409 results about "Sample pool" patented technology

The invention discloses a traceable cavity ring-down spectrometer system with a self-calibration function, and is applied to the technical field of gas concentration detection. The device comprises an optical cavity ring-down main body, a dynamic generation module and a reference analysis module which are connected in sequence, wherein the optical cavity ring-down main body comprises a laser light source, a beam splitter, an optical fiber isolator, a lens, a first reflecting lens, a first aluminum ring, a sample cell, a second aluminum ring, a cross shaft type universal coupling control device and a second reflecting lens which are arranged along a laser light path; the dynamic generation module comprises a high-purity gas inlet, a gas purifier, a dynamic generation device, a standard gas outlet and a three-way electromagnetic valve which are arranged along a gas path; the reference analysis module comprises an instrument standard gas inlet, a reference analyzer and a standard gas storage box which are arranged along the gas path. According to the method, the influence of instrument drift on analysis data caused by the use environment and instrument part loss of the online instrument is avoided, and the accuracy and the quantity value traceability of the instrument analysis data are guaranteed.

The invention discloses an industry instruction fine tuning data set automatic generation method and system, and belongs to the technical field of artificial intelligence, S10: selecting industry standard document data, and analyzing the data into a plurality of paragraph texts as data sources according to chapters; s20, generating different types of questions for the paragraph text by using a large language model; s30, scoring the generated questions according to a scoring rule, and filtering the questions with scores lower than a preset threshold value; s40, the solvability and difficulty of the question are adjusted through a large language model, and the quality of the question is optimized, and S50, a first version answer of the question is generated by using the large language model, and the answer is optimized by using a hierarchical region optimization search algorithm; s60, constructing a sample pool through global and local selection methods, calculating a sample compression ratio and screening a data set; the method has the beneficial effects that through a highly automatic process and an optimization algorithm, the data set generation efficiency and quality are improved, the labeling cost is reduced, and support is provided for rapid deployment of a large language model in industry application.

The invention relates to the technical field of laser particle size analyzers, in particular to a laser particle size analyzer with a double-ultrasonic structure, which comprises a particle size analyzer body, a sample pool is arranged on the particle size analyzer body, a sample dispersion mechanism for crushing particle blocks is arranged in the sample pool, and a stirring auxiliary mechanism for preventing suspended particles from settling is arranged above the sample pool. The stirring auxiliary mechanism can push the particles to a sound field strong area of the sample dispersion mechanism so as to improve the dispersion efficiency, and a splash-proof covering mechanism is arranged at the upper end of the sample pool. By arranging the sample dispersion mechanism and adopting a mode of combining phase orthogonality and material orthogonality layout, a stable three-dimensional acoustic vortex can be formed in the sample pool, and the coverage range of the ultrasonic cavitation effect can be obviously improved, so that the particle dispersion uniformity is improved, and the particle dispersion efficiency is improved. A detachable low-frequency ultrasonic vibrator is adopted at the bottom, so that the penetrating power of a high-concentration and high-viscosity sample is further enhanced, and dispersion dead angles can be effectively eliminated.

A method for detecting circularity of silicon carbide micro powder relates to the technical field of silicon carbide detection, and comprises the following steps: S1, turning on a flowing particle imager, preheating and then carrying out background detection; s2, adding standard particles into the sample pool, and enabling the flowing particle imager to perform automatic focusing; s3, adding the prepared silicon carbide micro powder sample into a sample pool, and detecting by using a flowing particle imager; and S4, storing the detection result, checking the silicon carbide micro powder particle image, and recording the detection result. The preparation of the silicon carbide micro-powder sample comprises the following steps: S31, adding the dried silicon carbide sample into a dry and clean beaker; s32, sequentially adding an impregnating compound, a dispersing agent and deionized water into the beaker; s33, carrying out ultrasonic dispersion; and S34, taking out the uniformly mixed sample. The method can solve the technical problem of detection failure caused by easy agglomeration of the silicon carbide sample when the flowing particle imager is used for detecting the circularity of the silicon carbide micro powder.

