procures sufficiently sparse and more accurate estimates; although the peak relative error of the identified impact force using l_p-norm regularization has a decreasing tendency as p is approaching 0, the results of l_p-norm regularization with 0 ≤ p ≤ 1/2 have no significant differences.

• 전기전자학회논문지
• /
• 제20권3호
• /
• pp.226-234
• /
• 2016

전기 저항 단층촬영법(ERT)은 대상체 내부 단면의 저항률 분포를 추정하고 이를 영상화하는 기술이다. ERT의 영상복원은 매우 비정치성이 강한 역문제의 일종으로 의미있는 영상을 얻기 위해서는 조정기법이 사용된다. 대표적으로 l2-norm 조정기법, l1-norm 조정기법, Total Variation 조정기법이 사용되며, 조정기법에 따라 ERT의 영상복원 성능이 달라진다. 즉, 상황에 맞는 적절한 조정기법의 사용은 ERT 영상 복원을 개선할 수 있다. 따라서, 본 논문에서는 모의실험을 통하여 상황에 따른 세 가지 조정기법의 영상복원 성능을 비교하였다.

• 말소리와 음성과학
• /
• 제7권3호
• /
• pp.131-138
• /
• 2015

In this paper, we propose L1-norm regularization for state vector adaptation of subspace Gaussian mixture model (SGMM). When you design a speaker adaptation system with GMM-HMM acoustic model, MAP is the most typical technique to be considered. However, in MAP adaptation procedure, large number of parameters should be updated simultaneously. We can adopt sparse adaptation such as L1-norm regularization or sparse MAP to cope with that, but the performance of sparse adaptation is not good as MAP adaptation. However, SGMM does not suffer a lot from sparse adaptation as GMM-HMM because each Gaussian mean vector in SGMM is defined as a weighted sum of basis vectors, which is much robust to the fluctuation of parameters. Since there are only a few adaptation techniques appropriate for SGMM, our proposed method could be powerful especially when the number of adaptation data is limited. Experimental results show that error reduction rate of the proposed method is better than the result of MAP adaptation of SGMM, even with small adaptation data.

• 전기전자학회논문지
• /
• 제20권2호
• /
• pp.152-162
• /
• 2016

전기 저항률 단층촬영법(ERT)은 표면 전극으로부터 측정된 전압을 사용하여 물체 내부의 임피던스 분포를 영상화하는 기술이다. ERT 역문제는 비정치성(ill-posedness)이 매우 심하여 영상복원의 수렴성을 확보하기 위해 조정방법이 사용된다. 사용된 조정방법에 따라 영상복원 성능이 달라지므로 상황에 따라 보다 강건한 영상 복원 성능을 얻기 위해, 서로 다른 영상복원 특성을 나타내는 L1-norm 조정방법과 Total Variation (TV) 조정방법의 두 개의 모드가 상호작용하는 상호작용 이중-모드 조정방법을 제안하였다. 제안한 이중-모드 조정방법은 실제 상황에 따라 달라지는 모드 확률을 계산하고 이에 근거하여 적합한 모드를 선택하거나 두 개의 모드를 결합한다. 모의실험을 수행하여 제안된 기법의 영상 복원 성능을 평가한 결과 비교적 양호한 성능을 나타내었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권7호
• /
• pp.3194-3216
• /
• 2018

Slow Feature Discriminant Analysis (SFDA) is a supervised feature extraction method inspired by biological mechanism. In this paper, a novel method called Two Dimensional Slow Feature Discriminant Analysis via $L_{2,1}$ norm minimization ($2DSFDA-L_{2,1}$) is proposed. $2DSFDA-L_{2,1}$ integrates $L_{2,1}$ norm regularization and 2D statically uncorrelated constraint to extract discriminant feature. First, $L_{2,1}$ norm regularization can promote the projection matrix row-sparsity, which makes the feature selection and subspace learning simultaneously. Second, uncorrelated features of minimum redundancy are effective for classification. We define 2D statistically uncorrelated model that each row (or column) are independent. Third, we provide a feasible solution by transforming the proposed $L_{2,1}$ nonlinear model into a linear regression type. Additionally, $2DSFDA-L_{2,1}$ is extended to a bilateral projection version called $BSFDA-L_{2,1}$. The advantage of $BSFDA-L_{2,1}$ is that an image can be represented with much less coefficients. Experimental results on three face databases demonstrate that the proposed $2DSFDA-L_{2,1}/BSFDA-L_{2,1}$ can obtain competitive performance.

