685 results about "Compensation algorithm" patented technology

The invention relates to a concrete temperature-stress two-parameter cooperative monitoring and early warning method and system, and belongs to the technical field of concrete structure health monitoring, and the method comprises the steps: activating a composite sensor network disposed at a preset position of a concrete structure, and executing a sensor self-calibration mode; the method comprises the following steps: synchronously acquiring temperature and stress original monitoring data of each node, processing based on an altitude air pressure compensation algorithm, and outputting a temperature gradient field matrix and a stress tensor sequence with aligned time domains; the method comprises the following steps: separately calculating a thermal stress component and an effective stress component, performing integral calculation on hydration reactivity, generating a multi-dimensional feature vector set, inputting a pre-constructed prediction model, executing time sequence evolution prediction, calculating a crack probability value, performing environment correction in combination with real-time environment parameters, outputting a risk level identifier, and matching a preset regulation and control strategy. And generating an equipment control instruction set and executing corresponding regulation and control actions. According to the invention, the accuracy and timeliness of crack risk prediction can be improved.

The invention relates to the technical field of production line control, and discloses an intelligent control method and system for production line mold machining. According to the method, a pressure distribution state is determined by obtaining initial cutting parameters of a cutter and surface pressure data of a mold, mold node stress distribution is calculated by utilizing a three-dimensional finite element model, and a stress concentration position is identified. And predicting a mold deformation trend through a linear regression model according to the thermal expansion coefficient and the environment temperature data. Meanwhile, the initial cutting parameters are optimized by combining a cutting parameter optimization objective function, and the objective function comprehensively considers factors of tool wear, surface quality and residual stress. And on the basis of a die deformation prediction result, a dynamic compensation algorithm is adopted to adjust supporting structure parameters and optimize cutting parameters, and finally optimal cutting parameters and a target tool path are generated. According to the method, precise coordination control over pressure distribution and mold deformation under the complex working condition is achieved, the residual deformation risk is reduced, and the mold machining precision is improved.

The invention relates to the technical field of forklift driving stability control, in particular to a forklift driving stability control system based on ultrasonic and laser composite ranging, which comprises a composite ranging module, a multi-modal data fusion module, a stability prediction and control module, an execution mechanism control module and an online learning system. The composite distance measuring module collects surrounding environment information and corrects motion errors through a dynamic compensation algorithm; the multi-modal data fusion module dynamically adjusts the weight of the sensor based on an environmental interference factor, and realizes space-time calibration in combination with extended Kalman filtering; the stability prediction and control module is fused with the vehicle dynamic characteristics to predict the rollover risk and generate a hierarchical control instruction; the actuating mechanism control module drives actuating components such as an active suspension and an electronic differential mechanism through a layered framework. The method can adapt to complex working conditions, the prediction precision and fault-tolerant capability are improved, safety and efficiency are balanced, the maintenance cost is reduced, and the method is suitable for various forklifts.

The invention discloses a wind profile radar radial speed quality control and horizontal wind field inversion method. According to the method, through the steps of multi-mode detection splicing, signal-to-noise ratio threshold value quality control, beam consistency inspection, horizontal wind component inversion, space and time continuity inspection and the like, the quality control process of the wind field data is optimized, the precision of the horizontal wind field data is remarkably improved, and particularly, the root-mean-square error and deviation in high-level data are remarkably reduced. The core innovation comprises a mode splicing strategy based on sounding data comparison, dynamic signal-to-noise ratio threshold calculation, threshold setting of beam consistency check and a wind component compensation algorithm when a vertical beam is missing. Through experimental verification, compared with a traditional wind profile radar data processing method, the wind profile radar data processing method has remarkable advantages in data accuracy and stability.