The invention discloses a dual-lens laser particle analyzer with an automatic calibration function. The dual-lens laser particle analyzer comprises a lead-in assembly, a shaking assembly and a mist brushing assembly, the introduction assembly is used for mixing and introducing to-be-measured particles and a water source, the shaking assembly is used for continuously shaking the particles of a to-be-measured mixed solution so as to ensure that the particles are uniformly dispersed, and the mist brushing assembly is used for brushing away mist which is generated on a lens due to too high humidity and interferes with the laser measurement degree. Through the combination of the first fixing plate, the first fixing shaft, the fixing ring and the torsional spring, the first sample pool can flexibly shake within a certain range, the motor drives the reciprocating lead screw to drive the toothed plate to move, the first sample pool can shake and swing, and through the continuous and diverse movement mode, the detection accuracy is improved. The uniform distribution of the particles in the solution is ensured, the interference of particle precipitation or aggregation on the measurement result is avoided, and the accuracy of measuring the particle size by the laser particle analyzer is obviously improved.

The invention discloses a detection device and a detection method for dissolved gas in insulating oil. The detection device comprises a rapid degassing module and an analysis module, the rapid degassing module comprises a sample bottle, a heating oil sample pool, a stirring structure, a high-purity gas storage tank and a cold trap; the sample bottle is used for injecting an insulating oil sample to be detected; the heating oil sample pool is used for heating the insulating oil sample in the sample bottle; the stirring structure is used for stirring the insulating oil sample in the sample bottle; the high-purity gas storage tank is used for introducing inert gas into the sample bottle and accelerating discharge of dissolved gas; the cold trap is used for enriching the mixed gas discharged from the sample bottle and selectively adsorbing and enriching the target gas to be detected; and the analysis module is used for detecting the gas analyzed from the cold trap. According to the detection device disclosed by the invention, the insulating oil sample to be detected is simultaneously subjected to heating stirring, gas purging replacement and cold trap enrichment, so that the dissolved gas in the insulating oil can be quickly removed, the degassing time is shortened to be within 5min, and the degassing treatment efficiency is greatly improved.

An MIG welding parameter optimization method based on a composite heat source simulation model comprises the steps that firstly, a geometric model and a physical model for simulation are determined, then grid division work is carried out on the simulation model, reasonable boundary conditions are designed for the simulation model, and then a Gaussian heat source model and a cylinder heat source model are imported into the models; the method comprises the following steps of: setting a time step, starting a simulation process, monitoring, comparing a simulation result with an actual experiment result, verifying, and continuously and repeatedly optimizing the model until the simulation result and the actual experiment result have good consistency. A composite welding heat source model is embedded in a traditional heat source simulation model, so that real data such as the shape and the forming size of a welding pool under the welding wire filling condition can be more accurately simulated, the simulation precision is high, the data are reliable, and the size of the actual welding sample pool can be well fit.

The invention relates to a device and a method of a high-sensitivity suspension single-droplet sensor for monitoring PM2.5 pollution, belongs to the technical field of aerosol microdroplet spectrum trace analysis, and solves the problem that ppb-magnitude gas cannot be monitored in the prior art. The method comprises the following steps: pumping aerosol into a sample cell of a droplet suspension system in a single droplet form; introducing required reaction gas into the sample pool through an atmosphere control system; scattered light signals of the single liquid drops are respectively input into the dynamic signal processing system and the microdroplet imaging system; acquiring the characteristic peak displacement change of the metal ion sensing microdroplet; establishing a gas particle distribution Mie scattering inversion model based on micro-droplet dynamic response; inverting the radius of the liquid drop according to the stimulated Raman scattering / stimulated elastic scattering signal; according to the quantitative relationship of the reaction, calculating the generation of sulfate by the droplet suspension system (sensor); and judging which gas in the actual atmosphere is a main oxidation pathway for oxidizing sulfur dioxide to generate sulfate pollutants according to a calculation result.

