501results about How to "Reduce bias" patented technology

The invention discloses a multi-working-condition-oriented equipment fault diagnosis method based on unbalanced learning, and the method comprises the following steps: collecting the operation data of equipment under different working conditions through a plurality of sensors, carrying out the preprocessing, feature extraction and feature selection of the collected original operation data, and obtaining a feature set; processing the feature set by adopting an unbalanced learning technology and a cost-sensitive learning algorithm to obtain a balanced data set; constructing a multi-working-condition fault diagnosis model based on learning, and optimizing model parameters by adopting a method of combining cross validation and grid search according to a balanced data set in a model training process; deploying a multi-working-condition fault diagnosis model obtained through training, carrying out the online diagnosis of equipment faults, and continuously updating and optimizing the model through a continuous learning mechanism. According to the method, the unbalanced learning technology is applied to multi-working-condition fault diagnosis, the accuracy and robustness of diagnosis can be remarkably improved, and real-time monitoring and predictive maintenance of the equipment health state are achieved.

The invention provides an intelligent post matching and employee growth path planning system and method and a storage medium. The system comprises a data acquisition and processing module, an employee portrait construction module, an intelligent post matching module and an occupational path planning module. The data acquisition and processing module collects employee performance data, education background, work experience and other information from internal and external data sources. The employee portrait construction module constructs a comprehensive portrait of each employee by using a clustering analysis and factor analysis technology; the intelligent post matching module performs intelligent matching by applying a recommendation system algorithm according to post demands and employee portraits, and finds the posts most suitable for the current capabilities and development desires of the employees; the occupational path planning module analyzes the industry development trend and future skill requirements based on the prediction model, and provides personalized occupational development paths and training suggestions for employees. According to the invention, historical performance evaluation and industry trend analysis of employees can be integrated, and comprehensive post matching and occupational planning solutions are provided.

The invention discloses an automatic fine ore sorting method and system based on XRT (X ray tomography) and hyperspectrum, which are used for realizing comprehensive fusion of ore surface and internal information by combining X-ray tomography and hyperspectral imaging technologies so as to improve the recognition precision and the sorting efficiency. XRT is used for extracting the internal structure and density characteristics of the ore, hyperspectral imaging is used for identifying the mineral composition characteristics of the ore surface, and information complementation is achieved. The system adopts a multi-modal convolutional neural network to perform independent feature extraction on XRT and hyperspectral data, and the extracted features are fused through a deep residual network. And the fused features are processed by a cross-modal attention mechanism to ensure that key features of different modals are fully integrated and optimized, and finally high-precision ore classification and sorting are realized. According to the method, the limitation of a single data source on feature extraction is overcome, the ore recognition and sorting precision is greatly improved, and the refined and intelligent development of ore sorting is promoted.

The invention belongs to the field of fault diagnosis, and particularly relates to a main shaft bending common characteristic diagnosis system based on a stacking integration method, and the system comprises the following modules: a sensor module which is used for collecting a vibration signal of a main shaft in real time; the data acquisition and processing module is used for converting an analog signal into a digital signal and preprocessing the signal acquired by the sensor; the feature extraction module is used for extracting features from the preprocessed vibration signals; the machine learning diagnosis module analyzes and classifies the extracted features by using stack integration based on a plurality of trained machine learning models, and judges the health state of the main shaft; and the output module is used for outputting the diagnosis result and sending out an alarm signal when abnormality is detected. Real-time monitoring and automatic diagnosis of the main shaft bending fault are achieved, compared with a traditional method depending on expert experience and manual detection, the fault detection efficiency and accuracy can be greatly improved, and misjudgment and missed judgment risks caused by human factors are reduced.

