319 results about "Reflection loss" patented technology

Some implementations described herein provide techniques and apparatuses provide a semiconductor device including a photonics device having a backside transmissive region and methods of manufacturing. The semiconductor device includes a first semiconductor device stacked over a second semiconductor device, where the first semiconductor device includes a photodiode structure and the second semiconductor device includes the backside transmissive region. The backside transmissive region, which is below the photodiode structure of the first semiconductor device, includes a trench structure having highly reflective structures and / or properties to maintain an optical power of light waves propagating through the backside transmissive region. An absence of structures within the trench structure lessens a likelihood of interferences which may cause a transmission loss (e.g., a reflection loss, an absorption loss, a scattering loss, and / or a mode mismatch loss) relative to another transmissive region that is adjacent to the photodiode structure.

The invention provides an image text recognition method and system based on a mask diffusion model, a storage medium and equipment, and belongs to the technical field of image or video recognition or understanding. According to the method, multi-scale visual features of an image are extracted through a visual encoder, a mask diffusion decoder is combined, a diversified mask strategy and random character replacement disturbance are adopted in a training stage, denoising loss and auto-reflection loss are calculated respectively, and a model is optimized in a combined mode; in the reasoning stage, starting from a full mask state, a complete text sequence is recovered through multi-round iterative denoising. According to the method, the one-way modeling limitation of a traditional autoregression model is broken through, all-around context-dependent modeling is achieved, an autoreversion error correction mechanism and a block low-confidence mask strategy are introduced, and the recognition accuracy and reasoning efficiency in complex scenes such as shielding and fuzzy scenes are remarkably improved. The method provided by the invention reaches a leading level on a plurality of public data sets, and has the advantages of high precision and high speed.

The invention discloses a broadband wave-absorbing material based on an MAX-phase metasurface structure and an impedance matching optimization method of the broadband wave-absorbing material. The wave-absorbing material comprises an MAX-phase ceramic substrate and periodic metasurface structures which are arranged on the surface of the MAX-phase ceramic substrate in an array mode and composed of conductive resonance units. The resonance units are isolated through gaps, an equivalent R1L1C1 series resonance circuit is formed in an electromagnetic field, and a base body is equivalent to a parallel resistor R2. By establishing an equivalent circuit model and optimizing parameters R1, L1, C1 and R2, impedance matching of the wave-absorbing material in a target frequency band can be regulated and controlled, so that the wave-absorbing material at least has an absorption peak of which the reflection loss is less than-10 dB. According to the optimization method, electromagnetic simulation and circuit simulation are combined, optimized circuit parameters are reversely mapped into structure geometric parameters, material preparation is guided, and accurate design and broadband absorption of the wave absorbing performance are achieved.

The invention provides an electromagnetic metamaterial design method based on a deep neural network and a genetic algorithm, and relates to the technical field of metamaterial design. Firstly, an initial structure sample for generating the electromagnetic metamaterial is obtained, and a training set is constructed. And constructing and utilizing the training set to train an electromagnetic metamaterial forward prediction network model. And taking a preset electromagnetic metamaterial structure parameter and set frequency information as the input of the model, and utilizing the electromagnetic metamaterial to forward predict the predicted reflection loss of each frequency point in the frequency information output by the network model. The method comprises the following steps: constructing a structure parameter original population of the electromagnetic metamaterial, solving an optimal solution of the structure parameter original population based on a genetic algorithm of neighborhood dynamic evolution regulation, and in the solving process, obtaining an optimal structure parameter of the electromagnetic metamaterial by predicting reflection loss and calculating an individual fitness value so as to design the electromagnetic metamaterial. According to the design method provided by the invention, the design efficiency and precision of the electromagnetic metamaterial are improved.

The invention relates to the technical field of data processing, and provides a data processing method for testing avionics equipment, which comprises the following steps of: testing a communication optical cable of an aircraft communication system through an optical time domain reflectometer; analyzing the deviation between the loss value of the reflection intensity of the optical signal at each moment and the reflection loss threshold value to obtain the fault probability of the optical cable in the monitoring time period; analyzing the difference between the moment when the optical signal reaches the bending position of the optical cable and the moment corresponding to the peak point, and obtaining the reflection unpredictability of the optical signal in the monitoring time period; combining the light source power difference to obtain an optical cable performance error in the monitoring time period; obtaining the extra loss degree of the optical cable in the monitoring time period by combining the change condition of the temperature data; determining a test result correction coefficient of the monitoring time period; and correcting the reflected signal oscillogram, and testing the communication optical cable. The invention aims to solve the problem that the actual performance judgment of the optical cable is influenced by the fine change interference of various factors in the optical signal propagation process of the communication optical cable.