An online soft measurement method for dioxin emission concentration in MSWI process includes: performing principal component analysis on process data according to a historical process data set of the MSWI process to obtain a drift index control limit; constructing an offline model based on FTBL, and inputting the process data and historical DXN true value data of the MSWI into an off-line model; performing principal component analysis according to online data, judging whether the online data is drift data according to the drift index control limit; if the typical sample pool is the drift data, constructing an online model based on FTBL, and inputting the process data and the drift data of the typical sample pool and output data of the incremental layer of the offline model into the online model; and determining a DXN emission concentration predicted value according to an offline calculation result and an online calculation result.

The invention relates to the technical field of soil treatment, and particularly discloses a soil micro-plastic electrostatic separation device and method.The soil micro-plastic electrostatic separation device comprises a sample pool, an electrostatic adsorption mechanism is arranged on one side of the sample pool, and the electrostatic adsorption mechanism is connected with an electrostatic discharge generator through a wire; the electrostatic adsorption mechanism comprises an adsorption plate and a connecting handle; the bottom box and the heightening frame bodies are made of wood, the height of the sample pool can be adjusted by installing the heightening frame bodies in an assembling mode, a nylon coating is arranged on the outer side of an adsorption plate and used for adsorbing micro-plastics and eradicating the risk of high-voltage electric leakage, and the height of the adsorption plate can be adjusted through an arranged supporting assembly; the particle size range of the micro-plastic covered by electrostatic adsorption is wide, compared with a traditional method, the recovery rate is remarkably improved, and the original form of the micro-plastic and the natural structure of soil are protected through normal-temperature physical separation and no reagent addition.

Disclosed in the present invention are an apparatus for measuring the refractive index of a liquid on the basis of a transmission optical path, and a measurement method therefor. The apparatus comprises a laser, a sample cell, a thermostatic bath, a photoelectric imaging device, a control unit and a PC, wherein the sample cell is light-transmissive at the front and rear, and one or both of walls of a light-transmissive face is / are of a right-trapezoidal wedge-shaped structure; the thermostatic bath can perform temperature control on a liquid sample; and laser light emitted by the laser penetrates the liquid sample in the sample cell, and finally forms an image on the photoelectric imaging device placed in a transmission direction. The method comprises: when laser light passes through a sample in a sample cell, the laser light being refracted at an interface between different media, such that the direction of an emergent optical path changes, wherein the variable of the change is related to the refractive index of the sample; the laser light forming an image on a photoelectric imaging device after penetrating the sample; and calculating the refractive index of the sample on the basis of the relationship between the position of an imaging point and the refractive index. The present invention achieves measurement of the refractive index of a liquid with an accuracy better than 0.05%.

The utility model discloses a modularized fluorescence analysis and detection device, which relates to the technical field of fluorescence analyzers and particularly comprises a first shell which is a control part, a display screen is arranged at the top of the first shell, a control mainboard is arranged in the first shell, and the modularized fluorescence analysis and detection device further comprises a second shell connected with the first shell; the second shell is a detection part, and a frame body for placing a sample cell is arranged in the second shell; a fluorescence receiving part is arranged at the top in the second shell, an irradiation part is arranged at the bottom in the second shell, the fluorescence receiving part is located above the sample cell, and the irradiation part is located below the sample cell. During use, the first electric actuator can be controlled to move by operating the keys, so that the positions of the fluorescence receiving part and the irradiation part are adjusted, and a proper distance is kept between the optical filter group at the bottom of the fluorescence receiving part and the sample cell; and the light beams emitted by the irradiation part can be driven by controlling the first electric actuator to be completely concentrated on the sample cell.

