Search

Two-dimensional (2D) irregular layout is widely applied in various manufacturing processes, such as sheet metal cutting, shipbuilding, and electronic component placement. An efficient layout algorithm can effectively improve the material utilization, thereby reducing manufacturing cost. But the free-form shape layout problem is very challenge as it is difficult to exactly represent a free-form shape. There is not an efficient method currently available for the 2D free-form shape layout. This paper proposes a method based on the geometric similarity feature searching and fuzzy matching for the 2D free-form shape layout. The freeman chain code is developed to describe the contour information of shapes and forward-lines to form the basis of the layout strategy. A strategy based on fuzzy matching is proposed for the layout, which includes searching geometric similarity features using the longest common subsequence and the proposed placement algorithm to complete the collision. Three computational experiments are conducted to analyze the performance of the proposed method. Experimental results show that the proposed method is feasible and effective with the good applicability to achieve a high filling rate in reduced time.

A support vector machine (SVM) is an important method in the detection and classification of the color difference on a polysilicon wafer. However, the accuracy of a SVM is affected by its feature vector and parameters. Owing to the complex color information and random texture features on the wafer surface, the feature design is extremely complicated. Meanwhile, a SVM optimized using a popular intelligent algorithm easily falls into a local optimum, and the convergence of the algorithm needs to be improved. Therefore, a classification method is proposed for detecting the color difference from multi-scale features in polysilicon wafer images. First, to extract the features, an image segmentation method is devised based on the maximum region contrast, which effectively applies a threshold segmentation of the wafer images. Second, the multi-scale features and color representations in different color spaces are used to construct a nine-dimensional feature vector that sufficiently describes the surface characteristics of the wafer. An approach to optimize the SVM is finally proposed using a magnetic bacteria optimization algorithm based on an elitist strategy for parameter optimization. The optimum individual of each generation is used to adjust the magnetic moment such that the solution approaches the optimal direction and enhances the global search ability. A fitness function is also introduced to improve the diversity of the solutions through a cross-validation method. The experiment results show that the proposed algorithm achieves an accuracy of 98.3% with a better classification performance than the other methods and that the color difference of polysilicon wafers can be effectively detected.

Chatter is a self-excited vibration that frequently occurs in thin-walled parts milling, which has become a major limitation to productivity and quality. Additionally, convolutional neural network (CNN) has been widely used in detection and classification, but the accuracy and convergence are affected by the initial weight and hyperparameters. Therefore, based on CNN, a method of chatter detection for the milling of thin-walled parts is proposed, which is realized by recognizing the image of the machined surface. First, aiming at the challenges of neural networks in which the weight is randomly initialized, a weight initialization method is proposed based on an improved magnetic bacteria optimization algorithm. The optimal magnetosome of each generation is used to adjust the magnetic moment of the next generation so that the population approaches the optimal solution direction to search for the global optimal value. Second, an improved genetic algorithm is proposed to optimize the network structure and improve the optimization efficiency of hyperparameters. The tabu list and the hill climbing algorithm are introduced into the improved genetic algorithm to avoid repeatedly counting the fitness of the same point and solving the oscillation problem near the optimal solution. The experimental results show that the accuracy, the value of the Matthew correlation coefficient, and the F1 Score value of proposed CNN are 98.3%, 95.5%, and 98.8%, respectively. Compared with other algorithms, the proposed method, which has outstanding recognition performance, is competent in contactless chatter detection.

Dental restoration is a complex process to recover patients' teeth using artificial materials. It is important to improve the utilization of the expensive material for cost saving. The tooth side surface is projected as a two-dimensional (2D) contour in the actual processing, which forms a packing problem of irregular contours. The 2D irregular packing has received a lot of attention in research to solve this NP-hard problem. Contours of the dental restoration are more complicated than those in traditional 2D irregular packing problems. The existing packing methods cannot obtain good results for these complex contours. In order to solve the contour packing problem of restoration models using the NC machining technology, a dental restoration contour packing algorithm is proposed based on the contour similar feature matching and collision. The algorithm uses dynamic programming for the longest common subsequence to search the solution of 2D free contour packing. Comparing with the existing packing methods, our proposed method can significantly reduce the processing time and improve the filling rate.

