40 results about "Experimental data processing" patented technology

The invention relates to an experimental data processing method and device, an AI analysis module and computer equipment, and belongs to the field of data processing.The method comprises the steps that multi-dimensional original data are partitioned according to types, and formats are unified; generating a similarity matrix based on time and space neighborhood information, and marking abnormal fluctuation points; effective signals are separated through a time-frequency feature matching noise library; extracting multi-layer features of basic statistics, time sequence correlation and trend change; and dynamically screening core features to update the tracking type experimental model. The matched AI analysis module integrates hardware circuits of data partitioning, similarity calculation, anomaly marking, noise matching, feature extraction and model updating, and whole-process acceleration is achieved. According to the method, through multi-dimensional data compatibility processing, accurate anomaly detection, multilayer feature fusion and model adaptive optimization, the experimental data processing efficiency and conclusion reliability are remarkably improved, and the method is suitable for real-time analysis of multiple scenes such as scientific research and industry.

The invention relates to the technical field of clinical experiment data processing, and discloses a clinical experiment data analysis method based on reinforcement learning, and the method comprises the steps: obtaining time sequence physiological parameters and treatment scheme execution state parameters; performing hierarchical feature extraction on the time sequence physiological parameters, constructing a dynamic response feature model, and generating an initial treatment response evaluation model in combination with the execution state parameters; loading risk constraints and curative effect indexes to generate a comprehensive evaluation model, and simulating to obtain strategy regulation and control data; constructing and training a reinforcement learning decision model, and generating a treatment path correction model and initial treatment parameters; obtaining dynamic strategy adjustment information, and constructing a multi-stage optimization model for iterative optimization to obtain optimal treatment path parameters; reasoning a curative effect standard reaching probability based on a real-time model, and adjusting a strategy to realize dynamic curative effect matching. According to the method, the clinical data dynamic analysis and treatment scheme optimization capability is improved, and the method is suitable for precise medical scenes.

The embodiment of the invention provides an electroencephalogram experiment data processing method and device and electroencephalogram equipment, and relates to the technical field of electroencephalogram equipment. According to the specific implementation scheme, in the process of continuously displaying multiple experimental stimulation contents, electroencephalogram experimental data of an experimented object are collected; acquiring experimental stimulation content to obtain acquired content; determining a content switching time point in the collected content, and generating content identification data based on the content switching time point, the content switching time point representing a time point between two adjacent experimental stimulation contents corresponding to the collected content; and fusing the content identification data with the electroencephalogram experiment data to obtain fused electroencephalogram experiment data. In the scheme, the electroencephalogram equipment can automatically generate the content identification data, and the dependency on the equipment is reduced. Compared with wireless connection, the accuracy of the content identification data can be improved. Besides, stimulation in a specified environment such as a natural environment can be directly collected for experiment without display equipment, so that the application range of the scheme is expanded.

The invention discloses a method for calculating the diffusion coefficient of an electrode material, which comprises the following steps: establishing a Fick diffusion law-based constant current gap titration (GITT) theoretical model, and describing the concentration distribution of ions in the electrode material; solving a partial differential equation in the GITT theoretical model through Laplace transformation to obtain an analytical expression of ion concentration; obtaining a diffusion coefficient calculation formula suitable for a full time range; verifying the accuracy of the diffusion coefficient calculation formula through numerical simulation; and a concentration-based diffusion coefficient expression is converted into a voltage-based diffusion coefficient expression, so that processing and analysis of experimental data are facilitated. The method for calculating the diffusion coefficient of the electrode material, provided by the invention, is based on a constant current gap titration technology, is suitable for full-range data, extracts experimental data, calculates a voltage difference, selects an optimal formula and generates a visual image through an automatic program, and remarkably improves the calculation precision of the diffusion coefficient and the processing efficiency of the experimental data.

The invention discloses a multi-workstation collaborative chemical experiment robot control system, and relates to the technical field of robot control. According to the system, through closed-loop collaboration of the chemical knowledge model module, the robot experiment execution module, the equipment management module and the multi-workstation collaborative scheduling module, intelligent design of an experiment process, full-process automatic execution, standardized integration of data and accurate optimization of parameters are realized, and the efficiency and repeatability of a chemical experiment are greatly improved; errors caused by manual intervention are reduced, the research and development process of chemicals is accelerated, meanwhile, the product performance standard-reaching rate is increased by means of a spectrum structure-activity relationship prediction model, efficient overall utilization of resources of multiple workstations is achieved, and the working efficiency is improved. The problems that in a traditional chemical experiment, cooperative scheduling of multiple workstations is not smooth, experiment process design depends on artificial experience, parameter adjustment lacks scientific data support, experiment data processing is fragmented, and product performance prediction precision is insufficient are solved.