The invention provides a micro robot motion control method and system, and the method comprises the steps: carrying out the synchronization and calibration processing of environment state information, and generating a comprehensive perception image; analyzing the comprehensive perception image to obtain a classification result, and predicting the environment state change influencing the motion track of the micro robot in a period of time in the future based on the classification result; generating a candidate path by using a fast random tree, and optimizing the candidate path in combination with an AI search algorithm to obtain a target path; constructing a three-dimensional dynamic probability graph, continuously updating the three-dimensional dynamic probability graph by using a reinforcement learning algorithm, and generating an instant decision map; and estimating the position of the micro-robot by using a Kalman filter to obtain the position state of the micro-robot, adjusting the speed and steering angle of a driving motor of the micro-robot, and generating trajectory tracking of the micro-robot. According to the technical scheme provided by the invention, the flexibility and intelligence of the micro robot are improved.

The invention belongs to the technical field of welding seam detection, and particularly relates to a laser welding seam defect detection machine which comprises support plates, adjusting assemblies arranged between every two adjacent support plates, driving assemblies installed on the support plates and balancing assemblies arranged on one sides of the support plates. Wherein three groups of support plates are arranged in a circumferential array, a front-end support is fixedly mounted on one side of the long edge of each group of support plates, an ultrasonic detection probe, a dot matrix laser and a camera are sequentially mounted on the three groups of front-end supports, an electric control module is mounted on the top surface of each support plate, and a power supply module, a data analysis processing module and a control system are arranged in each electric control module. The device can automatically move on a reducing pipeline, perform positioning detection on a welding seam, and perform circumferential uninterrupted detection in the detection process, so that the detection precision is higher, a mode of combining laser detection and ultrasonic detection is utilized, displacement deviation is automatically corrected after reducing, the error probability is greatly reduced, unnecessary workload is avoided, and the detection efficiency is improved. The detection efficiency is improved.

The invention relates to a wind power cabin equipment quality evaluation and maintenance decision-making method based on reinforcement learning, and belongs to the field of wind turbine generator operation and maintenance. According to the method, the state degradation model of the wind power cabin is established by using the Markov decision process, and the state transition probability is estimated in combination with the dynamic integrated probabilistic neural network, so that the prediction accuracy is improved. Weibull distribution parameter estimation is carried out on the fault data of the subcomponents of the wind turbine cabin through the grey model, life reliability analysis of the components is completed, then a reliability function of the wind turbine cabin is obtained, and equipment quality evaluation is achieved. According to the method, a maintenance method combining post maintenance and opportunity maintenance is adopted, an objective function of a maintenance decision model is established according to a component reliability function and a maintenance cost model, and an optimal maintenance strategy is solved by using an AD3QN algorithm, so that the minimum maintenance cost is realized, and the operation efficiency and economic benefits of the wind power plant are improved.

The invention discloses a classroom student attention assessment method and system based on multi-source feature fusion. According to the invention, face recognition is carried out through the face detection unit of the system, occurrence frequencies of different emotional states of students are obtained through a trained student expression recognition neural network model containing four expression discrimination blocks in the expression recognition unit, and fatigue times of the students are obtained through the fatigue detection unit. The number of times of abnormal head postures of the students is obtained through the head posture estimation unit, finally attention evaluation parameters of the students are obtained through an extreme learning machine model of the attention evaluation unit, and then the attention states of the students in the class are evaluated and displayed. According to the method, a more efficient dynamic student expression recognition method can be provided, real-time and accurate classroom feedback is provided for teachers, optimization and implementation of personalized teaching strategies are supported, longitudinal extension of teaching of the teachers and transverse extension of learning modes of the students are facilitated, and new talents are expanded to cultivate a new normal form.

