74 results about "Computational physics" patented technology

An optical scanning apparatus includes light sources, an aperture having, and coupling lenses. Numbers of the light sources and the coupling lenses are equal to each other and the apparatus satisfies an inequality 12.7 <(D×β₀×m₀×xdpi)/(2×NA) <38.1 and an equation β₀=(β₊−β₋)/(n−1), wherein D represents a width of the slit of the aperture, β₀represents a beam open angle, mo represents a total lateral magnification ratio, dpi represents a number of dots per inch, NA represents a numerical aperture, β₊ represents a positive open angle, β₋ represents a negative open angle, and n represents the number. Further, an image forming apparatus using this optical scanning apparatus and an image forming method are provided.

The invention discloses a GEO target autonomous segmental arc association and orbit determination method for space-based optical surveillance, relates to the technical field of spaceflight. The methodhas high association accuracy and orbit determination precision. The method comprises the following steps: firstly, solving a Lambert equation by utilizing position vectors corresponding to any two short arc initial rails; obtaining a high-precision initial orbit semi-major axis, performing distance assignment on angle measurement data points by utilizing a near-circular orbit hypothesis, furthercompleting initial orbit improvement under a perturbation condition, comparing the difference between an initial orbit improvement result and an original observation angle, and selecting an observation residual error slope as a judgment threshold to carry out initial orbit inter-arc-segment correlation; taking the associated initial orbit result between the two arc sections as the initial numberof the cataloguing library, carrying out the autonomous distance prediction of a new observation arc section, combining with the actual angle measurement data, carrying out the orbit improvement and correlation judgment, and achieving the steady GEO target autonomous cataloguing orbit updating and maintaining.

An optical device may include a sacrificial limiter filter including at least one layer of graphene disposed on a substrate. The at least one layer of graphene may be configured to absorb and scatter at least a portion of electromagnetic radiation incident on the at least one layer of graphene.

The invention discloses a device and method for detecting the optical detector working point of a particle detector. The device comprises a laser, an external modulation device, a cylindrical lens group, a sample cell, a focusing lens and an optical detector which are distributed according to preset positions, wherein the external modulation device is used to generate a detection signal and load the detection signal on the laser beam sent by the laser; and the device also comprises an oscilloscope and a processing circuit, wherein the optical detector is connected with the oscilloscope and the processing circuit through signal wires respectively to display the detection signal from the external modulation device on the oscilloscope and process the detection signal in the processing circuit. By using the device and method of the invention, the optical detector can work in a linear work area and the particle detection can be performed more accurately.

The invention discloses a method for a reactive ion beam etching (RIBE) of a blazed convex grating, which comprises the following steps: step one: placing a masking quartz substrate of the convex grating; step two: ensuing an angle of the ion beam direction and a groove at the peak of the grating to be theta if the spherical radius of the masking quartz substrate of the convex grating is R and the blazed angle of the blazed convex grating is theta; step three: manufacturing an etching slit; step four: placing the slit; and step five: etching the convex grating. The technical scheme of the invention adopts a subsection etching method, namely, a subsection ion beam etching is carried out on the masking quartz substrate of the convex grating with a sine groove shape, so that the relative error of the blazed angle of the blazed convex grating which is etched is 0.1 in theory.

The invention relates to a separation device making use of battery impedance spectrum, provides a process simplifying battery impedance spectrum measurement and further provides a battery separation method according to the battery impedance spectrum. Batteries to be measured are connected to a circuit through a special four-wire clamp, control and data processing (DP) first controls a high speed sampling unit analogue digital (AD) to start high speed sampling, then constant flow source horizontal line (HL) current is controlled to charge the batteries (British telecom) BT to be measured, and after being conditioned through a voltage current processing module VI, the voltage and current signals of the batteries BT to be measured are processed in a digital mode through the high speed sampling AD and are obtained by the control and data processing DP. After the measurement is completed, the control and data processing DP closes the output of the constant flow source HL, then the high speed sampling AD is stopped, then the obtained voltage and current data are uploaded to storage and separation system file protection (SFP), the same batteries are organized in a category by the storage and separation SFP according to separation conditions set by a user, and thereby the battery separation is completed.