The invention discloses a sample screening method and system based on weight clustering and dynamic interval reduction. The method comprises the following steps: generating an initial sample pool through a Monte Carlo method; based on the initial sample pool, utilizing a sample screening process to carry out multiple rounds of iterative screening to obtain a screened sample pool; the sample screening process comprises the following steps: acquiring uniformly distributed candidate samples from a current sample pool by adopting a weight clustering method; predicting an upper threshold value and a lower threshold value of a limit state function in the sample pool through an agent model by utilizing an interval reduction method; and selecting the candidate samples of which the predicted values of the limit state functions are between the upper threshold value and the lower threshold value, and updating the sample pool. According to the method, candidate samples are dynamically screened in combination with a weight clustering method, interval reduction is carried out by utilizing upper and lower thresholds of a limit state function predicted by an agent model, the screening precision is ensured, meanwhile, the scale of the candidate samples is effectively compressed, the redundant calculation amount is reduced, and efficient iterative optimization is achieved.

The invention discloses an extraction-free in-situ intelligent detection system and method for furfural in transformer oil. According to the method, the in-situ surface enhanced Raman detection method containing the liquid core optical fiber is designed, coupling of a transformer oil sample pool can be achieved by introducing the liquid core optical fiber into a detection system, oil way circulation is achieved, an analysis model of the furfural content and the furfural Raman characteristic peak in sample oil is established through a deep learning convolutional neural network, and the detection accuracy is improved. High-stability, high-sensitivity and high-speed extraction-free in-situ detection of furfural in transformer oil is realized.

The invention discloses a time-varying reliability analysis method based on a double-layer BP neural network, and relates to the field of time-varying structure reliability analysis, and the method comprises the steps: generating a sample pool based on a joint probability density function which affects a random design variable of a to-be-analyzed time-varying structure; calculating weights of sample points, screening part of the sample points and establishing a first training data set; an inner-layer BP neural network model is constructed and trained, the minimum value of the time-varying performance function corresponding to each sample point in the first training data set about time is calculated, and a second training set is formed; an outer-layer BP neural network model is constructed, and a second training set is adopted for training; the reliability of a time-varying structure is calculated by adopting a trained outer-layer BP neural network model, the minimum value of a time-varying performance function about time is solved through an inner-layer model, and the outer-layer model is constructed based on parameters of sample points and the minimum value solved by the inner-layer model, is used for calculating the failure probability and has good applicability. The method can be used for time-varying reliability analysis of complex structures.

The invention relates to the technical field of fluorescence detection, in particular to a three-dimensional fluorescence detection correction method based on double light sources, and the correction method comprises the following steps: obtaining double light source optical parameters, sample cell geometric parameters and original fluorescence intensity data, the geometric parameters of the sample pool comprise the length, width, height, pool wall reflection coefficient and sample attenuation coefficient of the sample pool; and constructing a light intensity distribution model in a three-dimensional space of the sample cell based on the optical parameters of the double light sources and the geometric parameters of the sample cell. According to the invention, accurate characterization of the light intensity of any spatial point in the sample cell is realized, light intensity deviation caused by various factors is quantified from the source, personalized correction is carried out on each spatial point, the influence caused by optical path difference at different positions, light source fluctuation and element error is effectively offset, and the accuracy and accuracy of the detection result are improved. The consistency of detection values of standard substances with the same concentration in different areas of the sample pool is greatly improved, and the linearity of a standard curve is remarkably optimized.

The invention discloses a multi-dimensional spectrum detection device based on combination of a transmission surface and a scattering surface, and belongs to the technical field of spectrum detection.The device comprises a camera obscura, a light source part, a detection part and a sample pool, the detection part and the sample pool are arranged in the camera obscura, the light source part comprises a light source and an angle fixing base, and the light source is connected to the angle fixing base through an optical fiber; an achromatic collimator is arranged on the end surface of the angle fixing seat and is arranged below the sample pool; the detection part comprises an emergent light detector arranged right above the midpoint of the sample cell and a plurality of scattered light detectors arranged on the side surface of the sample cell, and the emergent light detector collects transmission light spots generated after the sample is irradiated and obtains transmission information; the scattered light detector collects scattered light spots generated after the sample is irradiated and obtains a scattered spectrum image. According to the invention, the limitation of a single detection mode is effectively reduced, the distribution characteristics of scattered light at different angles and background information of transmission light are utilized to form complementary calibration, and the spectral quantitative analysis precision of the substance component content is improved.

