407results about How to "The principle of the method is simple" patented technology

The invention discloses an existing line reinforcing method in under-crossing railway frame bridge small-included angle large-height inclined jacking construction. The method comprises the following steps of: reinforcing an existing line by using an overhead rail (9) along the line, and supporting the existing line by using a plurality of horizontal lifting beams (5) which are vertical to the line direction at certain intervals below the existing line, and using manual hold digging piles (1) on the lower left and right of the horizontal lifting beams (5) as support points of the horizontal lifting beams on two sides of the line; laying a steel slideway (6) between a frame top (8) and the horizontal lifting beams (5) along the jacking direction; meeting the requirements of a longitudinal slope and a cross slope of the line by using wedge-shaped wood (12) between a sleeper (20) and the horizontal lifting beams (5); and before the hole digging piles are broken, transferring the load of the pile top to the frame top (8) through the steel slideway (6), and circularly breaking the piles and jacking. By the method, the vertical and horizontal rigidity of the line are enhanced, the stability of the line during jacking is ensured, the workload for manually maintaining the line is reduced, and the driving safety is ensured.

The invention relates to an aligning method of the concentricity of a coupling. The traditional method for aligning by using three gauges has the problems that: some equipment is limited by a space position and is not capable of being provided with three dial gauges and enabling the three dial gauges to rotate along with the coupling; or the end surface of the coupling is machined to be rough andis not capable of setting the dial gauges, and therefore the aligning of the concentricity of the coupling cannot be carried out by using the method for aligning through three gauges. According to the invention, the method comprises the following steps of: measuring and calculating the radial nonconcentricity of the coupling; and measuring and calculating the axial nonconcentricity of the coupling. The method provided by the invention has the advantages of concise principle, convenience for operation, simpleness in calculation and high accuracy, thereby being capable of being widely applied.

The invention relates to a background modeling method for movement target detection, and belongs to the technical field of photoelectric image processing. In order to establish an accurate background model in a complex scene, the method provides background statistical modeling methods utilizing inter-frame differential information, namely a median method and an average method, based on the problems that a background model is difficult to obtain for the background differential method and the background differential method is easily affected by external interference, the influence of a foreground target is removed, so that the established background model is more accurate and has high anti-interference performance. The median method and the average method have simple principles and high anti-noise performance and are easy to implement. In addition, according to the method, adjacent frames of video images are differenced, and an appropriate threshold value is selected, so that an accurate background model is guaranteed.

The invention provides a rail transit full-road-network passenger flow prediction method, which comprises the steps of: establishing a corresponding probability tree set according to the history origination and destination (OD) data, origin stations, inbound time periods and occurrence probabilities of outbound positions; continuously obtaining passenger inbound information (O data) and predicting passenger outbound information (D data) corresponding to each O data according to a prediction tree in historical statistic learning results; and overlaying influences of all OD data on full-road-network passenger flow to obtain information of the stations and the traffic zone passenger flow of the entire road network. The method solves the problem that the rail transit full-road-network passenger flow cannot be displayed and predicted in real time because the rail transit full-road-network passenger flow cannot be directly measured. Besides, the method has the advantages that the prediction result is good, the expansion is easy and the short-term passenger flow prediction and abnormal passenger flow prediction can be supported.

The invention discloses an island detecting method based on positive feedback of voltage harmonic distortion in the technical field of a distributed power generating system. In the invention, based on the grid connection control principle of the distributed power generating system, a voltage harmonic distortion positive feedback circuit is additionally arranged, the curve of the voltage harmonic distortion H of a microgrid and the area S surrounded by time shafts are calculated by utilizing a sliding data window, and whether an island is caused is judged according to the magnitude of the areaS. The island detecting method has a simple principle, has no effect on damaging the power quality, can be used for detecting the island quickly and effectively in the worst condition defined in the IEEE Std.1547, is applicable to various types of open circuit conditions, is not interfered by pseudo islands, and has good engineering practicality.

The invention relates to a low-contrast image enhancement method based on YCrCb. The low-contrast image enhancement method comprises the following steps: (1) performing image de-noising on an obtainedoriginal image by utilizing a trapezoidal low-pass filter; (2) performing conversion between a RGB color space and a YCrCb color space on the de-noised image to obtain an image brightness component;(3) performing piecewise linear transformation on a brightness component image by adopting a traditional spatial domain linear transformation method; (4) performing Gamma correction on the transformedimage; (5) performing Gaussian filtering on the corrected image by utilizing a single-scale Retinex algorithm and estimating illumination component and reflection component images; (6) performing image sharpening by utilizing a Gaussian sharpening algorithm aiming at the reflection component image; and (7) performing conversion between the YCrCb color space and the RGB color space on the sharpened image.

