33 results about "Single sample" patented technology

The application discloses a water quality and soil detection and analysis method and system based on multi-standard automatic evaluation and belongs to the technical field of environmental monitoring data processing. The system comprises a data import module, an intelligent identification module, a batch sample processing module, a multi-standard evaluation module, a comprehensive analysis module, a water chemistry verification module and a result output module. The system automatically analyzes the Excel table structure through an intelligent identification algorithm, extracts index names, units, detection results and background values, supports batch sample format identification, can analyze multiple samples at a time, integrates three environmental quality standards of underground water GB / T 14848-2017, surface water GB 3838-2002 and soil GB 36600-2018, automatically performs quality classification evaluation, calculates single pollution index P i , exceeding multiple, Nemerow comprehensive pollution index P N , performs water chemistry anion and cation balance verification, Shukarev classification, TDS and hardness checking, and supports single sample and batch sample export and summary table generation. The application solves the problems of weak Excel data recognition ability, batch sample processing difficulty, lack of multi-standard automatic switching, single comprehensive evaluation method and insufficient data quality checking in the prior art, and greatly improves the efficiency and accuracy of detection data analysis.

This invention, entitled "A Method and Application for Target Gene Quantification in qPCR Using a Quantitative Internal Standard and a Comprehensive Correction Coefficient," belongs to the field of nucleic acid detection technology. The technical problem it addresses is that existing qPCR techniques require the establishment of a standard curve using a reference sample of known concentration, increasing reagent costs and experimental complexity, and resulting in poor reproducibility. The key technical solution is to provide a method for quantitative viral load qPCR analysis of a single sample. Specifically, it provides an internal standard with a known copy number, i.e., a quantitative internal standard, and establishes a correction method. Through a systematic error compensation mechanism, it achieves accurate quantification of the target gene. This method ensures both accuracy and reproducibility while simplifying detection design in clinical applications.

This invention relates to a lightweight adaptive method for testing non-stationary data streams. Its core lies in constructing a two-branch adaptive architecture that decouples observation and execution statistics. A shadow observation branch, independent of the main inference branch, maintains a statistical reference baseline in real time, unaffected by adaptive calibration actions, thus suppressing the closed-loop coupling problem between drift perception and statistical updates at the mechanism level. Based on this decoupled architecture, the shadow branch calculates a composite drift surrogate metric to map and generate dynamic momentum parameters, and introduces relaxed cumulative variables to perform three-level state determination and spatiotemporal misalignment mask generation. Finally, during inference forward propagation, dynamic online calibration of statistics is performed only for the key levels selected by the mask. This invention alleviates the adaptation lag and statistical oscillation risks caused by fixed momentum, suppresses the statistical degradation risk caused by single-sample noise, and is suitable for real-time model optimization in single-sample, non-stationary streaming scenarios.

The present application provides a course learning model training method, a course learning method and related devices, the course learning model training method comprises: using the problem data set corresponding to the target course to perform reinforcement learning training on the large language model, in each training process, the current problem data set is filtered using a question classification threshold, and the filtered problem data set is used as the training sample of the next round of iteration training of the large language model, until the course learning model is obtained under the condition that the large language model meets the target iteration condition; the question classification threshold is determined based on at least one of the correctness of the reply data corresponding to different questions and the course category. The method and device of the present application can provide an adaptive course learning scheme with single sample granularity, based on real-time capability feedback and without additional strategy network training, thereby improving the reliability of the course learning result.

This invention belongs to the field of testing and analysis technology, and discloses a method for rapidly extracting and detecting the content of various terpenoids and sterols in rubber grass. The specific steps are as follows: Step 1: Preparation of standard solutions of terpenoids and sterols. Detection is performed using a general-purpose high-performance liquid chromatograph, eliminating the need for large-scale precision mass spectrometry equipment, resulting in lower purchase and maintenance costs. The operation process is simple and easy to learn, meeting the needs of high-throughput rapid detection of batch samples. A single sample test only takes 35-60 minutes, significantly improving detection efficiency. It can simultaneously complete the qualitative identification and content determination of five target components, including lupeol and taraxasterol. The method has good repeatability and high accuracy, with excellent linearity of the standard curve, and stable and reliable detection results. By optimizing the extraction solvent, temperature, time, and isocratic elution chromatographic conditions, the target components are fully extracted and separated effectively, allowing for accurate acquisition of the content data of each component.

