133results about How to "Avoid error effects" patented technology

The invention discloses an electric power load prediction method based on a long-short-term memory neural network. The method comprises the following steps: inputting power load data of historical moments, regional characteristic factors and a specified time period required to be predicted; carrying out training modeling on the power load data and the regional characteristic factors at historicalmoments by adopting an LSTM network, and generating a neural network model of power load prediction; predicting the electric load through the established neural network model; and finally, outputtingan electric load prediction result of the region at the appointed time through an output unit. The method has the advantages that the neural network model can be built according to the data over the years and the building characteristic factors, the power consumption load in the appointed time period can be accurately predicted, and the power consumption load prediction accuracy is effectively improved.

The invention discloses a straight line detection method and device and belongs to the field of image processing. The straight line detection method comprises the steps that gradient histogram feature sets of all pixel points in an image are obtained, wherein the gradient histogram feature sets are used for reflecting straight line features of the part where the pixel points exist; according to the gradient histogram feature sets of all the pixel points, at least one alternative direction of a straight line to be detected is determined; according to the alternative direction, the accurate direction and position of the straight line to be detected are determined. The problems that according to a straight line detection method in the prior art, when edge strength is weak or noise interference exists, a straight line is broken, and a detection result is inaccurate are solved. According to the straight line detection method, Hough transfom is not adopted, and image binaryzation is not needed, so that the influence of errors produced by binaryzation is sufficiently avoided, and straight line detection accuracy is improved.

The invention discloses a trans-scale geometrical parameter measurement method based on microscopic visual sense, belongs to the technical field of computer vision measurement, and relates to a measurement method suitable for the spatial position, physical dimension and form and location tolerance of three-dimensional trans-scale parts of a graph type. The method employs a binocular microscopic visual system to collect a characteristic image of a to-be-measured part, and finally achieves the quick measurement of geometrical parameters through three steps: microscopic visual calibration, image characteristic extracting and matching, and local coordinate system and machine coordinate system resolving. The method employs a mode of combining a numerical control machine tool with microscopic visual sense, and solves problems that a measurement process is tedious and the measurement is difficult because of large size span of parts. The method employs non-contact measurement, is simple in operation, cannot damage a workpiece, and achieves the on-machine measurement of the workpiece. The method prevents a secondary clamping error caused by the disassembly of the workpiece from affecting measurement results, finally achieves the accurate and quick on-machine measurement of trans-scale parts.

The invention provides a greenhouse information automatic monitoring method based on a suspension-type slide rail platform. The method realizes online patrol monitoring of greenhouse crop comprehensive information of image and infrared temperature characteristics of plant nutrition, water, diseases and diseases and pests as well as plant crown breadth, plant height and fruit growth characteristicsand the like through multi-sensor information fusion by adopting a multi-sensor system, utilizing a binocular vision multifunctional camera and combining with a laser ranging sensor and an infrared temperature sensor. By mounting the multi-sensor system on the suspension-type slide platform and combining an elevating system and an electric control rotation pan-tilt, not only accurate positioningand stationary point detection can be realized in the detection heading direction, but also multi-sensor information patrol detection at different detection distance, different pitching view fields and different detection angles is realized; and by adjusting the detection range and pan-tilt azimuth angle, the method not only can detect big plants, such as tomatoes and cucumbers, but also can meetdetection requirements of lettuce and medium and small plants in different growth periods.

The invention relates to a physical model apparatus and a physical model apparatus method for simulating inclined rock stratums with different inclination angles. The physical model apparatus comprises a reaction force system, a loading system, a preparation system and a compaction system, wherein the reaction force device is formed by connecting a six-sided reaction force frame through high-strength bolts and supplies reaction force to the loading system, the loading system loads on a model material by using a hydraulic jack to simulate ground stress, the preparation system is used for holding and preparing the inclined rock stratum of the whole test, and the compaction system uniform compacts each layer of the similar material in the preparation system by the hydraulic jack of the loading system. According to the present invention, with the apparatus and the method, the physical model test of the horizontal rock stratum can be performed, the physical model tests of the inclined rockstratums with different inclination angles can be met, the influence of the artificial material spreading and compacting on the test result can be reduced, the layered compaction preparation of the inclined rock stratum is achieved, the influence of the test position does not exist, the requirements of different sites can be met, and the apparatus and the method can be used for large-scale three-dimensional physical model tests.

