Search

Phase matching quantum key distribution is a promising scheme for remote quantum key distribution, breaking through the traditional linear key-rate bound. In practical applications, finite data size can cause significant system performance deterioration when data size is below 1010. In this paper, an improved statistical fluctuation analysis method is applied for the first time to two decoy-states phase matching quantum key distribution, offering new insights and potential solutions for improving the key generation rate and the maximum transmission distance while maintaining security. Moreover, we also compares the impacts of the proposed improved statistical fluctuation analysis method to those of the Gaussian approximation and Chernoff-Hoeffding boundary methods on system performance. The simulation results show that the proposed scheme significantly improves key generation rate and maximum transmission distance compared to the Chernoff-Hoeffding approach, and approaches the results obtained when the Gaussian approximation is employed. At the same time, the proposed scheme retains the same security level as the Chernoff-Hoeffding method, and is even more secure than the Gaussian approximation.

Objective: Proton therapy offers an advantageous dose distribution compared to the photon therapy, since it deposits most of the energy at the end of range, namely the Bragg peak (BP). Protoacoustic technique was developed to in vivo determine the BP locations. However, it requires large dose delivery to the tissue to obtain an averaged acoustic signal with a sufficient signal to noise ratio (SNR), which is not suitable in clinics. We propose a deep learning-based technique to acquire denoised acoustic signals and reduce BP range uncertainty with much lower doses. Approach: Three accelerometers were placed on the distal surface of a cylindrical polyethylene (PE) phantom to collect protoacoustic signals. In total 512 raw signals were collected at each device. Device-specific stack autoencoder (SAE) denoising models were trained to denoise the input signals, which were generated by averaging 1, 2, 4, 8, 16, or 32 raw signals. Both supervised and unsupervised learning training strategies were tested for comparison. Mean squared error (MSE), signal-to-noise ratio (SNR) and the Bragg peak (BP) range uncertainty were used for model evaluation. Main results: After SAE denoising, the MSE was substantially reduced, and the SNR was enhanced. Overall, the supervised SAEs outperformed the unsupervised SAEs in BP range verification. For the high accuracy detector, it achieved a BP range uncertainty of 0.20 +/- 3.44 mm by averaging over 8 raw signals, while for the other two low accuracy detectors, they achieved the BP uncertainty of 1.44 +/- 6.45 mm and -0.23 +/- 4.88 mm by averaging 16 raw signals, respectively. Significance: We have proposed a deep learning based denoising method to enhance the SNR of protoacoustic measurements and improve the accuracy in BP range verification, which greatly reduces the dose and time for potential clinical applications., Comment: 15 pages, 5 figures

Strong noise, nonlinearity, and complex frequency are the remarkable symbols of vibration signals when rolling bearing failure. So a new method that improved the accuracy of rolling bearing fault diagnosis is proposed based on variational mode decomposition (VMD) and grey wolf algorithm optimized extreme learning machine (GWO-ELM). The bearing vibration signal is decomposed by VMD, and decomposed Intrinsic Mode Component (IMF) is used for fuzzy entropy calculation to construct a multi-scale complexity eigenvector. The fault pattern recognition is carried out in the extreme learning machine afterward. The parameter selection problem in VMD and fuzzy entropy algorithm is analysed through the experimental fault data of the rolling bearing, and the noise robustness of the experimental data is verified. Through fault data of rolling bearings on site, the comparison of the results shows that the GWO-ELM model can effectively identify the abnormal conditions, and has a faster training and learning speed in diagnosing the faults.

In this paper, we study the following diffusion model of Kirchhoff-type driven by a nonlocal integro-differential operator where [u] s is the Gagliardo seminorm of u, is a bounded domain with Lipschitz boundary, , is a nonlocal integro-differential operator defined in (1.2), which generalizes the fractional Laplace operator , is the initial function, and is a continuous function and there exist two constants m 0 > 0 and such that As is well-known, the nonlocal Kirchhoff problem was first introduced and motivated in Fiscella and Valdinoci (2014 Nonlinear Anal. 94 156–70) and the above problem was studied by Xiang et al (2018 Nonlinearity 31 3228–50), the main results of Xiang et al (2018 Nonlinearity 31 3228–50) are as follows: The local existence of nontrivial, nonnegative weak solution for , where . The blow-up conditions for nontrivial, nonnegative weak solution when J(u 0) < 0, where J(u 0) denotes the initial energy. The main purpose of this paper is to extend the above results and we get: The global existence of nontrivial, nonnegative weak solution for any . The global existence and blow-up conditions for nontrivial, nonnegative weak solution when for the case , where d is a positive constant given in (1.13).