The utility model discloses a millimeter wave chip packaging structure based on a substrate integrated waveguide and an artificial magnetic conductor, which relates to the technical field of millimeter wave chip packaging, and comprises a chip, a cover plate and a base, the cover plate is in sealing fit with the base, the chip is arranged in the base, and the cover plate comprises a metal flat plate and a plurality of metal columns connected below the metal flat plate and distributed in an array. The metal column is not contacted with the base; the base comprises a frame, an SIW substrate, a direct current feed terminal and a heat sink, a first groove is formed in the center of the upper surface of the heat sink, and second grooves are formed in the two sides of the first groove and communicated with the first groove. According to the utility model, the SIW substrate is adopted to replace a traditional strip line terminal to serve as the millimeter wave input end and the millimeter wave output end of the shell, and the conductive materials on the upper surface and the lower surface of the SIW substrate are titanium tungsten / nickel / gold in sequence, so that the structure is simple, the processing technology is simple, the assembly technology is simple, the price is low, and the like; and the cable also has the advantages of wide working frequency band, large bearing power, small reflection loss, small transmission loss and the like, and also has the advantages of static electricity prevention and lightning protection.

The invention relates to a single-layer waveguide antenna structure which is formed by splicing a PCB main board and a waveguide board, a stand column-mushroom pin composite structure is arranged in the single-layer waveguide antenna structure, mushroom pins are arranged in the PCB main board, stand columns are arranged in the waveguide board, and a transmission waveguide wall, a radiation waveguide wall and a grounding layer are further arranged in the PCB main board; a transmission waveguide cavity and a radiation cavity are also arranged in the waveguide plate; the transmission waveguide wall, the mushroom pin, the transmission waveguide cavity and the stand column form a closed waveguide cavity, and the transmission waveguide wall serves as a bottom H face of the waveguide cavity, so that a single-layer waveguide antenna structure is achieved. By the adoption of the single-layer waveguide antenna structure, the bottleneck problems of high-frequency signal reflection loss, adjacent channel coupling crosstalk and the like caused by a heterogeneous material interface and sensitive assembly tolerance between a traditional waveguide system and a PCB are solved, the number of stand columns in a waveguide plate can be effectively reduced, and the manufacturing difficulty of the waveguide plate can be simplified.

The invention discloses a frequency modulation matching microwave plasma generator, which is characterized in that a circulator is connected between a microwave source and a plasma resonant cavity, microwaves are input from a port 1 of the circulator, and are output to the plasma resonant cavity from a port 2, so that the reflected microwaves are output from a port 3. Meanwhile, a matched load and a power meter are connected to a port 3 of the circulator, microwaves output by the port 3 of the circulator are absorbed by the matched load, most returned microwaves are absorbed by the matched load, a small part of the microwaves are reflected back to the port 3, the microwaves input by the port 3 are output at a port 1, the microwave power borne by the microwave source is reduced, and the stability and the service life of the microwave source are improved. The control system adjusts the output frequency of the microwave source according to the measured power, so that the output frequency of the microwave source is improved, and the output impedance of the microwave source is matched with the working impedance of the plasma resonant cavity again, so that the back-and-forth reflection loss of microwaves is avoided; and the impedance matching adjustment speed is greatly improved.

The invention discloses a laser wireless energy transmission system based on a vertical tracking structure and double-stage global optimization and a control method, and belongs to the technical field of photoelectric energy conversion and intelligent control. Aiming at the technical bottlenecks of high reflection loss of a photovoltaic array and MPPT failure caused by multi-peak characteristics due to non-uniform irradiation in the existing laser wireless transmission system, the invention provides three innovations: (1) designing a two-degree-of-freedom vertical tracking type photovoltaic array, and adjusting the pitch angle and azimuth angle of a photovoltaic panel in real time through a servo motor driving holder; a silicon nitride anti-reflection coating (the refractive index is smaller than or equal to 1.8) and a laser positioning sensor are combined to construct an incident angle deviation dynamic compensation model, and the reflection loss rate is reduced to be lower than 5%; (2) putting forward a global MPPT dual-stage optimization algorithm, dividing a voltage interval based on irradiation non-uniformity in a coarse positioning stage and generating an optimal candidate solution, and adopting an improved particle swarm algorithm (introducing an inertia weight dynamic adjustment mechanism and a local extreme value escape strategy) in a fine search stage to realize that the global maximum power point tracking precision reaches 99.2% in a multi-peak scene; and (3) establishing a reflection loss-power coupling control model, and solving the problems of tracking delay and energy oscillation of a traditional system through angle adjustment of the photovoltaic panel and closed-loop feedback of MPPT (Maximum Power Point Tracking) output. Compared with the prior art, the comprehensive efficiency of the system is improved by 18.6% under the dynamic irradiation condition, and the system is suitable for space laser power supply and mobile equipment wireless charging scenes.