The invention discloses a system and method for testing the breaking tenacity of a material based on a temperature-controllable electrochemical hydrogen charging environment. The method comprises the following steps: fixedly connecting a target material testing component, adding electrolyte, and inserting a temperature sensor, an auxiliary electrode, an extensometer and the like, and respectively connecting a temperature control assembly, an electrochemical working station, a stretcher and a computer; starting the electrochemical working station, carrying out electrolysis to produce a large amount of hydrogen on the surface of the target material testing component, and introducing hydrogen into the testing component through permeation and diffusion, so as to realize the hydrogen charging of the testing component; after reaching a set hydrogen charging time, starting the stretcher to carry out tensile experiment operation on the testing component, and calibrating breaking tenacity of a target material. According to the system, the performance parameters such as the breaking tenacity of the target material in a given hydrogen charging environment are synchronously measured.

The invention relates to an analog device for on-line monitoring of dissolving gas in transformer oil and a detection method thereof. The analog device comprises a hydraulic circulatory system, a fuel tank temperature control system and a gas collection detection system; the hydraulic circulatory system comprises a fuel tank and an oil pump which connected through a circulation pipeline, a gas inlet port is arranged on the circulation pipeline, and a flow control valve is also provided; the fuel tank temperature control system comprises a heating pipe and a temperature sensor arranged in the fuel tank, a temperature controller respectively connected with the heating pipe and the temperature sensor; the gas collection detection system comprises at least one group of a gate valve, an air chamber and a gas detector which are connected through the pipeline in order, the gate valve is connected with the fuel tank through the pipeline, an oil gas separating membrane is arranged between the gate valve and the air chamber, and the analog device detects oil gas separating effects by changing difficult conditions to screen the optimal condition. Compared with the prior art, the analog device has the advantages of simple structure, convenient operation, and accurate simulation result.

The invention discloses an automatic take-off control strategy design for a small unmanned helicopter, comprising steps of 1 determining whether a take-off order is responded, if yes, automatically starting up an engine and applying rotation speed stable control, 2 recording information of current longitude, height and yaw angle and solving mean values,3 if the step 2 determines that the rotation speed of the engine is stable, performing self-increscent on the collective pitch, 4 if the step 3 determines that the collective pitch reaches the pre-off-land collective pitch, performing plane position and posture control on the helicopter and continuously performing self-increment on the collective pitch until the helicopter is off the land, 5 if the step 4 determines that the helicopter is off the land, updating the collective pitch balancing value and invoking a vertical channel control law until the preset take-off height is approached, and if the step 5 determines that the take-off height is reached, and invoking a spot hovering control law to finish the take-off. The automatic take-off control strategy design for the small unmanned helicopter can improve the automatic take-off capability of the miniaturized unmanned helicopter and enhances the usage range, and is small in complexity and high in reliability.

The embodiment of the present invention discloses a method and a device for correcting CQI value to avoid the influence of error of the CQI measured value. The method for correcting the CQI value mainly comprises the steps of obtaining information representing wireless channel characteristics, determining a target value of the BLER according to the information representing the wireless channel characteristics, and correcting the CQI measured value according to the BLER measured value and the determined BLER target value. The method and device can determine the BLER target value according to obtained information representing the wireless channel characteristics, and correct the CQI value according to the BLER target value, taking different channel conditions into account, thereby avoiding the influence of the error of the CQI measured value and improving the transmission rate and idle port capacity.

The invention discloses a heating control method for a soya-bean milk machine and a soya-bean milk making method. The temperature variation value of a soya-bean milk material is used for judging whether the soya-bean milk reaches a boiling point so as to control the working mode of the heater. Compared with the prior art, the method disclosed by the invention is characterized in that error influence by the temperature sensor can be avoided, and the variation of the boiling points at different altitudes can be automatically adapted.

The invention relates to an automatic adjusting device and discloses a sample position automatic adjusting device in a product quality detecting system. The sample position automatic adjusting device comprises an objective table on which samples are placed, a sample positioning unit arranged above the objective table and used for detecting whether a sample is positioned in an assigned position or not, and a sample position adjusting unit connected with the objective table and used for adjusting the height and the horizontal position of the objective table so as to enable the sample to reach the assigned position. According to the invention, by arranging the sample positioning unit and the sample position adjusting unit that are mutually matched, the automatic adjustment of the height and the horizontal position of the sample can be achieved, the detection of the sample can be accurately performed, the error influences caused by that the height cannot be adjusted in real time due to thickness difference during sample detection can be avoided, the accuracy of the detection result is enhanced, the automatic control is achieved, the experimental accuracy is improved, the results are enabled to be more reliable, the operations are effectively reduced, and the operation method of the detecting system is enabled to be easier for common technical personnel to master.