The invention provides an edge real-time video analysis-oriented resource efficient continuous learning method and system. The method comprises the steps of real-time video analysis and model retraining; an online model precision reduction predictor oriented to the RoI granularity is designed; continuously collecting the precision reduction degree of the lightweight model compared with the high-precision model; then designing a double-layer mixed sample pool, and based on the designed RoI granularity-oriented online model precision reduction predictor; randomly introducing a certain proportion of RoI historical samples to relieve the problem of disastrous forgetting; and finally, developing a deep reinforcement learning DRL-based retraining progress controller. Based on the evaluation result of each round, dynamically estimating a model convergence state to determine a retraining termination opportunity; by introducing a behavioral clone BC loss function, the controller converges the optimal strategy more quickly in an offline training stage. The technical scheme of the invention has an important positive significance for improving the SP income in the MEC environment.

The invention discloses an adaptive PC-Kriging reliability analysis method and system based on active learning, and the method comprises the steps: obtaining uniformly distributed first candidate samples from a pre-generated MC sample pool through employing a weight clustering method, and constructing an initial PC-Kriging model; selecting samples by using an interval reduction method to construct a new sample pool, and dividing the new sample pool into two sub-sample pools of a security domain and an invalidation domain; aiming at the two sub-sample pools, adopting the weight clustering method again, and respectively selecting uniformly distributed second candidate samples from the two sub-sample pools; constructing a crossing point by using the second candidate samples distributed in the failure domain and the security domain; taking the constructed crossing point as a newly added experimental point, and carrying out iterative updating on the PC-Kriging model; and judging convergence and outputting a result. According to the technical scheme, through interval reduction of the focusing key area and model updating based on the crossing points, the limit state surface can be more accurately approached, the number of experimental points is reduced, the prediction precision is improved, and the calculation resource utilization is more efficient.

The invention relates to the technical field of livestock measurement, in particular to a beef cattle body size intelligent measurement device and method, in the beef cattle body size intelligent measurement device and method, hoof joint point pixel coordinates are converted and angle differences are calculated in combination with a lexicographical sequence block reordering algorithm, identity binding codes are accurately generated, uniqueness and stability of beef cattle individual recognition are ensured, and the accuracy of beef cattle body size measurement is improved. The method effectively reduces the identification confusion risk when multiple beef cattle are measured at the same time, generates a simulation sample through local pixel block rendering and generative adversarial network, screens a sample according with a threshold according to authenticity scores, improves the sample richness, optimizes the coverage of a model training set, extracts body size fields and position coding values, and calculates correlation coefficients between fields. Low-correlation field pairs are removed, only high-correlation fields are reserved for cross combination, a modeling sample pool is constructed, the representativeness of samples and variable combination diversity are enhanced, and the fitting precision and popularization capability of a body size estimation model are improved.

The invention discloses a method for measuring the molar concentration of nano-particles. The method comprises the following steps: diluting a nano-particle suspension with a high-molecular polymer solution or dispersing nano-particle powder into the high-molecular polymer solution; transferring the suspension liquid to a spacing net on the side of a sample pool in an electrophoresis tank; nano-particles migrate from the sample pool to the buffer pool under the driving of the electric field and are uniformly distributed on the spacing net in a single-particle manner; performing single-particle imaging on the spacing net by using an optical microscope; and counting the number of the nanoparticles in the image so as to measure the molar concentration of the nanoparticles in the sample. The method can be used under the condition that the properties of the nanoparticles are unknown or no standard substance exists, the detection object has no additional condition, and the accuracy is high.