The sensitivity of Goos-Hänchen (GH) shifts on the geometric parameters of the structures suggest their great application prospects in sensing and detection. However, most of the enhanced GH shifts are achieved under either the transverse electric (TE) wave or transverse magnetic (TM) polarized wave. Here, we theoretically demonstrate that the well-designed metal-dielectric grating structure with monolayer graphene has the potential for realizing the enhanced GH shifts under both the TE and TM polarized waves at the same specific wavelength, which is dramatically different from the previous works and suggests that the enhancement of the GH shift is polarization independent. In particular, the enhancement of the GH shift obtained in this structure under the TE polarized wave is caused by the excited guided mode resonance in the dielectric layers of the grating strip. Moreover, the enhancement of the GH shift under the TM polarized wave is mainly due to the excited surface plasmon polariton at the interface between the dielectric layer and the metal layer in the top of grating strip. We also find that the size and sign of the GH shift can be controlled by the chemical potential of monolayer graphene and the geometric parameters of this structure. The highly controllable and polarization independent GH shift in the metal-dielectric grating structure with monolayer graphene paves the way for the future applications in the polarization independent devices, such as, optical sensors, optical switches and so on.

The research of high-performance polarization controllers is of great significance for expanding the application field of polarization optics. Here, a polarization switch is demonstrated by using a dual-core photonic crystal fiber (DCPCF) with four symmetrical air holes, placed above and below each core, filled with magnetic fluid (MF). The switch, which utilizes a magnetic field to change the coupling length ratio of the x and y polarization modes, enables dynamic tuning of the polarization state and extinction ratio. Numerical results show that when the working length is 36.638 mm, the magneto–optical polarization switch can operate in four communication bands, i.e., 1509 nm to 1520 nm, 1544 nm to 1556 nm, 1578 nm to 1591 nm, and 1611 nm to 1624 nm. Moreover, the extinction ratio (ER) is greater than 20 dB in the fiber length range of 38.5 mm to 38.7 mm, indicating that the device has a good fault tolerance for the interception of the fiber length.

Infrared bubbles provide a unique opportunity to study the interactions between massive stars and surrounding material. We conduct a multi-wavelength study on the environment and star formation around an infrared bubble N 13. Three dust clumps and two molecular clumps are identified around N 13, which are all distributed on the layer. Young stellar objects (YSOs) are carefully searched using infrared colors and YSO candidates of WISE and Gaia DR2, and three Class I/II YSOs are found in N 13. In addition, four O-type stars identified in N 13 are probably the exciting stars. The dynamical and fragmentation ages of N 13 are 0.32–0.35 and 1.37–2.80 Myr respectively, which suggest that the radiation-driven implosion model may be dominant in N 13. By comparing the small-size bubble N 13 (R ∼ 1.9 pc) and the larger-size bubble G15.684-0.29 (R ∼ 15.7 pc) we found that star formation activity is more active in the large-size bubble. Brief comparisons of ten bubbles show that small-size bubbles have a small ratio of kinetic age versus the fragmentation time. Triggering star formation may be more active in bubbles with larger ratio between kinetic and fragmentation ages. Furthermore, the collect and collapse mechanism may play the dominant role in the large-size ones.

The vibration signals of rolling bearings are mixed with a variety of noises and a variety of faults which are coupled to each other, and which make the fault frequencies interfere with each other. This in turn leads to difficulty of accurate extraction of the features of composite faults. Therefore, it is necessary not only to suppress the noise of vibration signals but also to separate and extract the features of compound faults. A compound fault feature extraction method based on improved particle swarm optimization (IPSO) algorithm optimized maximum correlation kurtosis deconvolution (MCKD), improved variational mode decomposition (IVMD) and cyclic autocorrelation function (CAF), named IPSO-MCKD-IVMD-CAF is proposed. Firstly, the parameters of MCKD are selected adaptively by the IPSO, and the original signal is pre-processed by the parameterized MCKD. Secondly, the IVMD is utilized to decompose the signal, several intrinsic mode functions (IMFs) are obtained, the kurtosis and correlation coefficients of each IMF component are calculated. Finally, appropriate IMF components are selected by kurtosis and correlation coefficient to be superimposed into the reconstructed signals, and the reconstructed signals are analyzed by CAF. The proposed method has been successfully applied in simulation and measured vibration signal analysis, and the features of compound faults can be separated and extracted with high accuracy. The IPSO algorithm shortens the optimization time, and the IVMD solves the parameter selection problem. The analysis results of simulation case and measured data show the advantages of the proposed method IPSO-MCKD-IVMD-CAF. Compared to existing methods, the proposed method not only increases the accuracy of compound fault feature extraction, but also has a very good performance on the separation and extraction of compound fault features. More importantly, the improved algorithm reduces the computational complexity.

