83 results about "Experimental research" patented technology

The invention provides a bio-oil yield prediction method based on interpretable machine learning, and relates to the field of biomass pyrolysis technology and machine learning, and the method comprises the following steps: 1, obtaining a biomass pyrolysis original data set, and carrying out the preprocessing; 2, enhancing the preprocessed data by applying a condition table generative adversarial network CTGAN; 3, dividing the enhanced data set into a training set and a test set, performing training by adopting an extreme gradient enhanced XGB model, optimizing hyper-parameters of a BO model through Bayesian, and constructing a CTGAN-XGB-BO prediction model; and 4, performing model interpretation on the trained and optimized bio-oil yield prediction model. Compared with the prior art, the machine learning algorithm based on the CTGAN-XGB-BO model provided by the invention has the advantages that the prediction precision of the bio-oil yield is improved, the time and the cost of experimental research are reduced, and researchers can better understand and optimize the production process and improve the process control level due to explanation and analysis of the model.

The invention provides an optimal design method of an ultralow-temperature multi-load heat insulation support, a support assembly and an experimental research method, and relates to the field of large superfluid helium low-temperature distribution systems. Comprising the following steps: selecting thickness parameters of a casing pipe in each stage of temperature zone of a supporting assembly and a supporting panel as optimization design parameters to carry out parametric modeling, and introducing a stress maximum value and a heat leakage value as target values; and based on the thickness parameters of the cold shield sleeve and the supporting panel of each stage of temperature zone, taking a genetic algorithm as an optimization function, performing iterative calculation in combination with simulation software, obtaining the thickness parameters of the cold shield sleeve and the supporting panel under the optimal boundary, determining the heat leakage value and the maximum stress value of the supporting assembly according to the optimal boundary, and then processing the supporting assembly. By combining various simulation technical means and experimental research, the problems of poor performance of the support assembly, lack of process verification and the like caused by the adoption of a classic model are solved, and the rigidity and robustness of the large-scale superfluid helium transmission and distribution system are effectively optimized and improved.

The invention discloses an artistic painting analysis method based on a multi-modal large model, which comprises the following steps: firstly, acquiring an image from an artistic painting library, screening out an image with a title and an artist name by using a closed source large language model, and then marking a corresponding overall artistic analysis paragraph only paying attention to visual features, and annotation of art analysis is carried out on professional levels such as composition, color, light and shadow and the like, so that art painting analysis data is formed. And training and finely adjusting a multi-modal large model by utilizing the collected art painting analysis data to obtain an art painting generation type pre-training model for generating art painting analysis. Through experimental research, the art painting analysis capability of the multi-modal large model is remarkably improved.

The invention provides a method for identifying a microscopic association body structure in a heavy oil molecule simulation trajectory, which comprises the following steps of: analyzing a molecular dynamics trajectory, identifying an aromatic ring structure in a heavy oil molecule component, calculating a geometric center of the aromatic ring structure, combining a periodic boundary condition, and applying an improved DBSCAN clustering algorithm to realize automatic identification and clustering of an association body; and performing post-processing optimization on the clustering result, merging the shared clusters to generate a final association body set, and analyzing the number, size distribution, morphological characteristics and dynamic evolution information of the association bodies based on the structural characteristics of the final association body set. The method can accurately capture pi-pi accumulation behaviors of heavy oil molecules, further reveals morphological characteristics and dynamic evolution laws of heavy oil microassociates and influences of the heavy oil microassociates on the apparent viscosity of heavy oil, provides theoretical support for heavy oil viscosity-causing mechanism analysis and viscosity reducer design, can complement experimental research, and has a wide application prospect. And efficient development and green utilization of a complex thickened oil system are promoted.

