The magnetohydrodynamic (late-time) electromagnetic pulse (E3) generated by high-altitude nuclear explosions has a serious impact on national key infrastructures such as the power system. Due to the complex mechanism of late-time electromagnetic pulse generation and many dependent factors, including explosion yield, explosion height, explosion orientation, explosion time, observation point position and soil conductivity, there is no available mature code that can simulate the whole generation process of late-time electromagnetic pulse. This paper introduces the generation mechanism of late-time electromagnetic pulse, discusses the relationship between the electric field of late-time electromagnetic pulse and the change of explosive yield of nuclear devices, explosive height, and atmospheric conditions. The electric field peak of E3A increases linearly with the explosion equivalent, while the electric field peak of E3B shows obvious saturation effect with the explosion equivalent increase. The current status of the simulation code of late-time electromagnetic pulse is analyzed, and it can provide a reference for further research on the numerical simulation method and code development of late-time electromagnetic pulse.

To improve the low frequency radiation characteristics of the radiating-wave simulator based on transverse electromagnetic (TEM) horn, a novel movable simulator made up of exponential-type TEM horn, two vertical perfect electric conductor (PEC) plates at its aperture, two sloping PEC plates and parallel resistance is designed firstly. The effect of different exponential tapered rates, the height of the two vertical PEC plates, the width of the source port and the parallel resistance at the end of the two sloping PEC plates to the near-field radiation performance of the novel simulator is simulated and optimized by finite-difference time-domain (FDTD) method. The radiation characteristics of the optimized novel simulator and its arrays is also given. The calculation results show that, the full width at half maximum (FWHM) of the electric field at the testing point which is 3 m away from the optimized novel simulator’s aperture center reaches 18.95 ns, and the optimized novel simulator’s sizes are 6 m×6 m×6.24 m while those of the normal simulator must be 9 m×12 m×6.8 m to get the same low-frequency radiation performance as that of the optimized novel simulator. And higher peak-value of the electric field at the testing point of the optimized novel simulator can be got compared with the normal simulator. In addition, the ratio of the delayed oscillation’s amplitude of the electric field in time-domain at the testing point of the optimized novel simulator to its peak-value is significantly reduced compared with that of the previous studies, while the peak-value of the testing point of the optimized novel simulator keeps high. The electric field’s peak value at the center point in the testing plane of the optimized novel simulator’s 2×2 array model is the largest, and the effective region meeting the requirement of field 6dB uniformity on the testing plane of 2×2 array model has the largest domain; The effective region on the 2×2 array model’s testing plane has the largest horizontal range, followed by 2×1 array model; The effective regions on the testing planes of 2×2 array model and 1×2 array model have the largest vertical range.

The existing heterodyne power combiners are not suitable for applications where input and output signals need to be in the same direction with limited space. To solve the problem, this paper designs a high-power and miniaturized heterodyne power combiner operating at frequencies of 9.3 GHz and 9.7 GHz. Based on the traditional filter-based heterodyne power combiner, the proposed design utilizes an over-mode rectangular waveguide E-plane power combiner. The waveguide filters are parallel and the input ports are also located on the same plane, so that the combiner is suitable for the specific applications. The sizes of the rectangular waveguide are reduced to suppress higher-order modes. Besides, the distance between mode strips is decreased in integer multiples of half-wavelength of the waveguide to compresses the overall length with high power capacity. The combiner has a length of 9.2λ, a width of 1.5λ and a height of 2.8λ, while λ is the wavelength corresponding to the frequency of 9.5 GHz in free space. At 9.3 GHz and 9.7 GHz, the return loss of the combiner is more than 20 dB, its combining efficiency is more than 98% , and the isolation between input ports is more than 20 dB. At microwave pulse breakdown threshold of 80 MV/m, the combiner provides a power capacity of 310 MW.