Reversible solid oxide fuel cells (RSOC) can realize energy storage and power generation in a compact structure, and their high-temperature operation characteristics provide the possibility for combined heat and power generation. Therefore, RSOC can be used as the node of the multi-energy flow intersection of the integrated energy system. The flexible operation mode of RSOC will greatly reduce the operation cost. However, the optimal operation of an integrated energy system including RSOC is very rare at present, so this paper establishes an integrated energy system of the electric gas network including RSOC. On this basis, this paper constructs an RSOC model, a DC power flow model and a linearized gas power flow model considering multiple operating states. Then, based on GAMS, the economic cost of a 24-20 electric gas composite network is optimized. After verifying the reliability of the model, the optimization results considering two working conditions and four working conditions are compared. It is concluded that the model considering multiple operating states of RSOC will reduce the operating cost of the system by 24%.

Objective. Computed tomography (CT) to material property conversion dominates proton range uncertainty, impacting the quality of proton treatment planning. Physics-based and machine learning-based methods have been investigated to leverage dual-energy CT (DECT) to predict proton ranges. Recent development includes physics-informed deep learning (DL) for material property inference. This paper aims to develop a framework to validate Monte Carlo dose calculation (MCDC) using CT-based material characterization models. Approach. The proposed framework includes two experiments to validate in vivo dose and water equivalent thickness (WET) distributions using anthropomorphic and porcine phantoms. Phantoms were irradiated using anteroposterior proton beams, and the exit doses and residual ranges were measured by MatriXX PT and a multi-layer strip ionization chamber. Two pre-trained conventional and physics-informed residual networks (RN/PRN) were used for mass density inference from DECT. Additional two heuristic material conversion models using single-energy CT (SECT) and DECT were implemented for comparisons. The gamma index was used for dose comparisons with criteria of 3%/3 mm (10% dose threshold). Main results. The phantom study showed that MCDC with PRN achieved mean gamma passing rates of 95.9% and 97.8% for the anthropomorphic and porcine phantoms. The rates were 86.0% and 79.7% for MCDC with the empirical DECT model. WET analyses indicated that the mean WET variations between measurement and simulation were −1.66 mm, −2.48 mm, and −0.06 mm for MCDC using a Hounsfield look-up table with SECT and empirical and PRN models with DECT. Validation experiments indicated that MCDC with PRN achieved consistent dose and WET distributions with measurement. Significance. The proposed framework can be used to identify the optimal CT-based material characterization model for MCDC to improve proton range uncertainty. The framework can systematically verify the accuracy of proton treatment planning, and it can potentially be implemented in the treatment room to be instrumental in online adaptive treatment planning.

A bidirectional quantum operation teleportation scheme is proposed by utilizing a six-qubit composite Bell-GHZ state. In which, two users Anne and Benson can perform their respective concerned single-qubit operations on each other’s target states. Compared with the other existing schemes of quantum operation teleportation (QOT), the present scheme can realize the remote implement of the positive and negative two-way quantum operations, which not only improves the communication capacity, but also expands the application scenarios of QOT. Additionally, it is deterministic and feasible with the present experimental techniques.

We establish a model based on the dispersion of the diffractive grating to investigate the influence of the spectral envelope and wavelength stability on the combined beam quality M 2 of the spectral beam combining (SBC) system. We obtain the analytic expressions of combined beam quality corresponding to the second-moment-spectral width of the beamlets, which can be used to forecast the combined beam quality for any integrable spectral envelope shape of beamlets in an SBC system. Additionally, we find that the center wavelength shift of beamlets has a significant impact on the combined beam quality. The second-moment-based width of the center wavelength shift distribution among all beamlets should be less than 0.5 times the spectral width. Here, the spectral width is the weighted root mean square of all beamlets’ spectral width, where the weight is the power of each single channel.