The utility model belongs to the technical field of experimental research, and particularly relates to an experimental device for researching a subsurface zone pollutant migration rule, which comprises a water injection component, a column body, a filling medium and a detection component, the column body is of a cylindrical cavity structure, the upper end and the lower end of the column body are each provided with a water injection opening, and the two water injection openings are each communicated with one output end of the water injection assembly; the filling medium is arranged in the column body and is used for simulating a soil layer structure from surface water to underground water; the detection assembly comprises a processor, an environment monitoring module and a pollutant collection module; the environment monitoring module is used for monitoring different depth environment indexes in the filling medium; and the pollutant collection module is used for collecting pollutants at different depths in the filling medium at regular time and transmitting the pollutants to the processor for analysis. The device can avoid the interference of external environmental factors, and provides reliable reference data for accurately analyzing the migration rule of the subsurface zone pollutants in the vertical direction.

The invention discloses a similarity and popularity combined collaborative filtering method. The method comprises popularity calculation and popularity and similarity combined collaborative recommendation. The invention provides an article popularity iterative calculation method introducing high-order information, and constructs a contrast loss function to realize effective fusion of similarity and popularity. Experimental studies performed on three common data sets show that compared with a mainstream collaborative recommendation method, the model provided by the invention has obvious advantages in the aspects of recommendation performance and operation efficiency.

The invention belongs to the field of teaching experiments, and particularly discloses a safety culture basic teaching experiment research method and system. The method comprises the steps of adopting a literature review method to conclude and organize existing definitions about safety culture and materialize safety culture elements; selecting a security culture measurement questionnaire as a security culture quantitative measurement method, and selecting a corresponding security culture definition and a related security culture element setting questionnaire; according to the specific personnel composition and the corresponding proportion of the unit to be measured, selecting a proper sampling method to calculate the sample capacity, and collecting and sorting questionnaires after quantitative measurement of the safety culture of teachers and students of the school; and carrying out reliability and validity test, independent sample test and single variance factor analysis on the recovered effective questionnaire, and proposing corresponding improvement measures of a tested unit according to an analysis result. According to the invention, a mixed teaching mode is designed, so that the teaching means are improved and perfected, and the innovation and practice capabilities of students for safe culture learning are improved.

The invention relates to the technical field of biology, and discloses an RNA (Ribonucleic Acid) evolution precursor screening method based on evolutionary biology and computational biology, which comprises the following steps: constructing a phylogenetic tree, screening an amino acid sequence, and predicting a three-dimensional structure of a dimer formed by bacterial transcriptase. Constructing a nucleotide analogue small molecule database; docking the predicted transcriptase dimer with a nucleoside analogue, and screening small molecules with an excellent binding mode; molecules similar to original nucleotide-protein interaction are screened out to serve as potential candidate molecules; and carrying out double-chain structure prediction on the screened candidate molecules to obtain the RNA small molecule evolution precursor. According to the method, nucleoside analogues are systematically screened out by combining evolutionary biology and computational biology methods to serve as precursor candidate small molecules of RNA in the RNA world, compared with a traditional RNA precursor experimental research method, the method is more systematic, and compared with life origin exploration purely based on evolutionary biology, theoretical result support of computational biology is provided.

The invention relates to a reservoir compressibility evaluation method and system considering shale formation development characteristics, and belongs to the field of shale oil and gas development geology. Comprising the following steps: 1, performing experimental research on the influence of shale formation development characteristics on mechanical properties; 2, researching the influence of the shale formation on the tensile property; 3, researching the influence of the shale formation on the compressive property; step 4, performing a longitudinal fracturing fracture network layer-crossing physical simulation experiment considering the strata stress barrier belt; 5, longitudinal fracturing crack net layer-penetrating finite element numerical simulation considering the strata stress barrier belt is carried out; 6, researching the influence rule of the lamina development characteristics on the hydraulic fracture crossing behavior; 7, establishing a fracturing performance evaluation index of the shale reservoir; and step 8, evaluating the combination distribution position of the dominant lamina of the target block and the compressibility of the reservoir. According to the method, the reserve utilization rate, the recovery rate and the yield rate of shale gas development are maximized, and theoretical basis and technical support are provided for efficient development of shale gas reservoirs.