The invention relates to the technical field of foundation pit engineering, and discloses a foundation pit enclosure structure soil pressure calculation method and system under bilateral coupling deformation, and the method comprises the steps: obtaining deformation correlation parameters of an enclosure structure of a target foundation pit; initial deformation fitting is conducted on the enclosure structure, active and passive initial soil pressure of the enclosure structure is obtained, and the initial soil pressure is associated with the embedding depth of the enclosure structure; the deformation correlation parameters and the initial soil pressure are input into the bilateral coupling deformation calculation model, and horizontal deformation of the two sides of the enclosure structure is obtained; inputting the horizontal deformation of the two sides into a non-limit soil pressure calculation model to obtain the soil pressure of the enclosure structure; and if the horizontal deformation on the two sides and the soil pressure do not meet the convergence requirement, inputting the soil pressure into the bilateral coupling deformation calculation model for iteration. According to the method, the stress conditions of the two sides of the enclosure structures can be considered, so that the mutual influence of stress between the enclosure structures on the two sides is coupled.

The invention discloses a high-voltage cable outer metal sheath grounding defect on-line monitoring method and monitoring system, which can work under the condition that a cable is not powered off, realizes on-line monitoring, and realizes on-line monitoring through the cooperation of a current sensor, a voltage sensor, a single-chip microcomputer and an upper computer. And the equipment can quickly detect when a fault occurs and is convenient for maintenance workers to timely eliminate obstacles, so that the risk of major electric power accidents is reduced. The algorithm is similar to an improved neural network algorithm and is used for diagnosing the defects of the outer metal sheath of the cable, the weight is obtained through training learning, and finally the defects of the outer metal sheath of the high-voltage cable are determined. The protection content covers each component of the technical scheme, and comprises a monitoring method, a device structure, a sensor technology, a data processing algorithm, a neural network learning algorithm and the like.

The invention discloses a direct visual field testing device, which relates to the technical field of testing and comprises a video auxiliary device and a marker device, the video auxiliary device realizes three-dimensional positioning and real-time video acquisition of preset eye spots through a height adjusting mechanism, an angle adjusting mechanism and a holder adjusting mechanism; the marker device realizes accurate adjustment and stable positioning of a marker in a test area through a height adjusting rod, a knuckle bearing and laser-assisted positioning; the direct visual field testing device is suitable for accurate measurement of the visual field range of a driver, can efficiently collect and record testing data, has the advantages of being convenient to operate, high in adaptability, high in precision and the like, and is widely applied to the fields of vehicle design, driving training, safety research and the like.

The invention relates to the technical field of flood prediction, in particular to a non-consistent regional flood frequency analysis method based on the Bayesian theory. The method comprises the following steps: acquiring flood sequence data, meteorological data and drainage basin characteristic data of hydrological stations in a drainage basin; performing regional flood dynamic change analysis on the flood sequence data and the meteorological data to generate regional flood time sequence dynamic change data; on the basis of the regional flood time sequence dynamic change data, hydrological station flood characteristic parameters in the drainage basin are extracted from the drainage basin characteristic data, and hydrological station flood characteristic parameters in the region are obtained; and performing flood distribution parameter regression coefficient prior information derivation on the flood characteristic parameters of the hydrological stations in the region to obtain a maximum likelihood estimation value of a single hydrological station distribution parameter. According to the method, the Bayesian theory, the spatial correlation analysis and the spectral analysis are introduced, so that the accuracy and the applicability of flood frequency analysis based on the Bayesian theory are improved.

The invention relates to the field of valve control, in particular to a valve control method insensitive to temperature fluctuation, a controller and a vaporization device. The valve is arranged on a liquid inlet flow path of the vaporization part. The method comprises the steps that a current flow signal fed back by a flow sensor of the liquid inlet flow path is received; deriving the current flow signal, and gaining to obtain a current derivative compensation signal; pID control is carried out based on the flow set value and the flow measurement value corresponding to the current flow signal, and a current valve control basic signal is obtained; estimating a current temperature control disturbance signal caused by temperature fluctuation of the heating vaporization cavity based on the current flow signal and the valve control signal at the previous moment; and superposing the current valve control basic signal, the current derivative compensation signal and the current temperature control disturbance signal to obtain a current valve control signal. The multi-level control strategy combining PID control, derivative compensation and disturbance compensation is adopted to control the valve so as to cope with periodic disturbance of a heating system, and stability and control precision of liquid flow can be ensured.