The application discloses a double-machine clutching type testing device and method for mechanical loss measurement of a variable-frequency motor, which comprises two identical variable-frequency motors, a rotating speed measurement unit, a power measurement unit and a driving control unit, wherein any one of the variable-frequency motors is used as a driving motor, and the other one is a load motor and is always powered off during the whole testing process; the driving motor is separated from or connected with the load motor through a shaft coupling to realize no-load or load; when the driving motor is connected with the load motor, the input active power of the driving motor under different rotating speeds is measured through the rotating speed measurement unit and the power measurement unit; when the driving motor is separated from the load motor, the input active power of the driving motor under different rotating speeds is measured; and the change of the input active power measurement value of the driving motor under the same rotating speed and no-load operation is the mechanical loss of the corresponding rotating speed. The application has the advantages of relatively simple measurement process and experimental data processing process, high measurement efficiency and high measurement precision.

The utility model discloses a device for measuring two-dimensional magnetic induction intensity on an axial plane of a Helmholtz coil, and belongs to the technical field of electronic components. The device comprises a two-dimensional Hall probe, a stepping motor guide rail moving circuit and a high-precision voltage acquisition module, wherein the two-dimensional Hall probe consists of a two-dimensional Hall probe mounting base and two linear Hall sensors which are orthogonally mounted on the two-dimensional Hall probe mounting base; the stepping motor guide rail moving circuit and the high-precision voltage acquisition module are connected with the two-dimensional Hall probe; two ends of the high-precision voltage acquisition module are respectively connected with the two-dimensional Hall probe and the computer. According to the utility model, the symmetrical distribution characteristic of the Helmholtz coil magnetic field is utilized, the two-dimensional Hall probe is designed by using the two mutually orthogonal Hall sensors, and the three-dimensional space magnetic field distribution is reflected by measuring the two-dimensional magnetic field of the 1 / 4 axis plane of the Helmholtz coil; the high-precision voltage acquisition module is designed based on the single-chip microcomputer, so that measurement errors are effectively Compared with an existing device of the same type, measurement is easier and more convenient, experimental data processing is easier and more intelligent, and the expandability of the device is higher.

The present invention discloses a method for establishing an operating parameter combination model of a corn grain harvester based on an orthogonal experiment, comprising: step one, analyzing the entire operating process of the corn harvester, and preliminarily determining key operating factors that affect the operating quality of the corn harvester; step two, determining orthogonal experimental factors and levels; step three, formulating an orthogonal experimental plan; step four, carrying out a corn harvester harvesting experiment according to the orthogonal experimental plan determined in step three, and analyzing the experimental data; step five, performing experimental data processing and analysis; and step six, applying the optimized parameter combination to the actual operation of the harvester for verification. The present invention obtains an optimized operating parameter combination through this model, so that the harvester works under the optimized operating parameter combination, and can obtain better operating quality and operating efficiency, thereby providing a new solution for improving the operating quality and operating efficiency of the harvester. The operating parameter combination model can be used for the design and development of an intelligent control system for the harvester.

The invention relates to the technical field of nuclear fusion experiment data processing and numerical simulation, and discloses an automatic concurrent physical analysis method and system for a magnetic confinement fusion device, and the method comprises the steps: S1, receiving input parameters including a shot number list and a time slice list, and constructing a directed acyclic graph; s2, scheduling and executing based on the directed acyclic graph: S2.1, requesting diagnosis and balancing original diagnosis data from a plurality of data source interfaces in parallel; s2.2, converting the coordinate system of the obtained original diagnostic data into a unified normalized small-radius magnetic flux surface coordinate, and performing profile fitting to obtain profile data; s2.3, executing multi-physical coupling calculation with a dependency relationship to generate convergent plasma balance and profile information; s3, verifying the data in the process and performing grading processing on exceptions; s4, writing the final result and the intermediate result into an HDF5 file; and S5, recording full-link metadata and a running log from data acquisition to result output. According to the invention, one-button analysis is realized, and the analysis efficiency is improved.