The invention relates to a method for judging a direct current arc fault based on a real-time measured multi-band frequency component rate. Along with the large-scale application of photovoltaic generation, and particularly the wide application of a photovoltaic solar panel to the roof and the external wall of a building, fires caused by a photovoltaic generation system happen in many places. Through analysis, people find that the fires are basically caused by the direct current arc fault of the photovoltaic generation system, and therefore, detecting and isolating the direct current arc fault for causing equipment damage and fire hidden danger become problems needed to be solved, however, due to the specificity of the direct current arc fault of the photovoltaic generation system, the conventional fault waveform based arc fault detection principle is not suitable for the direct current arc fault of the photovoltaic generation system. The method has the core thought that the overlarge energy is contained in the arc to cause the fire damage caused on the basis of the fault arc, and by using a method for detecting the energy source ratio of the frequency spectrum component energy of multiple bands in a voltage and current waveform to the total spectrum, the direct current arc faults possibly causing fires can be rapidly and reliably detected without being affected by other noises in the system, and the safety of the photovoltaic generation system is ensured.

The invention discloses an insulator abnormal automatic detection method based on ensemble classifier online learning. The method comprises the following concrete steps of carrying a high-definition PTZ camera by using a unmanned aerial vehicle / pole tower as a carrier; shooting a power transmission line visible light insulator image; performing preprocessing, image division, image multi-feature extraction and the like on the obtained image to obtain various typical features of abnormal isolator images and to perform normalization processing; next, selecting decision trees to be used as a weakclassifier; using the normalization features as the classification attribute to train the weak classifier; performing repeated training to obtain a plurality of decision trees and the classification weight; finally, generating a strong classifier with high classification accuracy through weighted voting; meanwhile, timely updating the strong classifier by combining the online learning technology;and separating each type of abnormal insulator images for subsequent processing. The method has the advantages that the principle is simple and visual; the implementation is easy; the image processingtechnology and the machine learning algorithm are combined for intelligently recognizing the insulator abnormality; and a novel idea and method is provided for the insulator abnormal operation monitoring.

The invention discloses an automatic solid grain sorting system which comprises a grain queuing device for sequentially transferring the solid grains and transmitting single grains to a signal collecting area, a signal collecting device connected with the grain queuing device and used for collecting information of the solid grains in the signal collecting area, an information processing device which is used for carrying out information processing on solid grain information data collected by the signal collecting device to judge the attributes of the grains and sending the attribute signals to a grain sorting device, and the grain sorting device connected with the information processing device and used for acquiring the attribute signals and sorting the solid grains according to the attribute signals.

The invention relates to an AC / DC (Alternating Current / Direct Current) mixed micro-grid optimized operating method based on a main-slave game model. The method comprises the following steps: acquiring weather data of an area in which a micro-grid is located such as illumination intensity, temperature and cloud cover, counting historical data of a photovoltaic power generation amount in the micro-grid, and predicting power outputted by a photovoltaic cell in an AC / DC mixed micro-grid in a next day by adopting a prediction method based on a typical trend; acquiring transmission power data at two sides of a bidirectional converter and a tendency controller in the AC / DC mixed micro-grid, and establishing a mathematic model of various devices in the AC / DC mixed micro-grid; dividing a day into various dispatching time periods, establishing a main-slave game model of cooperation of a photovoltaic operator and a grid company at each dispatching time period, wherein the main-slave game model comprises a photovoltaic utilization rate model maximizing the benefit of the photovoltaic operator and an AC / DC mixed micro-grid loss model minimizing the loss of the representative grid company; and improving the global searching performance of an algorithm by adopting a two-order oscillation particle swarm algorithm of an asynchronous variation learning factor. The AC / DC mixed micro-grid optimized operating method is high in applicability and high in prediction precision.