The invention discloses a high-precision linear vibration feedback motor system, and relates to the technical field of precise electromagnetic driving, and the system comprises a core driving module which is a moving magnet type linear motor of a Halbach array magnetic circuit structure, and is integrated with a multi-physics field sensing unit; the dynamic parameter tracking module is used for acquiring a reed rigidity coefficient, a damping coefficient and a magnetic constant in real time when the motor runs, and establishing a dynamic displacement model based on a recursive least square method; a double-closed-loop control framework is adopted, an inner loop adopts field-oriented control, and an outer loop is based on model prediction control; the aging prediction unit activates a temperature rise compensation algorithm when the accumulated number of vibration times is greater than a set threshold value; the self-adaptive calibration engine is used for injecting a sweep frequency excitation signal when the system is started, and automatically compensating individual difference parameters according to the harmonic peak offset; and the digital twin mapping unit is used for establishing a real-time mapping relationship between the physical parameters of the motor and the virtual model, and dynamically correcting the model parameters by comparing the actual displacement with the model displacement.

The invention discloses a force control joint torque self-correction method based on digital twinborn mapping, which comprises the following steps of: acquiring and preprocessing multi-source operation data of a force control joint to generate a standardized input data set; establishing a three-layer digital twinning mapping model to realize physical, simulation and cognitive layer mapping; a bias load device is arranged, and angle, torque, current and temperature signals are collected; space-time tensor compensation is executed on the simulated twinborn layer, and a dynamic compensation value is generated; bootstrap consistency correction is executed on the cognitive twin layer, and comprehensive correction parameters are fused; and inputting comprehensive parameters to adjust a simulation layer structure, outputting a correction torque and performing closed-loop updating. By constructing the three-layer digital twinning mapping model and introducing the torque compensation algorithm, torque dynamic error identification, adaptive compensation and real-time self-correction of the force control joint under complex working conditions are realized, and the measurement precision and the system stability are improved.

The invention discloses a ground wire multi-machine synchronous traction pay-off scheduling method and system, relates to the technical field of high-voltage transmission line construction, and is used for solving the problem of insufficient multi-machine real-time tension synchronization precision in the existing method. A tension sensor and a speed sensor which are arranged on each traction device collect a tension value, a traction speed value and position coordinate data of a ground wire in real time, and transmit the data to a central scheduling controller; establishing a multi-machine unified time synchronization reference and tension reference model, calibrating a timestamp through a GPS module, and initializing a threshold value based on historical parameters; comparing tension values one by one to calculate a deviation vector sequence; activating a compensation algorithm to execute a speed fine adjustment and tension compensation instruction; iteratively calculating the feedback adjustment value until the deviation is stable; and integrating the synchronous data to the overall paying-off scheduling model to dynamically update the path planning and speed curve. According to the method and the system, the problem of insufficient multi-machine real-time tension synchronization precision in the prior art is effectively solved.

The invention relates to a course axis drift suppression method and device, a holder and a medium, and the method comprises the steps: receiving original data of a current course, the original data comprising an absolute angle outputted by a course encoder, and an acceleration and an angular velocity outputted by an IMU module; the original data are converted into a unified coordinate system through a rotation matrix, and mechanical installation deviation is compensated; performing data fusion on the compensated original data through a preset fusion algorithm; and outputting the fused absolute angle to a servo mechanism of the holder, and by adopting the method, through real-time sensor data alignment, a dynamic compensation algorithm and a fusion algorithm, a course drift suppression mechanism which is resistant to dynamic interference and low in cost is constructed, and static drift lt can be realized; the dynamic tracking error is 1t, 0.01 degree / h and 1t; and the technical index is 0.05 degree.

The invention relates to the technical field of chip test engineering control, in particular to an active temperature control method and system for a chip aging tester, and the method comprises the steps: analyzing a test process according to test task parameters, and evaluating the expected thermal load change rate of a test chip, so as to dynamically generate a chip target junction temperature control curve; based on the chip target junction temperature control curve, the actual chip junction temperature and the current actual furnace temperature, the theoretical furnace temperature is obtained through a junction temperature closed-loop control algorithm, the furnace temperature compensation amount is calculated through a feedforward compensation algorithm in combination with the test state parameters, and the ideal furnace temperature is obtained according to the theoretical furnace temperature and the furnace temperature compensation amount; the actual temperature control quantity is obtained according to the basic temperature adjusting quantity and the temperature control compensation quantity, a preset chip target junction temperature control curve can be tracked more accurately, transient and steady state errors caused by test program switching and environment fluctuation can be more effectively restrained, and control over the chip junction temperature with higher precision and faster response is achieved.