The invention belongs to the technical field of short-wave communication, and particularly relates to a short-wave sky wave channel propagation parameter optimization method based on antenna directivity, which comprises the following steps: acquiring directivity diagrams of a transmitting antenna and a receiving antenna; determining a propagation path of sky wave rays from a transmitting antenna to a receiving antenna by adopting a numerical ray tracing algorithm; according to the propagation path, calculating free space propagation loss, ionosphere loss and ground reflection loss; performing directional gain correction on the signal intensity on the propagation path according to the directional diagrams of the transmitting antenna and the receiving antenna; calculating the total propagation loss according to the free space loss, the ionosphere loss, the ground reflection loss and the antenna directivity correction value; outputting propagation parameters including the total propagation loss, the propagation path, the antenna directivity correction value and the time delay; according to the method, antenna directivity parameters and propagation path loss are dynamically coupled, and the defect that a traditional model ignores the influence of antenna radiation directivity on signal strength is overcome.

The invention discloses a knowledge-driven intelligent metasurface unit rapid design method based on double neural networks. Aiming at the problems that a traditional intelligent metasurface unit design process depends on a large amount of electromagnetic simulation and is complex in calculation and low in efficiency, the invention provides a dual-artificial neural network (ANN) proxy model optimization framework combined with physical knowledge guidance. The method comprises the following steps: firstly, respectively training two neural network models to predict the amplitude and phase response of a metasurface unit of a tuning element in different extreme working states; and then, according to the physical prior knowledge of monotonic change of tuning element parameter change on electromagnetic response, the maximum reflection loss and phase difference under any geometric parameter combination are quickly and accurately deduced, so that double-target optimization design assisted by the agent model is realized. By means of the method, the design efficiency and performance of the metasurface unit can be greatly improved, dependence on traditional full-wave electromagnetic simulation is remarkably reduced, and the effectiveness and superiority of the method are verified through simulation and experimental results.

According to the topological defect cavity based on the valley photonic crystal and the multi-channel logic device comprising the topological defect cavity, an edge mode, protected by topology, in the topological photonic crystal can resist the influence of backscattering, and a brand new thought is provided for solving the problem of high reflection loss of a defect cavity-waveguide direct coupling system under non-resonant frequency. The topological defect cavity is designed based on the valley photonic crystal, and compared with a traditional photonic crystal defect cavity, low reflection loss can be achieved when the topological defect cavity is directly coupled with the waveguide. Besides, the resonant frequency and mode symmetry of the topological defect cavity have flexible tunability, and the multi-channel logic device based on the topological defect cavity and the harpoon type energy beam splitter is provided by utilizing the characteristics of low reflection loss and tunable resonant frequency of the topological defect cavity. According to the logic device, the output result of the harpoon beam splitter can be reserved, meanwhile, more diversified output results are achieved, and the important application value of the topological defect cavity in the integrated micro-nano photonics is reflected.

The invention relates to the technical field of electromagnetic wave absorption, in particular to a composite wave-absorbing heterostructure plate for high-power electronic shelter equipment and a preparation method of the composite wave-absorbing heterostructure plate. The composite wave-absorbing heterostructure plate is of a gradient composite structure, the outer layer is a carbon fiber composite material plate, the middle layer is a composite honeycomb plate filled with carbon nano composite slurry and aerogel, the inner layer is a dielectric material composite plate with a carbon nano composite slurry composite layer on one side, and the carbon fiber composite material plate and the honeycomb plate are connected through a bonding layer. And the cellular board and the dielectric material composite board are connected through a bonding layer. Through the heterostructure design of the carbon nano composite wave-absorbing slurry and the silicon dioxide aerogel, the effective wave-absorbing bandwidth covering 16 GHz with the reflection loss larger than 40 dB and smaller than-15 dB in the ultra-wideband of 2-18 GHz is achieved, the wave-absorbing material is remarkably superior to a traditional wave-absorbing material, and the electromagnetic compatibility requirement of high-power electronic equipment is met.