Charge transport in insulating composites is fundamental to designing high performance in electrical breakdown strength processes. A fundamental understanding of the charge transport at nanoscale in insulating composites remains elusive. Herein, we fabricate two types of interfaces in epoxy (EP) composites (Al2O3/EP and bubble/EP, respectively). Then the local dynamic charge mobility behavior and charge density are explored using in situ Kelvin probe force microscopy. After the external voltage in the horizontal direction is applied, significant differences are demonstrated in the evolution of charge transport for epoxy matrix, filler/bubble, and their interface, respectively. The interface between Al2O3 and epoxy is easier to accumulate the negative charges and introduce shallow traps. Lots of positive charges are located around a bubble where deeper traps are present and could prevent charge migration. Thus, this work offers extended experimental support to understanding the mechanism of charge transport in dielectric composites.

Uncapacitated Facility Location Problem (UFLP) is a NP-hard Problem that to determine the optimal Location of facilities. In this paper, the general Genetic Algorithm (gGA) is first introduced and adopted to solve a small case of UFLP (case1). Then an improved genetic algorithm (iGA) based on real coding is proposed to solve the UFLP problem. This paper mainly makes appropriate adjustments in the selection of fitness function, crossover operator and mutation operator to be more suitable for UFLP. Facilities thus can be roughly allocated according to the cost of facilities and the demands of customers. Case2 was calculated by several CAP data (CAP101, CAP103, CAP104, CAP131, CAP133, CAPB and CAPC) in OR-LIBRORY. The results proved that the iGA is feasible and effective, and it is found that 80% of the results obtained by the iGA are within 0.05% of the optimal solutions. Compared with several other common algorithms, the advantages of iGA increase as the scale of calculation increases. Finally, applying iGA to an on-site oilfield pipeline network, it was found that it can find the optimal solution in a short time.

Due to the development of communication technology, mobile devices continue to increase. As one of the critical technologies of 5G, D2D communication is an up-and-coming technology. In this paper, multiple D2D pairs usually multiplex the same channel, which will cause serious channel interference. To solve this problem, a distributed deep reinforcement learning framework is proposed, which can well adapt to the power allocation in a dynamic environment. The simulation results show that compared with other benchmark methods, the proposed scheme can improve the overall D2D rate.

The dependence of Brillouin gain spectrum (BGS) characteristics, including the Brillouin frequency shift (BFS) and the BGS bandwidth, on germanium concentration in large-mode-area Ge-doped passive fibers is investigated theoretically and experimentally. The simulation results show that the BFS is inversely proportional to GeO2 concentration, and the BGS bandwidth initially increases with the augment of GeO2 concentration, and then decreases. The BGSs of four fibers with core diameters of 10μm and 20μm for different GeO2 concentrations are compared experimentally. Experimental results demonstrate that with the same core diameter, the variations of BFS and BGS bandwidths with GeO2 concentration accord with the simulation results. Additionally, the BGS characteristics of three large-mode-area passive fibers with diameters of 10μm, 25μm, and 30μm are measured, which confirm that the increasing of the fiber diameters will cause the BGS bandwidth to broaden. We believe that these results can provide valuable references for modulating the high-power narrow-linewidth fiber lasers and Brillouin fiber amplifiers.

Hydrogels with intrinsic high stretchability and flexibility are extremely attractive for soft electronics. However, the existing complicated and laborious methods (such as mold curing) to fabricate microstructured hydrogel (MH) still limit the development of hydrogel-based sensors for flexible devices. Herein, we use digital light processing 3D printing technology to rapidly construct double-network (DN) ionic conductive hydrogel, and then design and print fingerprint-like MH film to manufacture an iontronic pressure sensor. In particular, the DN hydrogel consists of acrylamide/acrylic acid to form a covalently cross-linked network, and magnesium chloride is introduced to form an ionic cross-linked physical network in the hydrogel. The printability (with resolution 150 μm) and mechanical property tunability of DN hydrogel enable the convenient fabrication of sensors. With the biomimetic fingerprint MH film, the iontronic pressure sensor not only has a high sensitivity (0.06 kPa−1), but also has a large detection range (26 Pa–70 kPa) and good stability (200 cycles of pressure loading). We demonstrated that our sensor can be applied to realize tactile sensing in a prosthetic application and detect human motion. With the easy strategy of constructing DN hydrogel with microstructures by 3D printing technology, hydrogel-based sensors are anticipated to be employed in more smart electronics.