The invention discloses a uniform-section continuous beam damage identification method based on support influence line for reaction curvature difference. The method comprises the steps of setting a support reaction measurement point at each support of a continuous beam, and obtaining a support influence line for reaction of each support of the continuous beam through applying a moving load; solving the curvature of the support influence line for reaction, further performing difference, and performing damage positioning through sudden change of a curvature difference curve of the support influence line for reaction; and further carrying out damage degree quantification through the difference change of the curvature of support influence line for reaction of the continuous beam. According tothe method, information of the continuous beam before damage is not needed, only the measurement points need to be arranged at the support, the requirement for the number of the measurement points islow, the number of measurement sensors is reduced, damage of the uniform-section continuous beam can be accurately positioned and quantified, and the method is applied to damage evaluation of the uniform-section continuous beam.

The invention discloses a local erasing voltage control method and system for a liquid crystal writing device. The control method comprises the steps of defining the process from entering a writing area to leaving the writing area of the erasing device as an erasing period; defining a voltage value loaded on each conductive region on the two conductive layers in the erasing preparation state; wherein the voltage value enables an electric field formed between the two conductive layers to be smaller than an erasing electric field; in an erasing period, controlling the voltage values applied to the two conductive layers to return to an erasing preparation state every time an erasing action is executed; and when executing the next erase action, controlling the voltage value to be converted from the erase preparation state to a set voltage value required for realizing local erase. Compared with the prior art that the voltage applied by the two conductive layers needs to be reduced to zero after each erasure action is executed, the time required for executing the erasure action each time is greatly shortened, and the erasure efficiency is improved.

The invention discloses a circuit for reducing the power consumption of a TOSA. The circuit comprises a DC-DC power module. The feedback end of the DC-DC power module is connected with the ground through a third resistor and connected with a microcontroller through a second resistor. A first resistor is connected between the output end of the DC-DC power module and the second resistor in series. The output end of the DC-DC power module is connected with a collector of a triode. A base electrode of the triode is connected with an APC circuit, and an emitting electrode of the triode is connected with a laser diode of the TOSA through a fourth resistor. The microcontroller is connected with the two ends of the fourth resistor in parallel and used for monitoring the difference of the voltage across the two ends of the fourth resistor, and the output voltage of the DC-DC power module is gradually reduced according to the difference of the voltage. According to the circuit for reducing the power consumption of the TOSA, due to the fact that the output voltage of the DC-DC direct-current power module is gradually reduced, under the condition that the TOSA works stably, the power consumption of the TOSA is effectively reduced.

A container liquid level monitoring method comprises the steps that a liquid level height value, a liquid pressure value and a liquid temperature value in a container are collected in real time through a liquid level sensor, a pressure sensor and a temperature sensor, and the liquid level sensor, the pressure sensor and the temperature sensor are arranged in the container; g the liquid level height value, the pressure value and the temperature value are input into a liquid level monitoring model so that a liquid level height accurate value is obtained; the liquid level monitoring model is an RBF neural network model; the RBF neural network model is trained through a training algorithm and an input training sample to optimize a target function, the training sample comprises a liquid level height value, a pressure value, a temperature value and a liquid level height actual value, and the target function adopts an error quadratic sum minimum function. According to the invention, the accuracy of liquid level monitoring is improved, and the monitoring efficiency is improved.

The invention relates to a pixel matrix driving device, a pixel matrix driving method and a display thereof. The pixel matrix driving device comprises a pixel matrix, a time schedule controller and adriving module; the pixel matrix comprises a plurality of sub-pixels, and different polarity inversion positions are arranged between the i-th column of sub-pixels and the (i + 1)-th column of sub-pixels; the time sequence controller is used for obtaining an initial pixel value and obtaining a first gray-scale value and a second gray-scale value according to the initial pixel value; and the driving module is used for obtaining a first loading voltage according to the first gray-scale value, obtaining a second loading voltage according to the second gray-scale value, and loading the first loading voltage and the second loading voltage to the pixel matrix. According to the invention, the polarity reversal positions of two adjacent columns of sub-pixels are different; and the voltage is alternately loaded to the sub-pixels by the first loading voltage or the second loading voltage in a certain row of sub-pixels and a certain column of sub-pixels, so the horizontal equidistant cross grainscan be improved, the whitening phenomenon of the display panel in side view can be improved, and the display panel has good display quality in front view and side view.