The invention discloses a preparation method of a beaded cobalt-doped Bi25FeO40 / BiFeO3 nanofiber wave-absorbing material, which comprises the following steps: obtaining a precursor by a standing method, and adding the precursor, bismuth nitrate, ferric nitrate and a spinning aid into a mixed solvent compounded by N, N-dimethylformamide and absolute ethyl alcohol to obtain a spinning solution; spinning the spinning solution by adopting an electrostatic spinning technology to obtain a nanofiber membrane; and finally, drying and calcining to obtain the wave-absorbing material. According to the wave-absorbing material, multiple reflection and scattering of electromagnetic waves are increased, the absorption performance of the electromagnetic waves is improved, due to the fact that a heterogeneous interface is generated after two phases are compounded, the interface polarization loss is further increased, when the thickness of the wave-absorbing material is 2.33 mm, the minimum reflection loss reaches-56.01 dB, the effective absorption frequency band width is 6.32 GHz, and the whole Ku wave band is covered.

This invention discloses a patterned solar glass with a neutral colour anti- reflection coating (ARC) designed to enhance light transmission and optical performance The dual-side structured glass features a smoother side coated with a two-layer anti-reflection coating. The bottom layer is formulated from a silane precursor, a hydrolyzing agent, an acid catalyst, porosity-enhancing agents, and a combination of solvents, achieving a refractive index of 1.4 to 1.5. The top layer is a nano-composite-based AR coating with a refractive index of 1.25 to 1.35 is applied on the bottom layer. The layers are chemically bonded via a roll-coating process, with precise thickness control wherein, the bottom layer ranges from 180–210 nm and the top layer ranges from 140–160 nm. This gradient refractive index structure results in a transmission gain of 2.3% and minimizing Fresnel reflection losses. The coating ensures neutral colour appearance and uniform light transmission, improving omnidirectional performance under varying light angles.

The invention discloses an electric vehicle wireless charging system and an optimization method thereof. The electric vehicle wireless charging system comprises a transmitting end optimization module, a receiving end optimization module, a dynamic matching control module and an intelligent charging management module. The transmitting end optimization module is used for reducing magnetic field leakage in the energy transmission process through magnetic field focusing; the receiving end optimization module is used for detecting relative position deviation of the transmitting end and the receiving end and reducing external magnetic field interference; the dynamic matching control module is used for matching the impedance of the transmitting end and the receiving end in real time to reduce reflection loss; and the intelligent charging management module is used for implementing a segmented charging strategy according to the battery state and executing safety protection. By applying the technical scheme of the invention, the problems of high reflection loss caused by serious magnetic field leakage and impedance mismatch and low overall charging efficiency caused by lack of intelligent charging control in the existing wireless charging system of the electric vehicle can be solved.

The invention discloses an acoustic resonance assisted nano-SiO2 coated flaky carbonyl iron powder as well as a preparation method and application thereof, the flaky carbonyl iron powder is taken as a main raw material, an acoustic resonance assisted modification technology is adopted, a compact and uniform SiO2 coating layer is constructed on the surface of the flaky carbonyl iron powder, and an FCIP coated SiO2 composite material with a core-shell structure is prepared. According to the coating treatment, the oxidation resistance and corrosion resistance of the carbonyl iron powder are remarkably improved, electromagnetic parameters of the carbonyl iron powder are regulated and controlled, and particularly the wave absorbing performance of the material in a low-frequency wave band is enhanced. An electromagnetic parameter test and reflection loss calculation show that the obtained composite material has better impedance matching characteristics and lower reflection loss value. In practical application, a coated sample shows higher thermal stability and lower surface corrosion rate, and the service life of the material in a complex service environment is effectively prolonged. The method is simple and convenient in process, low in cost and suitable for being applied to functional material systems such as electric conduction, wave absorption and electromagnetic interference prevention on a large scale.