The power consumption readings of sub meter and total meter of distribution transformer of low-voltage users follow the law of conservation of energy. The meter power loss rate of abnormal low-voltage users must also be abnormal. This paper studies the solution of the meter power loss rate under the four abnormal power consumption scenarios of single (multi) user and full (partial) period. The traditional linear solution method has accurate identification effect for the abnormal power consumption scenario of full period, but it cannot identify the abnormal power consumption scenario of partial period. In this paper, an improved artificial fish swarm algorithm is proposed. By adjusting the fixed step to the adaptive step, the power loss rate of each sub meter is obtained, and the abnormal power users are pinpointed. The research results are verified by simulation examples on IEEE European Low Voltage Test Feeder. The results show that the improved artificial fish swarm algorithm in this paper can identify abnormal power users for the above four abnormal electric field scenarios. The algorithm provides a new alternative for the identification of abnormal low voltage users.

The three-dimensional elevated warehouse is a production equipment, and it is inevitable that it will malfunction during the operation. If it is not handled properly, there will be inconsistencies between information and practice, resulting in inconsistency between inventory information and the actual product. In order to solve the above problems, this project team carried out functional research on various SQL tables of multi-dimensional data collaboration, and constructed a virtual inventory method based on Markov multi-dimensional data collaboration. The behavioral characteristics of the cargo location information in the cargo location table TWMS_LOC and the pallet status table TWMS_PLT are extracted, and the two are compared. If the deviation between the two exceeds the threshold, the status is judged to be no cargo information. Experiments are conducted to test the feasibility and effectiveness of the sentences of the proposed virtual inventory. The results show that the inventory method can better describe the status of the cargo location and information, and effectively realize the virtual inventory at the database level. The research results of this method will completely replace the complicated manual inventory business. It can provide accurate and efficient virtual inventory solutions for similar logistics elevated warehouses.

Device-to-device (D2D) communication is regarded as a key technical component of the fifth-generation (5G), D2D communication usually reuses spectrum resources with cellular users (CUs). To mitigate interference to cellular links and improve spectrum efficiency, this paper investigates a sum-rate maximization problem in the underlay of D2D communication. Particularly, a joint channel selection and power allocation framework based on multi-agent deep reinforcement learning is proposed, named Double Deep Q-Network (DDQN). It can adeptly select the channel and allocate power in a dynamic environment. The proposed scheme only requires local information and some outdated nonlocal information, which reduces signaling overheads significantly. Simulation results show that the proposed scheme can improve the D2D sum rate and ensure quality-of-service (QoS) of CUs compared with other benchmarks.

Aiming at the problem of path planning for transportable charging stations, a site selection model and path planning model for dedicated routes for transportable charging stations are proposed. The path planning problem model includes the vehicle path problem and the dynamic path problem, and the path planning model is emphatically studied. Considering the network optimization model of chargers and transportable charging station paths, by establishing a path planning model, setting parameters and models, using particle swarm algorithm to calculate the path planning model of chargers and mobile charging stations (Vehicle Routing Problem Considering Stationary and Movable Charging Points (referred to as “VRP-SMCP”) is solved, and the feasibility and effectiveness of the proposed model and algorithm is verified through comparative analysis of calculation examples, and the optimal path planning under the condition of minimizing cost is obtained.

Acoustic emission signal denoising is the premise of acoustic emission signal identification. The traditional filter and wavelet analysis have the problems of relying on prior information and poor adaptability in acoustic emission signal denoising. Therefore, a noise denoising model of acoustic emission signal based on noise reduction autoencoder is proposed. By unsupervised learning training, the noise reduction autoencoder has more stable invariant characteristics, so that the error between the reconstructed signal and the original signal converges to a minimum, thus achieving the purpose of denoising. The denoising experiments are carried out on the basis of processing 3000 corrosive acoustic emission signal samples. The experimental results show that the model has better denoising effect when the number of hidden layer neurons is 400. The proposed DAE model has better performance and robustness than Donoho wavelet threshold denoising method.

