Aiming at the parameter optimization problem of deep learning network in the process of network traffic prediction modeling, in order to improve the network traffic prediction results, we proposed a network traffic prediction model based on improved gray wolf algorithm to optimize the deep learning network. Firstly, the historical data of network traffic was collected and preprocessed by phase space reconstruction and normalization. Secondly, gray wolf algorithm was introduced to quickly search the relevant parameters of the global optimal deep learning network, the preprocessed historical data of network traffic was learned according to the optimal parameters, and a prediction model that could mine the change law of historical data of network traffic was established. Finally, the network traffic prediction model of deep learning network optimized by other algorithms was compared and analyzed. The experimental results show that the network traffic prediction accuracy based on improved gray wolf algorithm to optimize deep learning network is more than 90%, which is much higher than other comparison models, and the modeling time of prediction modeling process is less than that of comparison model, which can meet the requirements of high accuracy and real-time of network traffic management.

By constructing the cohomologies complexes of Rota-Baxter Lie algebra homomorphisms, we discuss the formal deformation of Rota-Baxter Lie algebra homomorphisms and prove that Rota-Baxter Lie algebra homomorphism is rigid when the 2th-cohomology group of the deformation complex is zero.

We considered the linearized compressible Navier-Stokes equations in a one-dimensional bounded interval. By using Hormander theorem, we first proved the unique continuity of the dual problem, and then obtained the approximate controllability of the Navier-Stokes equations through the unique continuity of the dual problem.

We discussed the E-total coloring of complete bipartite graphs K_4,n which were vertex-distinguished by multiple sets by using
 the method of proof by contradiction, the method of pre-assignment of color sets and the method of constructing specific coloring, and determined E-total chromatic numbers of K_4,n which were vertex-distinguished by multiple sets.

Aiming at the problem that the current data compression algorithm of sensor network had the defects of low compression ratio and serious data deformation, in order to improve the realtime performance of data transmission of sensor networks, we proposed a data compression algorithm based on spatiotemporal correlation. Firstly, the original data of sensor network was collected, and the spatial transformation technology was used to analyze the correlation between the data of sensor network in space, and then the noise was removed to reduce the spatial resource occupied by noise. Secondly, according to the temporal correlation of the data of sensor network, the compression sensing algorithm was introduced to compress the spatial coefficient to reduce the data redundancy of sensor network. Finally, the performance of data compression algorithm of sensor network was analyzed by simulation experiment. The simulation results show that compared with other sensor network data compression algorithms, the proposed algorithm can improve the data compression ratio of sensor network without losing the data information of sensor network, obtain faster data compression speed of sensor network and reduce the communication pressure of sensor network.

Aiming at the problem that the traditional neural network proportionalintegraldifferential (PID) controller and linear quadratic regulator (LQR) optimized PID controller had long recovery time and poor antiinterference for speed control of brushless direct current motor， we proposed a BP neural network PID controller optimized by LQR for speed control of brushless direct current motor. Firstly, BP neural network was used to adjust the PID gain to improve the dynamic adaptability and robustness of the controller. Secondly, LQR was used to optimize the optimal output of BP neural network, which was closer to the target PID gain. The simulation results show that the controller can effectively improve the response speed, reduce the steadystate error and enhance the antiinterference ability.

In order to realize the high sensitivity detection of glassy carbon electrode in the field of electrochemical sensors,  the glassy carbon electrode was electrochemically activated by constant potential method in phosphate buffer solution at pH=5.0. The effects of pH value,  accumulation potential and accumulation time on the detection effect of Pb²⁺ were investigated, and the detection performance of activated electrode for Pb²⁺ was investigated under the optimal experimental conditions. The experimental results show that the activated electrode can enhance the response current of electrochemical detection of Pb²⁺, and has the characteristics of short detection time and high sensitivity. The detection linear range of Pb²⁺ is 1×10^-10—5×10^-6 mol/L,  the minimum detection limit is 3×10^-11 mol/L,  and the limit of quantification is 1.0×10^-10 mol/L. The activated electrode has a high recovery in the determination of  Pb²⁺ in tap water, and can be used in the actual water quality detection.