This invention relates to an experimental data processing method, apparatus, AI analysis module, and computer equipment, belonging to the field of data processing. The method includes: segmenting multi-dimensional raw data by type and unifying the format; generating a similarity matrix based on temporal and spatial neighborhood information to mark abnormal fluctuation points; separating effective signals by matching a noise library with time-frequency features; extracting multi-layer features of basic statistics, time-series correlation, and trend changes; and dynamically selecting core features to update a tracking experimental model. The accompanying AI analysis module integrates hardware circuits for data segmentation, similarity calculation, anomaly marking, noise matching, feature extraction, and model updating, achieving full-process acceleration. This method significantly improves the efficiency of experimental data processing and the reliability of conclusions through multi-dimensional data compatible processing, accurate anomaly detection, multi-layer feature fusion, and adaptive model optimization, and is suitable for real-time analysis in various scenarios such as scientific research and industry.

The invention discloses a flow field reconstruction and PIV (particle image velocimetry) measurement method based on a physical information neural network, which belongs to the technical field of fluid mechanics, deep learning, computer vision and PIV crossing and comprises the steps of low-cost experimental system construction, experimental data processing and period identification, simulation synchronization and data set construction and physical information neural network flow field reconstruction. According to the invention, the problems of high cost and non-real time of the traditional PIV technology are solved, a novel low-cost solution is provided for flow field measurement, and the popularization and application of the PIV technology in basic scientific research and teaching are promoted.

The application provides a kind of space science experimental data online collaborative analysis method and system, it is related to scientific experimental data processing and analysis technical field.The method comprises: in response to the project creation request sent by the first device, create space science data online research project;In response to the analysis request of space science data online research project sent by the second device, according to the running environment information scheduling environment mirror and computing resources, create running environment;In response to the online analysis operation sent by the second device, based on the plurality of functional components configured by running environment, determine target sub-result, in response to the result update request sent by the second device, the target sub-result identification information is carried in result update request, according to the identification information of target sub-result, update target result.The application can quickly and accurately complete the online processing and analysis of complex scientific tasks in the scene of multi-person collaborative processing, efficiently utilize space science experimental data, effectively improve data processing and analysis efficiency.

The invention provides a full-scene interactive middle school physics virtual experiment teaching system. The system comprises a core experiment simulation module, the module utilizes a physical engine, a three-dimensional modeling technology, a virtual reality technology and an image and sound processing technology, a physical experiment process can be reproduced truly, a correct result can be provided for correct operation of students, and phenomena conforming to a real experiment can be presented for wrong operation; and a matched experimental data processing module supports acquisition, analysis and result presentation of experimental data. The system also integrates comprehensive teaching management functions of user management, course management, experiment management, teaching condition statistics, authority management and the like. On the basis, students can carry out high-quality experiments in a safe and convenient virtual environment, and understanding of physical theories is deepened through operation by themselves; meanwhile, the dependence of schools on entity laboratories and equipment is effectively reduced, the utilization efficiency of teaching resources and the flexibility of teaching are improved, and a powerful auxiliary tool is provided for physics teaching.

This invention claims protection for an improved wavelet thresholding denoising method. The method decomposes a noisy signal through several layers of wavelet decomposition, obtaining the corresponding high-frequency wavelet coefficients for each layer. The amplitude of the corresponding wavelet coefficients is analyzed; those for useful signals are lower, while those for noise are higher. An improved threshold function is applied to process the wavelet coefficients at each layer, retaining coefficients below the threshold and discarding those above. These wavelet coefficients are then reconstructed using inverse wavelet transform, outputting the denoised signal. This invention constructs a new threshold function to address the shortcomings of traditional wavelet denoising algorithms in noise reduction. Experimental data processing results show that the new threshold function effectively suppresses the influence of random errors in the inertial measurement unit. Compared to the original compromise threshold algorithm, the signal-to-noise ratio is improved, the root mean square error is reduced, and the scheme of this invention achieves better signal smoothness and excellent denoising effect compared to traditional denoising schemes.