The invention provides an automatic bait casting metering device of an offshore fish cage. The device comprises a shell, a center rotary shaft, a rotary wheel, a torque measuring device, a drive device and a rotation speed measuring device, wherein the upper end and the lower end of the shell are provided with a feeding inlet and a discharge outlet respectively, the center rotary shaft is vertically arranged in the shell, the rotary wheel is horizontally installed on the center rotary shaft and can rotate along with rotation of the center rotary shaft, the torque measuring device is used for measuring the torque of the center rotary shaft, the drive device is connected with the center rotary shaft and is used for driving the center rotary shaft to rotate, and the rotation speed measuring device is fixed to an output shaft of the drive device. The automatic bait casting metering device is simple in structure, accurate in metering, good in reliability and capable of achieving real-time feedback and control, the metering influencing factors are few, the device cannot be influenced by gravity changes such as mechanical vibration in the conveying process of granulated feed, and accordingly error influences caused by static measurement such as weighing are avoided.

The invention relates to the field of inertia detecting instruments. In order to provide an accelerometer having the advantages of high measuring precision, high resolution ratio, wide dynamic range, simple and compact structure, electromagnetic interference resistance and the like, the technical scheme includes that a quartz flexible accelerometer based on laser self-mixing interference comprises a meter and a digital rebalance loop, wherein the meter comprises a shell, a pendulum component, a torquer and a laser self-mixing interference signal sensor, and the digital rebalance loop comprises a laser device driving power supply, a photoelectric detection circuit, a data acquisition circuit, a digital signal processing circuit and a torque current generator. When acceleration is inputted,the laser self-mixing interference signal sensor detects displacement of the pendulum component and converts the displacement into electric signals, the electric signals are converted into current signals in proportion with the acceleration by the aid of the digital rebalance loop, the current signals are fed back to the torquer and lead the torquer to generate rebalance torque, and the acceleration can be obtained by detecting the current of the torquer. The quartz flexible accelerometer is mainly applied to acceleration measurement.

The invention relates to a speed spectrum automatic picking method based on a convolutional neural network, which belongs to the field of seismic data processing. The method first randomly generates aspeed picking curve and a horizontal gather, a reaction correction method is then used to generate a simulated gather, Radon transform is made based on the gather to generate a speed spectrum, and the process is repeated to generate an arbitrary number of training labels. The neural network with 4 convolutional layers and 2 fully connected layers is designed, a maxpoling layer and a dropout layerare added to reduce the calculation amount, and the occurrence of overfitting is prevented to a certain extent; and finally, by using the randomly generated labels, the neural network is trained, andthe network is used to recognize a speed spectrum not participating in training. The convolutional neural network can effectively pick up the position of a speed energy mass and generate a corresponding speed curve, the precision is higher in comparison with a theoretical curve, the process of manually picking energy masses is avoided, the manpower is saved, and the error influences introduced byhuman factors can be avoided.

The invention relates to a method for testing a capacitive touch chip. In the method, a numerical value output by a comparer is compared with a theoretical voltage value to form a parameter, and then the parameter is compared with a preset parameter, therefore, the influence of errors can be overcome, and the defect of incapability of testing a chip because a direct output value is not matched with the preset parameter when the initial of a binary numerical value appearing in a test period is a symbol also can be overcome, and finally whether a chip is qualified or not is accurately judged and rapidly.

The invention provides a method for measuring an atmospheric refraction value at a large zenith distance. The method comprises the following steps: S0, providing a telescope, an image acquisition terminal, a processor and a corner reflector, and measuring the actual included angle L0 between the two reflectors, wherein the corner reflector is provided with two reflectors to enable the telescope has first and second fields of view; S1, observing two different zenith areas positioned at the small and large zenith distances by virtue of the first and second fields of view, and extracting star images acquired by the image acquisition terminal; S2, distinguishing which zenith area the observation star image in each star image comes from, and acquiring the corresponding celestial coordinates of each observation star image; S3, marking the zenith as Z, and solving horizontal coordinates (A0, h0) and (A1, h1) corresponding to the actual field-of-view centers sigma0 and sigma1 in the first and second fields of view according to the corresponding celestial coordinates of each observation star image; and S4, acquiring the atmospheric refraction value at the actual field-of-view center sigma1 in the second field of view according to Z, A0, A1 and L0. According to the method disclosed by the invention, the influence of place parameters, instrument parameters and other system errors is avoided, the accumulative measurement error is eliminated, and the measurement accuracy is improved.