The invention provides a Steel / CFRP hybrid square tube bending performance prediction method based on different loading positions, and the method comprises the following steps: 1, carrying out the experimental research of a Steel / CFRP hybrid square tube under bias loading through arranging a bias loading experiment, forming a final deformation comparison, and achieving the analysis of the comparison of the mechanical properties of the Steel / CFRP hybrid square tube under different loading positions; step 2, constructing a finite element model of the Steel / CFRP hybrid square tube under bias loading; 3, a theoretical prediction model of the Steel / CFRP hybrid square tube under bias loading is constructed, and the influence of the bias distance on theoretical prediction energy absorption is obtained; 4, analyzing a multi-parameter coupling effect under bias loading; 5, analyzing bending characteristic influence factors; comprising the influence of different laying layer angles, the influence of different CFRP wrapping lengths and the influence of different pressing head shapes.

The invention belongs to the technical field of mining engineering safety, particularly relates to an experimental research method for coal rock impact dynamic disasters of a filling prevention and control working face, and aims to solve the problem that the sudden fracture process of overlying strata is difficult to reproduce in the traditional coal mine impact dynamic disaster research. A PCL material is heated, an asbestosed wire gauze is drawn out, a drop hammer falls in an iron pipe, a two-dimensional similar simulation experiment platform for instantaneous fracture of the overlying hard roof of the coal seam is constructed, gravitational potential energy is converted into impact kinetic energy through impact of falling of the drop hammer, and coal rock stress mutation is induced. The method comprises the following steps: simulating the instantaneous caving condition of a top plate in an overlying strata structural instability state, and capturing the whole process of dynamic interaction of a filling body-coal rock system under a coal rock impact dynamic disaster by combining a sensor and a monitoring system; meanwhile, a filling material is introduced to support the goaf, coal rock bearing pressure experimental data before and after goaf filling are compared, and the protection effect of the filling material on the coal rock under the condition of instantaneous collapse of the overlying roof is studied.

In order to solve the problem of accurate simulation of sandstone fracture evolution and dynamic crystallization in geotechnical engineering experiment monitoring, the invention provides a sandstone internal dynamic crystallization controllable system and method based on force-flow coupling driving, and four defects of out-of-control crystallization position, confining pressure distortion, transportation blockage and monitoring limitation are solved. The system integrates core packaging, solution conveying, stress applying, crystallization restraining and fracture monitoring functions. A core packaging unit adopts a molecular sieve composite membrane to reduce interference, and an annular ERT electrode realizes millimeter-level fracture positioning; the capillary transport unit drives the solution to be directionally transported by using gradient ceramic layer aperture gradual change; the axial loading module realizes uniform confining pressure through cooperation of a weight and an elastic film; dry nitrogen is introduced into the evaporative crystallization module to constrain the crystallization range. The method comprises the steps of packaging, communication, solution injection, stress application, crystallization control and crack monitoring. A reliable platform is provided for complex engineering scene fracture evolution research, the experimental research range is expanded, and the correlation research comprehensiveness of sandstone is improved.

The invention relates to a hydraulic experiment table for simulating lifting of a lifting tail board, and aims to provide a practical platform for hydraulic teaching and experimental research in related fields. The hydraulic experiment table comprises a tail plate device, a table body and a hydraulic loop. The tail plate device is composed of a tail plate, a connecting rod mechanism, a connecting rod mechanism fixing support, a portal frame and angle steel. The platform body is provided with a horizontal working platform surface, a vertical working platform surface and a hydraulic station, so that different pressure and flow output can be realized; the hydraulic loop mainly comprises a hydraulic element and a pipeline, and a sensor is installed at a key part and used for monitoring parameters of the hydraulic experiment table. The experiment table further comprises an electromagnetic valve control unit, a PLC control unit, a sensor interface and a display screen. The experiment table can simulate falling, rising and overturning of the lifting tail plate, provides visual and practical experiment demonstration for experimenters, helps students to deeply understand the principle of a hydraulic system, and improves the actual operation capability of hydraulic loop construction.