The application discloses a fireproof mortar pyrolysis reaction mechanism obtaining method and system based on a multi-peak pyrolysis curve, and comprises the following steps: S1, collecting non-isothermal pyrolysis experimental data of the fireproof mortar, i.e. thermogravimetric experimental data; S2, processing the thermogravimetric data to obtain a differential thermogravimetric curve of the fireproof mortar, and determining the number of reaction peaks; S3, fitting the differential thermogravimetric curve based on a deconvolution method to obtain a sub-reaction peak; S4, optimizing the sub-reaction peak by using a particle swarm algorithm to determine the combined kinetic parameters thereof; and S5, obtaining the pyrolysis reaction mechanism and kinetic parameters of each reaction peak. The application processes the multi-peak pyrolysis conversion rate data of the fireproof mortar by using the deconvolution method, overcomes the difficulty in obtaining the pyrolysis mechanism function of the sub-reaction peak by using the combined kinetics and the particle swarm algorithm, solves the problem that the mechanism function needs to be assumed in the traditional method, and obtains the reaction mechanism function based on the combined kinetics, which is more suitable for the actual pyrolysis process of the fireproof mortar.

The application discloses a kind of ring upsetting friction coefficient determination methods based on image recognition, belongs to upsetting friction coefficient determination technical field, S1: the image of ring sample after upsetting is collected;S2: the inner contour of ring sample after upsetting in image is extracted by image edge extraction method, and the edge contour is further extracted using subpixel edge extraction method to inner contour;The inner diameter of ring after upsetting represented by edge contour is obtained;S3: based on standard friction coefficient theory calibration curve, the inner diameter of ring after upsetting, original inner diameter before upsetting, original height before upsetting and target height after upsetting are used to calculate the friction coefficient after upsetting.The application can improve detection efficiency on the basis of ensuring measurement accuracy through two rounds of edge feature extraction, thereby solving the time-consuming and laborious problem in experimental data processing when measuring friction coefficient by ring upsetting method.

The invention discloses a five-hole probe calibration and data processing method based on deep learning, relates to experimental measurement of aerodynamic performance of a gas turbine, and is used for quickly and accurately processing data acquired by a five-hole probe in the gas turbine. The method comprises the following steps: carrying out a five-hole probe calibration experiment; constructing a five-hole probe neural network dimensionless calibration data set; constructing and training a five-hole probe neural network; carrying out turbine blade grid pneumatic experiment and five-hole probe experiment data acquisition; experimental data processing based on a neural network; and inverting aerodynamic parameters. The internal flow of a gas turbine is complicated and accompanied by a large airflow angle and a strong pressure gradient, fine measurement is often adopted for fully capturing flow characteristics, and the five-hole probe calibration and data processing method based on deep learning is constructed to be used for realizing rapid and high-applicability five-hole probe data processing.

The invention relates to the field of aero-engine heat transfer and mass transfer analysis, and discloses a method and a system for processing cooling effect heat exchange experimental data under a curved surface structure of an aero-engine. The method comprises the following steps: expanding a whole curved surface cloud picture into a two-dimensional image containing real arc length coordinate points, reconstructing the expanded two-dimensional image into a grid matrix, performing identification assignment operation on the grid matrix by using a respe function, and accurately mapping physical quantity data to corresponding positions of the grid matrix, thereby forming a two-dimensional image type grid cloud picture. According to the method, data are displayed in a more intuitive and ordered manner, a polygonal region is drawn to select an invalid data range, and a createMask function is used to generate a binary mask region, so that the range of invalid data is clearly identified, corresponding physical quantity data is marked as invalid data, and interference is eliminated for subsequent deep analysis of valid data.

This invention relates to the fields of structural engineering experimental data processing and machine learning technology, and discloses a hysteresis curve parameter recognition method based on machine vision phenomenology. It obtains initial estimates of hysteresis model parameters using traditional parameter recognition methods. These initial estimates serve as a benchmark for applying physical constraints to perform adaptive intelligent sampling in the parameter space, generating multiple parameter combinations. The force-displacement curves corresponding to each parameter combination are converted into standard grayscale images. A Siamese convolutional neural network model is constructed and trained to learn the mapping relationship between image similarity and curve fitting goodness. The similarity between the generated image and the original experimental image is evaluated using the trained model, and the parameter combination corresponding to the highest score is selected. Further multi-layer physical rationality verification is then performed to select the optimal hysteresis model parameters. This invention integrates the physical basis of traditional methods with the visual shape recognition capabilities of CNNs, significantly improving parameter recognition accuracy and robustness while ensuring that the recognition results conform to physical laws.

