In the master-slave collaborative navigation problem of UAVs in satellite rejection environment，it is difficult for the host to perform autonomous navigation and cannot meet the performance requirements of a single machine in collaborative navigation to provide accurate navigation information for the slave and the consistency of trajectories between multiple machines needs to ensure consistent states，and differences in the perception of the same state among multiple machines can lead to divergent trajectories. To address these issues，firstly，combining polarization orientation and optical flow positioning technology，a self navigation method that integrates polarized light/optical flow/IMU is designed to meet the performance requirements of a single machine in a multi machine system. Secondly，based on the particle filter，a relative state consistency filtering method is designed to improve the consistency between the slave and the host state，so as to improve the trajectory consistency between multiple machines. Finally，a simulation and outdoor real machine experimental platform is built. The experimental results show that，compared with the traditional Kalman filtering method，the root mean square error of position is reduced by 57%，the velocity error is reduced by 42.5%，and the course error is reduced by 73%. The outdoor experiment verifies that the method can effectively improve the consistency of the relative state and trajectory between multiple aircrafts.

A high-dynamic image generation algorithm based on multi-factor fusion of long and short exposures for the actual use scenarios of the camera is proposed in this paper，which solves the problem of narrow dynamic range in the actual shooting process of the camera and improves the measurement effect of the camera. Starting from the image data of long exposure and short exposure，the algorithm comprehensively considers multiple weight factors such as gradient，saturation，brightness，etc.，and generates and fuses a high dynamic image based on multi-scale sampling，which effectively overcomes the problem of excessive dynamic range and generates a more realistic and detailed image. Experimental results show that compared with the traditional algorithm，the proposed algorithm can effectively reduce noise and achieve better imaging results while maintaining image details，and the imaging effect can be improved by about 20%.

In order to solve the problem of low contrast，texture blurring and illumination uniform in the underwater images，a new visibility enhancement algorithm which is based on the underwater imaging model being different from the monochrome atmospheric scattering model is proposed in this paper. The algorithm works in the l color space that has little relevance among channels. Firstly，NSCT transform is used to decompose the l channel. And then the low-frequency band is proposed based on the dark channel prior，among which the key is to design a method for estimating the intensity of light sources under non-uniform lighting condition. The directional bandpass subbands are proposed by convolution with Laplace operator. Finally，the output image is obtained through anti-NSCT transform and RGB color space convertion. The experiment using the images captured under different conditons is conducted to compare the performace of the proposed algorithm with other ones. And the results show that the proposed algorithm in this paper has good performance in eliminating nonuniform brigntness，improving contrast and sharpening the targets，and has certain advantages.

Conventional speech signal keyword extraction often uses graph neural network algorithms，which achieves keyword extraction through the representation of key information feature vectors. However，this method lacks recognition of speech signal scale feature components，resulting in poor quality of final keyword extraction. A speech signal keyword extraction method based on autoregressive pre trained language model is proposed. Based on the waveform changes of the speech signal in the time domain，the signal is denoised，and the scale feature components are obtained by multi-modal decomposition of the signal. From this，the word vector features of the speech signal are constructed，and the semantic vector parameters of the entire speech signal are obtained by combining clustering algorithms. The semantic vector parameters of the initially selected keywords are then deduplicated. An autoregressive pre trained language model is introduced to calculate the similarity between candidate keywords and semantic vectors of speech signals. The keyword extraction of speech signals are achieved. The experimental results show that within the range of 5~30 keywords，the extraction recall of the proposed method remains above 80%. The proposed method can effectively improve the quality of extracting keywords from speech signals，is easy to implement，and can be widely applied in the field of speech signal processing.