The application discloses an infectious disease image data enhancement method and system, which performs single-sample data enhancement on each sample in an infectious disease image data set, and then randomly selects samples for multi-sample data enhancement. In the single-sample data enhancement aspect, multiple data enhancement strategies are summarized, and a random function is used to select the data enhancement strategies, and the possibility of each strategy selection is equal. The samples are enhanced through multiple random layers. The multi-sample data enhancement is enhanced through the single-sample data enhancement part after linear enhancement of multiple samples. The final result summarizes the results of the two kinds of enhancement, and the generalization and robustness of the method are ensured.

The application discloses a pangolin species identification method based on nail piece element characteristics, which comprises the following steps: step S1: collecting nail piece samples covering existing 10 kinds of pangolin species, and obtaining initial data of multi-element characteristics of the nail piece samples; step S2: identifying the species of the nail piece samples by a molecular biology method, and obtaining the species labels of each nail piece sample; step S3: constructing an "element characteristic-species label" standard data set; step S4: selecting a table prior fitting network as a core machine learning algorithm, training a model by using training set data, and constructing the model; and step S5: identifying whether a to-be-detected sample belongs to a pangolin nail piece according to morphological judgment. The method has a single sample data collection time of less than 60 seconds, does not need to damage the sample, is used for nail piece identification of samples without prior clues, and the overall accuracy of a global ten-species classification model reaches 0.93.

This invention provides a method, apparatus, device, and storage medium for inferring a single-sample ncRNA regulatory network, relating to the field of gene identification technology. The specific implementation includes: acquiring ncRNA and target gene transcriptome data of a matched sample; for each target sample in the matched sample, calculating a first regulatory relationship strength matrix and a second regulatory relationship strength matrix of ncRNA and target gene before and after removing the target sample using a preset statistical algorithm; obtaining the fusion regulatory relationship strength matrix of ncRNA and target gene corresponding to the target sample based on the first and second regulatory relationship strength matrices; and inferring the single-sample ncRNA regulatory network of the target sample based on the fusion regulatory relationship strength matrix of ncRNA and target gene corresponding to the target sample. This invention can reflect the regulatory relationship strength between ncRNA and target gene at the single-sample level by constructing statistical correlation values, thereby inferring the single-sample ncRNA regulatory network.

This invention provides a fully automated immunoassay analyzer, comprising: a base, a reaction disk for carrying multiple immunochromatographic test strips installed inside the base, a sample processing area for processing samples installed on the base, and a moving arm assembly movably installed on the base for driving a pipette arm to perform liquid pipetting, mixing, and spotting operations in the sample processing area. Compared with the prior art, this invention has the following advantages: by setting up the moving arm assembly, in conjunction with the pipette arm and negative pressure nozzle, its automated operation can simultaneously perform multiple sample single detection, single sample multiple detection, or multiple sample multiple detection, with high throughput, making up for the insufficient throughput of existing manual experiments. Automated sample pretreatment and automatic recycling of waste after experiments avoid manual errors and are convenient and pollution-free.

The application discloses a steel rail fatigue crack classification and identification method based on a CNN-BiTCN-CA, a hybrid network architecture combining CNN and BiTCN is constructed, local perception and spatial features of signals are extracted through CNN, and meanwhile, an expanding causal convolution mechanism of BiTCN is used to extract dynamic time sequence dependency of signals in front and backward two time dimensions; a cross attention mechanism is introduced to complementarily fuse spatial features and time sequence features extracted in a cross road; and a steel rail crack evolution state is specifically divided into three physical stages of crack initiation, crack propagation and fracture. Experimental data show that the accuracy of the CNN-BiTCN-CA model in the above multi-classification task can reach 99.83%, and the single sample processing time is less than 0.1 millisecond, which meets the high-precision classification requirement of high-speed railway steel rail fatigue cracks and can provide more detailed and phased evaluation of the fatigue state of the steel rail.

This invention relates to a self-optimizing prediction method for pavement service performance, comprising: acquiring pavement vibration data, extracting vibration features, constructing a time-series feature tensor as model input, using the root mean square of vibration in the time domain as model output, and constructing a feature-decoupled multi-head deep network model; dividing the current pavement vibration data into a teacher stage and at least one incremental stage, calculating the comprehensive drift score of a single vibration feature, and calculating the drift score of a single sample based on this; training the model using the teacher stage data as a basic teacher model, and then using the incremental stage data for incremental learning to obtain a student model, the loss function of incremental learning being constructed in conjunction with the drift score of a single sample; outputting prediction results using the student model; when accumulating new data, using the student model as a new basic teacher model, and using the accumulated new data for incremental learning to achieve continuous model evolution. Compared with existing technologies, this invention has the advantages of achieving long-term and accurate predictions.

