Search

The production requirements of multi-variety and high-quality products pose significant challenges for implementing intelligent production lines that integrate process sensing, adaptive processing, and precise decision-making. Digital twin is the key enabling technology for intelligent production lines, but it lacks the perception of multidimensional process information to enable adaptive production. Thus, this paper proposes the adaptive digital twin motivated by a knowledge model with process information interactions. First, the process information related to product quality is located, and the process information is digitally modeled to implement the virtual-real mapping. Second, multidimensional feature extraction and fusion of process information digital model are performed to form adaptive adjustment capabilities for multi-variety production. Finally, the formal modeling of the adaptive adjustment capability enables the construction of the knowledge model that drives the adaptive digital twin. In adaptive digital twin, the virtual and real mapping of process information facilitates the knowledge model to continuously update the domain knowledge, which enables the adaptive decision-making ability to dynamically adapt to different processing requirements in the physical model. As a fusion technique developed in intelligent production lines, the modeling of adaptive digital twin has been well applied and validated in commutator finishing lines. [ABSTRACT FROM AUTHOR]

The digital twin (DT) for application development in specific fields or specific factors has emerged as a foundational framework for intelligent manufacturing. The increasing demand for customized and diverse production poses enormous challenges for the effective evaluation of production line processing capability (PLPC). Therefore, this paper presents a multidimensional information model-driven digital twin specifically designed for PLPC evaluation. The information model describes the characteristics of PLPC in the structural, physical, and kinematic dimensions, and it provides multidimensional information for constructing a comprehensive evaluation model in combination with the fuzzy Delphi method. An information model-driven digital twin enables more efficient extraction of processing capabilities relevant to product manufacturing and provides a digital and visual assessment of PLPC. In the case of the commutator processing industry, this paper highlights the feasibility of the digital twin model and its potential impact on forward-looking guidance for product process planning. [ABSTRACT FROM AUTHOR]

Chatter is a common state in milling and turning, which will reduce the machining quality of parts. For taking adequate measures to avoid chatter, chatter detection is necessary. However, chatter recognition still has many difficulties for both turning and milling. Due to the material removal effect, the chatter frequency will change. The existing frequency band extraction methods, such as ensemble empirical mode decomposition (EEMD) and wavelet packet transform, can not fully reflect the chatter information. As a result, a novel online chatter recognition method is proposed based on spectrum characteristics, which can be applied to the chatter recognition for both turning and milling under the combination of multiple cutting parameters black simultaneously. Firstly, we carried out the hammering experiment to obtain the frequency response function and modal frequency of the machining system before and after machining. The signal distribution characteristics of turning and milling signals are analyzed. Four features sensitive to chatter are extracted, including spectral standard deviation, frequency spectral expectation (FSE), spectral skewness, and relative power spectral entropy (RPE). Taking these features as inputs, a chatter recognition model is established based on an extreme gradient boosting (XGBoost) algorithm. Finally, we conducted the turning and milling tests. The experimental results show that the proposed chatter recognition model can be effective under different cutting parameters for turning and milling. Besides, results also show the chatter recognition model has extraterritorial applicability. [ABSTRACT FROM AUTHOR]

The percent global glomerulosclerosis is a key factor in determining the outcome of renal transfer surgery. At present, the rate is typically computed by pathologists, which is labour intensive and nonstandardized. With the development of Deep Learning (DL), DL-based segmentation models can be used to better identify and segment normal and sclerosed glomeruli. Based on this, we can better quantify percent global glomerulosclerosis to reduce the discard rate of donor kidneys. We used 51 whole slide images (WSIs) from different institutions that are publicly available on the internet. However, the number of sclerosed glomeruli is much smaller than that of normal glomeruli in different WSIs, which can reduce the effectiveness of Deep Learning. For better sclerosed glomerular identification and segmentation performance, we modified and trained a GAN (generative adversarial network)-based image inpainting model to obtain more synthetic sclerosed glomeruli. Our proposed inpainting method achieved an average SSIM (Structural Similarity) of 0.8086 and an average PSNR (Peak Signal-to-Noise Ratio) of 22.8943 dB in the area of generated sclerosed glomeruli. We obtained sclerosed glomerular segmentation performance improvement by adding synthetic sclerosed glomerular images and achieved the best Dice of glomerular segmentation in different test sets based on the modified Unet model. [ABSTRACT FROM AUTHOR]