As the main load-bearing structure of heavy machine tools, cranes, and other high-end equipment, the large-scale box structures usually bear moving loads, and the results of direct topology optimization usually have some problems: the load transfer skeleton is difficult to identify and all working conditions are difficult to consider comprehensively. In this paper, a layout design method of stiffened plates for the large-scale box structures under moving loads based on multiworking-condition topology optimization is proposed. Based on the equivalent principle of force, the box structures are simplified into the main bending functional section, main torsional functional section, and auxiliary functional section by the magnitude of loads and moments, which can reduce the structural dimension and complexity in topology optimization. Then, the moving loads are simplified to some multiple position loads, and the comprehensive evaluation function is constructed by the compromise programming method. The mathematical model of multiworking-condition topology optimization is established to optimize the functional sections. Taking a crossbeam of superheavy turning and milling machining center as an example, optimization results show that the stiffness and strength of the crossbeam are increased by 17.39% and 19.9%, respectively, while the weight is reduced by 12.57%. It shows that the method proposed in this paper has better practicability and effectiveness for large-scale box structures.

At present, most of the geomagnetic navigation methods are based on the single geomagnetic scalar characteristics, and the iterative closest contour point (ICCP) algorithm is the most extensively utilized. But when there are several contour lines with the same scalar value in the matching area, or the scalar feature in this area is not obvious, navigation accuracy will be seriously affected. In this letter, a new geomagnetic navigation method based on vector matching algorithm [vector ICCP (VICCP)] is proposed, combining the searching principle of trusted points sets and tracks with the matching principle of geomagnetic vector correlation restriction. Consequently, navigation results of it will have greater accuracy, more reliable validity, and practicability compared with the traditional ICCP algorithm. The performance of the matching and the correction methods is analyzed by simulation and experiment. In simulation, the position error of VICCP is less than ICCP under the conditions of nonobvious scalar geomagnetic features, which are, respectively, reduced from 1340.0 to 72.8 m, from 1267.7 to 33.3 m, and from 14115.7 to 36.9 m. And the conclusion is also verified in the experiment. In addition, VICCP algorithm is not sensitive to initial position. Thus, the proposed VICCP algorithm can effectively improve the performance of geomagnetic navigation.

Cutting force can be used in monitoring machine processes and for guiding production. Therefore, the measurement of cutting forces has significant meaning in research. In this paper, an innovative force measuring tool holder system was presented, which integrated strain sensors and wireless transmission devices as well as created real-time measure axial force and torque in milling and drilling processes. In order to improve the sensitivity of the detection system, an annular groove was cut on the standard commercial tool holder. The Sensors of axial force and torque were arranged in the middle of the annular groove to perceive the minimal deformation in a corresponding direction. The structure of the tool holder was designed by optimization in which the required stiffness is used as the constraint and the detection accuracy is the highest as a target. All of the sensors and other electronic components, such as the data acquisition and the transmitter module and the power module, were integrated with the tool holder as a whole system. The equipment could be clamped on the spindle hole of the milling or drilling machine. A static and dynamic calibration test platform was built, and the static calibration test platform determined the testing performance with a wide range. Cutting tests were performed on a floor type boring-milling machine. The results were consistent and had a high waveform degree of coincidence on the axial force and torque trend compared with the Kistler dynamometer. The results also accurately reflected the axial force and torque changes in dynamic milling and drilling.

The interaction of pressure and flow in a hydraulic system with multiple working conditions, multiple actuators, and large flow limits action adjustment and control. Through a pilot pressure control circuit, hydraulically operated valves can adjust pressure or direction more effectively. A recent study proposed a two-stage pressure control method based on multistage orifices and solenoid valves. The requirements of the number and diameter ratio of short orifices in the series to realize the two-stage pressure control were theoretically analyzed. Scheme design and experiment were carried out. The influence of structures of complex flow channel and solenoid valve on the higher or lower pilot control pressure was considered in the experiment. The method was experimentally verified and successfully applied in a turbine electrohydraulic control system with lower maintenance costs, making the system more reliable in the case of electrical failure. Research results provide insight into pilot pressure control in fluid systems using multistage orifices to achieve either higher or lower pressure. In addition, it has important guiding significance for the design of valves or engineering systems based on pilot hydraulic pressure.