The invention relates to the technical field of three-dimensional visualization park management, and particularly discloses a three-dimensional visualization-based smart park management platform, which comprises a self-adaptive sensor calibration module for dynamically calculating a sensor data accuracy coefficient to ensure the precision of data acquisition; the multi-dimensional data fusion and three-dimensional modeling module is used for calculating an adaptation coefficient to evaluate the response capability of the system to environmental change by integrating multi-source data; the intelligent image data registration module optimizes a matching precision coefficient and improves the space docking precision of the image data and the three-dimensional model; the real-time error monitoring and feedback optimization module uses a machine learning algorithm to correct an error between the sensor and the three-dimensional model, dynamically calculates an error correction coefficient, and ensures data consistency; and the comprehensive performance evaluation module calculates a sensor data accuracy coefficient, an adaptation coefficient, a matching precision coefficient and an error correction coefficient through normalization to obtain a comprehensive performance coefficient, and optimizes a system operation strategy in real time.

The invention relates to the technical field of tracing, and discloses a towel heating robot capable of navigating and tracing and a control method thereof. Comprising a detection unit, a tracing unit, a judgment unit and a tracing adjustment unit, the detection unit obtains current environment information to construct a three-dimensional environment model, the tracing unit plans tracing path information for current position coordinates and target position coordinates according to the three-dimensional environment model, the judgment unit judges the tracing path information, and the tracing adjustment unit adjusts the tracing path information. When it is judged that the tracing path information exists in the historical tracing path information set, the tracing path similarity is determined according to the historical tracing path information, a first tracing path is determined according to the tracing path similarity, and when it is judged that the tracing path information does not exist in the historical tracing path information set, a second tracing path is determined according to the judgment result, the tracing adjusting unit corrects the robot according to the real-time environment information, the first tracing route or the second tracing route. The method has the characteristic of establishing the three-dimensional environment model to optimize the tracing route.

The invention relates to the technical field of medicine detection, and discloses medicine raw material detection equipment for improving the purity of raw materials, the medicine raw material detection equipment comprises a base and a supporting box, the bottom of the supporting box is fixedly connected to the outer side of the base, the top of the base is fixedly connected with a detection mechanism, and the top of the detection mechanism is fixedly connected with a detection plate; the top end of the supporting box is fixedly connected with a driving box, multiple filtering and impurity removing can be conducted on medicine raw materials through the impurity removing assembly, impurities in the raw materials can be screened out while the caked raw materials are screened out, the purity of the medicine raw materials is improved, the content of invalid or harmful ingredients in medicine can be reduced by removing the impurities, and therefore the medicine raw materials can be conveniently taken out. The purity and quality of the medicine are improved, the raw materials with higher purity can be used for preparing the medicine which is more efficient, safer and more stable, the treatment effect and the satisfaction degree of a patient are improved, and the content of active ingredients of the medicine can be improved by removing impurities, so that the curative effect of the medicine is enhanced.

The invention relates to the technical field of energy storage risk prediction, and particularly discloses an electrochemical energy storage risk prediction method, which comprises the following steps of: determining a risk characteristic value of an electrochemical energy storage system, obtaining an abnormal deviation value of an operation environment of the electrochemical energy storage system, and obtaining a risk judgment threshold value of the electrochemical energy storage system; according to the method, the real-time operation state of the electrochemical energy storage system is monitored, whether the electrochemical energy storage system has risks or not is judged by combining a risk judgment threshold value of the electrochemical energy storage system, and the risk level of the electrochemical energy storage system with the risks is determined. According to the method, the problems that the existing risk management system is not easy to carry out and is lack of comprehensive analysis are solved, rapid response and implementation of effective rectification measures are facilitated, the accident occurrence probability is remarkably reduced, the efficiency and pertinence of maintenance work are improved, a perfect risk judgment threshold system is established, potential risks can be systematically identified and evaluated, and the overall risk management capability is improved.