The invention discloses a NiZnCu ferrite composite metasurface with a P wave band broadband and a low RCS (Radar Cross Section) based on 1-bit coding. Firstly, the impedance matching condition of the ferrite is improved through Mn < 2 + > ion doping, so that when the thickness of the NiZnCu ferrite is 2 mm, effective wave absorption with the reflection loss smaller than-10 dB is achieved within the range of 0.25 GHz to 1.0 GHz. Furthermore, NiZnCu ferrite is combined with a sandwich structure (square metal sheet / FR4 / metal back plate, 6 mm) to construct a 1-bit coding metasurface structure unit, optimal array arrangement is obtained by adopting a genetic algorithm, and directional reflection is reduced to the greatest extent. And finally, full-wave simulation software is used for verification, and the result shows that the NiZnCu ferrite composite coding metasurface can further reduce the RCS by using the diffuse reflection effect on the basis of keeping the ferrite wave absorbing effect, the range of the RCS is 0.50 GHz-0. 75 GHz, and the maximum RCS reduction value can reach 29.4 dB.

The invention discloses a 16-in-one high-power combiner working in a radio frequency band, and belongs to the technical field of fusion radio frequency heating. The invention comprises a coaxial input hard feeder line, a radial transmission line (RTL), a gradient impedance matching coaxial line and a coaxial output hard feeder line. The physical size is designed through theoretical calculation, electromagnetic field distribution is optimized through simulation, 16 radio-frequency signal input ports are designed through radial symmetric topology, a radial transmission line serves as a main synthesis path of radio-frequency signal source power, the power input by the 16 radio-frequency signal source ports is subjected to one-step efficient synthesis in the center of the radial transmission line, and the radio-frequency signal source power is obtained. And the 50-ohm characteristic impedance standard output of the synthesized radio frequency signal power is realized through an impedance matching technology. The sixteen-in-one power combiner designed by the invention has the characteristics of high synthesis efficiency, symmetrical structure, good amplitude and phase consistency, high isolation, low return loss, low reflection loss, simple design structure, easiness in manufacturing and the like.

Liquid metal (LM) interconnects have angled features to reduce reflection loss and / or insertion loss of signals sent across the interconnects. The LM interconnects may be used to couple two electronic components, e.g., to physically and electrically couple an IC device and a circuit board together. One component has an array of LM wells, and the second component has a corresponding array of pins. The LM wells have tapered shapes, e.g., tapered sidewalls with LM within the sidewalls. A cap layer along seals the wells to hold the LM inside the wells. The pins on the second component have a flared portion that widens in the direction of the second component and tapers in the direction of the tip. When a pin is inserted into a corresponding LM well, the tip and straight (e.g., cylindrical) sections may be within the LM well, while the flared portion remains within the cap layer.

The invention relates to an underwater acoustic transducer which can resist long-term work under the condition of deep sea high water pressure, is high in sensitivity, wide in band, low in noise and high in electromagnetic interference resistance, is used in deep sea detection and communication sonar systems, and solves the problems that an existing underwater acoustic transducer is low in receiving sensitivity along with increase of water pressure, poor in long-term work corrosion resistance and low in cost. And noise of the hydrophone and external electromagnetic interference cause signal interference. A novel high polymer material is adopted to reduce acoustic impedance, enhance the coupling effect with seawater and reduce sound wave reflection loss. Meanwhile, by optimizing the internal structure design, the stability and durability of the device under the deep sea high pressure condition are ensured. And the hydrophone shell is made of an alloy material with good corrosion resistance, so that the hydrophone shell is prevented from being corroded in a marine environment for a long time, and the service life is prolonged. By improving the design and manufacturing process of the transducer, the transducer can cover acoustic signals in a broadband range, and the multi-spectrum requirement of a deep sea detection and communication system is met. Advanced noise suppression technology and shielding measures are adopted, noise of the hydrophone is effectively reduced, external electromagnetic interference resistance is improved, and it is guaranteed that high-quality acoustic data are obtained in a complex deep sea environment. In conclusion, the invention provides the hydrophone which is excellent in working performance in the deep sea high-pressure environment, has the characteristics of high sensitivity, broadband response and the like, can effectively resist the influence of a severe environment, and provides key technical support for deep sea detection and a communication sonar system.