The invention discloses an asymmetric decoupling structure and a base station antenna system. The asymmetric decoupling structure comprises an asymmetric metal microstrip structure and a dielectric support, wherein the asymmetric metal microstrip structure comprises at least two metal microstrips of different sizes; the metal microstrips are arranged in parallel and are erected on a substrate of abase station antenna through the insulating dielectric support; and the length of the metal microstrips is (1/2+/-10%) lambda, and lambda is the wavelength in an equivalent medium. Mutual inhibitionis realized by using a field cancellation method, the asymmetric decoupling structure has relatively low radiation capability, the radiation performance of the antenna is not deteriorated while the decoupling efficiency is maintained, the structure can be used for inhibiting mutual coupling between closely placed microstrip antennas, the problem of influence of a passive decoupling structure on anantenna radiation field is solved, mutual interference between antenna units is reduced, and the asymmetric decoupling structure is especially suitable for an array multi-antenna system of the base station.

The invention discloses a combined radiation source strength backward induction method and a combined radiation source strength backward induction system for a point source and surface source of a nuclear power plant. The method is characterized in that dosage rates at a plurality of positions are acquired through a detector; a normalized radiation source strength is diffused in a space; an optical distance is calculated by means of a ray tracking method; an equation set coefficient is calculated according to information such as material and accumulation factor, thereby performing backward induction for obtaining source strength; then dosage rate calculation is performed on detector position; through performing linear regression analysis on a measured value and a calculated value, key parameters such as standard deviation, gradient and intercept are calculated, and furthermore a quality factor is calculated for measuring acceptance degree of a calculation result at each time. Furthermore the invention provides a weighted iteration method. An error which is introduced by a detector with relatively high uncertainty is reduced. The steps are repeated by multiple times in an iteration manner until the quality factor reaches a preset fixed value, and anticipated radiation source strength information is obtained.

The invention belongs to the field of novel artificial electromagnetic materials, and particularly relates to an ultra-thin ultra-wideband checkerboard structure RCS reduction metasurface. Based on anartificial magnetic conductor AMC structure, through the design of top layer metal patterns of two super-surface basic unit composite structures and further selection of materials and sizes, basic units 0 and 1 are arranged in an array to form super units 0 and 1, and then the two super units are designed and obtained according to a chessboard structure. In the 8-12GHz wave band, the phase difference can meet the RCS reduction requirement in a wide frequency band, and the ultra-wideband antenna is suitable for a high-temperature environment. The metasurface thickness is small, the medium thickness is 2.3 mm, and the overall metasurface thickness is 2.37 mm and is reduced by 0.7 mm compared with the prior art; the ultra-wideband band-pass filter has the advantages that the ultra-wideband band-pass filter is wide in frequency band and good in reduction effect, the bandwidth of more than 10dB can reach 3GHz, the relative bandwidth can reach 26.9%, and the maximum RCS reduction value is 25dB when the ultra-wideband band-pass filter works at 8-12GHz; the technical problems that the thickness of the metasurface is reduced while the bandwidth of the metasurface is guaranteed and the metasurface can be applied to a high-temperature environment are solved.

The invention discloses a method for generating two-dimensional irregular virtual particles, which comprises the following steps of: S10, carrying out first processing on the particles to obtain a first plane projection image corresponding to the particles; S20, determining a quality center of the first plane projection image, and establishing a rectangular coordinate system by taking the qualitycenter as an original point; S30, constructing a first equivalent ellipse according to the coordinate values corresponding to the pixel points in the first plane projection image in the rectangular coordinate system; S80, calculating a second residual radius; S90, calculating a polar radius corresponding to the central angle; and S100, calculating a radius r and a radius corresponding to the central angle according to the polar radius and the second residual radius, and generating virtual particles according to the central angle and the radius. The method can be suitable for more particles, and the generated virtual particles are slightly different from real particles.

