Abstract:This study aims to improve the integrated testing of large-aperture telescopes to clarify the fundamental principles of an integrated testing system based on astrophotonics. Our demonstration and analyses focused on element-position sensing and modulation based on spatial near-geometric beams, high-throughput step-difference measurements based on channel spectroscopy, distributed broadband-transmittance testing, and standard spectral tests based on near-field energy regulation. Comprehensive analyses and experiments were conducted to confirm the feasibility of the proposed system in the integrated testing process of large-aperture telescopes. The results demonstrated that the angular resolution of the light rays exceeded 5 arcsec, which satisfies the requirements for component-position detection in future large-aperture telescopes. The measurement resolution of the wavefront tilt was better than 0.45 µrad. Based on the channel spectral method—which combined a high signal-to-noise ratio and high sensitivity, along with continuous-spectral digital segmentation and narrowband-spectral physical segmentation—a resolution of 0.050 μm and a range of 50 μm were obtained. After calibration, the measurement resolution of the pupil deviation improved to exceed 4% accuracy, and the transmission measurements achieved a consistency of over 2% accuracy. Regarding fringe-broadband interferometry measurements, the system maintained high stability, ensuring its operation within the coherence length, and robustly detected the energy without unwrapping the phase. The use of a projector for calibrating broadband-spectrum measurements led to a reduction in contrast from 0.8142 to 0.6038, which further validates the system's applicability in the integrated testing process of large-aperture telescopes. This study greatly enhanced the observational capabilities of large-aperture telescopes while reducing the integrated system's volume, weight, and power consumption.

Abstract:The measurement uncertainty analysis is carried out to investigate the measurable dimensions of cylindrical workpieces by the rotary-scan method in this paper. Due to the difficult alignment of the workpiece with a diameter of less than 3 mm by the rotary scan method, the measurement uncertainty of the cylindrical workpiece with a diameter of 3 mm and length of 50 mm which is measured by a roundness measuring machine, is evaluated according to GUM (Guide to the Expression of Uncertainty in Measurement) as an example. Since the uncertainty caused by the eccentricity of the measured workpiece is different with the dimension changing, the measurement uncertainty of cylindrical workpieces with other dimensions can be evaluated the same as the diameter of 3 mm but with different eccentricity. Measurement uncertainty caused by different eccentricities concerning the dimension of the measured cylindrical workpiece is set to simulate the evaluations. Compared to the target value of the measurement uncertainty of 0.1μm, the measurable dimensions of the cylindrical workpiece can be obtained. Experiments and analysis are presented to quantitatively evaluate the reliability of the rotary-scan method for the roundness measurement of cylindrical workpieces.

Abstract:The hydraulic actuator, known as the "muscle" of military aircraft, is responsible for flight attitude adjustment, trajectory control, braking turn, landing gear retracting and other actions, which directly affect its flight efficiency and safety. However, the sealing assembly often has the situation of over-aberrant aperture fit clearance or critical over-aberrant clearance, which increases the failure probability and degree of movable seal failure, and directly affects the flight efficiency and safety of military aircraft. In this paper, the simulation model of hydraulic actuator seal combination is established by ANSYS software, and the sealing principle is described. The change curve of contact width and contact pressure of combination seal under the action of high-pressure fluid is drawn. The effects of different oil pressure, fit clearance and other parameters on the sealing performance are analyzed. Finally, the accelerated life test of sliding seal components is carried out on the hydraulic actuator accelerated life test rig, and the surface morphology is compared and analyzed. The research shows that the O-ring is the main sealing element and the role of the check ring is to protect and support the O-ring to prevent damage caused by squeezing into the fit clearance, so the check ring bears a large load and is prone to shear failure. Excessive fit clearance is the main factor affecting the damage of the check ring, and the damage parts are mainly concentrated at the edge of the sealing surface. This paper provides a theoretical basis for the design of hydraulic actuator and the improvement of sealing performance.