The invention discloses a six-degree-of-freedom parallel robot motion analysis modeling and rapid solving method. A six-degree-of-freedom mechanical platform is composed of six supporting rods with linear actuators, an upper platform, a lower platform and steering devices, wherein six steering devices are correspondingly arranged on the upper part and the lower part, the lower platform is fixed onan infrastructural facility, and the upper platform is controlled to move in the space by six degrees of freedom through telescopic motion of the six supporting rods. The mechanical platform has a parallel structure, namely six drivers jointly act on one platform, and the six-degree-of-freedom mechanical platform specifically comprises the following steps of kinematic modeling and solving and motion controlling, wherein the kinematic modeling and solving comprise the following steps of moving coordinate transformation, rotating coordinate transformation and composite posture; and the motion controlling comprises the following steps of establishing a mechanical system model, numerical simulation results are calculated according to the established model, and finally better control of the six-degree-of-freedom parallel robot is achieved. The six-degree-of-freedom parallel robot motion analysis modeling and rapid solving method is used for helping the kinematic modeling of the six-degree-of-freedom parallel robot to be solved quickly and controlled accurately.

The invention discloses a ground feature classifying method based on simplified polarization synthetic aperture radar data, which relates to the synthetic aperture radar remote sensing information processing technology. The ground feature classifying method includes steps: a) obtaining Stokes parameters after filtering; b) extracting characteristic parameters including polarizability m, relative phase difference delta among orthogonal channels and total polarization power span for showing scattering mechanisms; c) calculating power ingredients corresponding to the basic scattering mechanisms, and determining main scattering mechanisms of pixels and dividing data into three types of pixels with identical scattering characteristics; d) further dividing each type of pixels with identical scattering characteristics into a plurality of types of pixels with equal sample numbers by the aid of the span; and e) realizing iterative Wishart classification. The ground feature classifying method is simple in principle and convenient in calculation, and can be used for analyzing and classifying ground feature scattering characteristics.

The invention belongs to the technical field of battery, and particularly relates to a method and device for measuring air permeability of a battery diaphragm at different pressures and different temperatures. The method for measuring the air permeability of the battery diaphragm comprises the steps of fixing and sealing the battery diaphragm; stetting a diaphragm temperature by a heating device; introducing air from an air inlet side and setting a pressure value; metering a time required by passing a certain volume of air from the air inlet side; and calculating the air permeability of the diaphragm. The device for measuring the air permeability of the battery diaphragm comprises an air supply system, a pressure stabilization system, a variable temperature measuring system and a metering system. The method and device for measuring the air permeability of the battery diaphragm have the advantages of simple operations, low cost, accurate data and high practicality. Meanwhile, the method and device can measure the air permeability of the diaphragm at different temperatures, ascertain variable relationships between the air permeability of the diaphragm and the temperature, and further study a hole closing temperature and a diaphragm broken temperature of the battery diaphragm, and can predicate battery performance changes caused by temperature variation.

The invention discloses a system performance assessment method of a polarized interferometric synthetic aperture radar and relates to the technology of the synthetic aperture radar. The method firstly builds a relationship between the system parameters of the polarized interferometric synthetic aperture radar and the characterization parameters of the system interference performance (interference coherence and interference phase estimated standard deviation); the parameters of a typical application context are given, and data are acquired through a simplified effective polarized interference data simulation method; then the data are used to calculate the parameters of the typical application context again; and through the comparison, the parameters are analyzed and the performance is estimated so as to assess the actual application performance of the system of the polarized interferometric synthetic aperture radar. The method disclosed by the invention is simple in principle, is convenient to calculate and can be applicable to the performance assessment and comparison as well as feasibility analysis of the typical application of the future satellite-borne or sky-borne polarized interferometric synthetic aperture radar.

The invention discloses a method for assembling a zinc oxide nano-chain by using DNA as a template, comprising the following steps of: (1) mixing and stirring an aqueous solution of DNA, an aqueous solution of zinc nitrate and an aqueous solution of hexamethylenetetramine; (2) heating to 60 DEG C-80 DEG C and keeping the temperature for 0.5-4h; and (3) reducing the temperature to room temperature of 20 DEG C-25 DEG C and carrying out ultrasound treatment in an ultrasound wave dispersion instrument for 10-20 minutes and sampling to obtain the zinc oxide nano-chain by using the DNA as the template. The method is simple, easily obtains raw materials, consumes less energy and reduces the pollution to the environment; the equipment is less and the cost is low, so the method is suitable for batch production; the reaction time is short, the yield is high and no toxicity and harm exist, so that the method provides the possibility for preparing the zinc oxide nano-chain on a large scale; and zinc oxide nano-chains with different shapes can be constructed by operating the space structure of biological macromolecule DNA. The invention can simultaneously solve the problems of the control of the size and the shape of nanometer materials and the dispersion stability.