The invention relates to a method for accurately testing the content of salt in seawater desalination liquid based on ion concentration and equipment special for the method. By the method, an activity coefficient is corrected, measuring accuracy is improved, and the application range is expanded. Values of standard potential E<0> and slope S can be obtained through correction, a functional relation between the concentration of sodium ions in the seawater desalination liquid and the electromotive force of NaCl in the seawater desalination liquid is deduced according to a Nernst equation. During use, two indicating electrodes, namely a chloride ion selective electrode and a sodium ion glass electrode are used as a positive electrode and a negative electrode and are connected with an instrument to form a primary battery, the electromotive force of the seawater desalination liquid with unknown concentration is measured and substituted into the functional relation, and the display instrument can directly display the accurate concentration of the sodium ions in the seawater desalination liquid. The method is suitable for measuring low concentration in a narrow range, and is also suitable for accurately measuring the content of the salt in the seawater desalination liquid with high salt concentration, the accuracy of the method is higher than that of the conventional method, and a standard buffer solution is not required to be added in a linear range.

The invention relates to a method for assisting with the parking of a vehicle (2) in a parking space (1) lateral to the vehicle (2), wherein a faulty length L+ of the parking space (1) is detected by a sensor arrangement of the vehicle (2) as the vehicle (2) passes the parking space (1) on a curved movement path (6). Additionally, to determine the actual length L of the parking space (1), a movement path of the vehicle (2) deviating from straight travel of the vehicle (2) while passing the parking space (1) is detected, for example based on the steering angle of the vehicle (2), and from the path a corrective value for the detected length L+ of the parking space (1) is determined. The actual length L of the parking space (1) is thus determined with high accuracy, without requiring additional measured values, such as the orientation of the parked vehicles (4, 5) or the depth of the parking space (1).

The invention provides a method for measuring the surface activation degree of a modified powder material. The method comprises the steps: adding water and an over dry modified powder material with a ratio of material to liquid of 0.02-0.05g / mL in a container, and standing after shaking; then adding liquid paraffin in the container according to the volume ratio of the liquid paraffin to the water of 1:(1-4), and standing until a water phase and an oil phase in the container demix after shaking; separating the water phase from the oil phase, and drying the modified powder material in the water phase to constant weight; and finally, calculating an activation index H of the modified powder material. According to the method, by using change of surface hydrophilicity and hydrophobicity before and after the powder material is modified, powder materials which are modified by a modifying agent and not modified by the modifying agent are dispersed in the water phase and the oil phase respectively for realizing separation, and thus the activation index is calculated. Compared with a conventional method for measuring the activation index, the method has the advantages that by using a water-oil two-phase measuring system, the applicable range is greatly widened, and the reliability and stability of a measurement result are improved.

The invention discloses a rock core capillary self-absorption simulation instrument. The instrument comprises a rock core holder, the rock core holder is connected to a confining pressure pump by a third electric control valve, and a second pressure meter is arranged between the rock core holder and the third electric control valve; one end of the rock core holder is connected to a back pressure value, and the other end of the rock core holder is connected to an intermediate container with scale through a second electric control valve and a fluid flowmeter in order; the intermediate containerwith scale is connected to a pressure apparatus though a first electric control valve, and a first pressure meter is arranged between the intermediate container with scale and the first electric control valve; the first electric control valve, the second electric control valve and the third electric control valve are connected to an apparatus for controlling system as well as acquiring and processing data. The instrument can simulate the capillary self-absorption process under the conditions of overbalance, underbalance, and near-balance of stratum; a computer is used in an experiment processfor realizing processing automation of test and data results, in order to obtain self-absorption speed and intrusion depth and other parameters of capillary, and better simulate self-absorption of rock core.

