48 results about "Compositional data" patented technology

Disclosed in the present invention is a deep learning-based power distribution network load prediction method, comprising: acquiring regional load data and renewable energy power generation data to form a data set, and preprocessing the data set to obtain a first data set; using a convolutional neural network to extract a time series feature in the first data set, and converting the time series feature into a data form of a deep learning model by means of an embedding layer to obtain one-dimensional time series data; performing fast Fourier transform on the one-dimensional time series data to obtain a frequency curve, extracting amplitude values on the frequency curve to calculate corresponding periods, and selecting the corresponding periods to slice the one-dimensional time series data and form same into two-dimensional matrixes; using a two-dimensional convolutional network to perform feature extraction, reshaping the two-dimensional matrixes that have undergone feature extraction into one-dimensional arrays, and performing adaptive fusion on the one-dimensional arrays to obtain a first time series feature; and inputting the first time series feature into a fully connected layer for weight calculation and linear transformation to obtain a power distribution network load prediction result, thereby improving the stability and reliability of electric power supply.

The invention relates to the technical field of engineering detection, and discloses a block chain-based engineering detection data tracing method, which comprises the following steps of: presetting a data tracing rule, and determining a triggering condition and a tracing condition type of a data tracing request; collecting engineering detection data in real time, and performing hash operation on the engineering detection data to generate a data hash value; forming a data tuple by the data hash value, the detection timestamp, the detection equipment identifier and the detection personnel identifier; storing the data tuple in a distributed account book of a block chain, and verifying the uplink operation of the data tuple through a smart contract; and when a preset data traceability request triggering condition is met, receiving a data traceability request, and retrieving a matched data tuple in a distributed account book of the block chain according to a traceability condition in the request. According to the invention, data full-process protection is realized through the block chain technology, and integrity, security and efficient traceability are ensured.

The invention discloses a construction waste treatment feeding control method and system, and the method comprises the following steps: carrying out the real-time material characteristic collection of entering construction waste, so as to obtain the material composition data, particle size distribution data and water content data of the construction waste; a construction waste feeding initial characteristic data set is generated based on the material composition data, the particle size distribution data and the water content data and transmitted to the central control unit for comparison, and a material-process matching scheme is generated; the control module is used for sending a control instruction based on a material-process matching scheme, realizing dynamic feeding adjustment parameter control of the construction waste, acquiring operation state data of the processing module in real time, and generating a feeding-operation state data set; and feeding back the feeding-operation state data set to the central control unit for dynamic correction, and adjusting operation parameters of a feeding adjustment execution mechanism to form a corresponding closed-loop feeding control flow. The feeding control efficiency in the construction waste treatment process can be improved.

The invention discloses a Mars soil simulant performance evaluation method based on spectrum-mechanics combined characterization, which comprises the following steps: acquiring surface soil chemical components and physical and mechanical reference data of a target Mars area, taking basalt, magnetite and hematite as raw materials, solving the optimal mass ratio by adopting an optimization algorithm, and preparing a sample. Then, systematic spectrum characterization is carried out on the sample, and components and spectrum characteristic data of the sample are obtained; and meanwhile, carrying out physical and mechanical property test on the sample to obtain mechanical property data. Further, spectrum and component data serve as independent variables, mechanical data serve as dependent variables, an associated data set is constructed, the influence weight is analyzed, and a component-structure-performance quantitative relation is established. And finally, based on all the data and the internal association thereof, calculating the comprehensive similarity between the simulant and the real Mars soil, and generating a comprehensive performance evaluation report. According to the method, multi-dimensional and standardized quantitative evaluation of chemical, mineral and mechanical properties of the Mars soil simulants is realized.

A computer-implemented method for machine learning based rig-site on-demand drilling mud characterization, property prediction and optimization is described. The method includes predicting rheological properties of drilling mud from compositional data based on a first empirical relationship across the compositional data. The method also includes predicting sensor based rheological performance of drilling mud based on a second relationship between the predicted rheological properties of the drilling mud and sensor responses from one or more sensors. Additionally, the method includes providing feedback to the one or more sensors, wherein the feedback comprises the sensor based rheological performance, sensor responses, and a formulation associated with the drilling mud, wherein the feedback enables updating the formulation associated with the drilling mud.