The invention discloses a method for detecting ionized layer TEC exception. The method comprises the following steps of: A, performing smoothing filtration on an ionized layer TEC time sequence; selecting the TEC time sequence of any point according to the space-time characteristics of ionized layer TEC distribution, and performing smoothing filtration on the ionized layer TEC time sequence by adopting a moving average method to obtain a smoothed TEC time sequence; B, calculating a difference between a TEC smoothing value and an observation value so as to obtain a time sequence of a smoothed residual error; C, determining the arc length of a sliding window, calculating a mean value and a standard difference, adding 2 times standard difference to the mean value to obtain an upper limit value, subtracting 2 times standard difference from the mean value to obtain a lower limit value, performing exception detection on the time sequence of the residual error, comparing the data with the upper limit value and the lower limit value, and judging that the data between the upper limit value and the lower limit value is normal; and D, storing the detection result. The method is simple, has obvious effect, creatively performs the exception detection of the ionized layer TEC based on the smoothed residual error data, deducts the long periodic tendency of the TEC, and greatly improves the reliability of the detection result.

The invention discloses an island detection method based on negative sequence power positive feedback, which belongs to the technical field of distributed generation system detection. A negative sequence power positive feedback loop is added in a distributed generation (DG) system, and when a microgrid is converted into island operation from grid-connected operation, microgrid voltage is acted with inverter output current through the positive feedback loop to increase negative sequence power, so an island can be detected by analyzing the square root H of the negative sequence power. The method comprises the following steps of: calculating the square root H of the DG output negative sequence power in the microgrid and setting the threshold theta 1 of H; and then judging whether the island occurs or not, namely judging whether the island occurs or not according to the set value of H. The method is simple in principle and wide in application range; the island can be effectively detected under various failure conditions and other terrible conditions; the method is not affected by a pseudo island; the requirement on application on site can be met; and the method has practical engineering significance.

The invention discloses a double-speed low strain method for detecting integrity of an existing pile foundation through carrying out excitation inside a pile body. The double-speed low strain method comprises the steps of location of a pile top position: finding the central position of a to-be-detected pile according to a construction drawing, and marking the central position; drilling of a foundation and a foundation pile: drilling a hole in the central marked position of the to-be-detected pile, and sampling; sensor mounting: mounting two sensors on the hole wall of the obtained drilled hole, wherein the two sensors are in different depths; vibration excitation seat mounting: mounting a vibration excitation seat below the pile top and at the upper end of the upper senor, wherein the vibration excitation seat is in close contact with the hole wall of the drilled hole and does not slide on the hole wall under the action of a vibration excitation hammer; vibration excitation of the vibration excitation hammer: selecting the vibration excitation hammer according to the pile diameter, the pile length and the defect position to perform vibration excitation for a plurality of times; and sampling and data analysis: carrying out synchronous sampling on the two sensors, and analyzing sampled signals. The double-speed low strain method has the beneficial effects that the principle is simple and clear, an acquisition device is simple, the result is more accurate and reliable, and relatively good engineering application and popularization prospect is achieved.

The invention discloses a processing method and system for safely opening an automobile door. When a driver pulls over an automobile and the driver and passengers in the automobile want to open the automobile door to get on or get off the automobile, a central processing unit in the automobile detects whether pedestrians, non-motorized vehicles and motor vehicles are behind the automobile or not according to four cameras on rearview mirrors and a back bumper and judges whether the automobile door can be opened or not according to distances of the pedestrians, the non-motorized vehicles and the motor vehicles; the central processing unit judges according to acquired information, and the automobile door can be opened if no pedestrians, non-motorized vehicles and motor vehicles are behind the automobile or the pedestrians, the non-motorized vehicles and the motor vehicles are far away from the automobile; if the pedestrians, the non-motorized vehicles and the motor vehicles are behind the automobile in short distances, the automobile door cannot be opened, and the central processing unit sends alarms to the passengers in the automobile. Therefore, collision of open automobile door with the pedestrians, the non-motorized vehicles and the motor vehicles approaching to the rear of the automobile can be avoided, safety is improved, and user experience is improved as well.

The invention discloses a closed-loop control distribution method of a multi-control surface unmanned aerial vehicle based on an angular acceleration sensor, which can be applied in flight control of an unmanned aerial vehicle with redundancy multi-control surface. The method comprises: obtaining an expected angular acceleration command through an angular acceleration command of the unmanned aerial vehicle and measured angular acceleration information, obtaining the angular acceleration of the unmanned aerial vehicle through the angular acceleration sensor, obtaining the error between an expected moment increment and an actual moment increment, and performing control distribution calculation on the moment increment error in real time to obtain the reflection angles of the control surfaces. The method provided by the invention obtains angular acceleration feedback information through the angular acceleration sensor of the unmanned aerial vehicle to form a closed-loop control distribution structure so as to achieve a good control distribution effect in the case of normal work and failure of the control surfaces, and the method is simple in calculation and convenient in realization.