The invention belongs to the field of equipment detection, and particularly relates to an image-identification technology-based non-contact displacement detecting device and a method for detecting displacement of a detected device through the device. The device comprises an image acquisition module (1), a laser generation unit (2), a data storage and transmission module (3), a power supply and control module (5) and an objective target (8); the image acquisition module (1), the laser generation unit (2) and the data storage and transmission module (3) are connected with the power supply and the control module (5) separately; the image acquisition module (1) is connected with the data storage and transmission module (3); the distance between the image acquisition module (1) and the laser generation unit (2) is fixed; the objective target (8) is arranged on the detecting device; at least one of the exposed surfaces of the objective target (8) is provided with at least two patterns with geometric centers. The device has the advantages that convenience is brought to carrying and installation; the application range is wide, the adaptability is high, and the service life is long; the measurement method is simple, and the result is accurate; deformation and displacement in three directions can be detected at the same time, and the efficiency is high.

The invention provides a fracture toughness testing device under a low temperature high pressure hydrogen charged environment. The device comprises a sealed box body and a tensile device. The tensiledevice is set on the sealed box body. The device also comprises a refrigeration assembly, a liquid pressurization assembly, an electrochemical assembly, a detection assembly and an industrial personalcomputer. The liquid pressurization assembly is used for supplying a high pressure electrolyte solution to the interior of the sealed box body. The refrigeration assembly is used for cooling the electrolyte solution. The electrochemical assembly is used for generating hydrogen on a surface of a tensile test sample. The detection assembly is used for detecting a pressure and a temperature in the sealed box body and an opening crack state of the tensile test sample. The industrial personal computer is used for controlling the refrigeration assembly, the liquid pressurization assembly, the electrochemical assembly and the tensile device through collection and analysis of information of the detection assembly. The device is used for testing fracture toughness under a low temperature high pressure hydrogen rich environment. A relatively systematical and complete testing scheme is formed. A low temperature high pressure hydrogen rich deep sea environment is effectively simulated. Reliability of final testing data is improved.

The invention relates to the technical field of hydraulic system part defect detection, in particular to a hydraulic gear pump defect detection method and system based on artificial intelligence. The method comprises the following steps: acquiring a tooth surface target area of a gear pump image; obtaining the complexity of the target area; obtaining the entropy of the target area under each color channel, obtaining a spatial distribution vector according to the entropy of the target area, and obtaining an abnormal index according to the difference between the spatial distribution vector and the spatial distribution vector of the normal tooth surface; obtaining the difference degree between any two pixel points in the target area, and obtaining the difference index of the target area according to the difference between the difference degree and the difference degree of the normal tooth surface; and obtaining a defect confidence coefficient according to the complexity, the abnormal index and the difference index, and determining that the gear pump has defects when the defect confidence coefficient is greater than a preset threshold value. The target area is comprehensively considered from multiple aspects, different characteristic parameters are obtained, and the accuracy of tooth surface defect detection is improved.

The invention discloses a high current injection probe comprising a body. The body comprises a first current injection clamp and a second current injection clamp. One end of the first current injection clamp is connected with one end of the second current injection clamp through a hinge. A magnetic connector is arranged on the other end of the first current injection clamp. The other end of the first current injection clamp is magnetically adsorbed on the other end of the second current injection clamp through the magnetic connector. The closing tightness of the high current injection probe can be effectively enhanced so as to enhance the accuracy of the electromagnetic compatibility test. The invention also provides an electromagnetic compatibility test device.

The invention discloses a four-electrode conductivity adjustable temperature sensor. A main structure of the four-electrode conductivity adjustable temperature sensor comprises two shells which are detachably assembled, wherein the shells are filled with a liquid with known conductivity (or resistivity), a rubber gasket, an adjusting rod, two pairs of electrodes and a wire. Relative positions of the two shells are adjusted through screws so that a distance between two voltage electrodes is changed, corresponding voltage drop is changed too, conductivity is adjustable, and high consistency between batches of the temperature sensors and high accuracy of a conductivity value are obtained. Compared with an existing temperature sensor, the temperature sensor is not affected by external and selfstresses, avoids temperature measurement errors caused by the stresses, avoids measurement errors caused by electrode polarization, enables temperature measurement to be more accurate, and is simplein structure, small in size, high in precision, good in long-term stability, good in integration, high in batch consistency and so on.