The application discloses a modeling and simulation method for surface morphology fractal growth of a nano-silver flexible film, and steps are as follows: 1, establishing a nano-silver flexible film surface morphology fractal growth model conforming to the surface morphology fractal growth of the nano-silver flexible film; 2, calibrating parameters of the model according to electron microscope scanning image data of the nano-silver flexible film obtained through experiments; and 3, verifying effectiveness and applicability of the model for simulating the surface morphology fractal growth of the nano-silver flexible film by using a morphology fractal dimension measurement technology.The model for the surface morphology fractal growth of the nano-silver flexible film can quickly simulate the surface morphology growth process of the nano-silver flexible film prepared by an in-situ reduction method, and can simulate the surface morphology fractal growth process of the film under different silver ion concentrations and reduction reaction time parameters.The simulation result of the model can guide experimental research on preparation of the nano-metal flexible film and accelerate the optimization speed of the preparation process.

The invention discloses a method for establishing a mouse radioactive jawbone injury animal model, which comprises the following steps of: performing tooth extraction operation, performing radiation treatment after one day, and performing Micro-CT (Computed Tomography) scanning, intraoral examination and histopathologic detection on the lower jawbone on the thirteenth day after radiation so as to determine that the radioactive jawbone injury mouse model is successfully constructed. The animal model has the advantages that the pathological process of human early radioactive jaw necrosis is accurately simulated; the model is stable, good in repeatability and suitable for large-scale experimental research; and a reliable, short-cycle and convenient-to-operate animal model is provided for the mechanism research of radioactive jaw necrosis and the development of new drugs.

This invention discloses a method for preparing a Ti-Mo-Cr alloy diffusion couple without stress-induced phase transformation, belonging to the field of titanium alloy diffusion technology. This invention utilizes hot isostatic pressing (HIP) technology, selecting appropriate process parameters to prepare diffusion couples with both ends being single-phase, exhibiting good adhesion and free from stress-induced phase transformation. On the one hand, this invention allows for the preparation of diffusion couples that meet experimental requirements by adjusting process parameters, thus enabling effective subsequent experimental research and avoiding blindly setting experimental parameters. On the other hand, it allows for precise control, greatly avoiding the occurrence of stress-induced phase transformation or gaps at the diffusion couple interface due to inaccurate parameter control during fixture preparation, which could lead to experimental failure. This invention enables the preparation of diffusion couples that conform to diffusion concentration gradient experiments, which can be used for subsequent electron probe microanalysis to further solve for the diffusion coefficient, facilitating the establishment of a titanium alloy diffusion kinetics database.

A quantitative experiment research analysis method for a nerve stem action potential non-stress period phenomenon comprises the following steps: step 1, selecting a plurality of different double-pulse wave interval time values t as experiment conditions, and carrying out an experiment sub-step 1, an experiment sub-step 2 and an experiment sub-step 3 aiming at each wave interval time value t; step 2, superposing the neural stem AP waveform data obtained in the experimental sub-step 1 and the experimental sub-step 3, dividing the superposed neural stem AP waveform data by 2, and representing the neural stem AP waveform data AP1 induced by the first pulse electrical stimulation in the double-pulse electrical stimulation in the experimental sub-step 2 by using the result; and step 3, subtracting the neural stem AP waveform data AP1 induced by the first pulse electrical stimulation in the experiment sub-step 2 from the mixed waveform data APall obtained in the experiment sub-step 2. According to the method, the two neural stem AP waveforms AP1 and AP2 induced by the first pulse and the second pulse respectively can be obtained, and the critical value ranges of the absolute refractory period, the relative refractory period and the refractory phenomenon-free period can be accurately determined.