The invention provides a reinforcement learning-based ship motion multi-step real-time prediction mixing method and system, and the method comprises the steps: obtaining original ship motion data through an attitude sensor disposed on a ship, dividing the original ship motion data into a training data set and a prediction data set, carrying out the RTWPD, and obtaining a prediction data set; decomposing a high-frequency component and a low-frequency component of the ship motion data into subsequences with a fixed number of layers; an ORELM basic prediction model is established for each sub-sequence obtained through decomposition, an AdaBoost.MRT reinforcement learning mode is introduced, iterative training is conducted continuously, and a plurality of trained ORELM models are combined together; and finally, reconstructing a prediction result of the sub-sequence to obtain a model multi-step initial prediction result, establishing an LSSVM error correction model for a multi-step prediction error, further extracting a change rule of a small part of ship motion data contained in a multi-step error sequence, and correcting to obtain a final ship motion multi-step prediction result and outputting the final ship motion multi-step prediction result. According to the invention, the stability and accuracy of ship motion attitude multi-step prediction are improved.

The invention discloses a field effect transistor source-drain resistance extraction method comprising the steps: 1, testing to obtain a first electrical characteristic curve formed by absolute valuesof drain current and gate voltage of a linear region of a field effect transistor, and extracting threshold voltage of the linear region; 2, selecting a plurality of sampling points in a high gate voltage region, and calculating reciprocals of absolute values of difference values between gate voltages corresponding to the sampling points and threshold voltages of the linear region to serve as first parameters; 3, calculating the ratio of the drain voltage to the drain current of each sampling point and taking the ratio as the total resistance; 4, forming a second relation curve according to the first parameter of each sampling point and the total resistance; and 5, extending the second relation curve and intersecting the second relation curve with the longitudinal axis, and taking the intercept as the source-drain resistance. Source and drain resistors can be extracted for a single transistor, the method is suitable for field effect transistors with uniformly or non-uniformly doped channels, the test area is saved, the test speed is improved, and the resistance extraction precision is improved.

The invention provides a Raman spectrum denoising method based on deep learning, which solves the problem that pure Raman spectrum information is easy to lose in the denoising process in the prior art, and input parameters are set by means of human intervention. The invention provides a deep learning network-based Raman spectrum denoising method, which comprises the following steps of: generating Raman spectrum data required by model training, and dividing the Raman spectrum data into a training set, a verification set and a test set; then, a U-Net Raman spectrum denoising model is established, the model is divided into an encoder network module and a decoder network module, feature extraction is performed on Raman spectrum data by using one-dimensional convolution in the encoder network, and an extracted feature spectrum is reconstructed and a pure spectrum is output in the decoder network. According to the method, noise of the Raman spectrum can be effectively removed, Raman spectrum information is well reserved, and accurate and reliable information is provided for further qualitative and quantitative analysis of the Raman spectrum. The method can be widely applied to the optical field.

The invention relates to a pipeline flow field pressure distribution evaluation method and device integrating measured data and relates to the technical field of pipeline flow field pressure distribution evaluation. The method comprises the following steps of: carrying out equal-difference splitting on a parameter floating threshold range so as to obtain a plurality of parameter floating sub-thresholds; obtaining a plurality of empirical parameter estimation values, and obtaining a plurality of corresponding empirical parameter correction values on the basis of each parameter floating sub-threshold; according to the plurality of empirical parameter correction values, combining the empirical parameter correction values to obtain a plurality of empirical parameter combinations; performing simulation according to each empirical parameter combination to obtain theoretical pressure distribution conditions; and comparing the theoretical pressure distribution situations with a monitored pressure distribution situation obtained by monitoring, and taking the theoretical pressure distribution situation corresponding to the minimum total error of the theoretical pressure distribution situation and the monitored pressure distribution situation as the actual pressure distribution situation. According to the pipeline flow field pressure distribution evaluation method and device integrating the measured data of the invention, the threshold range of each empirical parameter is split, the software simulation data is compared with the real-time measurement data, and high-precision flow fieldpressure distribution is gradually established for a pipeline, so that the phenomena of pipeline cracks and leakage can be analyzed more accurately.