We proposed a fast and stable numerical method to solve the two-dimensional Cahn-Hilliard equation with constant mobility. The second order finite difference method was used in spatial discretization and Crank-Nicolson method was used in time discretization. We proved theoretically that the discrete energy had the property of dissipation with time evolving. The fixed point iteration method was used to solve the nonlinear algebraic equations in the fully discrete scheme, and the fast discrete cosine transform (FDCT) was used to improve the computational efficiency. The numerical results show that the discrete free energy is non increasing with respect to time, and the method has the advantaes of good stability, small storage and fast computation speed.

Aiming at the problems that the classical deep learning algorithm was difficult to extract facial features effectively and the accuracy of character identification was difficult to reach the ideal accuracy due to factors such as illumination, shooting angle and image quality, we proposed an  age recognition algorithm based on strong facial semantics. Firstly, the feature weights of facial regions were enhanced by the attention matrix to achieve the purpose of extracting feature regions. Secondly, a cascaded bi-directional long short-term memory (Bi-LSTM) network was used to learn the feature dependency relationships between temporal frames 
and  compensate for the influence of missing features on recognition accuracy. When tested on IMDB-WIKI facial dataset and Adience dataset, the age recognition accuracy of the algorithm reached 78.34% and 77.89%, respectively. Experimental results show that compared with other methods based on deep learning algorithms, the proposed algorithm has higher accuracy in the task of person age recognition based on image datasets.

A grad intensity method is put forward to process edges’ intensity of an image, this measure mostly uses simple grad transform to potentate the image edges, at the same time, a dilation method of mathematical morphology is applied to achieve rapid orientation of the license plates. In the process of characters segmentation, we first proposed least area circumrotation means in order to ascertain the best rotation angle of the license plate, then we adopted the way of perpendicularity projection to segment the characters of the license plate.

To solve the problem existing in the application of binocular stereo vision imagepair, a threestages matching preprocessing method was proposed. This method combines the wavelet denosing algorithm and histogram equalization method effectively. The random noise and Gauss noise of image are filtered and luminance difference between stereo imagepairs is balanced at the same time. Finally the contrast of imagepairs was improved by Laplace sharpen processing. The effectiveness of threestages matching preprocessing algorithm was proved by experiment. The application of the algorithm could improve the matching accuracy of stereo imagepairs effectively.

Based on the characteristics of polyvinylidene fluoride (PVDF) piezoelectric materials, a pressure sensor was designed, which consists of probe, charge amplifier circuit, filter circuit and voltage amplifier circuit.  The vibrating elastic iron slice was deformed by the contact point displacement. The measurement of pressure can be obtained by measuring the change of piezoelectric signal of the PVDF material on the iron slice. The  sensitivity of sensor is 61.22 mV/N and may be applied to various pressure measurement.

This paper presents a multiscale saliencybased image fusion model. First, the source images were decomposed into subimage sets by contourlet transform,  then  the saliency of each piece of subimage was calculated  and  the coefficients were selected by maximizing saliency, and finally, the fused image was obtained by inversing contourlet transform. The saliency of the image was  calculated by spectral residual theory. Experimental results  show that the proposed model outperforms traditional models.

Niche technology was used to enhance the identification ability and clustering ability of the algorithm for local optimal solution, 
improve the local search ability of the artificial bee colony algorithm, and improve the characteristics of rapid search for global optimal value of the scout bee. We proposed an artificial bee colony algorithm based on niche technology, and applied it to extract the character image edge on the bogie wheelset axle end. This algorithm improved the disadvantage of show 
recognition speed and poor recognition effect of character edge image caused by the large amount of operation. Neighborhood gray gradient value was used as the fitness function value of the algorithm, and the niche radius was used to maintain the diversity of the population, so as to improve the local search ability of the characters, and finally a high quality edge image of the character image was obtained. The experimental results show that the number of local character edge points can be increased 1.66 times by using niche algorithm.