In order to solve the problems of low emissivity of the ground calibration blackbody，large differences in radiation characteristics between the far and near fields，the inability of the calibration body to adapt to the calibration of loads with different imaging resolutions，and the inconvenience of setting up and withdrawing from the field when the space-based infrared load carries out radiant temperature measurements of typical geographic targets under the conditions of the outfield，a high emissivity calibration plate and a low emissivity calibration plate are developed. A ground temperature difference type calibration blackbody with high emissivity，uniformity and stability is developed，and high emissivity calibration plates and low emissivity calibration plates are made. A method is proposed to realize the absolute radiation temperature calibration of infrared load based on the double temperature expansion source method，and the relative radiation calibration is realized by the image gray scale transformation of far and near ground field using the high emissivity calibration plate and the low emissivity calibration plate. In order to verify the validity of the method，the radiant temperature data measured by a high-precision radiation thermometer on a typical feature target in the near-field are used as the basis for comparing the present method with the existing method based on a large-size surface-source blackbody for realizing in-situ radiant calibration，and the evaluation of the measurement uncertainty is carried out. The results show that the maximum deviation of the radiant temperature calibration of this method is 0.31 ℃，and the measurement uncertainty evaluation result is 0.46 ℃（k=2），which is a good calibration accuracy. Meanwhile，the calibration plate has the features of customizable size，flexible deployment and withdrawal，low cost，etc. It can be applied to the on-site radiometric calibration of space-based infrared loads with different imaging resolutions，which can provide certain technical support for the assessment of target stealth effectiveness，infrared simulation modeling，target characterization and other work.

The current centering and edge grinding equipment did not have the ability to get the parameters online，and manual grinding could cause the verticality of the positioning platform to exceed the tolerance. Therefore，it is particularly important to control the center deviation of the optical components and the concave aperture accuracy after optical centering. This paper designed a visual online testing device for centering. By adding the functions of online measurement of caliber and calculation of dragging edge feed amount，a visual detection device was designed and installed on the equipment to solve the problems of damaged surface smoothness of components，excessive verticality and caliber of the positioning platform caused by manual modifications. The actual test results show that the measurement accuracy of eccentricity，verticality，and concave diameter of the parts reaches 0.01 mm，which meets the practical application requirements of optical processing. By the visual online testing，the machining of the outer circle and positioning platform of the parts can be completed in one clamping process，upgraded the concave aperture accuracy of the components，reduced the manual grinding processes，and improved the product qualification rate and machining efficiency. A reliable and efficient solution for the centering technology of optical processing was provided.

The traditional transmission line detection is easy to be affected by foggy weather，and the detection accuracy and efficiency are low. In this paper，a high-precision and high-efficiency detection method is proposed，which integrates multi-module de-fog network，network partitioning strategy and PowerNet network model. Firstly，based on the end-to-end learning，a multi-module defogging network is designed to solve the problem of low detection accuracy of transmission line defects caused by fog. Then，in order to improve the inspection efficiency of edge technology，a network partitioning strategy based on binary particle swarm is designed. On the basis of these，PowerNet network model is proposed to solve the problem of low accuracy of transmission line defect detection. Finally，the method proposed in this paper is validated and analyzed by experiments. The experimental results show that the proposed method has high accuracy and real-time defect detection，and its accuracy and efficiency can reach 99.3% and 28 ms photo，respectively. It can be seen that the method proposed in this paper has high engineering practical value.

In this paper，the distributed abnormal vibration monitoring array in the distribution network is constructed based on the low reflectivity grating，and the phase signals demodulated by adjacent sensors are decomposed respectively. Then，the dynamic detection eigenvalues of spatial scale are constructed by the Hilbert transform of each mode function. When the distribution of eigenvalues fluctuates greatly，the abnormal vibration location and characteristics can be quickly and accurately located. The results show that the proposed method can quickly extract abnormal vibration features and effectively monitor the occurrence of abnormal vibration events with a period of 0.5 s，so it has certain application value in the monitoring of distribution network lines.