The application provides a face clustering method and device. The method comprises: obtaining a target face sample to be processed; sequentially matching and determining the target face sample with formal clustering and candidate clustering; if the matching with the formal clustering or the candidate clustering is successful, merging the target face sample into the formal clustering or the candidate clustering matched successfully; if the number of face samples in the candidate clustering after merging is greater than a first quantity threshold, taking the candidate clustering as the formal clustering. The application realizes efficient, real-time and incremental clustering of face samples in a resource-restricted environment, realizes accurate sample shunting through a hierarchical processing mechanism, supports single-sample incremental updating, avoids high overhead of batch re-clustering, makes the clustering structure evolve dynamically with sample accumulation, and provides stable and reliable technical support for face recognition, intelligent album and other applications.

Embodiments of the present application provide a high-throughput high-temperature creep test device and method, relating to the technical field of material high-temperature performance testing. It aims to improve the problem of low test efficiency caused by the fact that traditional creep test can only obtain single-point temperature data at a time. A basic heat source is provided by a heating assembly, and a heat gathering element is used to feed the sample's spontaneous radiation heat energy back to the preset area directionally, so as to realize local heat enhancement without increasing additional energy consumption, thereby actively building a controllable axial temperature gradient on a single sample. The gradient makes different axial positions of the sample correspond to different test temperatures, providing a physical basis for multi-point data collection. A loading assembly applies a creep load to simulate real working conditions, and a deformation measuring device synchronously obtains deformation data of multiple positions in the gradient range, thereby realizing the acquisition of creep information at multiple temperature points in a single test and significantly improving the test efficiency.

Embodiments of the present application provide a model construction method, a malicious code identification method, a storage medium and a terminal, the method comprising: extracting static features and dynamic features of a plurality of samples, taking the plurality of samples as nodes, taking the static features as edges of related nodes, taking the dynamic features as attributes of related nodes, and establishing a sample relationship graph; labeling each sample according to maliciousness of each sample in the sample relationship graph to obtain a label of each sample, the label comprising malicious and non-malicious; inputting the sample relationship graph and the label of each sample in the sample relationship graph into a graph neural network model for training to obtain a malicious code identification model. Through the malicious code identification model, a single sample can be detected to identify whether the single sample is malicious code, errors caused by identifying malicious code only through static features or only through dynamic features are avoided, and the accuracy of malicious code identification is improved.

This application provides a stress corrosion testing device and method, relating to the field of stress corrosion testing technology. The testing device includes a testing drive mechanism and multiple testing units, which are connected sequentially along the testing direction of the sample. Each testing unit includes a first clamping part, a second clamping part, a connecting structure, and a locking mechanism. The first clamping part is configured to fix one end of the sample; the second clamping part is configured to fix the other end of the sample; the connecting structure connects the first and second clamping parts, restricting their relative movement only along the testing direction of the sample; the locking mechanism is configured to lock the first and second clamping parts when the sample reaches a preset condition. This avoids the problem of stress control drift in other testing units caused by the fracture of a single sample, improving the accuracy and stability of the test results for other testing units.

This invention relates to the field of water quality testing technology, specifically to a method for detecting suspended solids content in water, comprising at least the following steps: S1, collecting a water sample and obtaining a filter membrane carrying suspended solids through suspended solids separation; S2, drying and cooling the filter membrane to constant weight, and then using a beta-ray particulate matter analyzer to obtain the mass concentration Δ of suspended solids trapped on a unit area of ​​the filter membrane. mS S3. The mass concentration of suspended solids in the water sample is calculated using a formula. The detection method used in this application can effectively analyze suspended solids in the range of 0.05-1000 mg / L, and the analytical method is stable and reliable, with high accuracy of analytical results, short detection time for single samples, and high detection efficiency.

This invention provides a method for equipment risk classification based on historical failure frequency, specifically including the following steps: determining multi-factor application scenarios for the equipment; analyzing equipment failures based on these scenarios; calculating the single-sample equipment failure rate; calculating the frequency of equipment failure triggering events based on the failure rates under different differential factor conditions; determining the event consequence level by combining consequence level standards for personnel, property, social impact, and environment; determining the risk level of equipment failure risk events based on the risk matrix, relying on the triggering event frequency and consequence level of equipment failure risk accidents; determining the equipment failure risk level based on the determined risk level of the equipment failure risk events, and identifying the target equipment level. The technical solution of this invention overcomes the problems of existing equipment risk classification methods lacking universality and the variability in equipment failures under different scenarios.