A device (100) for characterizing particles using nanoparticle tracking analysis is provided. The device comprises: a sample cell (101) for containing a sample, said sample comprising a plurality of particles suspended in a fluid; a light source (102) configured to illuminate the sample; an imaging system (103) configured to collect light (112) scattered or fluoresced by particles moving within the sample cell (101) and within a detection zone (111) of the imaging system (103) and to capture a video of particles moving within the detection zone (111); and a computer (104). The computer is configured to process the video to automatically determine the depth of the detection zone (111). Determining the depth of the detection zone (111) comprises: tracking particles moving within the detection zone (111), thereby generating a trajectory for each particle; and analyzing how the measured attribute of the trajectory within the sub-segment of the video varies as the size of the sub-segment varies.

According to the target identification method and training method in the infrared image, the enhanced image features and the jigsaw image features are utilized to construct the sample pool, and the model is trained by optimizing the contrast loss between the enhanced image features and the jigsaw image features, so that the model can close different versions of the same image through contrast learning, and the target identification efficiency is improved. According to the method, the distance between the enhanced image and the jigsaw image in the feature space is determined, so that the dependence of target identification classification on annotation is reduced, and the feature extraction capability is enhanced and the classification robustness is improved by utilizing the comparative learning of the enhanced image features and the jigsaw image features.

This application belongs to the field of aircraft structure simulation calculation, and specifically relates to a method for constructing a high-dimensional output aircraft structure global proxy model, including: step 1, according to the probability distribution f of the structural model input variable X X (x), extract N input samples {x (1) ,…,x (N)} T , build a sample pool for global modeling. Step 2: Select N0 input samples from the sample pool and substitute each input sample into the finite element model for calculation to obtain the corresponding output response sample. Step 3: Perform PCA dimensionality reduction decomposition on the output response sample and convert the output response sample into the mean-centered output principal component sample. Step 4: Construct the Kriging model g with the input sample output principal component sample. K (X); Step 5, use the variance learning function to learn the Kriging model g K (X) is updated to obtain the Kriging model g for calculation K (X); Step 6: Calculate the Kriging model g K (X), perform matrix reconstruction to obtain a high-dimensional output prediction model.

The invention discloses an AI-enabled real-time electrochemical detection system, which is composed of a real-time electrochemical detection module and an embedded AI data processing and analysis module, the real-time electrochemical detection module comprises an electrochemical electrode system, a sample pool and an electrochemical workstation, the surface of a working electrode is modified with a probe of a specific recognition target, and when the probe is combined with the target, the probe of the specific recognition target can be detected by the embedded AI data processing and analysis module. The working electrode interface charge transfer impedance changes in real time to generate a dynamic electric signal, and the electrochemical workstation provides electric input and collects the signal; the embedded AI data processing and analysis module is used for denoising the dynamic electric signals through an adaptive filtering algorithm based on wavelet transform, extracting time-frequency domain joint features through a 1D-CNN and LSTM mixed deep learning model, and establishing a nonlinear mapping relation between the signals and target concentration. Through AI enabling electrochemical detection, the defects of low time efficiency, large error and low information utilization rate of the traditional technology are comprehensively overcome, and an innovative solution is provided for high-precision, rapid and anti-interference target substance detection.

The invention relates to the technical field of coal-fired power plant desulfurization system control, and discloses a dynamic load response desulfurization slurry circulating pump intelligent optimization regulation and control method comprising the following steps: S1, data acquisition; s2, data preprocessing; s3, constructing a model and training the model; s4, iteratively optimizing the model; and S5, optimal regulation and control of the circulating pump. According to the intelligent optimization regulation and control method for the desulfurization slurry circulating pump based on dynamic load response, big data transmitted by operation of desulfurization system equipment is comprehensively collected, a data cleaning method is adopted, and noise and abnormal values in the data are removed, so that the accuracy of model construction and the reliability of model training and prediction are guaranteed, and the method is suitable for large-scale popularization and application. The method comprises the following steps of: establishing a model, performing reasonable circulation and test division on data, constructing a sample pool, thereby improving the generalization ability of the model, establishing a prediction model by adopting a support vector machine, enabling the model to have higher fitting precision, and enhancing the optimization operation mode of the slurry circulating pump to adapt to the complexity of boiler load and working condition change.