In the field of computational electromagnetics, the discontinuous Galerkin time domain (DGTD) method typically relies on irregular grid partitioning in model space and high-order polynomial interpolation calculations on elements. When comparing two-dimensional spatial quadrilateral mesh partitioning to triangular mesh partitioning at the same interpolation order, quadrilateral meshing offers fewer degrees of freedom and higher computational efficiency. However, traditional basis function spaces, relying on isoparametric transformations and polynomial tensor product interpolation, only possess low-order completeness on quadrilateral elements. Consequently, their stability and accuracy are significantly influenced by grid distortion. To address this challenge, this thesis proposes a high-order B-spline interpolation DGTD method based on irregular quadrilateral meshes for solving Maxwell's equations. The advantage of B-spline interpolation lies in its high-order completeness on irregular elements, effectively eliminating internal degrees of freedom within the elements. Furthermore, the coefficient matrices of the discrete system for Maxwell's equations also possess exact analytical forms. Analyzing the eigenmodes of cavities and the electromagnetic scattering of wedge structures, thus the maximum allowable time step increasing by 2.5 times and reducing the required unknowns by 25% compared to COMSOL software, the proposed algorithm exhibits notable advantages in terms of higher stability and precision.

Aiming at the application requirements of array antenna with high-power capacity, high efficiency and low profile characteristics, a high-power capacity and high efficiency open waveguide array antenna is proposed and designed. The antenna consists of a compact 16-way waveguide power distribution network, 4×4 rectangular open waveguide unit cells and ceramic sealing radome. By designing the size of the open waveguide and loading E-plane metal bar on the surface of the open waveguide, the electric field distribution on the radiation aperture surface is more uniform, and the radiation gain of the unit cell is improved. The step matching structure is used to realize the size transformation from the output port of the waveguide power distribution network to the interface of the open waveguide unit cell, and the impedance bandwidth of the system is improved. The ceramic radome loaded on the array keeps the interior of the antenna in a vacuum state and improves the power capacity of the antenna. According to the application requirements of X-band high-power array antenna, a 16-element open waveguide array with a center frequency of 9.5 GHz is optimized and designed, the simulation results show that the aperture efficiency is greater than 90% and the reflection coefficient is less than -13.9 dB in the range of 9.25-9.65 GHz. The antenna is processed and tested, the measured antenna reflection curve and radiation pattern at the center frequency are in good agreement with the simulation results, the antenna gain at the center frequency is 21.7 dBi. The overall profile height of the antenna is twice the wavelengths at the central frequency, and the power capacity in vacuum obtained by simulation is 40 MW, the antenna has the characteristics of high power capacity, high efficiency and low profile.

During the commissioning of the primary helium circulator of the high-temperature gas-cooled reactor (HTGR), it could not complete the performance test of the full speed range, because the resistance of the primary loop was lower than the design condition. Based on the theoretical characteristics of the primary helium circulator and similar principles, a method for calculating the commission parameters of the primary helium circulator under different resistance conditions was developed. Combined with the monomeric test operating points of the primary helium circulator, the method was used to accurately calculated the operating point parameters of the cold and hot performance tests of the primary helium circulator, and guide the completion of the full speed and full power performance tests of the primary helium circulator in HTGR. By comparing and analyzing the commission and factory test results of the primary helium circulator, the feasibility of this calculation method is verified, and correction factors for the conversion of working conditions between air medium and helium medium is provided. Through comparing and analyzing the commission and operation data of the primary helium circulator, it can be seen that the commission conditions provided in this article have sufficient enveloping ability, which can cover all operating conditions of the primary helium circulator during the operation of HTGR. This proves that the variable resistance condition commission method of the primary helium circulator meets the performance verification requirements of HTGR, and it can be used to guide the primary helium circulator commission of subsequent HTGRs.

Accurate prediction of the occurrence time and location of coolant boiling is of great significance for safety assessment of Sodium Cooled Fast Reactors (SFR). Based on a two fluid six equation model, conservation equations are constructed for the gas-liquid two-phase flow of sodium. The evaporation-condensation model is used to characterize the interphase mass exchange, and explicit and implicit methods are used to calculate evaporation-condensation model. Constitutive relationships such as Sobolev resistance model, two phase flow heat transfer model, and phase momentum exchange are considered. A porous medium analysis approach suitable for simulating SFR coolant boiling was developed, and comparative verification was conducted using KNS-37 L22 loss of flow experiment data. L29 flow data is used to verify the applicability of the model. The results indicate that the established sodium boiling porous medium analysis approach can effectively simulate the boiling phenomenon. It predicts that the boiling time will be around 6.3 s, which is 0.2 s different from the result of experiment. The overall trend of temperature and flow rate changes are in good agreement with the experimental data.