Thickness，as a key parameter of semiconductor films，directly affects their working performance. However，for semiconductor substrates and wafers with micrometer thickness，which are more widely used in the market，the existing measurement methods still have problems such as insufficient accuracy，poor stability and slow measurement speed. In order to solve these problems，this study constructs a high-precision，miniaturized，and low-cost semiconductor thin-film thickness adaptive measurement and correction system based on the principle of infrared interferometric thickness measurement and the principle of laser surface shape detection. Experimental results show that compared with the traditional infrared interferometric thickness measurement system，the film thickness measurement correction system constructed using this paper can improve the accuracy of the measurement results by about 80% and the stability by about 60%，even when measuring the thickness of semiconductor thin films with a tilted component，the same effect can be achieved. The system achieves accurate measurement of the thickness of a single film layer and is able to adaptively correct errors introduced by mechanical vibrations or placement errors that result in small inclinations of the film，providing a solution to the challenge of measuring the thickness of micrometer-sized films with existing techniques.

The Rydberg atom has a wide frequency band and highly sensitive microwave reception capability. The microwave electric field measurement technology developes based on the electromagnetic induced transparency effect of Rydberg atom has become a hot technology at the forefront of electromagnetic wave electric field measurement. Different from the research frequency band（<10 GHz）of most experimental works，this paper explores the ability of the Rydberg atom to receive centimeter waves. In our experiment，three different microwave resonance frequencies are measured，and the intensities of electric fields are calibrated by microwave coupling Rydberg-states induce AT splitting，and the sensitivity and resolution are detected amplitude modulation scheme. According to the Allan variance curve calculated by continuous acquisition，the measurement sensitivity of better than 50 nV·cm-1/Hzand the field strength resolution better than 20 nV &#x22C5; cm-1 are evaluated. This work strengthens the further understanding of the detection capability of Rydberg atom in the full microwave band，and provides an important reference for the application of Rydberg atom in high-frequency microwave.

Surface defects are important criteria for evaluating the surface quality of optical components，mainly including scratches and pitting. To accurately evaluate the surface quality of optical components，this paper proposes a detection and evaluation method based on machine vision. Based on the principle of dark field scattering imaging，an image acquisition system is built，and a defect recognition system is designed using digital image processing technology. A digital evaluation system based on quality inspection standards has been developed to detect surface defects of optical components. The experimental results show that the evaluation results of the detection method on the surface quality of optical components are consistent with the manual evaluation results，and the matching degree reaches 100%，which verifies the reliability of the method.

Tactical laser weapon system is an effective method to destroy unmanned aerial vehicle（UAV）. The destruction efficiency simulation of tactical laser weapon system anti-UAV is presented in this paper. The research background of the tactical laser weapon system is proposed firstly，and then the modeling and simulation of tactical laser weapon anti-UAV destruction is derived by means of deduce the operation flow and destruction theory. Finally，the relationships between destruction distance of system and influence factors are discussed. The results show that the UAV flight altitude and visibility are seriously affected the destruction efficiency of the tactical laser weapon system，and simply increasing laser power has no apparent results on destruction efficiency，the coherent combining and spectral combining technique can improve the laser beam quality and correspondingly enhance the performance of the tactical laser weapon system，the laser weapon’s destruction distance increased by 24%. The study is useful to increase the combat effectiveness of the tactical laser weapon system.

Laser angle deception jamming is an effective means to anti semi-active against laser guided weapons. In order to improve the anti-jamming efficiency of the laser seeker，based on the study of the real-time wave gate technology of the laser seeker，two control strategies of symmetrical contraction and biasing contraction for the wave gate are introduced. A calculation model of the effective probability of the jamming laser signal which accepted by the laser seeker wave gate，the jamming laser signal appears in the wave gate which obeys the normal distribution is also taken into account. The influence of jamming laser signal advance time，the wave gate width，the control contraction ratio and biasing value on the effective jamming probability is studied by using the method of control variables. The simulation results show that when the advanced synchronizing interference is adopted，there is a corresponding optimal jamming lead for different wave gate widths which can achieve the maximum effective probability of jamming signal. By setting reasonable isometric contraction coefficient and biasing value for the wave gate can effectively reduce the jamming probability and improve the anti-jamming performance of the seeker，Which can provide technical support for the research on the effectiveness of laser angle deception jamming.