The invention discloses a linear calibration method. The linear calibration method comprises the following steps: when data of a system to be calibrated have curve characters in a two-dimensional coordinate system, firstly selecting a group of sampling data, determining a basic line with two points at the two ends of the group of data, then amending the intercept and gradient in sequence on the basic line, distributing the points of the group of sampling data to one side of the amended straight line as many as possible, and finally amending the point with greater deviation integrally. At the condition of the curve characters during calibration, the rated output value is approached by adopting a universal straight line manner directly, so that the curve character that numerical values are usually convex or concave in practice is solved, the precision of actual output value is improved, the control operation difficulty is reduced, and meanwhile the method is convenient to understand and realize; the method is simple in principle and higher in universality, and is suitable for all numerical values with linear transformation theoretically.

The invention provides a reflecting surface antenna double-beam forming design method. Through creatively introducing a double-beam covering area shape characteristic analyzing step in a designing process, a targeted initial beam acquiring manner and a corresponding antenna structure configuration strategy are made. Based on searching of a reflection plane scattered field focusing point under planar wave excitation according to a conjugate field matching theory, a rotation rule in long axis direction of a longitudinal movement beam of feed sources at two sides of a focusing point is utilized, and wave beam covering matching design is performed. Through feed source position adjustment, axial torsion of the whole reflection surface antenna system is performed, thereby improving beam covering matching property of double beams and performance of a restraining area, and furthermore improving gain of a double-beam coverage area after optimization. According to the reflecting surface antenna double-beam forming design method, according to a wave beam shape characteristic relationship, antenna layout and the like, and design principles such as a wave beam shape is adjusted according to the feed source position. A design order, a parameter selecting principle and technical details in double-beam design are described, and gain improvement of the double-beam coverage area is realized.

The invention relates to a turbulence degradation image semantic segmentation method based on boundary perception and adversarial learning, and the method comprises the following steps: (1), combininga simulated turbulence-degraded image with a real turbulence-degraded image based on a turbulence imaging physical model for a turbulence-degraded image, and constructing a turbulence-degraded imagesemantic segmentation data set; (2) for the data set obtained in the step (1), constructing a DeepLabV3 + semantic segmentation model based on boundary perception in combination with fuzzy and distortion characteristics of a turbulence degradation image; (3) taking the DeepLabV3 + semantic segmentation model based on boundary perception in the step (2) as a generator, and constructing a boundary perception generative adversarial network GAN model based on adversarial learning in combination with a discriminator composed of five convolutional layers; and (4) for the GAN model obtained in the step (3), performing model training on the turbulence degradation image semantic segmentation data set obtained in the step (1) to obtain a trained semantic segmentation GAN model, and performing semantic segmentation on the turbulence degradation image by using the trained semantic segmentation adversarial network GAN model to obtain a prediction segmentation graph.

The invention discloses an islanding detection method based on voltage harmonic fault component. Through detecting differences presented by voltage signals at specific frequencies of 150 Hz, 250 Hz and 350 Hz in the case of normal grid-connected operation and islanding operation, the islanding operation state is detected. A sliding data window is used for calculating absolute values |DeltaU3|, |DeltaU5|, and |DeltaU7| for voltage specific subharmonic fault components at the grid-connected inverter end, and whether islanding happens according to the sizes of the |DeltaU3|, the |DeltaU5|, and the |DeltaU7|. The quality of electric energy outputted by the grid-connected inverter is not influenced, the transient response of the system is not disturbed, the sampling frequency is lowly required, and hardware realization is facilitated. The method of the invention can quickly and effectively detect islanding effects in the most serious islanding condition defined by the IEEE Std.1547.1 standard, is applicable to various broken circuit conditions, and has good practicality.

