Unknown parameter’s variance-covariance propagation and calculation in generalized nonlinear least squares problem

The unknown parameter’s variance-covariance propagation and calculation in the generalized nonlinear least squares remain to be studied now, which didn’t appear in the internal and external referencing documents. The unknown parameter’s vari- ance-covariance propagation formula, considering the two-power terms, was concluded used to evaluate the accuracy of unknown parameter estimators in the generalized nonlinear least squares problem. It is a new variance-covariance formula and opens up a new way to evaluate the accuracy when processing data which have the multi-source, multi-dimensional, multi-type, multi-time-state, different accuracy and nonlinearity.

A CLASS OF FACTORIZED QUASI-NEWTON METHODS FOR NONLINEAR LEAST SQUARES PROBLEMS

This paper gives a class of descent methods for nonlinear least squares solution. A class of updating formulae is obtained by using generalized inverse matrices. These formulae generate an approximation to the second part of the Hessian matrix of the objective function, and are updated in such a way that the resulting approximation to the whole Hessian matrix is the convex class of Broyden-like up-dating formulae. It is proved that the proposed updating formulae are invariant under linear transformation and that the class of factorized quasi-Newton methods are locally and superlinearly convergent. Numerical results are presented and show that the proposed methods are promising.

ON THE SEPARABLE NONLINEAR LEAST SQUARES PROBLEMS

Separable nonlinear least squares problems are a special class of nonlinear least squares problems, where the objective functions are linear and nonlinear on different parts of variables. Such problems have broad applications in practice. Most existing algorithms for this kind of problems are derived from the variable projection method proposed by Golub and Pereyra, which utilizes the separability under a separate framework. However, the methods based on variable projection strategy would be invalid if there exist some constraints to the variables, as the real problems always do, even if the constraint is simply the ball constraint. We present a new algorithm which is based on a special approximation to the Hessian by noticing the fact that certain terms of the Hessian can be derived from the gradient. Our method maintains all the advantages of variable projection based methods, and moreover it can be combined with trust region methods easily and can be applied to general constrained separable nonlinear problems. Convergence analysis of our method is presented and numerical results are also reported.

Solving method of generalized nonlinear dynamic least squares for data processing in building of digital mine

Data are very important to build the digital mine. Data come from many sources, have different types and temporal states. Relations between one class of data and the other one, or between data and unknown parameters are more nonlinear. The unknown parameters are non random or random, among which the random parameters often dynamically vary with time. Therefore it is not accurate and reliable to process the data in building the digital mine with the classical least squares method or the method of the common nonlinear least squares. So a generalized nonlinear dynamic least squares method to process data in building the digital mine is put forward. In the meantime, the corresponding mathematical model is also given. The generalized nonlinear least squares problem is more complex than the common nonlinear least squares problem and its solution is more difficultly obtained because the dimensions of data and parameters in the former are bigger. So a new solution model and the method are put forward to solve the generalized nonlinear dynamic least squares problem. In fact, the problem can be converted to two sub problems, each of which has a single variable. That is to say, a complex problem can be separated and then solved. So the dimension of unknown parameters can be reduced to its half, which simplifies the original high dimensional equations. The method lessens the calculating load and opens up a new way to process the data in building the digital mine, which have more sources, different types and more temporal states.

Selective ensemble modeling based on nonlinear frequency spectral feature extraction for predicting load parameter in ball mills

Strong mechanical vibration and acoustical signals of grinding process contain useful information related to load parameters in ball mills. It is a challenge to extract latent features and construct soft sensor model with high dimensional frequency spectra of these signals. This paper aims to develop a selective ensemble modeling approach based on nonlinear latent frequency spectral feature extraction for accurate measurement of material to ball volume ratio. Latent features are first extracted from different vibrations and acoustic spectral segments by kernel partial least squares. Algorithms of bootstrap and least squares support vector machines are employed to produce candidate sub-models using these latent features as inputs. Ensemble sub-models are selected based on genetic algorithm optimization toolbox. Partial least squares regression is used to combine these sub-models to eliminate collinearity among their prediction outputs. Results indicate that the proposed modeling approach has better prediction performance than previous ones.