This article introduces the one-button sequence control operation of the isolating switch in the intelligent substation, as well as the sequence control system architecture and different kinds of identification technology operation applications. Combined with actual engineering, different recognition technologies have been proven practical and innovative.

Based on the data of residents' income and consumption from Sichuan Statistical Yearbook and China Statistical Yearbook from 2014 to 2018, this paper makes a descriptive analysis on the current situation of residents' consumption in Sichuan Province from three aspects: income expenditure comparison, Engel coefficient and the proportion of each consumption spending. Based on the income and consumption data of urban and rural residents in Sichuan Province, the consumption structure of urban and rural residents in Sichuan Province is analysed from two aspects of marginal propensity to consume and income elasticity of demand by using the results of ELES model. At last, the paper puts forward relevant suggestions from four aspects: exploring ways to increase residents' income, reducing residents' housing cost, strengthening infrastructure construction and improving rural medical insurance coverage.

This paper mainly introduces the key information extracting method from User Equipment (UE) test data, the synchronous algorithm of various interfaces X Detailed Record (XDR) data, the key information filling technique of multi interfaces X Detailed Record (XDR), the analytical procedure based on big data is descript in details. An example is given to explain that more accurate reason for Voice over Long-Term Evolution (VoLTE) call not connected can be found with big data from core network, compared with UE data only. With this method VoLTE problem can be quickly solved, it is very useful for wireless operator to improve the user voice service awareness rapidly.

Social networks of engagement sometimes dramatically collapse. A widely adopted paradigm to understand this catastrophe dynamics is the threshold model but previous work only considered the irreversible K-core pruning process and the resulting kinetic activity patterns. Here we study the network alliance problem as a simplified model of social engagement by equilibrium statistical mechanics. Our theory reveals that the surviving kinetic alliances are out-of-equilibrium and atypical configurations which may become highly vulnerable to single-node-triggered cascading failures as they relax towards equilibrium. Our theory predicts that if the fraction of active nodes is beyond a certain critical value, the equilibrium (typical) alliance configurations could be protected from cascading failures by a simple least-effort local intervention strategy. We confirm these results by extensive Monte Carlo simulations., Revised with modified title, supplementary technical details included, to be published in Europhysics Letters (EPL)

The construction stage of the building project is an important link that takes the longest time, the most complex technology and the greatest impact on the quality level of the green building in the whole life cycle of the building. Therefore, the first step is to start from the construction stage in order to improve the quality management level of green building and realize the maximization of green building quality and environmental, economic and social benefits. This paper focuses on the construction quality management in the whole process of green building planning, design, construction, operation and scrapping system, introduces the idea of green building engineering, studies the construction quality problems of green building, and puts forward a new construction quality management strategy and mode. The new construction quality management strategy and mode can control the quality of green building construction stage, and promote the development and promotion of green building action.

The microscopic mechanism of thermal transport in liquids and amorphous solids has been an outstanding problem for a long time. There have been several approaches to explain the thermal conductivities in these systems, for example, Bridgman’s formula for simple liquids, the concept of the minimum thermal conductivity for amorphous solids, and the thermal resistance network model for amorphous polymers. Here, we present a ubiquitous formula to calculate the thermal conductivities of liquids and amorphous solids in a unified way, and compare it with previous ones. The calculated thermal conductivities using this formula without fitting parameters are in excellent agreement with the experimental data. Our formula not only provides a detailed microscopic mechanism of heat transfer in these systems, but also resolves the discrepancies between existing formulae and experimental data.

As carbonate reservoirs in deeper strata continue to develop, reservoir closure stress has significantly increased, where conventional acid fracturing technology cannot maintain acid-etched fracture conductivity and single well production rates are decreasing more quickly. This study proposes a rock self-supporting, highly conductive acid fracturing technique, where the shielding materials cover a portion of the primary hydraulic fracture surface to block the acid rock reaction. After acid injection, the unetched part will be a bearing surface, which serves as a large area and self-supporting high strength rock. This technique fundamentally changes the existing point support pattern of acid-etched fractures. Experimental results demonstrate that when the closed stress is 25%. When the rock is in the form of a dislocation support, the fluid disperses in the larger void channels, thus effectively maintaining fracture conductivity. The self-supporting acid fracturing technique is useful for increasing the utility of acid fracturing stimulation in deep and ultra-deep wells.