The invention relates to the technical field of computers, and discloses a multi-mode audio and video synchronous processing method and system based on a distributed architecture. The method comprises the following steps: generating a unique logic sequence identifier consisting of a device code, a media type and a frame number for each frame of audio and video data; distributing the frame carrying the identifier to a distributed node for processing and adding a local timestamp; and the sink node reconstructs an original frame sequence according to the identifier, calculates a synchronous offset by combining high-precision clock calibration, and realizes dynamic reordering and output through a double-buffer structure and a self-adaptive drift compensation algorithm. The system comprises a source end acquisition module, an identifier generation module, a task scheduling module, a distributed processing cluster module, a convergence synchronization module and a synchronous output module. According to the scheme, the system effectively guarantees the audio and video synchronization precision in a high-concurrency heterogeneous environment, the synchronization error is controlled within 20 milliseconds, and the playing quality and the user experience of scenes such as live broadcast and cloud games are remarkably improved.

The invention discloses an infrared CO2 high-precision detection system and method based on a self-adaptive balance signal, and belongs to the field of gas detection. The invention discloses an infrared CO2 high-precision detection system and method based on a self-adaptive balance signal. The infrared CO2 high-precision detection system comprises a sensing module, a signal conditioning and collecting module and a self-adaptive filtering processing module. According to the method, the problems of contradiction between response speed and stability, discrete signal tracking lag and sensitivity to environmental interference in the prior art are solved, the weight of a filtering channel is adaptively and dynamically adjusted, and the advantages of a fast filtering channel and a slow filtering channel are combined, so that balance between dynamic response and steady-state precision is realized; the problem of tracking lag of discrete signals is solved by analyzing the change rate of the signals in real time and dynamically adjusting a filtering strategy; and through a hybrid filter and an environment compensation algorithm, the influence of temperature and humidity changes on a detection result is reduced, so that the long-term stability of the sensor is improved.

The invention discloses an automatic identification method and system for power transmission and transformation equipment based on visible light and infrared dual-light fusion. The method comprises the following steps that a visible light image and an infrared image of the same scene are synchronously collected through a dual-light camera carried by an unmanned aerial vehicle; preprocessing the visible light image and the infrared image, including denoising, geometric correction and spatial alignment, in the spatial alignment, feature points are extracted through an SURF algorithm, and image registration is optimized through an RANSAC algorithm; and feature extraction is carried out on the fused image by using a deep convolutional neural network OfficientNet-Lite, and a multi-scale feature map is generated. The method has the beneficial effects that high-precision identification is realized; a dynamic fusion strategy and an attention mechanism enable the insulator crack (gt; 0.1 mm), the detection rate reaches 92%; the anti-interference capability is high, the false alarm rate is reduced from 15% to 4% through a dust compensation algorithm, and the influence of emissivity errors is reduced through material self-adaptive calibration; the whole process from data acquisition to report generation is free of human intervention, and the labor cost is reduced by 70%.

The invention provides a signal decoding method based on a PSI5 interface, and relates to the technical field of signal decoding, and the decoding method comprises the following steps: processing and adjusting a received PSI5 signal to obtain an accurate identification signal; and extracting a clock signal from the PSI5 signal, designing a clock compensation algorithm according to the quantum neural network, and dynamically adjusting the clock signal to obtain clock synchronization data. And determining a data frame boundary, and decoding the accurate identification signal to obtain original data. And performing comparative analysis on the original data and the PSI5 signal to obtain comparative difference data, judging whether the comparative difference data accords with a preset error type or not, and performing corresponding processing. According to the invention, fractional calculus processing is carried out on the received signal, then signal distortion is corrected by using a channel equalization method, and the quantum neural network is constructed to dynamically adjust the clock signal, so that clock synchronization is realized, a reliable time reference is provided for accurate processing of the signal, and the accuracy of signal decoding is improved.