The invention discloses a magnetic control chained polyaniline-based broadband bionic wave-absorbing material and a preparation method thereof, and relates to the technical field of electromagnetic wave absorbing materials and space functional composites.The magnetic control chained polyaniline-based broadband bionic wave-absorbing material achieves multi-band impedance matching through the cooperation of orientation arrangement of magnetic nano chains and the polarization effect of a polyaniline interface, is extremely low in average reflection loss at 8-18 GHz under the thickness of 1.5 mm, and can be applied to the field of electromagnetic wave absorbing materials and space functional composites. The composite material covers an X-Ku wave band, the material density is less than or equal to 1.6 g / cm < 3 >, the mechanical strength is high, the composite material can be directly compounded on a flexible film antenna substrate, the satellite load mass is saved by more than or equal to 40%, and the wave absorbing performance attenuation is less than or equal to 5% after 100 times of alternating circulation from-180 DEG C (liquid nitrogen) to + 120 DEG C (vacuum), which is superior to that of a traditional carbon-based material (the attenuation is more than or equal to 20%); magnetic field induction and chemical synthesis are combined, the process temperature is smaller than or equal to 250 DEG C, and a flexible processing production line of an existing satellite film antenna is compatible.

The invention discloses a laser-induced lignin-derived porous carbon wave-absorbing material and a preparation method thereof, and belongs to the field of material preparation and microwave absorbing materials. The method comprises the following steps: dissolving lignin in a sodium hydroxide aqueous solution, and uniformly mixing with a polyethylene oxide aqueous solution to obtain a lignin ethylene oxide solution; the method comprises the following steps: preparing a lignin membrane by using a lignin ethylene oxide solution, performing laser irradiation, washing, performing suction filtration, and performing vacuum drying to obtain the laser-induced lignin-derived porous carbon wave-absorbing material. The lignin-derived porous carbon material prepared by taking industrial waste lignin as a raw material has the performance characteristics of being harmonious, low in density, high in reflection loss and wide in wave band absorption. The preparation method is green, environment-friendly, simple to operate and low in cost, and the waste lignin in the papermaking industry is successfully converted into a high-value adjustable functional material. The method meets actual application requirements and has huge application potential in the technical field of microwave absorbing materials.

The invention discloses a muffling device, an air conditioner and a muffling control method.The muffling device comprises a shell, an acoustic metamaterial structure and a sound conduction interface, the acoustic metamaterial structure is arranged in a muffling cavity of the shell and is constructed to be a continuous swirling flow channel capable of prolonging a sound wave propagation path in the muffling cavity, and the sound conduction interface is arranged in the shell; the sound conduction interface is arranged at the joint of the shell and the valve pipe. Noise firstly enters the shell through the sound conduction interface, so that sound wave reflection loss is avoided; in the silencing cavity, the continuous swirling flow channel of the acoustic metamaterial structure forces the propagation path of sound waves to be greatly prolonged in a limited space, so that low-frequency resonance noise is effectively inhibited; meanwhile, a physical closed environment provided by the shell isolates external aging factors, and the problems that a traditional rubber damping material loses efficacy due to hardening and the performance of sound insulation cotton is degraded are solved; according to the invention, the risk of new resonance caused by frequency deviation is fundamentally avoided, and synchronous control and long-term reliable noise elimination of low-frequency noise are realized.

The present invention discloses an electromagnetic absorption composite material, its preparation method, and its application. The composite material comprises graphene; metal nanoparticles coated with carbon shells are interspersed within the graphene's layered structure. The electromagnetic absorption composite material of the present invention has an intercalated structure, with metal nanoparticles interspersed within the graphene. The surface of the nanoparticles is coated with carbon shells to form a core-shell structure, which improves the electromagnetic absorption performance of the material by enhancing multiple reflection losses.

The invention provides a novel perovskite cell and a preparation method thereof, the novel perovskite cell comprises an antireflection film stack layer, a glass substrate, an equivalent refraction layer, a hole transport layer, a perovskite layer, an electron transport layer and a top electrode which are arranged in sequence, the antireflection film stack layer is used for constructing refractive index transition between air and the glass substrate, and the hole transport layer is used for transmitting the hole transport layer. The absolute amount of effective light entering the battery is increased; the target effective refractive index of the equivalent refraction layer is associated with the first refractive index of the glass substrate and the second refractive index of the hole transport layer, and is used for constructing refractive index transition between the glass substrate and the hole transport layer so as to reduce the interface reflection loss of the perovskite cell; it is ensured that incident light passing through the glass substrate penetrates to the perovskite layer without significant loss. Through the superposition design, the utilization rate of light in the battery can be effectively improved, and the power generation efficiency of the battery is further improved.