This application discloses a dataset conflict detection method, apparatus, electronic device, and storage medium. The method includes: constructing a classification annotation ontology; defining an attribute set; acquiring the sample data to be detected and the annotation results of the sample data to be detected, and converting the annotation results into sample instance representations; generating a machine-readable constraint set based on the classification annotation ontology, the attribute set, and attribute inheritance rules; performing ontology structure conflict detection on the classification annotation ontology to obtain ontology-level conflict results; performing intra-sample annotation conflict detection on the sample instance representations of a single sample data to be detected to obtain annotation-level conflict results; performing inter-sample annotation conflict detection on multiple versions of sample annotation results and / or equivalent sample annotation results in a historical dataset to obtain dataset-level conflict results; and performing conflict detection on the output application results of the sample data to be detected to obtain application-level conflict results. This application can perform multi-level conflict detection on datasets.

This invention proposes a multi-source precipitation reconstruction method and storage medium based on multi-temporal-scale collaborative back-learning, belonging to the field of hydrological and meteorological data processing and remote sensing big data fusion technology. It acquires daily dynamic meteorological features and static surface features of the study area, constructs single samples based on learning objectives, and integrates them into sample units according to time span. A U-Net encoding / decoding model is constructed, and the sample units are input into the model. After the decoder outputs reconstructed values ​​at 9km, 3km, and 1km resolutions, they are first aggregated along the time dimension, and then the loss functions at each temporal and spatial scale are calculated to construct the total loss. The Adam optimizer is used for iterative optimization until convergence to obtain the trained model. Precipitation reconstruction can be achieved by inputting the feature data of the year to be reconstructed. This method can overcome the spatiotemporal alignment constraints, achieve multi-resolution collaborative reconstruction, and solve the problem of multi-temporal-scale fusion.

The application discloses a kind of based on heat dissipation's ultra-high cycle fatigue stress life curve evaluation method.This method is loaded by stepwise increment fatigue stress amplitude to single sample, until the sample breaks;Real-time monitoring test sample surface temperature rise under different stress amplitude and corresponding cycle loading cycle;By analyzing the heat dissipation of single sample under different stress amplitude, the fatigue life limit under each stress amplitude is calculated, and the complete stress life curve is constructed.The number of fatigue test samples is greatly reduced, the ultra-high cycle fatigue stress life curve of the material is quickly obtained, the experimental cost is effectively reduced, and reliable data support is provided for material design and service safety evaluation.

This utility model provides a sampling auxiliary device suitable for chemical products, including: a top beam, a movable seat, and a lower positioning seat. The front side of the top beam is provided with a set of movable seats for adjusting the left and right positions of the sample suction head. The movable seat includes a main body, a negative pressure pump, and a sample suction head. The right side of the main body has a right-angled trapezoidal cross-section. The inner rear end of the main body is provided with a set of inserts for movable limiting engagement with the upper guide rail. Compared with the prior art, this utility model has the following beneficial effects: The operator places multiple samples to be collected on the upper end of the sample carrier plate. Then, the operator starts the lead screw motor to control the main lead screw to rotate, which drives the movable seat to move left and right, so that the position of the sample suction head corresponds to the longitudinal height position of a single sample, and the position of the sample suction head is adjusted according to the sample. Then, the height position of the sample suction head is adjusted by using an electric cylinder, which makes it easier for the operator to collect the sample.

This invention discloses a method for diagnosing hidden damage in marine bearings using a hybrid attention and Inception graph convolution approach, belonging to the field of marine rotating machinery fault diagnosis technology. It aims to solve the problem of low accuracy in existing bearing diagnosis methods for identifying hidden damage. This invention employs a Min-Max normalization algorithm to process vibration signals, converting one-dimensional vibration signals into two-dimensional images through Gram angle difference fields, Markov transfer fields, and continuous wavelet transforms. A Hatt-GCN-Inception model is constructed, embedding a differential attention module within the model to enhance the sensitivity of damage features. The model's end fuses GCN to capture feature topological correlations. A dual optimizer combined with gradient accumulation and FP16 mixed-precision training techniques are used to achieve efficient model training. Pre-training is conducted using the CWRU dataset, and transfer validation is performed on the MFPT variable load dataset containing four types of noise. This invention offers high diagnostic accuracy and fast detection speed, with a single sample detection time of only 0.005 seconds. The accuracy under variable load and noisy conditions still reaches 99.95%, making it suitable for real-time and accurate fault diagnosis in marine bearings.