The invention discloses a variant reconstruction AUV shape optimization design method based on knowledge migration. The method comprises the steps that a historical database is constructed through historical design and optimization of a single aircraft; carrying out data selection so as to construct a target domain database in a target domain task of shape design optimization of the multi-body structured aircraft; generating a target domain sampling data set through sampling; establishing an intermediate domain task and a corresponding intermediate sample set for shape design optimization of the dual-body aircraft; constructing an agent model of the target domain task and the intermediate domain task; establishing a migration integration model based on a three-domain task; carrying out iterative optimization on sample points in the target domain sampling data set by utilizing the integrated migration model, and constructing a sample pool; and constructing a multi-body structured aircraft appearance hydrodynamic simulation calculation framework, performing hydrodynamic simulation based on the sample points in the sample pool, and taking the design parameters contained in the finally obtained sample point with the minimum sailing resistance as a multi-body structured aircraft appearance design scheme.

The embodiment of the invention relates to an annotation quality control method and device with active learning ability, and the method comprises the steps: constructing and training a difficulty evaluation model and a quality prediction model before executing a first document annotation task; in the labeling process of each piece, labeling personnel are dynamically distributed based on the evaluation result of the difficulty evaluation model and the personnel information base, the labeling quality is monitored in real time by using the quality prediction model in the labeling process, the labeling result is subjected to quality evaluation according to rules when labeling is finished, and the personnel information base and the labeling sample pool are updated; after the first document task is executed, whether active learning needs to be carried out or not is confirmed regularly according to the recent evaluation condition, if yes, typical records are selected from the labeled sample pool to update the verification data set, and the difficulty evaluation model and the quality prediction model are upgraded based on the updated verification data set. According to the invention, the efficiency can be improved, the cost can be reduced, real-time early warning and dynamic task allocation can be realized, and the quality control quality can be actively improved.

The invention belongs to the technical field of optics, and discloses an experimental device, experimental method and demodulation method based on tilted fiber bragg grating refractive index sensing, the experimental device comprises a broadband light source, a polarization controller, a tilted fiber bragg grating, a liquid sample pool, a data acquisition module, a temperature automatic compensation module and a data processing module, the light output end of the broadband light source is connected with the light input end of the polarization controller, the light output end of the polarization controller is connected with the light input end of the tilted fiber bragg grating, the light output end of the tilted fiber bragg grating is connected with the data acquisition module, and the data acquisition module is connected with the automatic temperature compensation module. The temperature automatic compensation module is connected with the data processing module, and the tilted fiber bragg grating is arranged in the liquid sample pool. The experimental device is relatively simple in structure, good in experimental real-time performance and capable of meeting the requirements for high sensitivity, microscale and dynamic monitoring.

The invention discloses a security detection method and device for network transmission data, and belongs to the field of data security. The method comprises the following steps: performing security detection on network transmission data by using a security detection model and a clustering model; and outputting a detection result based on comparison of detection results of the security detection model and the clustering model. When the comparison result is inconsistent, calculating a comprehensive safety coefficient, and storing the network transmission data into an experience sample pool; and respectively carrying out model optimization on the security detection model and the clustering model by utilizing samples in the empirical sample pool. According to the scheme, the supervised security detection model and the unsupervised clustering model are used for parallel detection, and bimodal high-precision detection is generated; through cross-modal conjoint analysis of two modal models, doubtful network transmission data is stored in an experience sample pool, parallel optimization of a supervised security detection model and an unsupervised clustering model is realized by using a sample of the experience sample pool, online closed-loop hot update of the models is realized, dynamic perception of novel attacks is improved, and the misjudgment rate is reduced.