The invention discloses a pneumatic joint sealing performance intelligent test method and system based on dynamic pressure spectrum and multi-parameter fusion, and belongs to the technical field of pneumatic element sealing detection. According to the method, a dynamic pressure spectrum simulating an actual working condition is applied to a tested joint, pressure, temperature and ultrasonic signals are synchronously acquired, environmental thermodynamic interference is eliminated by adopting a temperature real-time compensation algorithm, and pressure change purely caused by leakage is separated by deducting system baseline data; and finally, fusing double characteristic parameters of the pressure leakage rate and the ultrasonic energy, and comprehensively evaluating the sealing performance through an intelligent judgment model. The system comprises a gas circuit tool module, a multi-sensing module and a control processing module, can effectively solve the problems that a traditional static pressure maintaining method is greatly influenced by temperature, transient leakage cannot be detected and the like, has the advantages of being high in detection precision, high in anti-interference capability, capable of simulating real working conditions and the like, and remarkably improves the test reliability and efficiency.

The invention discloses a duct grouting fullness detection method and device, and relates to the technical field of duct grouting fullness detection.The duct grouting fullness detection method comprises the following steps that an ultrasonic probe is arranged at an inlet of a duct and transmits an ultrasonic signal, and a receiver is arranged at an outlet of the duct and used for receiving reflection and transmission signals and collecting ultrasonic propagation data in the duct; and performing frequency decomposition on the acquired signals, and analyzing energy attenuation conditions of the signals in different frequency bands by using penetration characteristics of ultrasonic waves with different frequencies. According to the method, multi-band ultrasonic detection, self-adaptive signal compensation and machine learning technologies are combined, so that the accuracy and robustness of duct grouting fullness detection are remarkably improved. Multi-band detection covers different depths, and an adaptive compensation algorithm maintains signal stability and reduces misjudgment. The machine learning optimizes data fusion, ensures that the system adapts to different construction conditions and outputs stable results, and the generated grouting map provides clear visualization to help engineers to quickly identify and maintain defect areas.

The invention provides a fair-faced concrete forming quality control method, belongs to the technical field of fair-faced concrete, and realizes real-time monitoring of a template deformation state by establishing a component geometric dimension reference coordinate system and a template deformation monitoring network. A deep learning model based on a ResNet residual network architecture is adopted to predict dynamic lateral pressure distribution in a concrete pouring process, a deformation compensation algorithm is utilized to accurately calculate a deviation value between an actual size and a design size of a component, a concrete surface quality evaluation system is established to identify surface defects, and a defect distribution matrix is established; and a quality control feedback adjustment mechanism is executed to dynamically optimize and adjust process parameters of subsequent pouring operation, and the technical problem that the geometric dimension precision of a component is difficult to control due to the fact that the formwork deforms under the lateral pressure effect of concrete is solved.

The invention discloses a joint motor control method and system based on an angle sensor, a medium and a product, and relates to the field of general control or adjustment systems.The method comprises the steps that real-time angle data and temperature data of the angle sensor are collected, and a real-time data packet is generated; determining a rapid compensation parameter corresponding to the temperature value, performing preliminary compensation on the angle value to obtain a preliminary compensation angle value, and generating a motor control parameter to control a joint motor to operate; storing a plurality of continuous real-time data packets into a compensation cache region; when the number of the data packets in the compensation cache region reaches a preset number threshold value, temperature change features and angle drift features are extracted, and accurate compensation parameters are calculated; and performing secondary compensation on the preliminary compensation angle value based on the accurate compensation parameter to obtain an accurate compensation angle value, and updating the motor control parameter based on the accurate compensation angle value. According to the invention, response delay caused by a compensation algorithm can be reduced.