A system is configured for automatic recognition of data store architecture and tracking dynamic changes and evolution in data store architecture. The system is a complementary system, which can be added onto an existing data store system using the existing interfaces or can be integrated with a data store system. The system comprises three main components that are configured to compose an approximation of the data store architecture. The first of these components is adapted to execute an analysis of the architecture of the data store; the second of the components is adapted to collect and compile statistics from said data store; and the third of the components is adapted to compose an approximation of the architecture of said data store.

The application discloses an iterative classification matching data association method suitable for complex environment, and comprises the following steps: performing primary association; the characteristic group of primary association success constitutes a data set Z t + and F t + ; the characteristic group of primary association failure constitutes a data set Z t ‑ and F t ‑ ; solving a least square matching vector Θ according to the data set Z t + and F t + ; updating the data set F t ‑ by the least square matching vector Θ to obtain an updated data set, combining the data set Z t ‑ and F into new input, and performing iteration until a mean square error detection is satisfied. The application optimizes the data association method, improves the consistency of algorithm estimation, and reduces the algorithm calculation complexity, and makes up for the deficiency of ICNN and JCBB algorithms in large-scale underwater environment.

The application relates to a commodity title abstract generation method and device, equipment and medium in the computer technology field, the method comprising: acquiring a title text of a commodity; extracting knowledge entries belonging to product words and belonging to attribute words from the title text; determining information scores of each knowledge entry according to statistical features of the knowledge entries; selecting corresponding product word and attribute word combination texts to construct a first candidate abstract set; calculating the similarity between multiple long texts combined by part word elements in the title text and the title text; selecting long texts with high similarity to construct a second candidate abstract set; inputting data pairs composed of the title text and each candidate abstract in the first candidate abstract set and the second candidate abstract set into a pre-trained text classification model to the convergence, predicting the quality scores corresponding to each candidate abstract, and selecting a candidate abstract with a high quality score as the abstract of the title text. The application can generate high-quality abstracts.

The application discloses a fly ash treatment production line real-time monitoring system and method based on multi-sensor fusion, relates to the technical field of fly ash treatment production line monitoring, and comprises the following steps: based on a fly ash treatment process, stages of a fly ash treatment production line are divided, after the division is completed, input parameters of each stage are obtained, the input parameters are output by an upstream stage, and output parameters of the input parameters are transmitted to a downstream stage; data information of each stage of the fly ash treatment production line is collected through a sensor, and data of each stage is respectively composed into a data set; by comprehensively considering weights, real-time values, target values and upper limits of running deviations of monitored physical parameters in the stage, a safety risk assessment value of the stage under a current running state is calculated; a risk transmission coefficient is calculated, a dynamic adjustment model is established by analyzing a relationship between the calculated risk transmission coefficient and a product failure rate, and fly ash treatment production line real-time monitoring is realized.

This invention discloses a dynamic error control method for automatic drilling and riveting equipment based on an error mechanism model, comprising: acquiring production data and classifying the production data according to the causes of errors to construct a processing dataset; constructing a mathematical model based on the error generation mechanism, inputting the processing dataset into the mathematical model, and obtaining optimal parameters affecting processing quality based on the principle of minimum parameters; labeling the optimal parameters based on processing quality, and combining the labels and optimal parameters into a dataset; constructing an initial model and training it using the dataset to obtain a predictive model for predicting the processing quality of the automatic drilling and riveting equipment. This invention also provides a dynamic error control device for automatic drilling and riveting equipment. The method provided by this invention can solve the problem of poor product processing quality caused by processing errors.

The invention discloses a data screening method and device, a storage medium and a program product. The method comprises the following steps: a data consumption network element sends a first request to a data screening device, the first request indicates to perform data screening on data in a collected data set, and the first request comprises a method for screening the data in the data set; the method for screening the data in the data set comprises the following steps: data types contained in each piece of data forming the data set in the data set; the data screening device sends a first response to the data consumption network element, the first response comprises first data, and the first data is a data set composed of data meeting the data screening method in the data set. The data consumption network element instructs the data screening device to screen the data in the collected data set, and provides a method for screening the data in the collected data set, so that the data screening device screens the data in the data set according to the method, and effective data meeting the requirements of the data consumption network element is provided.