Light radiation is a crucial component of the energy produced in nuclear explosion, making the study of its space distribution highly significant. This paper presents the derivation of a formula for computing thermal energy induced by the light radiation of a nuclear explosion. The derivation integrates the fireball development laws with the transient energy dynamics of light radiation. The resultant formula exhibits a dependency on several factors, including the height of the explosion, the yield of the explosion, atmospheric attenuation coefficients, as well as the radius and temperature of the fireball. By creating diverse maps and adjusting pertinent parameters, simulating calculations are conducted to elucidate the distribution patterns of the transient thermal energy from nuclear explosion light radiation. Furthermore, the burn injury grading standards are incorporated into the simulation by introducing a search function that autonomously categorizes the injury grading zones on the virtual map. What’s more, neural networks are employed to train the numerical models, aiming to discern the correlation between the parameters associated with nuclear explosions and the injury grading radius on the map. This innovative approach enables direct prediction of the injury grading radius based on nuclear explosion parameters, thus significantly shortens the calculation process.

Linac control network is the reference for optical axis transmission. According to the layout of key equipment in linac tunnel, the control network is laid out on the ground and wall of the tunnel. Tracker is the main instrument for measuring linac control network, and the angle measurement error is a key factor affecting the accuracy of the tracker. Based on the layout of wall and ground network points and the measurement plan, laser tracker’s angles were decomposed as horizontal and vertical angles in all measurement states. Then, through the linkage testing of high-precision CMM and laser trackers, all the decomposed angles were calculated, and the calculated values are used to correct the tracker’s measurement angles. Based on the test results, regardless of the measurement state, measured angle of the tracker is larger than the calculated value. The ground network points’ vertical angle deviation and horizontal angle deviation almost equal to the nominal accuracy of the tracker. When the horizontal angle of wall points exceeds 15°, deviation increase significantly, and the deviation should be corrected while measuring the wall network points.

The Institute of High Energy Physics of the Chinese Academy of Sciences completed the research and development of the high quality factor 1.3 GHz superconducting cryomodule in June 2023, taking the lead in the world to realize the technical route of the medium temperature baking. Eight 1.3 GHz 9-cell superconducting cavities with the medium temperature baking process are integrated. During the integration test of the cryomodule, the temperature of the high-order mode (HOM) coupler of the superconducting cavity was abnormal, which made the superconducting cavity unable to work stably under high gradient. In this paper, the electromagnetic analysis of the high-order-mode coupler is carried out by the HFSS software and eigenmode Solver in CST software and the thermal analysis of the high-order-mode coupler is carried out by theory and Ansys Workbench software. Combining with the high-power experiment of cavity, the cause of the abnormal performance of the superconducting cavity was found. Also, the cooling structure of the HOM coupler in the superconducting cavity was further optimized to solve the instability of the superconducting cavity under high gradient in the module.

A fluorescence target historical image data storage system based on MongoDB database was constructed to address the issues of historical image data storage: continuously increasing data generated by the system, and slow historical data retrieval speed of the Heavy Ion Research Facility in Lanzhou (HIRFL) fluorescence target. To save, observe and analyze fluorescence target beam images, this article establishes an EPICS based historical data archiving system to obtain PV (Process Variable) data of fluorescence target images. The obtained data is stored using MongoDB database sharding technology, and the image conversion and web page implementation are achieved through the Django framework. Image classification algorithms are applied in the system to improve data read and write speed. This system can stably obtain, store, and observe fluorescence target beam history images on HIRFL, providing convenience for beam analysis and tuning work.

Compared with two-dimensional fan-beam and parallel-beam CT systems, cone-beam X-ray CT has advantages such as fast scanning speed, high X-ray utilization efficiency, consistent axial and horizontal resolution of reconstructed images, and is the focus of current industrial CT technology development. However, the imaging quality is affected by the presence of scattered radiation. To reduce the impact of scattered radiation on image quality, this paper proposes a new scatter correction method based on a slanted-hole scatter correction plate. The principle and implementation of this method are thoroughly investigated. By acquiring the raw scan data and the scan data after using the slanted-hole scatter correction plate, scatter field data is obtained using interpolation and smoothing techniques. Then, by subtracting the scatter field data from the original data and performing reconstruction, scatter-free CT images can be obtained. Compared with the grating-based scatter correction plate method, the results show that in cone-beam CT scans of turbine blades, the contrast-to-noise ratio of typical regions (the cooling channels within the blades and the inner walls of the blades) is improved by 14.2% and 56.8% respectively with the slanted-hole scatter correction plate method, whereas with the grating-based scatter correction plate method, the same positions only show an improvement of 5.6% and 27.6% respectively. This validates the superiority of the slanted-hole scatter correction plate scatter correction method.

