36results about How to "Not affected by accuracy" patented technology

The invention provides a fault location method based on single-terminal location. The method is characterized by using a transmission line travelling wave fault location device to measure the power frequency quantity of a home terminal bus of a transmission line when short circuit happens to the transmission line, determining the fault point range by an impedance method and estimating the transition resistance; then analyzing the transient voltage/current travelling waves through wavelet transform and adopting a single-terminal travelling wave method combined with the length of line to directly compute the positions of the fault points if the faults are judged to be common short circuit faults; and adopting a transmission line single-terminal fault location method combining the impedance method and the travelling wave method to compute the positions of the fault points if the faults are judged to be special short circuit faults. The method has the following beneficial effects: the accuracy of fault location is ensured by utilizing the single-terminal travelling wave method, can be within 500m and is not affected by such factors as the length of line and the like; and the method can adapt to fault location under the condition of different fault types and is basically not affected by such factors as the transition resistance, the line structure, the system operation mode and the like.

The invention relates to a groove wheel turnover device with an auxiliary positioning function. The groove wheel turnover device comprises a box body, a transmission mechanism driving a workpiece to turn over and an auxiliary positioning mechanism; the transmission mechanism is an external gearing groove wheel mechanism, and the mechanism comprises an outer groove wheel with four pull grooves and a pull disk with two pull pins; and the auxiliary positioning mechanism is a direct driving push rod cylindrical cam mechanism, the cam mechanism comprises a cylindrical cam and a driven body, the cylindrical cam and the pull disk are arranged coaxially, the peripheral face of the cylindrical cam is provided with a curved concave groove, the driven body is driven by the cylindrical cam to conduct reciprocating motion in the axial direction of the cylindrical cam, the driven body can be driven to conduct auxiliary positioning of the groove wheel at a suitable position by the synchronous rotation of the cylindrical cam and the pull disk rotate, and the groove wheel turnover device has the beneficial effects that the positioning precision is high, and the positioning is reliable.

The invention discloses an airborne photoelectric system target motion vector estimation method based on aerial photography measurement. The method is based on linear motion compensation and target tracking functions of an airborne photoelectric system, and comprises: using an aerial photography airborne photoelectric system oblique photogrammetry principle, and obtaining a motion vector of a ground target in an image space coordinate system in real time by performing automatic target tracking under forward motion compensation on the ground moving target; and then by establishing a corresponding relation between the image space coordinate system and the ground photogrammetry coordinate system, according to the motion vector of the target in the image space coordinate system and the azimuthangle of the aiming line of the airborne photoelectric system, solving the motion vector of the target in the ground northeast coordinate system, including the motion rate and direction. The airbornephotoelectric system target motion vector estimation method based on aerial photography measurement does not depend on real-time positioning of the target, and has the characteristics of high precision, high calculation speed and the like.

The invention provides an auto-calibration platform orienting a modular joint toque sensor, belongs to the field of joint torque sensor calibration, and solves the problem that the manual torque loading method has less data acquisition and cannot perform dynamic test and the problem that the accuracy of the input torque value is low through the motor stalling input torque. A brake, a primary coupler, a torque checking sensor, a secondary coupler and a cylindrical closed box are arranged on a base in turn. A transmission shaft rotates to drive the brake to rotate through the secondary coupler,a torque checking sensor connecting shaft, the torque checking sensor, a hollow slip ring connecting shaft and the primary coupler. The brake can provide the constant torque load. Multiple sets of input instructions are given to the brake, the torque size measured by the torque checking sensor is acquired and the value is assigned to the torque sensor of the modular joint so that auto-calibrationand dynamic calibration can be realized. The auto-calibration platform is mainly used for calibration orienting the modular joint toque sensor.

The invention discloses a wind energy indirect prediction method suitable for multivariable input of a wind power plant. The method comprises the steps of 1) predicting a wind speed; 2) predicting an average trend component by adopting a polynomial fitting method and forecasting a random component by adopting a frequency reduction prediction method; 3) integrating forecasting results of two independent components, namely, the average trend component and the random component to obtain a wind speed prediction value; and 4) making a historical wind energy and wind speed data distribution graph, performing polynomial fitting, generating a power curve, and calculating a wind energy prediction value according to the wind speed prediction value by utilizing the power curve so as to realize wind energy indirect prediction. The wind power plant with relatively high wind energy fluctuation also can be accurately predicted without depending on the influence on wind power plant modeling. In addition, the prediction method is a data-driven and adaptive method, and a prediction result of the prediction method does not depend on priori knowledge of a user.