The invention discloses an unsupervised monocular endoscope depth estimation method combined with a segmentation large model. A network framework comprises a depth prediction network phi D, a pose estimation network phi T, an intrinsic decomposition module phi I and a synthesis reconstruction module phi L. Wherein the depth estimation network adopts a deep large model fine tuning technology, is combined with rich semantic prior information in a segmentation large model, and is comprehensively trained by using decomposition synthesis loss, reflection loss, mapping synthesis loss and an edge perception depth smoothing loss function guided by a segmentation mask. The semantic enhancement branch network adopts a network architecture of a segmented large model, image semantic information is accurately extracted, structures such as different tissues and organs are distinguished, and a semantic cognition system for an operation image is established. And obtaining semantic features of a target image through a fine tuning encoder, and after the semantic features are fused with depth features, enabling a depth estimation model to be combined with context semantics to optimize depth prediction. Meanwhile, an automatic segmentation mask graph generated by a segmentation large model is utilized to calculate depth smooth loss guided by a mask, and the rationality of a depth estimation result is further constrained. According to the scheme, the encoder and the automatic segmentation mask in the large model are segmented, semantic prior information in the large model is fully combined, the depth estimation effect is improved, and the operation precision and the safety performance of the endoscope minimally invasive surgery are further enhanced.

The invention discloses a wave-absorbing material design method based on neural network guidance, and belongs to the field of intelligent material design. The method comprises the following steps: selecting actually measured magnetic conductivity frequency dispersion data of a magnetic material as a boundary condition, constructing a parameter space containing magnetic conductivity, dielectric constant, material thickness and frequency, and calculating a reflection loss RL value through a transmission line model; gPU acceleration tensor operation is adopted to generate a three-dimensional parameter space, and discretization processing is carried out on the parameter space; implementing a double-task screening strategy to extract an optimal dielectric constant combination and an effective region boundary; constructing a neural network model with physical constraints, and inputting thickness and frequency prediction dielectric constant parameters; and generating solution set distribution to guide material design through inversion verification. A physical model and deep learning are fused, an Einstein summation method and a broadcast mechanism are adopted to realize hundred million-level parameter parallel calculation, the problem of multi-parameter coupling optimization is solved, the physical rationality of output parameters is ensured through a softplus activation function, and the design efficiency and precision of the magnetic wave-absorbing material are remarkably improved.

The utility model provides an antireflection film, a perovskite / crystalline silicon laminated solar cell and a photovoltaic module, and relates to the technical field of solar cells. The antireflection film comprises first areas and second areas which are alternately distributed; the first region is a first silicon nitride layer; the second region is a silicon oxide layer; the edge end faces of the adjacent first silicon nitride layer and the silicon oxide layer are connected. According to the technical scheme, the refractive indexes and the reflectivity of the first silicon nitride layer and the silicon oxide layer forming the antireflection film are different, light absorption can be optimized, reflection loss can be reduced, and compared with a single-layer antireflection film made of a single material, the reflection loss of light on the surface of the solar cell can be greatly reduced, and the photoelectric conversion efficiency of the solar cell is improved.

The invention provides a micron rod-shaped Fe-Si-Al-based composite electromagnetic wave-absorbing material and a preparation method thereof, spherical FeSiAl alloy powder is subjected to ball milling to obtain micron rod-shaped powder, and the surface of the FeSiAl alloy powder is coated with TiO2 by utilizing hydrolysis reaction of tetrabutyl titanate, so that a FeSiAl-coated TiO2 core-shell composite electromagnetic wave-absorbing material is prepared; the FeSiAl-coated TiO2 core-shell composite electromagnetic wave-absorbing material is annealed in an argon or hydrogen-argon mixed gas atmosphere, the micron rod-shaped iron-silicon-aluminum-based composite electromagnetic wave-absorbing material is prepared, the shape anisotropy and the magnetic conductivity of FeSiAl are effectively improved, the oxygen vacancy concentration of TiO2 is regulated and controlled through annealing in a hydrogen-argon mixed gas reducing atmosphere, the polarization relaxation of dipoles is enhanced, and the electromagnetic wave-absorbing effect is improved. The core-shell structure enhances interface polarization, improves magnetic loss and dielectric loss, and effectively enhances the wave-absorbing bandwidth and reflection loss of the wave-absorbing material.