The invention relates to a method for utilizing a general instrument for testing a pulse doppler fuze starting characteristic. According to the method, a pulse emission signal of a fuze under test isdirectly utilized as a synchronizing signal for triggering a vector signal generator so as to produce a radio-frequency signal, no other instrument is needed, and the working frequency can reach to amillimeter wave, so that the cost is lower; and the method is simple and convenient to operate, adjust in parameter adjustment, good in universality and wider in application range. Particularly, the vector signal generator can be combined with the synchronizing signal of a pulse doppler fuze so as to carry out delay adjustment for adapting to synchronizing datum drift of the pulse doppler fuze. According to the method provided by the invention, data can be conveniently adjusted according to the selected parameters such as a trajectory and speed, the vector signal generator emits vector echo data consistent to the current trajectory parameter, and a pulse doppler fuze starting parameter is acquired and recorded. The method can realize zero delay and can particularly adapt to a pulse synchronizing signal clock of the pulse doppler fuze, and the error influence caused by clock asynochronization of the vector signal generator and the fuze under test is avoided.

The invention relates to a sow estrus analysis method and device, a terminal and a computer readable storage medium. The method includes: judging whether the standing times and standing time of a sowat night exceed an estrus threshold or not, and if so, judging that the sow is in estrus; acquiring the standing times and standing time of the sow in estrus, and removing disturbance data caused by the collective standing of sow around the sow in estrus to obtain effective estrus data; performing piecewise linear regression on the effective estrus data to obtain an estrus trend line; determiningthe estrus ending time and optimal insemination time of the sow according to the estrus trend line. The sow estrus analysis method and device, the terminal and the computer readable storage medium have the advantages that the estrus state of the sow is precisely judged through scientific quantitative calculation, the estrus ending time and optimal insemination time of the sow in estrus is precisely speculated, and mating precision and quality are increased effectively.

The invention discloses an adjustable microwave circuit test fixture. The fixture of the invention is a fixture which can fix microwave circuit boards of different shapes and sizes; the fixture can be divided into a bottom sheet layer, a gasket sheet layer and a pressing sheet layer which are distributed sequentially from bottom to top; the gasket sheet layer includes two metal sliders and an SMA connector; slide slot structures are formed in a bottom plate and pressing sheets; and fixing holes are formed in the metal sliders and are used for fixing the SMA connector. According to the adjustable microwave circuit test fixture of the invention, the positions of the metal sliders and the pressing sheets are adjusted through the slide slots and are fixed through screws, and therefore, the microwave circuit boards of different shapes can be well fixed, and the microwave circuit boards can be made to be in multi-point grounding states; since the SMA connector is fixed on the metal sliders through screws, and therefore, problems caused by poor welding or poor grounding in a circuit test can be solved, and the problem of the falling of the SMA connector caused by cable connection can be also solved.

The invention discloses a transformer bushing damp state evaluation method based on a high-voltage frequency domain dielectric spectrum, and the method can achieve the quantitative evaluation of the average water content of a transformer bushing through building a quantitative diagnosis model between the ratio of the real parts of complex capacitors with the frequency points of 1mHz and 10kHz, and the average water content. Therefore, the technical problem that the damp state of the oil paper insulation sleeve can only be diagnosed qualitatively but cannot be diagnosed quantitatively is solved, the reliability of damp state evaluation is improved, meanwhile, the error influence of the geometric dimension of the transformer sleeve on damp state evaluation is avoided, and the accuracy is improved.

The invention discloses a sparse adaptive DOA estimation method and system based on a pseudo-inverse perception dictionary, and the method comprises the steps: receiving K narrow-band signal sources through an M-element uniform linear array, and obtaining an observation vector of a noise-free field; taking the observation vector as an initial residual error, adopting a sparsity adaptive orthogonalmatching pursuit method based on a pseudo-inverse perception dictionary, obtaining a residual error value corresponding to the current stage by continuously iterating the step length, obtaining a minimum residual error value according to the convergence comparison of the residual error values of different stages, and obtaining a corresponding sparse coefficient; and obtaining the directions of the non-zero elements in the K narrowband signal sources. According to the method, the convergence value of each step iteration is compared, the step corresponding to the minimum convergence value is used as the sparsity, the threshold error is effectively eliminated, and the application range is expanded; the advantages of the perception dictionary are combined, and a new design criterion of the pseudo-inverse perception dictionary is provided, so the error influence caused by atom strong correlation under the condition of high resolution is overcome, and the direction of the incoming wave is positioned more accurately.