A new type of composite insulator mechanical failure named decay-like fracture happens rather often in UHV AC transmission lines. However, the cause from the point view of electric field is not clear. To investigate the effect of electric field on the occurrence and development mechanism of decay-like fracture, the 3D tower-conductor-insulator models for ZB2-type and CZ34-type transmission tangent towers are established and analysed based on the finite element simulation. The conclusion can be drawn that the tower type has a significant effect on the electric field distribution along the axial direction of the insulator string, and the degradation is more likely to occur in high-voltage side of the B-phase string, which is consistent with the fracture failure investigation. Simulation results also show that the withstand electric field strength is 297kV/m. The distribution of surface potential and electric field under the actual operating voltage provide a new idea for the condition monitoring of the decay-like fracture of UHV AC composite insulators in transmission lines, as well as a reference for further research in the field of decay-like fracture failure.

Comparing with the advantages and disadvantages of the existing target tracking algorithms based on deep learning, a vehicle tracking algorithm based on Yolov2 and GOTURN algorithm is proposed, which is called YOLOv2-tracker vehicle tracking algorithm. The Algorithm is trained and tested by using the collected training set and test set. The results show that the YOLOv2-tracker vehicle tracking algorithm can achieve higher tracking accuracy and faster tracking speed, and can effectively overcome environmental interference. Further analysis of the test results, the algorithm found that there is “errof” phenomenon, the paper discusses and analyzes the causes of this phenomenon, and put forward a reasonable solution. In addition, a “dynamic save” method is proposed to solve the “lost track” problem.

In view of the large amount of geological data, dispersion and the lack of time-consuming and error-prone and other routine statistics, the scope of volcanic collapse in Changbai Mountain area is comprehensively calculated. Analysis of relevant impact factors, such as rock, slope height, slope, human factors, etc. Using UDEC to simulate collapse and then use SPSS software to predict the range of influence and to carry out linear regression. Then analyze its correlation and discreteness, draw the linear regression equation, establish the prediction system of the range of the volcano collapse of Changbai Mountain, and reach the database management and scope.

The battlefield environment that modern electronic information equipment faces during use is very complicated, which has become one of the main characteristics of the informationized battlefield. This article starts with the composition and characteristics of a complex electromagnetic environment, references relevant standards, introduces the definition and composition of a complex electromagnetic environment, and summarizes its characteristics. On this basis, this article introduces the general methods of complex electromagnetic environment tests, including digital simulation tests, semi-physical simulation tests, and full-physical simulation tests. Among them, in the full physical simulation test, an equivalent method is generally used to achieve it. At the same time, this article introduces test evaluation methods, including element-level evaluation and effectiveness-level evaluation. Combining engineering practice, this paper puts forward the difficulties of constructing a complex electromagnetic environment. It is an inevitable requirement for the development of weapons and equipment to carry out comprehensive experimental research on the complex electromagnetic environment of aircraft, and to be able to assess the combat capability of the product. It is also an objective requirement for the development trend of the information military.

Rain streaks degrade the quality of image. Many methods have been proposed to solve single image rain streaks removal recently. However, some methods over-smooth the recovered image. A deep network architecture called Multi-Scale Feature Attention Network (MSFA-Net) is proposed in this paper. We propose a novel basic block structure to exploit the image features, which consists of multi-scale residual learning block and feature attention block. Several basic block structures with a local residual learning compose a group architecture. The outputs of each group architecture are concatenated for final multi-scale feature fusion. Then the features are fed into feature attention block and reconstruction module. Finally, a global residual learning module restore the clean image. Besides, the feature attention block combines channel attention with spatial attention. The proposed MSFA-Net removes the rain streaks which study a non-linear mapping relationship between the rainy and clean image from synthesized dataset. Through comparing with other state-of-the-art algorithms, our algorithm performs better for both synthesized rainy image data and real rainy image data.

There are some problems in the recognition of financial aid objects in Colleges and Universities, such as being interfered by human factors and weak in quantification. The existing relevant research is mainly about policies, systems and measures, while the application of artificial intelligence in precise financial aid in colleges and universities needs further research. And the new generation of AI technology in deep learning can automatically extract data characteristics, and it can be used in the accurate identification of poor students, with the advantages of having no need for human intervention. This model designs and optimizes the precise identification model of university funding objects based on artificial intelligence technology, and achieves a satisfactory recognition rate, which explores a new technical way to improve the equity of education funding and its efficiency exploration.