To investigate the motion law of the plasma sheath in a dense plasma focus (DPF) device and the influence of related design parameters, this paper uses a self-developed FOI program to conduct two-dimensional magnetohydrodynamic simulation of the plasma sheath motion process and focus formation process in the Mather type discharge chamber structure, and obtains results similar to the visible light experimental images of the Lawrence Livermore National Laboratory in the United States. At the same time, the influence of different pressure, current, anode radius and cathode-anode gap on the motion law of the plasma sheath is explored. The calculation results show that the plasma sheath will compress the gas radially with a certain degree of curvature, which is one of the reasons for the instability phenomenon; the axial velocity of plasma sheath is inversely proportional to the square root of pressure, and is proportional to the current. The larger the anode size of the device, the smaller the axial velocity of sheath. To increase the current, it is necessary to extend the anode length to match the focusing time with the current peak. The gap between cathode and anode has little effect on the axial motion process of plasma sheath near the anode.

Based on the practical needs of the simulated hail impact test of photovoltaic panels, a matching study between the reluctance coil launcher and the pulse power supply was carried out. The applicability of the two basic pulse discharge circuits, the current-continuing front type and the current-continuing back type, is theoretically analysed, and the results show that the current-continuing front type pulse discharge circuit has better matching with the launcher. In view of the reverse braking problem of the residual pulse current during the launching process, an energy-discharge circuit that can quickly consume the residual pulse current is designed to improve the topology of the current-continuing front type pulse discharge circuit to further improve its applicability, and it is verified with the help of simulation. The results show that the energy-discharge circuit can quickly consume the reverse pulling force effect of the residual pulse current, and the launch performance of thrown body is significantly improved, the exit speed of the thrown body is 50.8% higher than that of traditional discharge circuits, and the system emission efficiency is increased by 127.5%. This study can provide reference for the application of the reluctance coil launch technology in the field of simulated impact test.

Fiber lasers in the 2 μm band are widely applicated in biomedicine, environmental monitoring, nonlinear frequency conversion, as well as in laser radar and laser communications. However, the power scaling of lasers in this band are limited by low quantum efficiency, high thermal loading, and several nonlinear effect. In this study, we propose a higher-order phase modulation technique based on inverted probability tuning sequences to suppress the SBS effect, a signal laser time-domain stabilization control technique with multi-stage link auto-feedback to enhance the SRS threshold, and a fiber-based transverse-mode purification technique to control the beam quality. Finally, a two-stage main amplification structure is used to achieve an ultra-narrow linewidth near-diffraction-limited output of 1 kW at 1 950 nm, with an optical-to-optical conversion efficiency of 55.6%, linewidth of 3.8 GHz, beam quality M2 factor of about 1.2, and no transversemode instability effect.

The combination of deep learning technology and adaptive optics technology is expected to effectively improve the wavefront correction effect and better cope with more complex environmental conditions. The research progress of applying deep learning in the direction of wavefront reconstruction and wavefront prediction is detailed, including the specific research methods and corresponding neural network structure design adopted by the researchers in these two research directions. The performance of these neural networks in different practical application scenarios is analyzed, the differences between different neural network structures are compared and discussed, and the specific impacts of the structural differences are explored. Finally, the existing methods of deep learning in these two directions are summarized, and the future development trend of the deep integration of deep learning and adaptive optics technology is also prospected.