Damage Identification under Incomplete Mode Shape Data Using Optimization Technique Based on Generalized Flexibility Matrix

A generalized flexibility–based objective function utilized for structure damage identification is constructed for solving the constrained nonlinear least squares optimized problem. To begin with, the generalized flexibility matrix (GFM) proposed to solve the damage identification problem is recalled and a modal expansion method is introduced. Next, the objective function for iterative optimization process based on the GFM is formulated, and the Trust-Region algorithm is utilized to obtain the solution of the optimization problem for multiple damage cases. And then for computing the objective function gradient, the sensitivity analysis regarding design variables is derived. In addition, due to the spatial incompleteness, the influence of stiffness reduction and incomplete modal measurement data is discussed by means of two numerical examples with several damage cases. Finally, based on the computational results, it is evident that the presented approach provides good validity and reliability for the large and complicated engineering structures.

Overlapped peaks resolution for linear sweep polarography using Gaussian-like distribution

A resolution method based on Gaussian-like distribution for overlapped linear sweep polarographic peaks was proposed to simultaneously detect the polymetallic components, such as Zn（Ⅱ） and Co（Ⅱ）, coexisting in the leaching solution of zinc hydrometallurgy. A Gaussian-like distribution was constructed as the sub-model of overlapped peaks by analyzing the characteristics of linear sweep polarographic curve. Then, the abscissas of each peak and trough were pinpointed through multi-resolution wavelet decomposition, the curve and its derivative curves were fitted by using nonlinear weighted least squares （NWLS）. Finally, overlapped peaks were resolved into independent sub-peaks based on fitted reconstruction parameters. The experimental results show that the relative error of half-wave potential pinpointed by multi-resolution wavelet decomposition is less than 1% and the accuracy of Ip fitted by NWLS is higher than 96%. The proposed resolution method is effective for overlapped linear sweep polarographic peaks of Zn（Ⅱ） and Co（Ⅱ）.

Modeling the dynamic behavior of a lithium-ion battery for electric vehicles using numerical optimization

In order to simulate the dynamical behavior of a lithium ion traction battery used in elec tric vehicles, an equivalent circuit based battery model was established. The methodology in the guide document of the ADVISOR software was used to determine the initial parameters of the model as a function of state of charge （ SoC ） over an experimental data set of the battery. A numerically nonlinear least squares algorithm in SIMULINK design optimization toolbox was applied to further op timize the model parameters. Validation results showed that the battery model could well describe the dynamic behavior of the lithinm ion battery in two different battery loading situations.

Modelling height-diameter relationships in complex tropical rain forest ecosystems using deep learning algorithm

Modelling tree height-diameter relationships in complex tropical rain forest ecosystems remains a challenge because of characteristics of multi-species, multi-layers, and indeterminate age composition. Effective modelling of such complex systems required innovative techniques to improve prediction of tree heights for use for aboveground biomass estimations. Therefore, in this study, deep learning algorithm (DLA) models based on artificial intelligence were trained for predicting tree heights in a tropical rain forest of Nigeria. The data consisted of 1736 individual trees representing 116 species, and measured from 52 0.25 ha sample plots. A K-means clustering was used to classify the species into three groups based on height-diameter ratios. The DLA models were trained for each species-group in which diameter at beast height, quadratic mean diameter and number of trees per ha were used as input variables. Predictions by the DLA models were compared with those developed by nonlinear least squares (NLS) and nonlinear mixed-effects (NLME) using different evaluation statistics and equivalence test. In addition, the predicted heights by the models were used to estimate aboveground biomass. The results showed that the DLA models with 100 neurons in 6 hidden layers, 100 neurons in 9 hidden layers and 100 neurons in 7 hidden layers for groups 1, 2, and 3, respectively, outperformed the NLS and NLME models. The root mean square error for the DLA models ranged from 1.939 to 3.887 m. The results also showed that using height predicted by the DLA models for aboveground biomass estimation brought about more than 30% reduction in error relative to NLS and NLME. Consequently, minimal errors were created in aboveground biomass estimation compared to those of the classical methods.