Abstract:Terahertz time-domain spectroscopy is a kind of far-infrared spectroscopy technology, and its spectrum reflects the internal properties of substances with rich physical and chemical information, so the use of terahertz waves can be used to qualitatively identify food additives containing nitrogen elements. Analytic hierarchy process (AHP) was originally used to solve evaluation-type problems, and this paper introduces it into the field of terahertz spectral qualitative analysis, proposes a terahertz time-domain spectral qualitative identification method combined with analytic hierarchy process, and verifies the feasibility of the method by taking four common food additives (xylitol, L-alanine, sorbic acid, and benzoic acid) and two illegal additives (melamine, and Sudan Red No. I) as the objects of study. Firstly, the collected terahertz time-domain spectral data were pre-processed and transformed into a data set consisting of peaks, peak positions, peak numbers and overall trends; then, the data were divided into comparison and test sets, and a qualitative additive identification model incorporating analytic hierarchy process was constructed and parameter optimisation was performed. The results showed that the qualitative identification accuracies of additives based on single factors, i.e., overall trend, peak value, peak position, and peak number, were 80.23%, 70.93%, 67.44%, and 40.70%, respectively, whereas the identification accuracy of the analytic hierarchy process qualitative identification method based on multi-factors could be improved to 92.44%. In addition, the fuzzy characterisation of the absorption spectrum data was binarised in the data pre-processing stage and used as the base data for the overall trend, and the recognition accuracy was improved to 94.19% by combining the fuzzy characterisation method of such data with the hierarchical analysis qualitative recognition model. The results show that it is feasible to use terahertz technology to identify different varieties of additives, and this paper constructs a hierarchical analytical qualitative model with better effect, which provides a new means for food additives detection, and the method is simple in steps, with a small demand for samples, which is suitable for the rapid detection of small samples.

Abstract:Aiming at the problems of low efficiency, poor anti-noise and robustness of transfer learning model in intelligent fault diagnosis of rotating machinery, a new method of intelligent fault diagnosis of rotating machinery based on single source and multi-target domain adversarial network model (WDMACN) and Gram Angle Product field (GAPF) was proposed. Firstly, the original one-dimensional vibration signal is preprocessed using GAPF to generate the image data including all time series. Secondly, the residual network is used to extract data features, and the features of the target domain without labels are pseudo-labeled, and the transferable features among the feature extractors are shared through the depth parameter, and the feature extractors of the multi-target domain are updated anatomically to generate the features that the discriminator cannot distinguish. The model t through adversarial domain adaptation, thus achieving fault classification. Finally, a large number of validations were carried out on the bearing data set of Case Western Reserve University (CWRU) and the gear data. The results show that the proposed method can greatly improve the diagnostic efficiency of the model, and has good noise resistance and generalization.

Abstract:In order to improve the performance degradation prediction accuracy of proton exchange membrane fuel cell (PEMFC), a fusion prediction method (CKDG) based on adaptive noise complete ensemble empirical mode decomposition (CEEMDAN), kernel principal component analysis (KPCA) and dual attention mechanism gated recurrent unit neural network (DA-GRU) was proposed. CEEMDAN and KPCA were used to extract the input feature data sequence, reduce the influence of random factors, and capture essential feature components to reduce the model complexity. The DA-GRU network helps to learn the feature mapping relationship of data in long time series and predict the changing trend of performance degradation data more accurately. The actual aging experimental data verify the performance of the CKDG method. The results show that under the steady-state condition of 20% training data prediction, the CKDA method can reduce the root mean square error (RMSE) by 52.7% and 34.6%, respectively, compared with the traditional LSTM and GRU neural networks. Compared with the simple DA-GRU network, RMSE is reduced by 15%, and the degree of over-fitting is reduced, which has higher accuracy. It also shows excellent prediction performance under the dynamic condition data set and has good universality.

Abstract:The complex permittivity of baijiu varies with frequency, and dielectric spectroscopy has been used to evaluate the quality. To simplify the analysis and reduce the number of the parameters, a dielectric relaxation model is often used to fit the permittivity data. However, existing fitting methods such as the least squares and particle swarm optimization methods are often computationally complex and require preset initial values. Therefore, a simpler calculation method of the relaxation parameters considering the geometric characteristics of the permittivity spectrum is proposed. It is based on the relationship between the Cole-Cole relaxation parameters and the Cole-Cole diagram, which is fitted by a geometric method. First, the concepts of the Cole-Cole parameters and the diagram are introduced, and then the process of obtaining the parameters from the complex permittivity measurement data is explained. Taking baijiu with 56% alcohol by volume (ABV) as an example, the fitting is better than the least squares method and similar to the particle swarm optimization. This method is then used for the parameter fitting of baijiu with ABV of 42-52%, and the average error is less than 1%, demonstrating its wider applicability. Finally, a prediction model is used for baijiu with 53% ABV, and the error is only 1.51%. Hence, the method can be applied to the measurement of ABV of baijiu.