The application discloses a target plate clamping device and relates to the technical field of aircraft anti-penetration performance experimental research, which comprises a supporting seat, at least one pair of oppositely arranged fixing bosses, at least one pair of oppositely arranged loading blocks and adjusting bolts. The supporting seat has a first surface. The fixing bosses are fixed to the first surface, and each fixing boss is provided with a first through hole in the direction of opposite arrangement. The opposite arrangement direction of each pair of loading blocks is parallel to the opposite arrangement direction of the corresponding pair of fixing bosses. Each pair of loading blocks is located between the corresponding pair of fixing bosses, and each pair of loading blocks is movably attached to the first surface in the direction of opposite arrangement. An installation hole is formed in the loading block in the direction perpendicular to the first surface. A threaded hole is formed in the end of the loading block close to the adjacent fixing boss. The adjusting bolt is screwed through the first through hole and is screwed with the threaded hole. The adjusting bolt is used for driving the loading block to move so as to apply a force to the target plate.

This invention discloses a method for eliminating stress-induced phase transformation during the preparation of diffusion couples in Ti-V-Mo alloys, belonging to the field of titanium alloy diffusion technology. This invention utilizes spark plasma sintering technology, selecting appropriate process parameters to prepare diffusion couples with both ends being single-phase, exhibiting good adhesion and free from stress-induced phase transformation. On the one hand, this invention allows for the preparation of diffusion couples that meet experimental requirements by adjusting process parameters, thus effectively facilitating subsequent experimental research and avoiding blindly setting experimental parameters. On the other hand, it allows for precise control, greatly reducing experimental failures caused by inaccurate parameter control during fixture preparation. This invention enables the preparation of diffusion couples that conform to diffusion concentration gradient experiments, which can then be used for subsequent electron probe microanalysis to further solve for the diffusion coefficient, facilitating the establishment of a titanium alloy diffusion kinetics database.

The invention discloses a grating sensor performance parameter experimental research device which comprises an aluminum alloy fixing plate and further comprises a first displacement table fixing base and a second displacement table fixing base which are located at the top of the aluminum alloy fixing plate, and a first translation assembly, a second translation assembly and a frame body assembly are arranged at the top of the aluminum alloy fixing plate; according to the grating sensor performance parameter experimental research device, CCD imaging changes after the distance of the structure is adjusted, the influence of grating sensor performance parameters on imaging results is researched by observing CCD image changes, and the problem that the grating sensor performance parameters are lack of experimental research is solved; a foundation is laid for research and development of high-precision, high-resolution and quick-response grating measurement products, and the practical requirements of grating sensor research and development are met.

The application provides a method for identifying micro-aggregate structure in a heavy oil molecule simulation trajectory, which comprises the following steps: analyzing a molecular dynamics trajectory, identifying aromatic ring structures in heavy oil molecular components and calculating geometric centers of the aromatic ring structures, combining periodic boundary conditions, and applying an improved DBSCAN clustering algorithm to realize automatic identification and clustering of the aggregates; the clustering results are post-processed and optimized, the final aggregate set is generated after merging and sharing clusters, and the number, size distribution, morphological characteristics and dynamic evolution information of the aggregates are analyzed based on the structural characteristics of the aggregates. The method can accurately capture the pi-pi stacking behavior of heavy oil molecules, and further reveal the morphological characteristics, dynamic evolution law of heavy oil micro-aggregates and the influence of the micro-aggregates on the apparent viscosity of heavy oil, thereby providing theoretical support for the analysis of the viscosity-causing mechanism of heavy oil and the design of viscosity reducers, and complementing experimental research, and promoting the efficient development and green utilization of complex heavy oil systems.