The invention discloses an embedded motor torque ripple real-time compensation system based on deep learning, and relates to the technical field of motor control and artificial intelligence crossing, the system comprises a hardware data acquisition module, a signal feature extraction module, a deep learning compensation model module, an embedded real-time compensation module and a compensation effect verification module; a motor position and a current signal are acquired in real time through the hardware data acquisition module, dynamic analysis and multi-dimensional feature construction are performed on torque ripple related features by combining the signal feature extraction module, and the deep learning compensation model module learns a fluctuation rule in historical data by using a space-time sequence model to generate a high-precision compensation current. The model adapts to an embedded environment through lightweight design, real-time deployment and dynamic optimization of a compensation algorithm are achieved, compared with traditional PID control, the scheme can adapt to changes of motor operation conditions, the influence of torque fluctuation on system stability is reduced, and the smoothness and efficiency of motor operation are improved.

The invention discloses a low-drift laser scanning control method and system based on dynamic error compensation, relates to the technical field of laser scanning control, and aims to solve the drift problem in the laser scanning process and improve the scanning precision. The system comprises a system initialization and calibration module, a data acquisition and storage module, an error modeling module, a route planning and correction module and a real-time compensation and closed-loop optimization module, wherein the error modeling module and the route planning and correction module are key modules. The error modeling module comprises a key route screening unit and a data preprocessing unit, and is responsible for constructing a space-time joint error correction model; the route planning and correction module comprises an initial route generation unit, an error prediction unit and the like, and dynamic correction of a pre-planned route is achieved. According to the system, real-time compensation and closed-loop optimization are achieved through cooperation of all the modules in combination with a dynamic error compensation algorithm, scanning drift is effectively reduced, low-drift and high-precision laser scanning is guaranteed, and meanwhile follow-up scanning precision is optimized.

The invention provides a data error compensation method for a biological net port flowmeter, and belongs to the technical field of ocean detection. A multi-dimensional sensor is arranged at a biological net port to collect fluid parameters, a deep learning model is used for recognizing multi-phase flow characteristics, and a corresponding compensation strategy is selected; based on a correlation analysis algorithm, compensation processing frequency is dynamically adjusted to achieve self-adaptive balance of precision and efficiency, the flow state change degree is monitored through the Euclidean distance, compensation parameters are hierarchically adjusted to ensure system stability, a parallel processing architecture is adopted to execute coarse compensation and fine compensation algorithms at the same time, and results are fused according to weights. Finally, dynamic and accurate compensation of the measurement error of the biological net port flowmeter in the complex multiphase flow environment is achieved, and the technical problem that in the prior art, the biological net port flowmeter cannot achieve dynamic and accurate error compensation in the complex multiphase flow environment is solved.

The invention relates to the technical field of laser displacement meters, discloses a calibration method of a laser displacement meter for a Beidou device based on multi-data coupling, and effectively solves the problems of time delay compensation and data synchronization of cooperative work of multiple devices in a dynamic environment. According to the invention, when a carrier moves at a high speed or environmental parameters suddenly change, submillimeter-level positioning precision can still be maintained, meanwhile, data processing delay is controlled within a millisecond-level range, the reliability of a measurement system under a complex working condition is remarkably improved, and through double marking of a GNSS timestamp and a local clock, the measurement precision is greatly improved. And an accurate time deviation observation value is provided for a subsequent time delay compensation algorithm, so that a time-space unified data basis is established in a data preprocessing stage, and the feasibility of multi-source heterogeneous data fusion is remarkably improved.

The invention relates to a robot multi-frequency vibration composite suppression method and system based on inertia disturbance analysis. The method comprises the steps of obtaining dynamic operation data in the operation process of a robot; the operation data are input into an inertia disturbance analysis module, and a low-frequency vibration source caused by a mechanical structure and a high-frequency vibration source caused by a driving system can be accurately recognized; a disturbance frequency distribution diagram is generated, according to the distribution diagram, a self-adaptive filtering algorithm is adopted to suppress a high-frequency vibration source, a compensation algorithm based on model prediction control is adopted to suppress a low-frequency vibration source, the two strategies are cooperatively processed to generate a comprehensive control instruction, the advantages of the two algorithms are fully played, and effective composite suppression of multi-frequency vibration is achieved.