The invention belongs to the field of medical detection, and particularly relates to a portable biological tissue detection device based on a space deviation Raman spectrum technology. The biologicaltissue detection device comprises a computer, a detection controller, a laser, a spectrometer, a three-freedom-degree mechanical arm, an SORS detection head, a detection platform and a sealed sample chamber. An SORS technology is used; a sample to be tested is sealed; during the storage in dark places, equipment can be directly used for detecting the Raman spectrum, so that the detection is performed in a targeted way; the SORS technology and an SERS technology are used in a combined way; the defect of lower SORS detection biological sample signals is overcome; meanwhile, the problem that theSERS detection sample can only be used for detecting the expression information is solved; a novel SORS detection head is designed; the detection head has the advantages that the detection sensitivityis high; the control is easy; the displacement control precision is high; the influence by the precision of the step motor is avoided. The novel biological sample storage device is designed; the sample can be stored for a long time; the SORS signal detection can be detected at any time.

The invention provides a pedestrian behavior intention prediction method based on multi-task learning. The method comprises the following steps: a training sample set is constructed; a pedestrian behavior intention prediction model is constructed by using the basic network, the attitude detection network and the intention recognition network, the basic network takes a single frame image in the training sample set as input, image features are extracted, and a feature map is obtained; an encoder part of the attitude detection network comprises a part intensity field sub-network and a part association field sub-network which respectively take the feature map as input and take the joint feature map and the skeleton feature map as output, and a decoder part of the attitude detection network obtains a pedestrian attitude image according to the joint feature map and the skeleton feature map; the intention recognition network takes the feature map as input and takes the pedestrian behavior intention image as output; and the pedestrian behavior intention prediction model is trained, and the pedestrian behavior intention is predicted by using the pedestrian behavior intention prediction model. According to the method, a pedestrian detector does not need to be adopted, each pixel is processed separately, the running time is constant, and the prediction effect meets the requirement.

The invention discloses a multi-step wind power forecasting method based on singular spectrum analysis and locality sensitive hashing. The method is charactierzed by decomposing historical wind powerdata of a wind power plant into two independent components through singular spectrum analysis, the independent components being a low-frequency average trend component for reflecting wind energy overall change trend and a high-frequency fluctuation component for reflecting intermittency and fluctuation of wind respectively; reconstructing the two components in a phase space to obtain an average trend section and a fluctuation component section; and finding similar average trend sections of an average trend section to be forecasted through locality sensitive hashing, and carrying out locality predication. To prevent accumulated errors brought by separate forecast of each component and fixed error brought by forecast of only one component, the forecast input is combination of the similar average trend sections and corresponding fluctuation component sections, and finally, a prediction result of wind power output power is obtained. The method is clear in physical significance in wind power plant generation power prediction, short in prediction time and accurate and stable in prediction results; and the prediction results do not rely on prior knowledge of users.

The invention discloses a multichannel liquid sample reagent rapid charging reactor which comprises two parts of a sample reagent charging device and a preset first reagent cuvette. A piece of plastic slab in the sample reagent charging device is provided with multipath sample reagent dosing holes with equal spacing. The holes are corresponding to a rotating shaft sample communication groove on a rotating shaft in the sample reagent charging device. A liquid passes through via holes on the rotating shaft and is introduced into end sample conducting holes. Through rotation of the rotating shaft and a centrifugation method, a sample reagent is rapidly introduced into the preset first reagent colorimetric ware through the end sample via holes. The sample and reagent are mixed and react. And results of each projects are determined through apparatuses. By the adoption of the charging reactor, the range of measured projects is increased. Sample application accuracy is not influenced by precision of a sample injector. The operation process is simple. The charging reactor provided by the invention is especially suitable for human emergency biochemical detection, basic medical unit biochemical detection and animal biochemical detection, and can also be applied in other fields such as immunology, microbiology, molecular biology and the like.