The high-speed, heavy-load and changeable triangle track wheel is a motion device that can carry out interchange between the track wheel and tire in an ordinary vehicle. The topology optimization for the support frame can reduce weight and improve the maneuverability of the vehicle. However, it is difficult to consider simultaneously its weight, stiffness and modal in the process of the structure optimization. Thus, a topology optimization method for multi-objective and multi-working-condition is proposed based on the AHP (analytic hierarchy process) and average frequency method. Firstly, considering the static multi-stiffness target and dynamic vibration frequency target, using the compromise programming method and average frequency method, the objective function of the multi-objective and multi-working-condition topology optimization is established. Then, based on the optimization target, design criteria and indexes, the lightweight hierarchical structure model of the support frame consisting of three levels and eight weight factors is established. Values of 8 weight coefficients of the multi-objective topology optimization are determined through solving the weight factor judgment matrix. Finally, considering the multi-working-condition, taking the minimum objective function of the static and dynamic characteristics as target, and the volume ratio is 50% as boundary, the mathematical model of the topology optimization is established. Simulation results show that the stiffness and strength of the support frame are improved respectively by 74.3% and 1.3% while its weight is reduced by 16.3%. This method also provides a new way to the lightweight design for other large, heavy and multi-condition equipment.

The selection of biomimetic prototypes mostly depends on the subjective observation of a designer. This research uses TRIZ to explore some inferential steps in bionic design of the heavy machine tool column. Conflict resolution theory of TRIZ is applied to describe improved and deteriorated parameters and a contradiction matrix is used to obtain recommended inventive principles. A reference table of solutions corresponding to the biological phenomenon and TRIZ solutions is formed to expedite retrieving the biomimetic object. Based on the table, herbaceous hollow stem is selected to imitate column structure. Four kinds of plant are chosen from the biological database. To select the best from four candidates, a bionic ideality evaluation index is proposed based on similarity analysis and ideality evaluation theory in TRIZ. Thus, the bionic effect can be described and compared quantitatively. Bionic configuration is then evolved concerning manufacturing requirements. Size optimization of stiffener thicknesses is implemented finally, and satisfactory results of the lightweight effect is obtained.

As a special kind of advanced aerospace sheet metal forging equipment, more and more attention has been paid to the design and development of isothermal forming hydraulic press. Failure Modes, Effects and Criticality Analysis (FMECA) mathod is an effective technique to improve the reliability of products. To improve the reliability of isothermal forming hydraulic press, this paper propose FMECA method to predict all potential failure modes, causes and effects in the design stage. Then improvement and preventive measures are accordingly put forward to alleviate or eliminate failure effects. Finally, the reliability of hydraulic press body is improved using the FMECA mathod, which also provides a basis for reliability analysis and design of other systems.

Machine learning algorithms have been widely used to deal with a variety of practical problems such as computer vision and speech processing. But the performance of machine learning algorithms is primarily affected by their hyper-parameters, as without good hyper-parameter values the performance of these algorithms will be very poor. Unfortunately, for complex machine learning models like deep neural networks, it is very difficult to determine their hyper-parameters. Therefore, it is of great significance to develop an efficient algorithm for hyper-parameter automatic optimization. In this paper, a novel hyper-parameter optimization methodology is presented to combine the advantages of a Genetic Algorithm and Tabu Search to achieve the efficient search for hyper-parameters of learning algorithms. This method is defined as the Tabu_Genetic Algorithm. In order to verify the performance of the proposed algorithm, two sets of contrast experiments are conducted. The Tabu_Genetic Algorithm and other four methods are simultaneously used to search for good values of hyper-parameters of deep convolutional neural networks. Experimental results show that, compared to Random Search and Bayesian optimization methods, the proposed Tabu_Genetic Algorithm finds a better model in less time. Whether in a low-dimensional or high-dimensional space, the Tabu_Genetic Algorithm has better search capabilities as an effective method for finding the hyper-parameters of learning algorithms. The presented method in this paper provides a new solution for solving the hyper-parameters optimization problem of complex machine learning models, which will provide machine learning algorithms with better performance when solving practical problems.