作者：
Liu, JieNanjing Vocat Univ Ind Technol
Jiangsu Prov Precis Mfg Engn andTechnol Res Ctr Sch Mech Engn 1 Yangshan North Rd Nanjing 210023 Jiangsu Peoples R China
To discuss the low convergence accuracy of whale optimization algorithm (WOA) and the problem of converging to local optima, we proposed using nonlinear convergence factors and introducing nonlinear inertia weights in...
详细信息
To discuss the low convergence accuracy of whale optimization algorithm (WOA) and the problem of converging to local optima, we proposed using nonlinear convergence factors and introducing nonlinear inertia weights in the WOA. The modified WOA was used to optimize the trajectory of a six-degrees-of-freedom (6DOF) industrial robot. To improve the convergence accuracy and the local and global search ability of the WOA, we first replaced the convergence factor with a nonlinear convergence factor and added a nonlinear inertia weight. The algorithm was used along with a quintic polynomial equation to develop a time-optimal trajectory, for the robot, for use in practical application scenarios. Simulation experiment results showed that the duration of a complete loading-unloading process was reduced by 30% after robot motion trajectory optimization compared with that before optimization, indicating the effectiveness of the improved WOA and its suitability for robot trajectory optimization.
In this work, we investigate a signal estimation problem which is common and critical for durability design of vehicle bodies. The relation between the frequency responses of accelerometers is the target to model so t...
详细信息
In this work, we investigate a signal estimation problem which is common and critical for durability design of vehicle bodies. The relation between the frequency responses of accelerometers is the target to model so that the ones of easy-to-measure accelerometers can estimate the responses of hard-to-measure accelerometers. A piecewise linear frequency-domain identification method relying on finite impulse response (FIR) models is proposed and performed to tackle the nonlinearity issue in the signal estimation problems: first, the interesting frequency range is segmented into three subranges which are clearly identified by peak histograms of frequency signals. Then, FIR models which provide a satisfactory description of the system are constructed to estimate the frequency responses of the interesting signals at subranges, one for each. The performance of the proposed approach is validated by using real-world data under multiple working conditions. The results show that the proposed method has a good estimation accuracy, and it brings the benefit that the number of accelerometers can be significantly reduced during the durability design of vehicle bodies.
Sensor selection is one of the key factors that dictate the performance of estimating vertical wheel forces in vehicle durability design. To select K most relevant sensors among S candidate ones that best fit the resp...
详细信息
Sensor selection is one of the key factors that dictate the performance of estimating vertical wheel forces in vehicle durability design. To select K most relevant sensors among S candidate ones that best fit the response of one vertical wheel force, it has (SK) possible choices to evaluate, which is not practical unless K or S is small. In order to tackle this issue, this paper proposes a data-driven method based on maximizing the marginal likelihood of the data of the vertical wheel force without knowing the dynamics of vehicle systems. Although the resulting optimization problem is a mixed-integer programming problem, it is relaxed to a convex problem with continuous variables and linear constraints. The proposed sensor selection method is flexible and easy to implement, and the hyper-parameters do not need to be tuned using additional validation data sets. The feasibility and effectiveness of the proposed method are verified using numerical examples and experimental data. In the results of different data sizes and model orders, the proposed method has better fitting performance than that of the group lasso method in the sense of the 2-norm based metric. Also, the computational time of the proposed method is much less than that of the enumeration-based method.
The nonparametric probabilistic method (NPM) for modeling and quantifying model-form uncertainties is a physics-based, computationally tractable, machine learning method for performing uncertainty quantification and m...
详细信息
The nonparametric probabilistic method (NPM) for modeling and quantifying model-form uncertainties is a physics-based, computationally tractable, machine learning method for performing uncertainty quantification and model updating. It extracts from data information not captured by a deterministic, high-dimensional model (HDM) of dimension N and infuses it into a counterpart stochastic, hyperreduced, projection-based reduced-order model (SHPROM) of dimension n << N. Here, the robustness and performance of NPM are improved using a two-pronged approach. First, the sensitivities of its stochastic loss function with respect to the hyperparameters are computed analytically, by tracking the complex web of operations underlying the construction of that function. Next, the theoretical number of hyperparameters is reduced from O(n(2)) to O(n), by developing a network of autoencoders that provides a nonlinear approximation of the dependence of the SHPROM on the hyperparameters. The robustness and performance of the enhanced NPM are demonstrated using two nonlinear, realistic, structural dynamics applications.
Process uncertainty can have negative effects on part quality and is, therefore, critical to the safety and performance of products. Those effects are manifested in the dimensional measurement uncertainty associated w...
详细信息
Process uncertainty can have negative effects on part quality and is, therefore, critical to the safety and performance of products. Those effects are manifested in the dimensional measurement uncertainty associated with those parts and products. To minimize the effects of process uncertainties, the sources of dimensional uncertainty must be identified and clearly communicated to collaborators and suppliers. A principal source of dimensional uncertainty is the measurement equipment itself. This article presents an activity model, rule types, and sample rules for selecting dimensional metrology equipment. The activity model represents key operations and information flows associated with the dimensional measurement. Analysis of the included activity model facilitates the development of rule types for measurement equipment selection as described in the Quality Information Framework (QIF) standard. Rule types are based on design information and measurement requirements. Standard rule types enable industrial metrologists to capture, exchange, and share equipment selection rules with their collaborators. Example QIF rules are defined for successful and cost-saving use in planning a measurement process with functionally complex and appropriate dimensional measurement equipment.
Wireframes have been proved useful as an intermediate layer of the neural network to learn the relationship between the human body and semantic parameters. However, the definition of the wireframe needs to have anthro...
详细信息
Wireframes have been proved useful as an intermediate layer of the neural network to learn the relationship between the human body and semantic parameters. However, the definition of the wireframe needs to have anthropological meaning and is highly dependent on experts' experience. Hence, it is usually not easy to obtain a well-defined wireframe for a new set of shapes in available databases. An automated wireframe generation method would help relieve the need for the manual anthropometric definition to overcome such difficulty. One way to find such an automated wireframe generation method is to apply segmentation to divide the models into small mesh patches. Nevertheless, different segmentation approaches could have various segmented patches, thus resulting in diversified wireframes. How do these different sets of wireframes affect learning performance? In this paper, we attempt to answer this research question by defining several critical quantitative estimators to evaluate different wireframes' learning performance. To find how such estimators influence wireframe-assisted learning accuracy, we conduct experiments by comparing different segmentation methods on human body shapes. We summarized several meaningful design guidelines for developing an automatic wireframe-aware segmentation method for human body learning with such verification.
Process control in manufacturing industries usually lacks flexibility and adaptability. The process planning is traditionally pursued within the production scheduling and then remains unchanged until a major overhaul ...
详细信息
Process control in manufacturing industries usually lacks flexibility and adaptability. The process planning is traditionally pursued within the production scheduling and then remains unchanged until a major overhaul is necessary. Consequently, no process knowledge acquired by the machine operators is used to adapt, and thus improve, the process control. In this paper, a fully automated process control solution for container logistics is proposed, which is based on deep neural networks and has been trained from process steering decisions made by employees. Further, a fully automated framework for the labeling of container images is introduced, making use of inherent properties of the logistic process. This allows to automatically generate data sets without the need for manual labeling by an operator.
暂无评论