In recent years, the field of quantum information technology has experienced rapid growth, with a particular focus on electromagnetic sensors that utilize Rydberg atoms. Rydberg atoms, characterized by their high energy states, have garnered significant attention due to their highly sensitive response to external fields. These atoms offer several advantages, including self-calibration capabilities and direct traceability to the International System of Units (SI), which make them exceptionally suitable for applications in radio sensing and detection. Since Shaffer and others made a breakthrough in measuring microwave electric field intensity using the electromagnetic induced transparency effect of Rydberg atoms in 2012, the sensitivity and uncertainty of measuring microwave electric field intensity have significantly surpassed those of traditional microwave measurement results. Over the past decade, research centered around new theories and technologies, such as Rydberg atom superheterodyne technology, has enabled the measurement of electromagnetic wave frequency, polarization, phase, amplitude, and other parameters. Related engineering technologies are also experiencing significant growth, expected to have a disruptive impact on traditional radio technology. This comprehensive review aims to summarize the research progress in the field of Rydberg atom-based radio technology over the past ten years. It will start by examining the underlying principles of detection and then proceed to outline the developmental trajectory of this domain. Finally, the review will provide insights into the future trends and potential directions for the evolution of this technology.

Scatter-shift ghost imaging edge extraction methods require multiple sampling of the object to obtain a high quality edge map. To solve the problem of many samples and long time when extracting the edge of the object by scatter-shift ghost imaging, convolutional neural network is adopted to the edge extraction experiment of ghost imaging. Firstly, the unknown image is irradiated by Walsh scattering, the sampled signal collected by the barrel detector is input to the ghost imaging edge extraction network as the image feature information, finally the edge information map of the detected object is directly outputted by the trained network, and the output of the convolutional neural network is optimized by using the non-maximum value suppression algorithm. The experimental results show that for the reconstructed object of 128×128 pixels, the signal-to-noise ratio and structural similarity index of the ghost imaging edge extraction network output edge pattern are 5 times and 2 times higher than that of the scatter-shift ghost imaging respectively when the sampling number is 1600, which successfully improves the quality of the ghost imaging edge extraction under the low sampling rate and reduces the sampling time. The ghost imaging edge extraction scheme using convolutional neural network is conducive to fast and high-quality edge detection of ghost imaging in practical applications of object recognition and security inspection.

Circularly polarized radiation in the X-ray region is required by many experimental applications. Such demands can be met at FEL facilities by mounting few helical undulators after the planar undulators. For obtaining high degree of circular polarized radiation, the reverse taper technique is often used, where in the planar section, significant bunching of the electron beam is accumulated while the output linear polarized radiation power is suppressed. This paper investigates the technique of spatially separating the background weak linear radiation by first transversely deflecting the electron beam with dipole kicker magnets in the planar undulator section and correcting its orbit and traveling direction before entering the helical undulators. The resulting linearly and circularly polarized radiations will propagate in slightly different directions and can be separated spatially. Numerical simulations are performed to analyze the impact of electron beam deflection on the radiation properties.

Among free electron laser (FEL) schemes, the Harmonic Lasing Self-Seeded (HLSS) scheme can reduce the radiation bandwidth of self amplified spontaneous emission (SASE) and improve the spectral brightness of FEL in X-ray band. The principle of HLSS has been verified at FLASH, PAL, European XFEL and other laboratories. The HLSS scheme can improve the coherence of a SASE FEL without additional hardware, so it can be easily applied to domestic facilities under construction or running in SASE scheme. In this paper, the principle of the HLSS scheme to achieve narrow bandwidth is summarized, and the quantitative conditions of the undulator parameters are given. Typical parameters of Shenzhen superconducting soft X-ray free electron laser are used to simulate the HLSS scheme. Simulation results show that the bandwidth in the HLSS scheme can be reduced to about 1/2 of the bandwidth in SASE, and the spectral brightness can increase to about 2 times, when the output wavelengths are 4.5 nm and 6.75 nm.

A circuit model is established to analyze the possible cascading failure process of group prefire in 28-stage linear transformer driver (LTD). The results show that when a 4-stage group is triggered by mistake, it will cause the entire 28-stage LTD chain action, and the generated fault voltage cannot be limited by the isolation switch at the output of the 28-stage LTD. In addition, when the faulty group is located downstream of the LTD, the voltage propagating upstream along the transmission line will produce a peak field strength on the transmission line of the lower impedance of the first group that is much higher than the normal operations, which may cause the insulation failure of the device. By appropriately increasing the gap between the inner and outer conductors of the transmission line of the first group, the peak field intensity caused by the cascading failure can be weakened without affecting the output amplitude and waveform of the load current, thus improving the reliability of the operation of the device.