The authenticity of laser interference effect testing is high，but the cost is high. Therefore，laser interference effect testing under different interference situations is meaningful for evaluating the anti-interference ability of infrared imaging systems and the interference ability of laser equipment. In this paper，laser interference experiment under far-field conditions is done. And then，the far field situation of laser jamming is analyzed. It is pointed out that the spot distribution on the detector surface of imaging system is plane wave diffraction. Then the near field situation of laser jamming is analyzed. The wavelength and repetition rate of interference laser are simulated by the same parameters laser. The incidence angles of interference laser are simulated by three-dimensional turntable. The attenuation of power of interference laser beam is realized with attenuator. The divergence angle of laser beam is controlled by laser beam expander optical system. The spot distributions on the detector surface of the near field situation and the far field situation are quite similar when the truncation ratios of Gaussian laser beam are greater than three times. Finally，the structure chart of measurement system of laser interference effects are given. The research is useful for the laser disturbing imagining system test. It also can be applied to design of laser jamming test system.

A compact erbium glass laser for laser ranging is designed and its output characteristics are experimentally studied. Experiment results show that using a 940 nm semiconductor laser pump，when the output power is 12 W，the pump pulse width is 1.6 ms，the passively Q-switched crystal T0 is 92%，the output mirror reflectivity R is 90%，the output will be the pulse laser of center wavelength 1 535 nm，the pulse width will be 5ns，the single pulse energy will be 502 J，the repetition frequency will be 10 Hz. The compact laser structure size can be controlled at 32 mm × 12 mm × 8 mm，weight< 10 g.

With the rapid development of renewable energy and the widespread application of electric vehicles，the application of battery energy storage systems in photovoltaic charging stations of electric vehicles plays a crucial role in improving charging efficiency，stabilizing the power grid，and achieving energy balance. This article conducts research on the evaluation methods of battery energy storage systems in photovoltaic charging stations of electric vehicles，and proposes a comprehensive evaluation method for battery energy storage systems based on fuzzy evaluation. Firstly，a series solar cell structure was proposed and a mathematical model was established to analyze the charging and health status of energy storage batteries. Based on the battery model of the charging station energy storage system，a future life prediction method for battery detection in the photovoltaic charging station energy storage system of electric vehicles is further proposed. Finally，using the battery detection results，a comprehensive evaluation of the battery energy storage system is conducted based on the fuzzy evaluation method. This method can provide comprehensive and accurate evaluation results for battery energy storage systems，helping decision-makers make scientific decisions.

To achieve real-time detection of optical cables and vibration risk assessment，it is necessary to reliably identify fiber optic vibration patterns. Therefore，a performance optimization method for fiber optic vibration deep learning algorithm in pattern recognition was studied. This method utilizes distributed fiber optic sensors to collect vibration signals，and obtains light intensity signals based on the phase information of the Rayleigh scattering light signals from the sensors，converting them into electrical signals；After selecting the windowing and framing method to process the signal，the local feature scale decomposition method is used to obtain multiple feature parameters of the signal，which are then input into a one-dimensional convolutional neural network. The model learns and outputs the fiber vibration pattern recognition results. The test results show that this method can obtain the frequency changes of different vibration signals. After windowing and frame processing，the frequency of the vibration signal fluctuates between 0~400 Hz，and the phase does not exceed 20 rad. It accurately extracts the peak to peak value and spectral characteristics of the signal’s main lobe，and presents its trend with frequency，clearly showing the intensity and distribution of different frequency components. The significance lies in the effective completion of fiber optic vibration pattern recognition，which can determine the depth of fiber optic cable vibration signals and assist in optimizing the effectiveness of fiber optic cable operation fault detection and risk assessment.