By using hierarchical decomposition of function and the weighted inequalities, and by means of the boundedness of Hardy operator on variable Lebesgue spaces and the properties of bounded mean oscillation function space (BMO) functions, the author gave the boundedness of the higher order commutators generated by fractional Hardy operators with variable index and BMO functions on weighted HerzMorrey spaces with variable exponent.

By coupling the total energy convective upwind and split pressure (E-CUSP) scheme with the WENO-η scheme, we gave a new scheme (E-CUSP-WENO-η） to solve the one-dimensional shock tube problem. The numerical simulation results show that the new scheme is more accurate in capturing contact discontinuities and shock waves, among which the new scheme (E-CUSP-WENO-η(τ^opt₇)) with higher order optimal global smoothness factor has the least numerical dissipation. The coupled new scheme can capture shock wave more steeply, and the calculation results are accurate and stable.

The effects of different linear functional forms of dielectric column dielectric constant of two-dimensional functional photonic crystals on transverse electric (TE) wave and transverse magnetic (TM) wave band structures were studied by using plane wave expansion method. The results of numerical calculation show that the band gaps of two-dimensional functional photonic crystals are wider than that of two-dimensional conventional photonic crystals. By changing the different functional forms of dielectric column dielectric constant, the number, position and width of band gaps of the two-dimensional functional photonic crystals can be changed, so as to adjust the band gap of two-dimensional functional photonic crystals.

Aiming at the problem  of the imperfect tagging mechanism in the open source npm (node package manager) community, we proposed a method to automatically recommend tags for open source third-party library npm packages. Firstly,  according to the association relationship between existing tags in the npm community, the tags were clustered  and a tag library was  established  while solving the problem of tag synonyms. Secondly,  the word vector technology was used to calculate the semantic correlation degree between the Readme document of the npm package and the tags in the tag library. Finally, the tags were sorted according to  the degree of correlation to generate a tag recommendation list and complete the tag recommendation. The experimental results show that this method can effectively recommend tags for npm packages, and the accuracy rate of Recall@3 is 49.1%, Recall@5 is 56.3%, and Recall@10 is 66.9%.

The finite element method was used to solve the probability density function with periodic properties to study the statistical properties of stochastic periodic solutions. We proved the integral conservation of the full space approximate solution and the spatial integral approximate conservation of the approximate solution truncated to a finite interval, and verified  the effectiveness of the method  by numerical experiments.

By using the transmission matrix method, the matching matrix and transmission matrix of one-dimensional functional composite structure photonic crystal were given. On this basis, we studied the interface state of one-dimensional functional photonic crystals, and studied the influence of refractive index endpoint value, thickness of medium and incident angle on the position of interface state. The results show that interface state appears at the position where the imaginary part of the total impedance is 0. For functional media, when the initial endpoint value of refractive index increases, the position of interface state shifts red with the band gap, when the endpoint value of refractive index increases, the position of interface state shifts red  with the band gap. When the thickness of the medium increases, the position of the interface state shifts red with the band gap. When the incidence angle increases, the position of interface state shifts blue with the band gap. Therefore, the position of interface states can be adjusted by functional photonic crystals.

The structural problem of simple weight modules with a one-dimensional weight space in the quantum Loop algebra U_q(L(sl₂)) was solved by using a construction method, and it was obtained that any simple weight module with a one-dimensional weight space must be  isomorphic to one of the four classes of simple weight modules of U_q(L(sl₂)). In addition, a class of simple weight modules of  the quantum Loop algebra U_q(L(sl₂)) with  weight space dimension of 2, which was neither the highest weight nor the lowest weight, was constructed.

We considered a class of fourth-order thin-film equation with singular term and logarithmic source. Firstly, we obtained the local existence of weak solutions to the equation by  combining truncation function and  Galerkin approximation. Secondly, by virtue of the potential well method and Rellich inequality, we proved the global existence and decay estimate of weak solution to the equation under certain conditions. Finally, we proved the blow-up result of the  solution to the equation at a finite time by using the convex method, and gave the lower and upper bounds for blow-up time.