The invention discloses a data analysis method for greenhouse gas emission flux on an aerated water surface, and belongs to the technical field of gas detection.The method comprises the steps that a floating static box is arranged on the aerated water surface in a floating mode, and two exhaust ports of the floating static box are communicated with an air inlet of an online gas detector and the atmosphere respectively; the floating static box is used for collecting gas brought by bubbles on the aerated water surface, and the gas online detector is used for detecting the gas in the floating static box; and correcting the data detected by the gas online detector. According to the data analysis method for the aeration water surface greenhouse gas emission flux, detection of the aeration water surface gas emission flux can be achieved, particularly, for sewage treatment plant biochemical pool aeration water surface greenhouse gas emission, through data correction and a special algorithm, deviation can be reduced, and the detection accuracy is improved. The accuracy and the application range of the detection method are improved.

The invention discloses a power semiconductor welding layer life evaluation and failure prediction method. The method comprises the following steps: S1, acquiring chip design parameters, power semiconductor module sealing and testing process parameters and power semiconductor module working condition monitoring data; s2, constructing a heat transfer model, a mechanical stress model and an electrical property model by adopting a finite element analysis method according to the data obtained in S1; relates to the technical field of power semiconductors, and aims to comprehensively consider chip design parameters, power semiconductor module sealing and testing process parameters and working condition monitoring data, construct a plurality of physical models, deeply analyze the influence on the service life of a welding layer from multiple dimensions of heat transfer, mechanical stress and electrical performance, and greatly improve the service life of the welding layer compared with single thermal fatigue. Or by mainly depending on specific precursor parameters and utilizing a neural network fitting method, the residual life of the welding layer can be evaluated more comprehensively and accurately, and more accurate life prediction information is provided for reliable operation of the power semiconductor module.

The invention discloses a day-ahead-day two-stage optimization scheduling method and system for an RSOC integrated energy system. The method comprises the steps that the RSOC integrated energy system is established based on running characteristics of the RSOC; respectively establishing an RSOC variable working condition operation model, a combined heat and power generation model and a multi-element load demand response model based on the RSOC comprehensive energy system; based on the RSOC variable working condition operation model, the combined heat and power generation model and the multi-element load demand response model, jointly establishing a day-ahead-intra-day two-stage optimization scheduling model; optimizing the day-ahead-intra-day two-stage optimization scheduling model based on a day-ahead plan scheduling target and a day-ahead constraint condition to obtain optimal operation state data of system equipment; and based on the optimal operation state data, the intra-day rolling optimization target and the intra-day constraint condition, performing rolling adjustment on the day-ahead-intra-day two-stage optimization scheduling model to obtain an optimal scheduling operation scheme. Under the condition of ensuring stable and reliable operation of the comprehensive energy system, the operation cost of the system is reduced, and the optimization control precision of the system is improved.

The invention discloses a multi-modal data enhancement method and system based on an AI training platform, and belongs to the technical field of multi-modal data processing.The method comprises the steps that firstly, a multi-modal data fusion pipeline is developed on the AI training platform, and data of different modals are collected through a data collection module for preprocessing; in the preprocessing stage, a multi-modal data fusion pipeline is used for carrying out feature extraction on data from different modals, and joint modeling is carried out at a specific level; then, automatically generating high-quality synthetic data by adopting a data enhancement pipeline; and finally, developing a multi-modal deep learning pipeline to fully learn and understand deep features of different modal data. According to the method, the problem of model learning bias caused by single-modal data in the prior art is solved, the generalization ability and the data processing efficiency of the model are improved, meanwhile, the processing difficulty of a multi-modal data task is simplified for an AI training platform user, and the method has important application value.