The invention discloses a digital processing method and system for Zeeman effect experiment data, and relates to the technical field of atomic physics experiment data processing, and the method comprises the following steps: S1, image reading and preprocessing: obtaining a Zeeman effect interference image, carrying out the self-adaptive zooming according to the image size, converting the Zeeman effect interference image into a gray image, and carrying out the self-adaptive zooming according to the image size; enhancing the contrast ratio through self-adaptive histogram equalization, and removing salt and pepper noise by adopting 3 * 3 median filtering; s2, interference ring identification and measurement: carrying out edge detection by adopting a Sobel operator, detecting an interference ring in the image through Hough transform, screening two maximum concentric circles closest to the center of the image, automatically calculating the diameter of the ring, and converting the diameter of the ring into a physical size; s3, key physical quantity calculation: calculating a wavenumber difference, substituting the wavenumber difference into the optimized formula to calculate an electron charge-to-mass ratio, and performing error propagation analysis. Through automatic image processing and a precise algorithm, manual measurement errors are avoided, and the relative error of the electron charge-to-mass ratio can be reduced to 1.07% or below.

The invention provides an online collaborative analysis method and system for space science experiment data, and relates to the technical field of scientific experiment data processing and analysis. The method comprises the following steps: in response to a project creation request sent by first equipment, creating a space science data online research project; in response to an analysis request of the space science data online research project sent by the second equipment, scheduling an environment mirror image and computing resources according to the operating environment information, and creating an operating environment; and in response to an online analysis operation sent by the second equipment, determining a target sub-result based on the plurality of functional components configured in the operating environment, in response to a result updating request sent by the second equipment, the result updating request carrying identification information of the target sub-result, and updating the target result according to the identification information of the target sub-result. According to the method, online processing and analysis of complex scientific tasks can be rapidly and accurately completed in a multi-person cooperative processing scene, space scientific experiment data is efficiently utilized, and the data processing and analysis efficiency is effectively improved.

Embodiments of the present application disclose an experimental data processing method and device; after obtaining an experimental index data set, according to the experimental grouping and data type of the experimental index data in the experimental index data set, an experimental data set to be processed for the experimental index is constructed, then, based on a preset individual difference parameter, a difference regression coefficient of each experimental index is identified in the experimental data set to be processed, panel experimental data is screened out in the experimental data set to be processed, and based on the panel experimental data, the variance of the difference regression coefficient is calculated to obtain difference detection data of each experimental index, then, when receiving a difference detection request for a target experimental index, difference detection is performed on the target experimental index according to the difference detection data, and the detection result is sent to a terminal; the scheme can improve the processing efficiency of experimental data processing.

This invention specifically relates to a method for testing and analyzing the noise of a mechanical hydraulic steering system. The method includes: S1, placing microphones at the driver's inner and outer ears in the vehicle under test, simultaneously reading the engine speed, and measuring the in-vehicle noise level and corresponding engine speed under typical operating conditions; S2, warming up the vehicle's engine; S3, measuring typical noise under idling conditions; S4, measuring typical noise under driving conditions; S5, analyzing the measured data according to experimental data processing methods to determine whether noise problems exist. This invention describes a method for testing and analyzing typical NVH problems that may exist in mechanical hydraulic steering systems under vehicle conditions, proposing a method for testing and analyzing the noise of mechanical hydraulic steering systems, and solving the problem of insufficient testing methods in this field for the high-quality development of commercial vehicles.

The present invention provides a method for processing quantum computing experimental data, which can be applied to the field of data management technology. The method includes: constructing a quantum computing experimental data management system; using the quantum computing experimental data management system to convert the quantum computing experimental data into a preset data format, and fusing the quantum computing experimental data to obtain a merged experimental data set; setting a data access mechanism for the merged experimental data set to obtain an index-based multidimensional index structure, and based on user search requirements, searching the merged experimental data set through the multidimensional index structure; storing the merged experimental data set in a distributed database, and setting user permissions for the merged experimental data set based on a fine-grained access control mechanism; controlling the version of the merged experimental data set based on a data version control mechanism, and recording historical change information of the merged experimental data set based on data consistency. The present invention also provides a quantum computing experimental data management system.

This paper relates to the field of automated data processing and provides an experimental data processing method, apparatus, and system. The method includes: sending a task preparation sequence instruction to a data computing system, causing the data computing system to create task-related computing processes according to the task preparation sequence instruction, create data objects within the computing processes, and assign initial values; sending a data acquisition instruction to a device control terminal, receiving data acquired by the device control terminal through operation of the experimental equipment; generating a data modification instruction based on the acquired data and sending it to the data computing system; sending a data calculation instruction to the data computing system, causing the data computing system to perform calculations on the data of the relevant task computing processes according to the data calculation instruction; and sending a data output instruction to the data computing system, causing the data of the relevant computing processes to be stored in a data object database according to the data output instruction. This paper can realize the entire business process and has the advantages of ease of use, ease of learning, and versatility.