• 산업공학
• /
• 제24권2호
• /
• pp.151-155
• /
• 2011

In this study, we propose a signomial classification method with ₀-regularization (₀-)which seeks a sparse signomial function by solving a mixed-integer program to minimize the weighted sum of the ₀-norm of the coefficient vector of the resulting function and the $L_1$-norm of loss caused by the function. $SC_0$ gives an explicit description of the resulting function with a small number of terms in the original input space, which can be used for prediction purposes as well as interpretation purposes. We present a practical implementation of $SC_0$ based on the mixed-integer programming and the column generation procedure previously proposed for the signomial classification method with $SL_1$-regularization. Computational study shows that $SC_0$ gives competitive performance compared to other widely used learning methods for classification.

• 한국전자통신학회논문지
• /
• 제17권1호
• /
• pp.41-48
• /
• 2022

본 논문은 통합 베이즈 티코노프 정규화 방법을 총변이 정규화에 대한 해법으로 제시한다. 통합된 방법은 총변이 항을 가중된 티코노프 정규화 항으로 변형하여 정규화 모수를 구하는 공식을 제시한다. 정규화 모수를 구하고 이를 바탕으로 새로운 가중인수를 구하는 것을 복원된 영상이 수렴하기까지 반복한다. 실험결과는 영상 복원 문제에 대하여 제안하는 방법의 효능을 보여준다.

• 전기전자학회논문지
• /
• 제19권1호
• /
• pp.33-40
• /
• 2015

전기 임피던스 단층촬영법(EIT)에서 역문제는 매우 높은 비정치성이므로 이것을 완화시키기 위해서 사전정보가 사용되고 EIT 역문제를 푸는 과정에서 만족스러운 복원성능을 갖기 위해 조정 기법은 적용된다. 반복적 Gauss-Newton 방법은 정확성과 빠른 수렴속도로 인해서 일반적으로 역문제를 푸는데 사용되지만 항상 좋은 성능을 내는 것은 아니며 조정 인자 선택에 따라 성능이 좌지우지된다. 비록 L-곡선과 같이 조정 인자를 결정하는데 이용할 수 있는 여러 가지 방법들이 존재하지만 이러한 방법들이 모든 경우에 적용할 수 있는 것은 아니다. 게다가 조정 인자는 스칼라이고 반복 연산동안 변하지 않는다. 그러므로 이 논문에서는 복원 성능을 향상시키기 위해서 조정 인자를 결정해주는 새로운 방법을 사용하였다. 각각의 반복 연산과정에서 도전율의 norm을 구하고 이것을 대각 행렬형태인 조정 인자를 구하는데 사용한다. 제안한 방법을 인체 흉부 영상 복원에 적용하였고, 기존의 방법들과 복원 성능을 비교하였다. 모의실험 결과, 기존의 방법들과 비교해서 개선된 성능을 확인할 수 있었다.

• Structural Engineering and Mechanics
• /
• 제85권1호
• /
• pp.119-133
• /
• 2023

Impact event is the key factor influencing the operational state of the mechanical equipment. Additionally, nonlinear factors existing in the complex mechanical equipment which are currently attracting more and more attention. Therefore, this paper proposes a novel hybrid-separate identification strategy to solve the force identification problem of the nonlinear structure under impact excitation. The 'hybrid' means that the identification strategy contains both l1-norm (sparse) and l2-norm regularization methods. The 'separate' means that the nonlinear response part only generated by nonlinear force needs to be separated from measured response. First, the state-of-the-art two-step iterative shrinkage/thresholding (TwIST) algorithm and sparse representation with the cubic B-spline function are developed to solve established normalized sparse regularization model to identify the accurate impact force and accurate peak value of the nonlinear force. Then, the identified impact force is substituted into the nonlinear response separation equation to obtain the nonlinear response part. Finally, a reduced transfer equation is established and solved by the classical Tikhonove regularization method to obtain the wave profile (variation trend) of the nonlinear force. Numerical and experimental identification results demonstrate that the novel hybrid-separate strategy can accurately and efficiently obtain the nonlinear force and impact force for the nonlinear structure.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제6권2호
• /
• pp.9-24
• /
• 2002

Total Variation(TV) regularization method is effective for reconstructing "blocky", discontinuous images from contaminated image with noise. But TV is represented by highly nonlinear integro-differential equation that is hard to solve. There have been much effort to obtain stable and fast methods. C. Vogel introduced "the Fixed Point Lagged Diffusivity Iteration", which solves the nonlinear equation by linearizing. In this paper, we apply multigrid(MG) method for cell centered finite difference (CCFD) to solve system arise at each step of this fixed point iteration. In numerical simulation, we test various images varying noises and regularization parameter $\alpha$ and smoothness $\beta$ which appear in TV method. Numerical tests show that the parameter ${\beta}$ does not affect the solution if it is sufficiently small. We compute optimal $\alpha$ that minimizes the error with respect to $L^2$ norm and $H^1$ norm and compare reconstructed images.