This invention discloses a two-phase flow identification system and method based on critical elevation maps, comprising the following steps: acquiring relative position data of all equipment, pipes, and valves in the pipeline system to be analyzed, from the starting equipment to the final equipment, as well as operating temperature, operating pressure, and medium composition data of each material flow point; drawing a critical elevation map based on the collected data; and analyzing the impact of pressure and temperature changes at each material flow point on the medium phase state using the critical elevation map, combined with the thermodynamic properties and flow characteristics of the medium, to identify locations prone to gas-liquid two-phase flow. This invention, by establishing critical elevation maps, can identify pipelines prone to two-phase flow in advance, preventing gas-liquid two-phase flow in process equipment pipelines, which can lead to long-term pipeline vibration, resulting in pipeline fatigue, support structure failure, weld cracking, and flange leakage. Furthermore, for pump inlet pipelines, it avoids the presence of two-phase flow that could cause pump cavitation and damage to the pump equipment.

The invention relates to a rendering scene division and storage optimization method and system oriented to a mass model, and the method comprises the steps: collecting geometric data of all components in a scene, obtaining vertex information, calculating the bounding box information of the components and primitives, carrying out the statistics of material information and reuse conditions, and obtaining the preprocessing information of the components; initializing a BVH root node based on the component preprocessing information and setting a global bounding box, and carrying out BVH division on the component based on the length of a diagonal line of a space of the bounding box to obtain a scene division result; according to a scene division result, organizing geometric data files according to BVH node numbers, classifying vertexes, indexes, normal lines and uv data, putting the vertexes, the indexes, the normal lines and the uv data into Blocks, and recording geometric body position information in the blocks; a streaming rendering Node structure directory is constructed based on a BVH tree, and component transformation information, material markers and independent rendering block composition data are recorded in nodes.

The application discloses a rib center line extraction method and system based on point cloud, and the method comprises the following steps: step 1: a rib original CT data, an independent rib label and an independent rib center line label are combined to form a data set; step 2: the rib CT data is converted into point cloud data; step 3: the rib point cloud data is down-sampled; step 4: the rib point cloud data is pre-processed; step 5: an independent rib segmentation model based on point cloud is constructed; step 6: the model of step 5 is trained and verified; and step 7: rib CT prediction and rib center line extraction. The application utilizes point cloud to efficiently use spatial convolution, which is beneficial to extracting multi-scale and multi-level local feature information. Meanwhile, the point cloud data operation has the advantage of high efficiency, and can effectively avoid the technical problems of long time consumption and insufficient precision in the previous rib center line extraction method.

The invention discloses a soil microbial community composition data batch processing system, method and equipment, and relates to the technical field of microbial community processing, the system comprises: a flora data acquisition module acquires a flora data file, the flora data file comprises original flora measurement data of each soil sample in a plurality of soil samples, and the original flora measurement data is stored in a database; the sorting rule configuration module generates sorting rule data, the flora data sorting module performs sorting processing on the flora data file based on the sorting rule data, determines a parent class of each subclass in each flora original measurement data in the flora data file, obtains a sorted flora data file, and sends the sorted flora data file to the data processing module; and the flora data statistics module calculates the content of each subclass and the content of each parent class in each piece of flora measurement data in the sorted flora data file to obtain a flora data file after statistics. According to the invention, batch processing of the soil microbial community composition data can be automatically completed, and the processing efficiency is improved.

The application relates to the technical field of data prediction, and discloses a marine foundation bearing capacity prediction method and system based on scour pit fractal reconstruction and deep learning, a terminal and a storage medium.The method comprises the following steps: acquiring measured three-dimensional point cloud data, performing interpolation and reconstruction, and generating a three-dimensional scour pit digital geometric model; based on the three-dimensional scour pit digital geometric model, the foundation limit bearing capacity under different scour pit morphologies is calculated, and a scour pit morphology quantitative descriptor is extracted to form a data pair sample library; the deep learning model is trained by taking the data pair sample library as a training set, and a trained deep learning prediction model is obtained; the newly detected scour pit morphology data is input into the deep learning prediction model after being reconstructed and quantitatively described, and the foundation prediction bearing capacity under the current morphology is obtained. The application realizes end-to-end rapid prediction from detection data to bearing capacity evaluation, and provides an efficient and accurate decision support tool for the safe operation and maintenance of marine structures.