During a long-term structural health monitoring (SHM), large-scale machine tool structural response observed from various sensors show obvious big data characteristics. However, serious "data island" issues existing in traditional SHM inevitably limit the efficiencies of sensor data analysis. Besides, most existing identification methods for these diverse and complex signal data are manual methods, which are extremely time-consuming and inefficient. In this paper, a novel method based on knowledge graph (KG) was proposed to deal with the fine-grained domain knowledge modeling and multi-source sensor data integration problems in the field of machine tool SHM. With the help of KG-based querying and reasoning, it is more intelligent and convenient to retrieve and analyze sensor data. To verify the effectiveness of the proposed method, a long-term condition monitoring experiment was conducted on a CNC machine tool in an automobile factory. The dynamic properties of the machine tool structure were automatically identified based on KG, and a condition monitoring indicator based on the similarity of dynamic properties was applied to monitor the health condition of the machine tool. The final result showed the effectiveness of the KG for health monitoring of the machine tool structure. [ABSTRACT FROM AUTHOR]

Production system flexibility has become an important factor to meet the needs of diversified markets in an ever-changing and turbulent market environment. Thus, this paper addresses the problems of flexibility evaluation for decision making by evaluation method driven by digital twin, which models, predicts, and evaluates the flexibility of a given production system design scheme (PSDS). Based on the definition and characteristic analysis of flexibility in the production system design phase, and combined with data modelling (static function model (SFM) and dynamic prediction model (DPM)) and data analysis (AHP-based and simulation-based analysis), the proposed framework evaluates system flexibility from the static and dynamic levels, helps enterprises quickly obtain the system's comprehensive flexibility evaluation index, realizing the rapid screening of the production system design scheme (PSDS). The successful application of the evaluation framework in case enterprise has fully proved the feasibility and effectiveness of the method. [ABSTRACT FROM AUTHOR]

Tool wear prediction is of critical importance to maintain the desired part quality and improve productivity. Inspired by the successful application of deep learning in many condition monitoring tasks. In this article, a novel modeling framework is presented, which includes multiple stacked sparse autoencoders and a nonlinear regression function for tool wear prediction. Multiple stacked sparse autoencoders consists of two main structures. One model is designed with multidimensional stacked sparse autoencoders, which can learn more features from different feature domains in the raw vibration signal, and another single-dimensional stacked sparse autoencoders is used for feature fusion and deeper features learning. And a modified loss function is applied that improves the learning ability. In addition, due to the good properties of tool wear process in nonstationarity and complex nonlinear, a nonlinear regression function is utilized to enhance the progressive tool wear prediction tasks. A dataset from a real manufacturing process is used to evaluate the performance of the proposed modeling framework. Experimental results show that tool wear can be predicted accurately and stably by the proposed tool wear predictive model, which outperforms the already developed methods. [ABSTRACT FROM AUTHOR]

Chatter is a kind of self-excited vibration, which always has a negative impact on production efficiency. In this paper, a novel online chatter detection method for milling processes is proposed. In this method, firstly, the spindle revolution period component is extracted by angular synchronous averaging (ASA) the vibration signals generated in different cutting conditions. Then, the residual part related to chatter is calculated by subtracting the periodic component. Subsequently, the filtered signal is decomposed into a set of intrinsic mode functions (IMFs) using variational mode decomposition (VMD) to obtain chatter information. Finally, the multiscale permutation entropy (MPE) and multiscale power spectral entropy (MPSE) of the selected IMFs are calculated, and Laplacian score (LS) for feature selection is applied to select the optimal sensitive scale features with generalization. Online chatter detection based on selected sensitive scale features by splitting signal up into overlapping frames in milling process. The analysis results show that the proposed method can effectively detect the chatter under stable cutting conditions and variable cutting conditions. [ABSTRACT FROM AUTHOR]

A high-speed reciprocating operating mechanism with fast running and frequent stopping produces a certain vibration due to inertia. The inherent vibration of this mechanism becomes a key factor affecting positioning accuracy and time. However, the existing vibration identification method is suitable only for static or low-speed structure operations. Therefore, a method of identification for operational modal parameters based on strain response is proposed. This method combines the strain mode with the operational modal analysis method, both of which are verified under the static and operational states, respectively. First, the theoretical background of the identification method is derived. Next, the proposed approach is used to identify the displacement mode shapes and the strain mode shapes of the beam structure with modal parameters that are obtained by experimental modal analysis. These results show that the beam structure presents different displacement mode shapes and strain mode shapes under different constraints. Finally, the displacement mode shapes of the sorting arm are obtained by the method under operation. The recognition results are compared with the computational modal analysis, and the two mode shapes are found to be highly consistent, which verifies that the method is equally reliable during operation. This method expands the application scope of the classic experimental modal analysis and reduces the error because of the additional mass of the actual sensor. [ABSTRACT FROM AUTHOR]