The variability and fluctuations in distributed photovoltaic power generation due to changes in weather，seasons，and pollution or shading of photovoltaic components pose challenges. Addressing the randomness and volatility issues in photovoltaic power generation，this paper proposes an optimization of the rescheduling strategy for a photovoltaic power generation system based on big data mining. The objective is to reduce the probability of branch flow overload，utilizing the overload probability index to characterize system risks. Data from photovoltaic sensors，including the all-day light sensor Lufft WS501 and precision spectral average radiometer Eppley，are collected and recorded. Leveraging big data mining techniques，an iterative optimization scheduling model is employed to mitigate the risk of power congestion. Simulation verification on the IEEE39 nodes demonstrates that，compared to traditional scheduling，the overload probability through branch 26 decreases from 0.444 0 to 0.013 8，and through branch 15 decreases from 0.447 0 to 0.032 7，significantly reducing the overload probability. This approach accurately collects photovoltaic generation data，improves the reliability of the power system，and provides an effective method for addressing the randomness and fluctuations in photovoltaic power generation，holding crucial significance for the stable operation of power systems.

In order to avoid the time-consuming，expensive equipment and complicated steps of traditional lithography and to solve the neck-pinching technical problems in the field of microelectronics fabrication，liquid-surface self-assembled polystyrene（PS）nanosphere is utilized to prepare nanostructured arrays in this paper. Low-cost，large-area，size-controllable nanostructured arrays on n-type silicon substrates are successfully prepared and successfully employed to photodetector. It is shown that the introduction of nanostructures can effectively improve the anti-reflective property of silicon surface，and the reflectivity is reduced by up to about 96% compared with that of planar silicon. The light/dark current ratio of the nanostructured device approximately reached one order of magnitude. At +5 V bias，the responsivity of the nanostructured photodetector is improved by 10.1-fold and 3.7-fold compared with that of the planar one at the wavelengths of 365 nm and 405 nm，respectively. At +5 V bias，the external quantum efficiency（EQE）of the nanostructured photodetector is 10.1 times higher than that of the planar one at 365 nm and 3.7 times higher at 405 nm. This provides a low-cost，efficient，reliable and simple way to prepare high-performance nanostructured optoelectronic devices.

Based on the theory of all-dielectric metamaterial absorber，a new type of micro-cavity structure absorber for refractive index sensor is designed. Numerical simulations show that the absorber achieves 99.91% and 99.96% absorption at 539 nm and 673 nm，respectively，and is well matched with the Impedance of free space. The full width at half height is 11 nm and at half width is 5 nm，respectively. The absorption mechanism based on Mie resonance is explained by the electromagnetic field distribution at resonance wavelength，and the characteristics of the absorber which is insensitive to polarization angle are analyzed. The changes of absorption peak value and absorption frequency band are studied by changing the structure parameters of the absorber，and the sensing performance of the structure is analyzed. The FWHM，sensitivity S，FOM and FOM * of the two-channel absorber as the refractive index sensor are calculated based on the two absorption peaks as a function of the width of the ring nano-cavity. When the refractive index of the surrounding medium is changed from 1.00 to 1.08，the sensitivity s value can reach 175 nm/RIU and 212 nm/RIU. The sensitivity s value of peak 1 increased from 146 nm/RIU to 201 nm/RIU，while the sensitivity S value of Peak 2 remained stable at about 200 nm/RIU when the width of the annular nano-cavity increased from 20 nm to 100 nm. The FOM value of peak 1 increased slowly from 8.2 to 23.6，while that of Peak 2 was 65.3 because of its narrow frequency band. In addition，the FOM * values of the dual-band absorber at 539 nm and 673 nm resonant wavelengths are calculated to be 17 265 and 59 902，respectively. The absorber can achieve double-channel high absorption in the visible band，and has the advantages of high sensitivity and narrow band when used as a refractive index sensor. The microcavity structure of the absorber has potential applications in future Environmental monitoring and photoelectric detection.