An iterative algorithm of NWTLS-EC for three dimensional-datum transformation with large rotation angle

The Gauss-Markov （GM） model and the Errors-in-Variables （EIV） model are frequently used to perform 3D coordinate transformations in geodesy and engineering surveys. In these applications, because the observation errors in original coordinates system are also taken into account, the latter is more accurate and reasonable than the former. Although the Weighted Total Least Squares （WTLS） technique has been intro- duced into coordinate transformations as the measured points are heteroscedastic and correlated, the Variance- Covariance Matrix （VCM） of observations is restricted by a particular structure, namely, only the correlations of each points are taken into account. Because the 3D datum transformation with large rotation angle is a non- linear problem, the WTLS is no longer suitable in this ease. In this contribution, we suggested the nonlinear WTLS adjustments with equality constraints （NWTLS-EC） for 3D datum transformation with large rotation an- gle, which removed the particular structure restriction on the VCM. The Least Squares adjustment with Equality （LSE） constraints is employed to solve NWTLS-EC as the nonlinear model has been linearized, and an iterative algorithm is proposed with the LSE solution. A simulation study of 3D datum transformation with large rotation angle is given to insight into the feasibility of our algorithm at last.

Kinematic calibration method with high measurement efficiency and robust identification for hybrid machine tools

Geometric error is the main factor affecting the machining accuracy of hybrid machine tools.Kinematic calibration is an effective way to improve the geometric accuracy of hybrid machine tools.The necessity to measure both position and orientation at each pose,as well as the instability of identification in case of incomplete measurements,severely affects the application of traditional calibration methods.In this study,a kinematic calibration method with high measurement efficiency and robust identification is proposed to improve the kinematic accuracy of a five-axis hybrid machine tool.First,the configuration is introduced,and an error model is derived.Further,by investigating the mechanism error characteristics,a measurement scheme that only requires tool centre point position error measurement and one alignment operation is proposed.Subsequently,by analysing the effects of unmeasured degrees of freedom(DOFs)on other DOFs,an improved nonlinear least squares method based on virtual measurement values is proposed to achieve stable parameter identification in case of incomplete measurement,without introducing additional parameters.Finally,the proposed calibration method is verified through simulations and experiments.The proposed method can efficiently accomplish the kinematic calibration of the hybrid machine tool.The accuracy of the hybrid machine tool is significantly improved after calibration,satisfying actual aerospace machining requirements.

Efficient Statistical Inference for Partially Nonlinear Errors-in-Variables Models

In this paper, we consider the partially nonlinear errors-in-variables models when the non- parametric component is measured with additive error. The profile nonlinear least squares estimator of unknown parameter and the estimator of nonparametric component are constructed, and their asymptotic properties are derived under general assumptions. Finite sample performances of the proposed statistical inference procedures are illustrated by Monte Carlo simulation studies.

Numerical partitioning of components for four-modal sedimentary grain-size distribution based on gradient descent method

The gradient descent(GD)method is used to fit the measured data(i.e.,the laser grain-size distribution of the sediments)with a sum of four weighted lognormal functions.The method is calibrated by a series of ideal numerical experiments.The numerical results indicate that the GD method not only is easy to operate but also could effectively optimize the parameters of the fitting function with the error decreasing steadily.The method is applied to numerical partitioning of laser grain-size components of a series of Garzêloess samples and three bottom sedimentary samples of submarine turbidity currents modeled in an open channel laboratory flume.The overall fitting results are satisfactory.As a new approach of data fitting,the GD method could also be adapted to solve other optimization problems.

Method for Visual Localization of Oil and Gas Wellhead Based on Distance Function of Projected Features

A localization method based on distance function of projected features is presented to solve the accuracy reduction or failure problem due to occlusion and blurring caused by smog, when dealing with vision based localization for target oil and gas wellhead （OGWH）. Firstly, the target OGWH is modeled as a cylinder with marker, and a vector with redundant parameter is used to describe its pose. Secondly, the explicit mapping relationship between the pose vector with redundant parameter and projected features is derived. Then, a 2D-point-to-feature distance function is proposed, as well as its derivative. Finally, based on this distance function and its derivative, an algorithm is proposed to estimate the pose of target OGWH directly according to the 2D image information, and the validity of the method is verified by both synthetic data and real image experiments. The results show that this method is able to accomplish the localization in the case of occlusion and blurring, and its anti-noise ability is good especially with noise ratio of less than 70%.