The invention provides a jelly foreign matter deep learning detection method based on multi-mode intelligent sensing data driving, and the method comprises the steps: obtaining an original image of jelly, and generating a cross-view light field matrix through a polarized light compensation algorithm and synchronous triggering time sequence control; a cross-view light field matrix is obtained, the attenuation path of each light ray is reversely tracked in the jelly volume direction, and the foreign matter shielding probability is calculated in combination with a sugar degree-absorbance curve in a material library; foreign matter existence probability thermodynamic diagrams are generated in the three physical levels respectively, and a three-dimensional centroid probability field is obtained; and dynamically compressing a three-dimensional centroid probability field analysis time window according to the passing speed of the current production line jelly, comparing misjudgment cases of the same kind of jelly in a historical library, and adjusting a judgment threshold value of a corresponding space level to obtain a real-time judgment instruction set with production line parameter adaptability. Through deep cooperation of multi-mode light field perception and cross-level probability modeling, the technical breakthrough of full-dimensional foreign matter detection of jelly products is realized.

The invention belongs to the technical field of intelligent robots and environment perception, and particularly relates to a multi-mode autonomous positioning and inspection method and system for a hazardous chemical substance warehouse environment. According to the method, robust continuous positioning in a high-pollution characteristic environment is realized through combination of a layered confidence scheduling mechanism, an observation judgment mechanism and a time sequence clustering compensation algorithm; through path local reconstruction based on task semantics, the inspection system has task understanding and autonomous adjustment capabilities and adapts to multiple types of inspection tasks; the autonomous sustainable operation capability of the system in continuous inspection is improved through energy consumption and risk double-constraint control; a multi-mode degradation identification and self-healing mechanism ensures long-term stable operation of the sensor, and manual intervention is reduced; the whole inspection system can realize all-weather and unattended multi-mode autonomous inspection in a dangerous chemical warehouse scene, and provides stable mobile platform support for subsequent dangerous symptom identification and automatic early warning.

The invention relates to the technical field of gas detection, in particular to a gas detection system and method based on an AI technology, and the system comprises a differential sensing module, a drift compensation processor, a dynamic risk assessment unit and a cloud-edge collaborative learning framework. Compared with the prior art which adopts a single-sensor independent compensation scheme and has the defects that the false alarm rate is high and long-term drift is difficult to correct under environmental interference, the scheme innovatively introduces a dual-mode differential sensing architecture and an adaptive drift compensation algorithm, a reference sensor is packaged in an inert gas environment, and the reference sensor and a main sensor are subjected to real-time collaborative analysis; the pre-trained lightweight neural network is combined to dynamically eliminate temperature and humidity cross interference, the detection precision and stability under complex working conditions are remarkably improved, and the aging distortion problem of the semiconductor sensor is effectively solved.

The invention relates to the technical field of intelligent household electrical appliance control, and discloses an Internet of Things linked toaster intelligent temperature compensation method and system. The method comprises the steps that an intelligent temperature compensation algorithm model is formed through historical baking data training, and the intelligent temperature compensation algorithm model at least comprises an environment temperature influence factor calculation unit, a cavity aging coefficient calculation unit and a bread category baking curve database; the environment temperature influence factor calculation unit is used for calculating an environment influence weight in combination with the real-time environment parameters obtained by the environment perception linkage module, and the cavity aging coefficient calculation unit is used for calculating a cavity aging coefficient according to the historical use duration of the toaster; and generating a temperature compensation value at the current moment based on the real-time data corresponding to the toaster, the environmental influence weight, the cavity aging coefficient and the standard baking curve corresponding to the bread category. According to the invention, breakthrough improvement of a traditional toaster temperature control technology is realized.