The invention relates to a QSFP-DD packaging 400G optical module structure, and belongs to the technical field of optical modules. Comprising a module upper cover, a module base and a pull ring, the module upper cover and the module base form a containing cavity, an MPO optical fiber adapter and a module circuit board assembly are arranged in the containing cavity, and first cooling fins and second cooling fins are arranged on the module upper cover and the module base respectively; the module circuit board assembly comprises a PCB, a heat dissipation block is arranged on the PCB, and a heat dissipation table is arranged on the module upper cover. A module upper cover and a rubber coating area are arranged on the inner side of the pull ring, a pull ring handle is arranged at one end of the pull ring, and the pull ring handle is flush with the bottom of the module base. According to the QSFP-DD packaging 400G optical module structure provided by the invention, the heat dissipation efficiency is remarkably improved, the MPO jumper wire insertion space is optimized, more internal space of the PCB is provided, the signal loss is reduced, the connection stability of the light-transmitting assembly and the silicon optical assembly is enhanced, and the optical signal loss is reduced.

The invention discloses a heat setting process parameter recommendation method and device, computer equipment and a storage medium. The method comprises the steps of obtaining to-be-predicted production order data; preprocessing the to-be-predicted production order data to obtain a preprocessing result; inputting the preprocessing result into a prediction model to perform process parameter prediction so as to obtain a prediction result; wherein the prediction model is obtained by preprocessing production order data, labeling the data as a sample set and training a multi-weight main model formed by stacking a plurality of base models with weights and meta-models, and the weights of the base models are determined by the prediction performance of the base models; a loss function in the training process is added into the quality score of each piece of preprocessed and labeled data; and outputting the prediction result. By implementing the method disclosed by the invention, the effect of enhancing stacking by multiple models can be exerted, the dependence on artificial experience is reduced, the accuracy of process parameters and the one-time success rate of production are improved, and the product repair and production cost is reduced.

The embodiment of the invention provides a display device. The display device includes a display panel and a timing controller. The display panel comprises a plurality of data lines and a plurality of pixel columns. One pixel column is located between adjacent data lines and comprises at least two pixels. And the time schedule controller is connected with the display panel. The time schedule controller is configured to determine coupling influence accumulation data of a data line adjacent to a to-be-compensated pixel in a current frame in a previous frame before the current frame and data voltage transmitted by the current frame to the data voltage of the to-be-compensated pixel in the current frame; determining a gray scale compensation value of the to-be-compensated pixel in the current frame based on the coupling influence accumulated data and the gain coefficient; and compensating the original gray-scale value of the pixel to be compensated in the current frame based on the gray-scale compensation value to obtain a target gray-scale value. Therefore, the problem of non-uniform picture brightness caused by the coupling capacitance during display of the display panel is solved.

The invention relates to a method, device and equipment for determining the safety coefficient range of a karst cave slope and a medium. The method comprises the steps that the maximum value and the minimum value of the slope in the X direction and the Y direction are obtained; based on the maximum value and the minimum value of the slope in the X direction, the slope is divided in the X direction according to the vertical division interval, and a plurality of vertical units are obtained; based on the maximum value and the minimum value of the slope in the Y direction, unit bodies with random heights are generated in all the vertical units; according to the unit bodies, determining the distribution condition of the rock-soil body and the karst cave in each vertical unit, and constructing a slope karst cave simulation model; performing assignment calculation on the slope karst cave simulation model to obtain a slope safety coefficient; and a plurality of groups of slope safety coefficients are obtained through iterative calculation, and the range of the slope safety coefficients is determined based on each group of slope safety coefficients. Reliable data support can be provided for engineering design of the karst cave slope, and safety and engineering economy are both considered.