The application discloses a method for constructing a digital rock gene library based on massive experimental data, and comprises the following steps: collecting different typical rock samples, carrying out relevant experimental research, obtaining relevant experimental data of the typical rock samples, and processing the relevant experimental data to form rock standardized data; collecting typical geological disaster data occurring in history and processing the geological disaster data to form historical disaster standardized data; collecting real-time dynamic monitoring data of the current geological disaster; and fusing and analyzing the three kinds of heterogeneous data sources and uniformly processing the three kinds of heterogeneous data sources to form the digital rock gene library. The application adopts the three kinds of heterogeneous data sources to construct the digital rock gene library, makes up for defects that data is incomplete and methods are single in the existing rock mechanics field, and leads to different analysis results, has strong broad-spectrum and high accuracy.

The utility model discloses a silage compacting device for an automatic test, and aims to solve the problem that no special experimental instrument for silage exists in the experimental research process of the silage. Comprising a tank body, a shell is clamped and embedded in the inner wall of the tank body, a wire passing pipe is fixedly arranged on the outer wall of the shell, clamping and embedding devices are evenly arranged on the outer wall of the shell, a telescopic device is arranged on the inner wall of the shell in a matched mode, a compaction device is arranged at one end of the telescopic device in a matched mode, and a connecting rod is arranged above the compaction device in a matched mode; according to the device, the research process of experimenters on the green feed is optimized, the experiment research process is more efficient, accurate in compaction and convenient to operate, the research process of the experimenters on the silage is optimized, the efficiency of the silage is improved, and meanwhile the quality of the silage is guaranteed.

The invention relates to an osmanthus fragrans radiation mutation breeding method, and aims to solve the technical problem that a high-quality and high-efficiency osmanthus fragrans mutation breeding system and method are lacked at present. On the basis of long-term experimental research, an appropriate radiation mutagenic agent amount of osmanthus fragrans under < 60 > Co-gamma rays is screened out, a to-be-germinated branch of a perennial osmanthus fragrans arbor which grows robustly is cut in spring to serve as a scion, and grafting cultivation is performed after top end wax sealing so as to construct a mutant library; by combining phenotypic character characteristics (including quantitative characters and quality characters) which are selected after survival of sweet-scented osmanthus and are easy to identify effectively, a rapid radiation mutation breeding system and method taking sweet-scented osmanthus branches as materials are constructed, and a technical guarantee is provided for creating new sweet-scented osmanthus germplasm by utilizing 60Co-gamma ray radiation mutation. Powerful support is provided for improvement and transformation and upgrading of the osmanthus fragrans industry. The method is rapid, safe and efficient, high in survival rate and success rate, low in cost and easy to popularize and implement when applied to new sweet-scented osmanthus strain breeding.

The invention belongs to the technical field of plant disease prevention and control, and particularly relates to application of epicatechin and protocatechuic acid in prevention and control of citrus huanglongbing. Through systematic experimental research and application verification, it is found that epicatechin and protocatechuic acid can remarkably inhibit the number of citrus huanglongbing pathogenic bacteria, have a good synergistic effect, can remarkably improve the overall health condition of citrus trees and have excellent application prospects and popularization value. As epicatechin and protocatechuic acid have the characteristics of being natural, environment-friendly and efficient, the epicatechin and protocatechuic acid show important application potential in the field of prevention and treatment of the citrus huanglongbing, and are expected to become novel key tools in the field. The discovery not only provides a new thought for green prevention and control of the citrus huanglongbing, but also provides powerful technical support for sustainable development of the global citrus industry, and is of great significance for guaranteeing stable development of the industry.

The present invention discloses a method for analyzing glass phase separation based on molecular dynamics simulation. The method comprises the following steps: constructing an amorphous initial structure model of the glass; selecting a potential function for the interaction forces between the atoms of the glass according to the types of the initial glass atoms, and performing correction processing to obtain a corrected potential function; using the corrected potential function to evolve the amorphous initial structure model into a glass structure, and collecting the atomic coordinate information of the glass structure; inputting the atomic coordinate information into the visualization software VMD to obtain a glass phase separation structure diagram, and further obtaining the microscopic structure information of the glass phase separation. Based on the molecular dynamics theory, the present invention obtains the structural phase separation information at the atomic level of the glass through the molecular dynamics simulation method, makes up for the deficiency that the experimental research cannot achieve the atomic-level structure analysis, helps to more accurately understand the glass structure phase separation behavior, and regulates the glass phase separation.

