Search

Weld defect recognition plays an important role in the manufacturing process of large-scale equipment. Traditional methods generally include several serial steps, such as image preprocessing, region segmentation, feature extraction, and type recognition. The results of each step have significant impact on the accuracy of the final defect identification. The convolutional neural network (CNN) has strong pattern recognition ability, which can overcome the above problem. However, there are two problems: one is that the pooling strategy has poor dynamic adaptability and the other is the insufficient feature selection ability. To overcome these problems, we propose a CNN-based weld defect recognition method, which includes an improved pooling strategy and an enhanced feature selection method. According to the characteristics of the weld defect image, an improved pooling strategy that considers the distribution of the pooling region and feature map is introduced. Additionally, in order to enhance the feature selection ability of the CNN, an enhanced feature selection method integrating the ReliefF algorithm with the CNN is proposed. A case study is presented for demonstrating the proposed techniques. The results show that the proposed method has higher accuracy than the traditional CNN method, and establish that the proposed CNN-based method is successfully applied for weld defect recognition.

Gas overruns and explosions often occur in coal mining. To deal with this problem, the spatial structure model of a new non-ventilation excavation system is constructed. Combined with the model, a simulation experiment is conducted and the variations in gas concentration in each area of the system with variations in the amount of gas emission, drainage, and nitrogen curtain outlet velocity are obtained. The effectiveness of the gas barrier by the combined action of the gas drainage and nitrogen curtain is verified. This study provides the experimental basis for solving the problem of gas overruns and explosions, and is of great significance for realizing safe, efficient, and rapid excavation.

Breast cancer is a major health threat for women. The drug responses associated with different breast cancer subtypes have obvious effects on therapeutic outcomes; therefore, the accurate classification of breast cancer subtypes is critical. Breast cancer subtype classification has recently been examined using various methods, and Raman spectroscopy has emerged as an effective technique that can be used for noninvasive breast cancer analysis. However, the accurate and rapid classification of breast cancer subtypes currently requires a great deal of effort and experience with the processing and analysis of Raman spectra data. Here, we adopted Raman spectroscopy and machine learning techniques to simplify and accelerate the process used to distinguish normal from breast cancer cells and classify breast cancer subtypes. Raman spectra were obtained from cultured breast cancer cell lines, and the data were analyzed by two machine learning algorithms: principal component analysis (PCA)-discriminant function analysis (DFA) and PCA-support vector machine (SVM). The accuracies with which these two algorithms were able to distinguish normal breast cells from breast cancer cells were both greater than 97%, and the accuracies of breast cancer subtype classification for both algorithms were both greater than 92%. Moreover, our results showed evidence to support the use of characteristic Raman spectral features as cancer cell biomarkers, such as the intensity of intrinsic Raman bands, which increased in cancer cells. Raman spectroscopy combined with machine learning techniques provides a rapid method for breast cancer analysis able to reveal differences in intracellular compositions and molecular structures among subtypes.

In order to effectively monitor the battery usage status of electric vehicle, this paper designs a high-precision voltage monitoring circuit applied to multi-series battery packs, which based on the 0.18μm, 40V high voltage BCD process. According to the characteristics of the voltage superposition and battery distinction in the series battery pack, this circuit uses the voltage conversion circuit with the matched negative feedback technology to achieve effective “scaling” of the battery's absolute high voltage to ground-based analog signals. Under the typical/slow/fast process corners, -40°C~85°C and1.5V~31.5V, the simulation results show that the monitoring error of single cell voltage is 1.92mV, and the relative error rate is less than 0.043%, showing out of the strong robustness.

In order to solve the influence of the input stage transconductance of the rail-to-rail op amp with the common-mode input voltage change, real-time common-mode voltage monitoring technology is used to dynamically track the transconductance change of the rail- to- rail op amp input stage through the bias current. High-precision quantitative compensation for constant control of the input stage transconductance. The design and implementation of the specific circuit based on 0.18µm and CMOS process show that the op amp gain is 148dB and the phase margin is 61° under the condition of power supply voltage 3.3V, load resistance 100Ω and load capacitance 1nF. The total power consumption is only 39.6 microwatts, the common mode input range is as high as 0-3.3V, and the transconductance change rate of the input stage is only 2.1 %.