In ultra-precision flying cutting (UPFC), a vibration is a unique dynamic response due to periodic intermittent excitation of cutting forces, which has a significant impact on surface generation. In UPFC, cutting forces act on the machine tool system intermittently and periodically, which may cause a self-excited vibration. This self-excited vibration will cause chatter, which maybe occurs during machining operations and even catastrophically degrades surface quality and limits productivity. This study has been motivated by the fact that chatter occurrence is generally in relation to the tool wear, which has seldom been investigated comprehensively in literature. In this paper, a chatter model for UPFC that can account for the effects of tool wear and process damping was presented. A number of experiments with different machining conditions were conducted to investigate the effect of the tool wear on the chatter stability. These experiments proved that the stability lobe is closely related to tool wear and with the tool wear increase, which increases the tendency for chatter in UPFC. Finally, it was found that increasing the tool clearance angle and reducing the spindle speed can improve the stability limit. [ABSTRACT FROM AUTHOR]

During the testing and sorting of LED chips, traditional methods do not exclude the polycrystalline and fragmentary LED chips from the normal chips well. The purpose of this paper is to propose a new algorithm to solve this problem. The algorithm consists of three steps. Firstly, present a simple but efficient image segmentation method to get blobs. Secondly, analyze the blobs to exclude abnormal blobs and predict the pose (position and orientation) of the potential object based on the pose of the minimum enclosing rectangle (MER) of each remained blob. Finally, according to the predicted poses, locate the LED chips precisely in the originally captured image based on gradient orientation features. Experiments show that the algorithm is not only robust to illumination variation but also can locate the LED chips and exclude the polycrystalline and fragmentary chips efficiently. [ABSTRACT FROM AUTHOR]

Heavy machine tool work under such high-load conditions that chatter vibrations are prone to occur, which significantly diminishes the machining efficiency and quality. Stability lobe diagrams are commonly used to select appropriate spindle speed and axial depth of cut to get rid of chatter and maximize the material removal rate. However, this needs precise identification of the dynamics of the entire machine tool structure, especially in the low-frequency range. Operational modal analysis has been the proven technique for estimating dynamic characteristics of machine tool structures in operation conditions. In this article, a complete methodology was presented for employing operational modal analysis for heavy machine tool in machining conditions. A random cutting exciting method originally presented by Minis is modified which generates pseudorandom impulse force to excite a heavy vertical lathe structure. And the excitation signal of random cutting force was modeled to analyze the effect of cutting parameters on energy and frequency band of the excitation. One operational modal analysis method, the pLSCF (referred to as PolyMAX) method, was employed to estimate modal parameters during machining. It was also observed in chatter tests that the operational modal analysis results are more accurate than the traditional impact test results in characterizing the dynamics of machine tool structure in machining. [ABSTRACT FROM AUTHOR]

The tool tip vibration measurement has a great significance for online surface topography prediction and active vibration control. Many measurement methods have been developed. However, most of these methods are difficult to be applied in actual machining process for the following problems: the chips and cutting fluid will cause damage to the sensor or distort its measurement results and the sensor must be installed again when replacing the cutting tool. In order to overcome the above problems, this paper proposes an indirect measurement method: the vibration of the tool tip is obtained from the vibration of the measurement point on the turret. The relationship between the tool tip vibration and the turret measurement point vibration is obtained through impact tests. The vibration of the measurement point on the turret is measured via an accelerometer, which is easy to install. The measurement point is so far away from the cutting area and chip evacuation zone that the accelerometer and its measurement results will not be affected by the chips and cutting fluid. The accuracy of the measurement method is checked by comparing the calculated tool tip vibration calculated from the vibration of the turret measurement point to the directly measured results, which is measured via an accelerometer installed below the tool tip. The comparison results show that the calculated tool tip vibration at main frequencies as well as low frequencies has a high accuracy. Moreover, the effectiveness of the measurement method is verified through machined surface profile. [ABSTRACT FROM AUTHOR]

The dynamics of the machine tool structure are significant in high-precision machining and are meaningful to be evaluated. This paper proposes a new approach based on the modal mass distribution matrix to evaluate the dynamics and identify the weak components of machine tool structure and then to investigate the various components' effects on the machining process in different frequency bands. First, the approach is developed and verified by a designed mass-spring model whose weak components are known. Then, the proposed approach is used to identify the weak components of a machine tool structure whose modal parameters are obtained by experimental modal analysis. The result shows that the same component presents different weakness levels at different modes and that the various components of the structure present different weakness levels at the same mode. Finally, a cutting experiment is then conducted to verify the identification result. The result meets the former weak components' identification result well. [ABSTRACT FROM AUTHOR]