This application discloses a method, apparatus, device, and storage medium for high-resolution phase information extraction. The method includes: deconstructing and separating overlapping positive and negative phase images of a sample using a deep convolutional neural network model based on an attention mechanism; establishing a differential phase image based on the deconstructed and separated positive and negative phase images; generating network training data using an X-ray phase difference microscopy physical model; and directly using the trained model for experimental data processing to extract the absolute and differential phase information of the sample. The above-mentioned solution provided by this application can quickly and accurately extract the absolute and differential phase information of a sample from overlapping positive and negative X-ray phase microscopy data. Furthermore, this invention uses numerical simulation to generate network training data, eliminating the need to collect training data on an experimental system, greatly reducing experimental difficulty and saving experimental time.

A method and system for image recognition of stripe spacing in liquid crystal textures, relating to the field of liquid crystal texture experimental data processing technology, includes: segmenting and grayscale processing a liquid crystal texture image to obtain multiple sub-images; calculating the gradient of each sub-image to obtain image gradient information; transforming the gradient direction matrix by treating gradient directions differing by 180° as the same direction to obtain a transformed direction matrix; statistically analyzing the gradient directions corresponding to gradients greater than a first multiple of the maximum gradient value in the gradient magnitude matrix in the transformed direction matrix to obtain an directional gradient histogram; and, when determining that stripes in a sub-image have a main direction, dividing multiple horizontal sampling lines and extracting their grayscale data signals to calculate the corresponding stripe spacing. This application only statistically analyzes the gradient directions corresponding to gradients greater than a first multiple of the maximum gradient value in the gradient magnitude matrix to obtain the directional gradient histogram, which can filter noise, eliminate brightness unevenness, and improve the accuracy of image recognition results.

The embodiment of the invention discloses an experimental data processing method, system and device, equipment and a medium. The method comprises the steps of obtaining a data set corresponding to an experiment for business development, wherein the data set comprises experiment observation data and an experiment index statistical result obtained based on the experiment observation data; executing the following steps through the language model: calling a data verification service, generating a data verification result aiming at the experimental observation data, and calling a result verification service, and generating a result verification result aiming at the experimental index statistical result; calling a causal analysis service, and generating a causal analysis result based on the data verification result and the result verification result; and generating an experiment analysis result of the experiment based on the data verification result, the result verification result and the causal analysis result. According to the scheme, the efficiency, accuracy, flexibility and the like of experimental analysis are improved, and the reliability of experimental data processing is high.

The present application belongs to the technical field of marine surveying and mapping, and is a surface current prediction method based on an attention mechanism TCN-LSTM model, comprising the following steps: (1) collecting real surface current data through a data acquisition module, (2) processing surface current data using an experimental data processing module, (3) constructing an attention mechanism TCN-LSTM surface current prediction model using a model building module, (4) verifying and testing the surface current data, and (5) predicting the neural network model. The present application method utilizes the characteristics of long short-term memory networks and the advantages of time convolution, and simultaneously adds an attention mechanism, and predicts after processing real historical marine data of a certain sea area measured by a WaMoSII system. The method proposed by the present application has a surface current flow velocity accuracy of 92.9% and a flow direction accuracy of 90.4%.

The application discloses an intelligent auxiliary system for medical scientific research training based on an artificial intelligence large model, and particularly relates to the field of artificial intelligence and scientific research data processing, and is used for solving the problems of multi-source dispersion, non-uniform structure, missing steps difficult to identify and dependence on manual verification of experimental records in the existing medical scientific research process; a digital experimental record file is constructed by image acquisition on an experimental table record carrier and in combination with target detection and text recognition, logical breakpoints are recognized based on numerical jump and semantic co-occurrence, and jump record recognition is realized by using phenotype feature distribution and optimal transport calculation; missing steps are further reasoned and completed by dynamic programming, and finally, a complete experimental process draft is generated in combination with a structured template, automatic analysis, abnormal identification, process completion and multi-modal data fusion processing of scientific research data are realized, and the intelligent level of experimental data processing and the standardization of the scientific research process are improved.