The invention belongs to the technical field of computer vision, and discloses an arm movement automatic evaluation system and method for dyskinesia of a cerebral palsy patient, and the system is characterized in that a data collection module carries out the movement data collection of the movement task of the arm of the patient through a smart phone camera; the posture estimation and key point detection module uses a posture estimation algorithm to automatically detect and track arm movement of a patient and recognize key points of a human body, and after coordinate data of the key points are obtained, data preprocessing is carried out; the muscular dyskinesia feature extraction module extracts muscular dyskinesia features from the motion data of the arm of the patient and constructs a corresponding feature set; the automatic scoring module uses a multi-layer feedforward neural network model to automatically calculate the score of the muscular dystonia based on the arm movement characteristics of the patient; the report generation module is used for generating an evaluation report. The method is suitable for arm movement evaluation of myodynia of cerebral palsy patients, and an objective, convenient and efficient evaluation tool is provided for clinicians.

The invention discloses a railway track vertex cloud data extraction method and system based on deep learning. The method comprises the following steps: classifying track point cloud data by using a point cloud network enhanced multi-scale classification model to obtain a track key component point cloud classification data set; rail head point cloud data are extracted from the rail key component point cloud classification data set, center line fitting is carried out, a rail head center line is generated, and a reference point is selected on the rail head center line; a rolling window method is adopted, and a plurality of slices are generated along the rail head center line according to the reference points; and carrying out rail top point extraction on the plurality of slices to obtain a continuous rail top coordinate point set. The system comprises a data classification module, a rail head center line generation module, a slice generation module and a rail vertex extraction module. The rail top data extraction efficiency can be improved, the data accuracy and the rail safety are ensured, and scientific maintenance and management of the railway rail are facilitated.

The invention discloses an electrical steel coating adhesiveness rating method. The method comprises the following steps: manufacturing an electrical steel sample coated with an insulating coating; performing a bending test on the sample; collecting a coating adhesiveness image; processing the coating adhesiveness image to obtain white pixel points representing a coating falling area and black pixel points representing a coating non-falling area; calculating a coating falling ratio, and rating the coating adhesiveness of the electrical steel sample. The method has the advantages that visualization and visual analysis technologies are introduced, quantitative data of the adhesiveness of the coating are obtained, the adhesiveness of the coating is graded according to the data, the grading method is objective, the result accuracy is high, and errors and deviations caused by manual visual evaluation are reduced.

The invention provides a time-space downscaling method and device based on deep learning, equipment and a medium, and relates to the technical field of electric digital data, and the method comprises the steps: obtaining to-be-processed forecast data, forecast aging information and geographic information; pre-processing the to-be-processed forecast data, and inputting the pre-processed to-be-processed forecast data, the forecast aging information and the geographic information into a time-airborne downscaling model obtained by pre-training; the method comprises the following steps: extracting a feature map of to-be-processed forecast data through a time-airborne downscaling model, improving the timeliness adaptability of the feature map by utilizing forecast timeliness information to obtain a target feature map, performing non-integer multiplying power up-sampling processing on the target feature map, and generating target forecast data after time-airborne downscaling in combination with geographic information. According to the method, the temporal-spatial resolution of the forecast data can be well improved, and meanwhile, the forecast deviation is effectively reduced, so that the forecast effect is remarkably improved.

The invention relates to a controllable variable-diameter stabilizer for petroleum drilling and an application method.The controllable variable-diameter stabilizer for petroleum drilling comprises an electric mechanical driving module, an intelligent control module, a sensor feedback module and a communication module, and the electric mechanical driving module receives an instruction from the intelligent control module through an electronic signal; adjusting the outer diameter of the driving stabilizer; the intelligent control module receives data from a ground sensor or an underground sensor, calculates underground working conditions in real time and automatically adjusts the outer diameter of the stabilizer; the sensor feedback module is used for monitoring key parameters such as the posture of a drilling tool, a well track and bit pressure; the hydraulic auxiliary module and the electric mechanical driving module work cooperatively to provide dual assurance; the communication module is responsible for data transmission between an underground control center and a ground control center. According to the stabilizer, the performance of the stabilizer in the oil and gas drilling process is remarkably improved, the efficiency in the oil and gas drilling process is greatly improved, accurate and efficient outer diameter self-adaptive adjustment is achieved, and the safety and efficiency in the drilling process are remarkably improved.