The invention discloses a conductor untwisted strand detection method based on region growth. The detection method comprises the following steps: 1) inputting an aerially photographed power transmission conductor a top-view image I to obtain a gray image A; and obtaining a conductor region image C through a region growth method; 2) performing color inversion on the conductor region image C to obtain a color-inverted image D, and then performing threshold segmentation by using a Bersen algorithm to obtain a binary image E; 3) performing morphological processing on the binary image E, firstly removing a small region less than a designated pixel area, and then performing the expansion processing to respectively obtain an image F and an image H; performing center axis extraction, and rotatingthe image F to obtain an image J if the image F and the image H are not parallel; and 4) judging an untwisted strand location according to the a conductor single strand slope obtained by fitting, andperforming the marking. The method disclosed by the invention is simple in principle, visual and practical, and capable of accurately positioning the untwisted strand location.

The invention discloses a multi-fault diagnosing method for an oil well with a sucker rod pump based on an indicator diagram and relates to the technical field of oil production. The multi-fault diagnose method comprises the following steps of transforming a collected ground surface indicator diagram into a down hole pump indicator diagram; setting up a standard feature library of every fault type according to existing indicator diagram data in oilfield production; adopting interval data to express an eigenvalue of every invariant curve moment eigenvector; then calculating the relevancy between a collected sample and every fault type in the standard feature library; judging the possible fault types of the collected sample, and thus improving the credibility of fault diagnosing; at last, conducting multi-fault combination on the possible fault types of the collected sample; and deciding the possible multi-fault type by calculating an F index value of each combination. The multi-fault diagnosing method is simple in principle, small in calculating complexity, easy to achieve, and high in accuracy degree of diagnosing.

The invention discloses an island detection method based on a negative sequence voltage divider in a distributed power generation system detection technology field. The method is characterized by: firstly extracting an effective value of the negative sequence voltage at a position of a micro-grid common connection point; then calculating a ratio of the effective value to the effective value of a negative sequence voltage source which is added on the common connection point; when the ratio is less than a set threshold, determining that the island is not generated; when the ratio is greater than or equal to the set threshold, determining that the island is generated. By using the method of the invention, under various kinds of fault conditions and other scurviness situations, the island can be effectively detected. The detection is only related to a network topology structure and a pseudo island can not influence the detection. An on-site application requirement can be satisfied and the method possesses actual engineering significance.

A space vector modulation method for dual three-level inverter system zero sequence voltage elimination belongs to inverter system space vector modulation methods. Action time of a zero vector in each switch period is adjusted to make an average zero sequence voltage of a dual three-level inverter system to be 0. A reference voltage vector is decomposed in to two equivalent reference voltage vectors. The magnitudes of the equivalent reference voltage vectors are the same and the angle between the directions of the equivalent reference voltage vectors is 60 degrees. The amplitudes of the two equivalent reference voltage vectors are 1 / the square root of 3 of the amplitude of the reference voltage vector, and are respectively and individually produced by two diode clamping type three-level inverters to realize elimination of an instantaneous value of a zero sequence voltage of the dual three-level inverter system. The method and the principle are simple. The structure is clear. The two three-level inverters in the three-level inverter system are made to work in a power supply mode of a single direct voltage source through a unified high-speed algorithm and space voltage vector decoupling, so that the topological structure can be effectively simplified and the cost can be reduced. The four-quadrant operation of open winding asynchronous motor can be realized.

The invention provides a method for measuring low concentration SO2 based on gas difference absorption cross section characteristics. The method comprises the following steps: firstly, primarily selecting a measuring spectrum scope of SO2 according to the SO2 ultraviolet absorption characteristics; secondly, calculating multiple difference absorption cross sections on the basis of multiple known SO2 concentrations according to a DOAS algorithm under the condition of unchanged temperature and pressure and gradually reducing the measuring spectrum scope according to the difference absorption cross section characteristics; thirdly, utilizing a statistic and standard deviation method to screen the spectrum sampling data point in a selected measuring spectrum scope and establishing an optimal sampling point data set, thereby acquiring an optimal difference absorption cross section data set; and lastly, utilizing the optimal difference absorption cross section data set and the DOAS algorithm to inverse the sulfur dioxide gas concentration. According to the method provided by the invention, the measurement principle is simple, the lower limit of the measurement is as low as 3ppm, the low-concentration gas can be accurately measured under a short optical path condition, the relative error can be controlled within + / -1.6% and the scale span error can be controlled within + / -1.2%.

The invention discloses a component automatic recognition method for a mechanical assembly process. The method comprises the following steps of: the construction of a component assembly sequence library; the acquisition of component pile images; the feature extraction of the component pile images; the three-dimensional construction of the component pile images; the feature extraction of the component pile images; and the registration recognition of components. The component automatic recognition method of the invention has the advantages of simple principle, high efficiency, high detection precision, high robustness and the like.