8results about How to "Reduce error interference" patented technology

A novel knowledge distillation method from binocular parallax to monocular depth comprises the following steps: S1, training a parallax depth conversion module: inputting first training data into a pre-trained binocular stereo parallax teacher model to obtain a parallax map; the first training data comprises a left eye color image and a right eye color image; inputting the disparity map and the original disparity reference map into a disparity depth conversion module to obtain a pseudo label depth map; calculating a loss function according to the pseudo-label depth map, and updating parameters of a parallax depth conversion module; s2, mixing the pseudo-label depth map output by the parallax depth conversion module with the original true label depth map to obtain a mixed label depth map; s3, training a monocular depth estimation student model: training the monocular depth estimation student model through the second training data; the second training data comprises a mixed label depth map and a corresponding original reference map, and the method can achieve the training on a data set with a lower data size, and improves the global precision.

The invention provides a sewage dephosphorization control system, and relates to the technical field of sewage treatment, the sewage dephosphorization control system comprises a sewage treatment device, a detection device and a control host; the sewage treatment equipment is used for carrying out sewage dephosphorization treatment according to the target control parameters and the corresponding target parameter values; the detection equipment is used for detecting sewage data parameters before and after sewage dephosphorization treatment of the sewage treatment equipment and sending the sewage data parameters to the control host; wherein the sewage data parameters comprise inlet parameters, outlet parameters, process parameters and control parameters; a target causal prediction model is stored in the control host. The sewage dephosphorization control system provided by the invention can effectively reduce error interference on causal correlation analysis caused by long-time sequence recording, accurately identify contribution degree influence among key parameters in a sewage treatment process, and provide decision support for equipment control for industry personnel so as to reduce the phosphorus content in effluent.

The application discloses a kind of leakage point positioning method, equipment and medium of water supply network, belong to water supply network operation and leakage control technical field, for solving in existing leakage point positioning, it is easy to rely on single data, equipment and algorithm complex, and also lack high-precision space linkage and to data adaptability and insufficient refinement, it is difficult to realize the technical problems of accurate positioning of water leakage point of water supply network.Method includes: the operating state of water supply network is processed to multi-source data acquisition, obtains water supply network core data;The standardization processing of feature vector is carried out to the corrected water supply network core data after pre-processing, obtains effective water supply network data;Effective water supply network data is screened under the risk value of leakage point area about leakage point, obtains high suspected leakage point area based on water supply network section;Leakage point spatial positioning calculation is carried out to high suspected leakage point area under local topology, obtains the final leakage point area position in water supply network grid.

This application relates to a method, apparatus, and measurement system for dynamic measurement of assembly data. The measurement system includes a moving crossbeam and multiple calibration targets, with a 3D scanner and a laser tracker mounted on the moving crossbeam. The method includes: controlling the moving crossbeam to move to a measurement pose; positioning multiple calibration targets using the laser tracker to construct a geodetic coordinate system and tracking and determining the real-time pose of the 3D scanner in the geodetic coordinate system; controlling the 3D scanner to move and scan to obtain reference point cloud data and key area point cloud data of the assembly workpiece, and establishing the point cloud correlation between the two, establishing a workpiece coordinate system based on the reference point cloud data; based on the workpiece coordinate system, real-time pose, and point cloud correlation, mapping the key area point cloud data to the geodetic coordinate system and performing error correction; and generating an assembly deviation map based on the deviation between the corrected actual measurement data and the expected measurement data. This method can improve the cycle time efficiency and assembly accuracy pass rate of a pulsed assembly production line.

The present application provides a battery pack collision test method, including controlling a battery sorting machine to divide the same batch of battery cells into three groups of battery cells, wherein the battery cells of the first group include multiple single battery cells, the battery cells of the second group are used to assemble into multiple multi-battery cell structures, and the battery cells of the third group are used to assemble into multiple battery packs; controlling a first extrusion tester to perform a first static extrusion test on at least one single battery cell, and / or controlling a first impact tester to perform a first dynamic impact test on at least one single battery cell; controlling a second extrusion tester to perform a second static extrusion test on at least one multi-battery cell structure; controlling a third extrusion tester to perform a third static extrusion test on at least one battery pack, and / or controlling a second impact tester to perform a second dynamic impact test on at least one battery pack. The method can fully identify the design defects and risk points of the battery pack and evaluate the anti-collision capability of the battery pack.

The invention provides a nuclear power DCS system time calibration method and system, an electronic device and a medium, in the method, when a time calibration task is responded, whether a reference clock is forcibly calibrated or not is judged through local time and first calendar time of a dynamic clock, and the reference clock is forcibly calibrated without forcibly calibrating; the crystal oscillator error is quantized based on the second calendar time and the local time of the reference clock so as to capture the actual condition of the crystal oscillator deviation, then the dynamic clock timing rate is corrected by combining the first calendar time, the local time and the quantization parameter, and the obtained crystal oscillator deviation compensation parameter can dynamically adjust the dynamic clock timing rate instead of direct hopping timing. Time deviation caused by crystal oscillator errors is eliminated, the time reference of lower-level equipment can be effectively unified, and the problem of time errors caused by the crystal oscillator errors among the modules in the nuclear power DCS system is effectively solved.

The invention relates to the technical field of precision angular displacement measurement sensors, and discloses a ball array type reluctance time grating angular displacement sensor and a displacement measurement method.The ball array type reluctance time grating angular displacement sensor comprises a stator and a mover, and the stator comprises a stator base body and a plurality of steel balls installed in the stator base body; the rotor comprises a rotor base body and a sine / cosine winding arranged along the annular outer surface of the rotor base body, the sine winding is provided with an S + excitation winding, an S-excitation winding and a sine induction winding, and the cosine winding is provided with a C + excitation winding, a C-excitation winding and a cosine induction winding. The sine induction winding and the cosine induction winding respectively generate two paths of electric signals U1 and U2, a traveling wave signal U carrying linear displacement information is output through the logic circuit and is compared with a same-frequency reference signal Uf in phase, and displacement decoupling is achieved. Error interference caused by the external environment can be effectively reduced, the measuring range of the sensor is increased, and the measuring precision is improved.

The invention relates to the technical field of magnetic field strength detection, and in particular to a system and method for detecting residual magnetic field strength of a concrete component; the system and method comprise: a track device, a traction device, an end face detection device for detecting the magnetic field strength of two end faces of a concrete component; an annular detection device for detecting the magnetic field strength of a part of the side surface of the concrete component located in a hollow measuring cylinder; and a main control system for receiving and processing magnetic field strength data of the annular detection device and the end face detection device. The main control system controls the traction device to pull the concrete component into the hollow measuring cylinder in sections, thereby realizing accurate section-by-section movement and pipeline detection of the residual magnetic field strength of the concrete component, greatly improving the efficiency and stability of magnetic detection of the concrete component, filling the industry gap in the field of engineering detection, having low economic cost, being able to detect in a prefabrication site, reducing the transportation time of concrete components, shortening the project construction period, and having high application value.