Recently, a new kind of distributed element superconducting resonators with granular aluminum (grAl) has been developed for circuit quantum electrodynamics. Given that lumped element resonators possess certain advantages over the distributed element ones, in this paper, we use a relatively simple micro-nanotechnology, without the etching process, to fabricate the designed lumped element superconducting grAl resonators. They are deposited on silicon substrates by DC magnetron sputtering of pure Al in an atmosphere of Ar and O2. By cryogenic microwave transmission measurements, at a temperature of 50 mK, we find that, compared with a usual Al resonator of the same size, such a device shows significantly strong nonlinearity and thus higher kinetic inductance. The noise features, quality factors, and temperature-dependence of the resonant frequencies have been investigated experimentally. It is believed that the lumped element superconducting resonators with high kinetic inductance demonstrated here could also be applied to superconducting quantum computation and photon counting in the future.

Objective To assess the efficacy and safety of Reduqing granules in patients with common cold with wind-heat syndrome (CCWHS). Methods A randomized, double-blind, double-dummy, parallel, positive-controlled trial included 72 CCWHS patients was performed. The participants were randomly assigned to two groups, Reduqing (RDQ) group and Lianhuaqingwen (LHQW) group, in a 1:1 ratio. Patients in RDQ group received Reduqing granules and dummy Lianhuaqingwen capsules three times a day and patients in LHQW group received Lianhuaqingwen capsules and dummy Reduqing granules three times daily. The duration of treatment and follow-up were four days. Results There were no statistically significant differences in total markedly effective rate and total effective rate between RDQ group and LHQW group after treatment. Traditional Chinese Medicine (TCM) symptom score was significantly reduced after treatment in RDQ group, as well as in LHQW group. However, the difference of change in TCM symptom score between two groups was not statistically significant ( P > 0.05). There were no significant differences between two groups in the median time to fever relief [RDQ group (4 ± 8) h vs LHQW group (4 ± 5) h] or the median time to fever clearance (RDQ group 47 h vs LHQW 36 h). No serious adverse events were reported during the study. Conclusion Compared with Lianhuaqingwen capsules, Reduqing granules achieved similar therapeutic effect in the treatment of CCWHS and no drug-related adverse events were reported during the study. Therefore, Reduqing granules might be effective and safe in the treatment of CCWHS.

Building 3D tunnel model is one of the important work in underground mining management, the methods to build 3D tunnel model by secondary development of commercial software or using OpenGL are not convenient and efficient. After analyzing the structure of DXF file, a method that building 3D tunnel model based on DXF file is proposed. On the basis of designing the tunnel section type and parameters, the modeling process is given in details. The result shows the method has a good efficiency and more convenient to realize.

Based on the analysis of the operating principle and the mathematical model of Permanent Magnet Synchronous Liner Motor (PMSLM), a model of PMSLM double loop control system based on vector control is designed with Matlab /Simulink, and with the model, a Current track-type PWM inverter model is given. Based on the double loop control system, an iterative learning control (ILC) method with a configuration of feedback is proposed to achieve precise tracking control of the PMSLM over a finite time interval. In the end, the ILC simulation and experiment research for the whole system are carried out. It is verified that ILC with composite control algorithm has a good performance in improving the position tracking ability of the PMSLM system by simulation and experiments results.

Traditional sensors have not been able to meet the test requirements. To realize the intelligent transformation of sensors, we have carried out thorough research into TEDS technology in IEEE1451 standard in this article. IEEE1451 provides a standard for intelligent sensors. After introducing the definition and format of the standard, this article analyzes several methods to realize TEDS and gives the methods that are used in this article. It describes the implementation of concrete intelligent technology according to the hardware module and the software module and gives the process of the two parts described, and realize the intelligent recognition of traditional sensors, which provides the foundation for further transformation of other sensors and test systems.

In this paper, an iterative learning control (ILC) method with a configuration of feedback is proposed to achieve precise tracking control of a permanent magnet linear motor(PLMM) system over a finite time interval. The feedback configuration could improve the learning speed and guarantee the stability of the system, while the ILC feedforward control enhances the trajectories tracking performance by capitalizing on the experience gained from the repeated execution of the same operation. The method is suitable for real-time and repetitive motion control. It is model-free. The various highlights of the proposed method will be elaborated by the simulation examples.