By using the harmonic analysis method and Lagrangian method, we studied the Cauchy problem for the multi-dimensional inhomogeneous incompressible heat-conducting equations and  proved the global well-posedness of strong solutions for the system under small initial data conditions in the critical Besov spaces.

Aiming at the problems of low accuracy, under-detection, and missed detection in the vehicle target detection algorithm at traffic intersections, we proposed a target recognition algorithm of traffic intersection based on improved YOLOv7.  Firstly, the algorithm  used the feed-forward convolutional attention mechanism CBAM to enhance the network’s  attention to key features from both channel attention and spatial attention, improve the network’s running  speed, and optimize the network’s feature extraction capabilities. Secondly, a new learning module was formed by connecting the  spatial layer to depth  layers to form a  full-dimensional dynamic convolution, which improved the YOLOv7 feature learning method and enhanced the feature expression ability. Finally, the experiments were conducted on the actual collected traffic intersection dataset. The experimental results show that the proposed method  achieves an average accuracy of 96.1% on the corresponding dataset, and the training time is reduced to 16.71 h. Therefore, it has obvious recognition advantages  for small target detection at traffic intersections.

In order to denoise impulse noises in images, an adaptive window switching median filtering method is proposed in this paper. First, each pixel is classified as signal or noisy point by the BPnet. Next, the image is filtered by an improved median filter. According to the result of noise detection, the filter can adjust adaptively windows width and sample choicely, each noisy point in image is denoised by filtering. In terms of suppressing impulse noise and preserving image details, the proposed method is better than that based on median filter. Even in the extreme case of 70% noise corruption, noisy images can be effectively recovered.

The time evolution of atomic population of the system, where the two identical two-level atoms in Bell states interact with two-mode entangled coherent field in a cavity filled with the Kerr medium, was investigatedby means of quantum theory and numerical method. The effects of the atomic initial state, mean photon numbers of field, the degree of the entanglement for the entangled coherent field, and the coupling strength between the Kerr medium and field on the atom’s population are discussed via numerical calculations. The results show that the atomic population is time-independent provided that the atoms are in |β₁₁〉 state. For the other three Bell states, there will appear the periodical collapse revival of atomic populationif the initial mean photon number reaches some special values. With the increase of the initial mean photon number, the oscillation frequency of the time-evolution of atomic population increases correspondingly, and the amplitude decreases. The degree of the entanglement of the two-mode entangled coherent field has no

In order to improve the precision of big visual field photo electricity measuring system, a set of device for detecting optical systematic distortion was designed to compensate for the detecting piece according to the characters of the photo electricity measuring system and the target. On the basis of the principle of the distortion correction, the course of distortion correction in the measuring system was introduced via the distortion correcting device. The proceeding based on the correction method was compiled to adjust the distortion of the photoelectric measuring system. And the experimental analysis was carried out to prove the process feasiblility. The experiment result shows that the method reduces the measurement error caused by the distortion of optical system. After cubic equation compensation, the measurement accuracy has reached 0.244 pixel in the whole visual field. 

The aim of this paper is to discuss the existence of global BV solutions for a class of systems of quasilinear hyperbolic equ ations with Radon measures as initial conditions. We firstly considered its regularized problem. With the aid of a series of critical analysis and by a limit process, we obtained the existence of global solutions for the regularized problem. In particular, we also obtained the uniform BV estimates for it. Then again, by a limit process, the existence of global BV solutions for the original problem was obtained.

Graphitic carbon nitride (g-C₃N₄) sheets were synthesized via the pyrolysis of melamine at 1 000 K under vacuum. X-ray diffraction patterns strongly indicate that the synthesized carbon nitride was g-C₃N₄. Transmission electron microscopic images indicate that the product was mainly composed of graphitic carbon nitride sheets. The chemical bonding of the sample was investigated by X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy. 

We mainly surveyed the approaches on operon prediction and summarized the current art of the state in bioinformatics. After presenting most datasets in operon prediction and introducing most important popular algorithms, some disadvantages were pointed out. Furthermore, some future challenges w ere discussed and some solutions were suggested which are helpful to the future researches.