The invention relates to a scanning device and a scanning method for recording character sequence. The method comprises the following steps that 1) when the scanning device moves, an image fragment sequence including idler wheel images and character sequence images is collected; 2) any two image fragments in the collected image fragment sequence are processed, image partition is conducted on the two image fragments, and two idler wheel image fragments and two character sequence image fragments are obtained; 3) the motion displacement of idler wheel displacement signs arranged on idler wheels is calculated according to the two idler wheel image fragments; 4) the motion displacement of the scanning device is calculated according to the mapping relation between the motion displacement of idler wheel displacement signs of the idler wheels and the motion displacement of the scanning device; 5) the two obtained character sequence image fragments are combined according to the motion displacement of the scanning device, and a complete character sequence image is formed. The scanning method is wide in application range, and has a better processing effect on data fragments which include more repeated information and are sparse in information.

The invention discloses a wind energy forecasting method based on a trend detector and a mathematical morphology operator. The concepts of oscillation element and weight barycenter are introduced in the method, on this basis, an average trend detector is designed in combination with a conventional mathematical morphology operator, and unstable historical wind energy data of a wind power plant are decomposed into two dependent components, namely a low-frequency average trend component and a high-frequency random component, the average trend component reflects the total variation trend of wind energy, and the random component is specific embodiment of mutability and randomness of wind energy; the average trend component is forecast by adopting a sine forecasting operator, and the random component is forecast by adopting a local mathematical morphology operator; the forecasting results of the average trend component and the random component are added to obtain the forecasting result of the wind electricity output power of the wind power plant. The wind energy forecasting method disclosed by the invention makes full use of the advantages of mathematical morphology such as simple operation and high speed, so that the forecasting efficiency is improved, and the method is stable in forecasting result and high in forecasting precision.

The invention relates to the field of industrial automation, in particular to a high-precision station switching device simple in structure and low in manufacturing cost. The high-precision station switching device comprises a working table, a station rotating plate, a servo motor and a constant-torque motor. A reduction box, the servo motor and the constant-torque motor are disposed on the working table. The reduction box comprises a body, an output end, a first input end and a second input end. The station rotating plate is connected with the output end of the reduction box. The servo motor is connected with the first input end of the reduction box. The servo motor provides drive torque for the first input end. The constant-torque motor is connected with the second input end of the reduction box and provides resistance torque for the reduction box. The drive torque output by the servo is more than two times of the resistance torque output by the constant-torque motor.

The embodiment of the invention provides an image processing method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the line-preserving distortion correction of a target image, and obtaining a target line-preserving distortion correction image corresponding to the target image; inputting the target line-preserving distortion correction image into a shape-preserving network to obtain a target shape-preserving distortion correction mapping graph corresponding to the target image; and performing shape-preserving distortion correction on the target line-preserving distortion correction image by using the target shape-preserving distortion correction mapping graph to obtain a corrected image. The obtained corrected image has a line-preserving effect and a shape-preserving effect, the target shape-preserving distortion correction mapping graph corresponding to the target image is obtained through the shape-preserving network, multiple iterations are not needed, the speed of the image processing process is improved, meanwhile, the position of an image human body does not need to be determined through a portrait segmentation model, and the image processing efficiency is improved. And the conformal effect is not influenced by the precision of the portrait segmentation model.

The invention discloses a weather-adaptive target detection network model and method, and the method comprises the steps: collecting sunny day video data and rainy day video data through a visual collection device, and writing a script to process a video into a picture; marking the sunny day image data, taking a marked sunny day data set as a source domain data set, and taking an unmarked rainy day data set as a target domain data set; putting the source domain data set and the target domain data set into a weather-adaptive target detection network model for training to obtain a weight parameter file of the model; loading the weight parameter file to obtain a detection network model; and extracting a picture from the video stream according to a set frame number, inputting the picture into the detection network model for prediction, and displaying a prediction result. The detection precision is guaranteed and even improved under the weather influence condition, and meanwhile the problem that the workload of manual data labeling is large is solved by the method.

The invention provides a position sensing mechanism, which comprises a coding element used as a sensing signal source, a reading element used for sensing a signal of the signal source, and a processing unit used for receiving and analyzing a sensing signal output by the reading element, the reading element is technically characterized in that a magnetoresistance unit used for sensing a signal source signal in the reading element is a tunneling magnetoresistance (TMR), the magnetic moments of a reference layer and a free layer of the tunneling magnetoresistance are perpendicular to each other, the magnetic moment of one of the reference layer and the free layer which are perpendicular to each other is parallel to a film surface, and the magnetic moment of the other layer is parallel to a film surface, and the magnetic moment of the other layer is parallel to the film surface. And the magnetic moment of the other layer is vertical to the film surface.