The application discloses a kind of metabolomics data analysis method and system, method includes: obtaining the metabolomics data of biological sample under multiple conditions, adopt the strategy of fusion data distribution characteristics to carry out missing value filling;The response intensity distribution difference value of each metabolite feature is calculated by sliding window local fluctuation analysis, and candidate differential metabolite feature is screened;The response value of each feature under different conditions is sorted by size and fitted as feature response curve, and clustering is carried out by calculating curve similarity to obtain feature cluster;Further based on the overall morphology and local synchronism fusion degree of measure between features, construct weighted network and carry out community division, from each subcommunity High coordination intensity feature pair is screened in community, and potential biomarker combination is formed.The application realizes the whole process analysis from data reconstruction, difference screening to collaborative marker identification, improves the accuracy, robustness and biological explainability of single sample metabolomics data mining.

The application provides a multi-source precipitation reconstruction method based on multi-time-space scale cooperative reverse learning and a storage medium, belongs to the technical field of hydrological meteorological data processing and remote sensing big data fusion, and aims at obtaining daily dynamic meteorological characteristics and static surface characteristics of a research area, combining a learning target to construct a single sample, and integrating the sample unit according to a time span; a U-Net encoding and decoding model is constructed, the sample unit is input into the model, after 9km, 3km and 1km resolution reconstruction values are output by the decoder, the time dimension is aggregated first, and then each time-space scale loss function is calculated to construct a total loss; an Adam optimizer is used for iterative optimization until convergence, and a training model is obtained. Feature data of a year to be reconstructed is input, and precipitation reconstruction can be realized. The method can break through the time-space alignment constraint, realize multi-resolution cooperative reconstruction, and solve the multi-time scale fusion problem.

The application discloses a self-adaptive method for SNN single-sample test based on heuristic verification and risk gating. The method is aimed at the problems that single-sample updating in unlabeled test stream is easily misled by noise, and internal activity of the spiking neural network is easy to go to the extreme and cause performance decline. Before each sample enters, the model parameters, buffer and random state are snapped, and the calculation group, verification group and review group are generated to enhance the view. First, the baseline consistency and risk indicators are established by the verification group, and when the abnormality occurs, the robust configuration is switched and restarted. When the baseline is stable, only the update direction of the updateable parameter subset is calculated once, the small step heuristic update is executed, and the gating screening is performed. Through the rear, the final step length that meets the yield and safety conditions is screened from the candidate step length set, and the independent acceptance view is retested. After outputting the recognition result, the snapshot is unconditionally restored to block the cross-sample accumulation, which is used for online robust calibration in deployment inference.

The application discloses a single sample part segmentation method based on adaptive feature selection, which realizes part segmentation by establishing the corresponding relationship between the context reference sample and the test image by using the feature representation extracted from DINOv2 and SD. The application contains a new adaptive channel selection module ACS to fuse the features of DINOv2 and SD. ACS can enhance the distinguishability of the feature representation without destroying the original generalization performance of DINOv2 and SD. The application does not need to train or fine-tune the model, has the characteristics of strong adaptability and generalization ability and high data efficiency, improves the segmentation performance, and promotes the development of related applications.

The present disclosure provides a sample processing method, a model training method, a text processing method and device, relates to the technical field of computers, and particularly relates to the technical fields of artificial intelligence, machine learning, large models, natural language processing and the like. The specific implementation scheme comprises the following steps: obtaining a training sample; obtaining an instability index of the training sample, so as to represent the fluctuation size of the semantic representation consistency of the training sample by a to-be-trained model under N different pre-training nodes; wherein N is greater than or equal to 2 and is an integer; obtaining a single-sample loss corresponding to the training sample, so as to represent the loss value of the semantic representation result of the training sample based on the to-be-trained model relative to the semantic true value of the training sample; and based on the instability index and the single-sample loss, obtaining an application indication for the training sample, so as to adjust the contribution degree of the training sample when training the to-be-trained model. The present disclosure can accurately locate the training sample with high learning value.

The application discloses a rotating speed measurement method and device, equipment and medium, and relates to the technical field of rotating speed measurement. According to a preset sampling frequency, the rotating speed pulse signal transmitted by the rotating speed sensor is AD sampled to obtain a rotating speed digital signal; the rotating speed digital signal is sequentially divided into a preset number of samples; the single-sample attributes corresponding to each sample are acquired respectively; and the current rotating speed value is determined according to the sample length of each sample and the single-sample attribute corresponding to each sample. It can be known that, by AD sampling the rotating speed pulse signal transmitted by the rotating speed sensor, dividing the obtained rotating speed digital signal into samples, and calculating the current rotating speed value by using the single-sample attribute of each sample, the original rotating speed pulse frequency data is avoided to be used for rotating speed calculation, so that the calculation amount is greatly reduced, and the memory and processor resources are saved.