The invention provides a voltage stability control method for a multi-output switching power supply, which comprises the following steps of: sampling a main output channel in real time to obtain a voltage signal; the voltage signal is processed through a wide bandwidth error amplifier, a main channel adjusting signal is generated, and the main channel adjusting signal is used for driving a main power switch device; receiving an input voltage fluctuation frequency band of an auxiliary channel, a load current transient change gradient and a frequency spectrum coupling characteristic of a main channel adjusting signal through a dynamic prediction model, and generating an auxiliary channel adjusting signal with a phase lead compensation characteristic; a cross coupling active offset mechanism is constructed, time domain-frequency domain characteristics of coupling interference are extracted through a high-precision differential detection circuit, amplitude-frequency adjustable reverse correction signals are generated in real time through an adaptive compensation algorithm and injected into control nodes of an auxiliary output channel, and dynamic decoupling and ripple suppression are carried out. According to the method, high-stability control of multi-channel voltage under all working conditions is realized, and the dynamic response speed and the anti-interference capability are improved.

The invention discloses a UPA pressure automatic compensation system and method for mask substrate polishing. The UPA pressure automatic compensation system comprises a pressure execution module, a distributed sensing feedback network and an intelligent control unit. Wherein the pressure execution module is used for directly polishing a mask base plate and comprises a force application shaft, a polishing head, an air source, a UPA air bag and an electric proportional valve, the distributed sensing feedback network comprises an air path pressure sensor and a tail end contact pressure sensor, and the intelligent control unit receives a collection signal of the distributed sensing feedback network and sends the collection signal to the polishing head. On the basis of a current process formula and process parameter coupling model, a data fusion algorithm, a feed-forward compensation algorithm and a fuzzy PID algorithm are operated to obtain a feed-forward compensation amount and a fuzzy PID output amount, the feed-forward compensation amount and the fuzzy PID output amount are superposed, a total control voltage is generated and output to an electric proportional valve, and then the input air pressure of a UPA air bag is adjusted. Therefore, closed-loop control over the polishing pressure of the mask substrate is achieved. Through multi-sensor data fusion, fuzzy PID self-adaptive control and intelligent feed-forward compensation based on a process parameter coupling model, real-time, accurate and constant control of the polishing pressure is realized, so that the surface precision and yield of the mask substrate are improved.

The invention relates to the technical field of antenna control, and discloses a control method for a leaky-wave antenna and a communication device. The method comprises the following steps: measuring the dispersion characteristic of the leaky-wave antenna, establishing a second-order nonlinear mapping relation between a propagation constant and a control voltage, and obtaining a dispersion control parameter group; reversely deducing the target beam directional angle to obtain a target propagation constant, and calculating an initial voltage value of each control unit in combination with the parameter group to form a voltage gradient control sequence; correcting the propagation constant of each unit by adopting a distance compensation algorithm, and recalculating to obtain distance self-adaptive voltage distribution data; and detecting a beam quality factor in real time, updating voltage distribution data through a gradient descent algorithm when the beam quality factor deviates from a target value, and outputting a control voltage sequence to drive each control unit. The leaky-wave antenna control method and device solve the problems that in an existing leaky-wave antenna control technology, the beam quality is reduced and the data rate is sharply attenuated during long-distance transmission, and the beam control precision and the long-distance transmission performance of the leaky-wave antenna in the terahertz frequency band are improved.

The invention discloses a multi-modal data fusion method in electric power overhaul, which comprises the following steps of: firstly, respectively acquiring long-sequence video data and audio data in an electric power overhaul process, and constructing a window-level multi-modal data fusion model; secondly, performing time window division and coarse alignment on the long-sequence video data and audio data by the window-level multi-modal data fusion model, generating a frame-level quality score, and performing fine alignment on multi-modal characteristic data in a time window by adopting a segmented time sequence alignment and compensation algorithm; and calculating a window-level quality score, a window-level weak supervision label and a confidence coefficient, and finally performing multi-modal fusion according to the multi-modal feature data after fine alignment and the window-level weak supervision label to generate a window-level semantic representation as a model output. According to the method, accurate time alignment of video and audio data and consistent fusion of multi-modal semantic features are realized, and more reliable alignment data and semantic input are provided for subsequent target recognition.