Abstract:As one of the main application directions of quantum technology, underwater quantum communication is of great research significance. In order to study the influence of marine planktonic algal particles on the communication performance of underwater quantum links, based on the extinction characteristics of marine planktonic algal particles, the influence of changes in the chlorophyll concentration and particle number density of planktonic algal particles on the attenuation of underwater links is explored respectively, the influence of marine planktonic algal particles on the fidelity of underwater quantum links, the generation rate of the security key, and the utilization rate of the channel is analyzed, and simulation experiments are carried out. The results show that with the increase in chlorophyll concentration and particle density of aquatic planktonic algal particles, quantum communication channel link attenuation shows a gradually increasing trend. In addition, the security key generation rate, channel fidelity and utilization rate are gradually decreasing. Therefore, the performance of underwater quantum communication channel will be interfered by marine planktonic algal particles, and it is necessary to adjust the relevant parameter values in the quantum communication system according to different marine planktonic algal particle number density and chlorophyll concentration to improve the performance of quantum communication.

Abstract:Reconfigurable modular robots feature high mobility due to their unconstrained connection manners. Inspired by the snake multi-joint crawling principle, a chain-type reconfigurable modular robot (CRMR) is designed, which could reassemble into various configurations through the compound joint motion. Moreover, an illumination adaptive modular robot identification (IAMRI) algorithm is proposed for CRMR. At first, an adaptive threshold is applied to detect oriented FAST features in the robot image. Then, the effective detection of features in non-uniform illumination areas is achieved through an optimized quadtree decomposition method. After matching features, an improved random sample consensus algorithm is employed to eliminate the mismatched features. Finally, the reconfigurable robot module is identified effectively through the perspective transformation. Compared with ORB, MA, Y-ORB, and S-ORB algorithms, the IAMRI algorithm has an improvement of over 11.6% in feature uniformity, and 13.7% in the comprehensive indicator, respectively. The IAMRI algorithm limits the relative error within 2.5 pixels, efficiently completing the CRMR identification under complex environmental changes.

Abstract:To solve the sparse reward problem of job-shop scheduling by deep reinforcement learning, a deep reinforcement learning framework considering sparse reward problem is proposed. The job shop scheduling problem is transformed into Markov decision process, and six state features are designed to improve the state feature representation by using two-way scheduling method, including four state features that distinguish the optimal action and two state features that are related to the learning goal. An extended variant of graph isomorphic network GIN++ is used to encode disjunction graphs to improve the performance and generalization ability of the model. Through iterative greedy algorithm, random strategy is generated as the initial strategy, and the action with the maximum information gain is selected to expand it to optimize the exploration ability of Actor-Critic algorithm. Through validation of the trained policy model on multiple public test data sets and comparison with other advanced DRL methods and scheduling rules, the proposed method reduces the minimum average gap by 3.49%, 5.31% and 4.16%, respectively, compared with the priority rule-based method, and 5.34% compared with the learning-based method. 11.97% and 5.02%, effectively improving the accuracy of DRL to solve the approximate solution of JSSP minimum completion time.

Abstract:In the municipal solid waste incineration process, it is difficult to effectively control the gas oxygen content by setting the air flow according to artificial experience. To address this problem, this paper proposes an optimization control method of gas oxygen content based on model predictive control. First, a stochastic configuration network is utilized to establish a prediction model of gas oxygen content. Second, an improved differential evolution algorithm that is based on parameter adaptive and t-distribution strategy is employed to address the set value of air flow. Finally, model predictive control is combined with the event triggering strategy to reduce the amount of computation and the controller's frequent actions. The experimental results show that the optimization control method proposed in this paper obtains a smaller degree of fluctuation in the air flow set value, which can ensure the tracking control performance of the gas oxygen content while reducing the amount of calculation.

Abstract:Breast cancer has become a killer of women's health nowadays. In order to exploit the potential representational capabilities of the models more comprehensively, we propose a multi-model fusion strategy. Specifically, we combine two differently structured deep learning models, ResNet101 and Swin Transformer (SwinT), with the addition of the Convolutional Block Attention Module (CBAM) attention mechanism, which makes full use of SwinT's global context information modeling ability and ResNet101's local feature extraction ability, and additionally the cross entropy loss function is replaced by the focus loss function to solve the problem of unbalanced allocation of breast cancer data sets. The multi-classification recognition accuracies of the proposed fusion model under 40X, 100X, 200X and 400X BreakHis datasets are 97.50%, 96.60%, 96.30 and 96.10%, respectively. Compared with a single SwinT model and ResNet101 model, the fusion model has higher accuracy and better generalization ability, which provides a more effective method for screening, diagnosis and pathological classification of female breast cancer.