The present invention discloses an immersion index method for comprehensive evaluation of liquid phase damage in oil and gas reservoir fractures. Cores of the oil and gas reservoir to be evaluated are prepared, and water phase spontaneous imbibition experiments, water phase displacement experiments, and stress sensitivity evaluation experiments are respectively carried out on the cores. Each parameter is calculated, and then the water phase trapping damage rate and immersion index of the cores are calculated based on the parameters obtained from the above calculations. The present invention takes into account the time effect of water phase diffusion, considers the permeability damage caused by the reduction of rock strength due to water-rock reaction, determines the evaluation index for water-rock reaction damage, makes up for the deficiencies of the evaluation methods for liquid phase damage in oil and gas reservoirs, can achieve more accurate prediction of the water invasion damage effect of oil and gas reservoirs and the stimulation operation effect of reservoirs, and has certain significance for indoor experimental research and field application.

The present disclosure provides a wellbore instability analysis method based on big data, which relates to the field of wellbore instability. The wellbore instability prediction based on big data analysis and deep learning, combined with existing wellbore instability theory and experimental research results, finds the main control factors and related factors of wellbore instability in the studied block. Based on data classification, a big data model is established to improve the accuracy of analysis and prediction through big data learning. Through the application of the big data model, the deep learning technology is used for regional wellbore instability evaluation and research, and finally the regional wellbore instability evaluation and prevention technology based on big data is formed. The wellbore instability analysis tool based on big data analysis mainly provides the safety density window for preventing wellbore instability, the control analysis of hole diameter change, the optimization of drilling fluid performance and additive dosage, adjacent well data analysis, ECD analysis and drilling time effect analysis through data analysis, so as to reduce the hydration swelling effect of the target formation rock and prevent wellbore instability. The present disclosure is suitable for wellbore instability prediction.

The invention relates to a bacterial colony intelligent analysis method and device based on deep learning and a storage medium, which are applied to the technical field of microbiological detection and image analysis, and comprise the steps of performing model optimization on the characteristics of a bacterial colony image based on a YOLOv11 deep learning model in combination with a transfer learning method, remarkably improving the accuracy and robustness of bacterial colony detection, and improving the accuracy and robustness of bacterial colony detection. The problems of missing detection and false detection caused by overlapping of bacterial colonies, various forms and complex backgrounds can be effectively solved; the method supports the simultaneous recognition and classification statistics of a plurality of bacterial colonies, automatically generates a structured Excel report containing rich information, does not need manual secondary data arrangement, improves the standardization and traceability of bacterial colony analysis, and provides reliable data support for experimental research and detection evaluation. The full-process automation of bacterial colony detection, classification, counting and report generation is realized, manual intervention is not needed, the labor intensity of workers is greatly reduced, the bacterial colony analysis efficiency is improved, and the high-throughput detection requirement is met.

The invention discloses a method for constructing a digital rock gene pool based on mass experimental data, which comprises the following steps: collecting different typical rock samples, carrying out related experimental researches to obtain related experimental data of the typical rock samples, and processing the related experimental data to form rock standardized data; collecting typical geological disaster data occurring in history, and processing the typical geological disaster data to form historical disaster standardized data; collecting current real-time dynamic monitoring data of geological disasters; and performing fusion analysis and unified standardization processing on the three heterogeneous data sources to form the digital rock gene pool. The digital rock gene pool is constructed by adopting three heterogeneous data sources, the defect that analysis results are different due to incomplete data and single method in the existing rock mechanics field is overcome, the broad spectrum is high, and the accuracy is high.