The application discloses a disinfection process risk dynamic early warning and tracing method and system based on a multi-source data chain. The method comprises the following steps: collecting multi-source sensor data of a disinfection process according to a first period to generate standardized data records; for each key parameter, a dynamic safety boundary is dynamically determined according to historical data, and when the average value of the parameter exceeds the boundary, the boundary breakthrough is determined and the risk integral is accumulated; the standardized data records and their statistical characteristics are combined into data blocks according to a second period greater than the first period, and the hash chain is used for storage; when the trigger condition is met, the tracing is triggered, the hash chain integrity is verified by reverse scanning from the latest data block, the risk source is located according to a preset risk propagation rule table, and the risk propagation path is output. The application separates real-time early warning from hash storage, realizes dynamic adaptive boundary, risk accumulation attenuation, anti-jitter triggering and credible tracing, can quickly output the risk propagation chain, and effectively improves the safety and fault troubleshooting efficiency of the disinfection process.

This invention relates to a ground identification method and apparatus, mobile tools, and related products. The method includes: performing ring-direction alignment processing on a single frame of multi-ring point cloud transmitted by a multi-line lidar to generate a multi-ring aligned point cloud; for each ring aligned point cloud, forming data elements from each data point in the ring aligned point cloud and its forward, backward, and left points and storing them in a queue, and sequentially calculating the connectivity relationships of the data points in each data element; generating a connectivity graph of the ring aligned point cloud based on the connectivity relationships of the data points in each data element; determining suspected ground points from the multi-ring aligned point cloud based on the connectivity graph corresponding to the multi-ring aligned point cloud; fitting a ground equation based on the suspected ground points; and determining ground points from the multi-ring aligned point cloud based on the ground equation. This invention reduces the algorithmic complexity of ground identification in point clouds and improves computational efficiency by designing the connectivity graph of each data point in the multi-ring point cloud.

The invention discloses a model training method and a contract prediction method. The model training method comprises the following steps: an obtaining step: obtaining historical contract data in a historical preset time range; a preprocessing step: preprocessing the historical contract data to obtain multiple groups of sample data; in the determination step, each group of sample data is calibrated to obtain calibration values, a data set is formed by multiple groups of sample data and respective calibration values, and the data set comprises a training set; an extraction step: extracting a plurality of feature values from each group of sample data in the training set; in the training step, a logistic regression model is determined, the logistic regression model is trained through the multiple feature values and the calibration values of the multiple sets of sample data, and the trained logistic regression model is used for predicting whether repeated ordering exists in the contract or not. According to the method, stock overstock of the residual material blanks can be reduced.

A method encodes compositional data structures by receiving component sparse distributed representation arrays having a predetermined array length and target sparsity level, applying position-specific permutation transformations to encode ordinal position information, combining position-encoded arrays through bitwise union to generate an intermediate array, and processing the intermediate array through a dual-phase sparsity reduction procedure comprising a coarse additive phase and a fine subtractive phase controlled by a sparsity overshoot threshold to generate a composite encoded array. The dual-phase procedure converges in substantially constant iterations for 4 or more components with final sparsity tightly controlled around the target. A decoding method applies inverse position-specific transformations to generate position-decoded arrays, computes overlap scores with candidate components through bit counting operations, and determines component identities based on threshold comparison. Scalable decoding may use triadic associative memory. Applications include searchable compression, privacy-preserving analytics, and efficient neural network embeddings.

The invention discloses a data enhancement method for a bearing fault signal, and the method comprises the steps: obtaining a fault signal, carrying out the labeling of a real fault signal according to a fault type, and enabling the fault signal and a label to form a data set; constructing an initial model, wherein the initial model comprises a layered GAN generator and a discriminator; training the initial model by using the data set to obtain a data enhancement model for expanding the fault data sample; and inputting fault information in the to-be-enhanced data set into the data enhancement model to obtain a sample signal and a fault type, and adding the sample signal and the fault type into the to-be-enhanced data set for data enhancement. The invention further provides a data enhancement device. The method provided by the invention can comprehensively reflect fault data of time domain fault features and frequency domain fault features in real samples, thereby providing high-quality data support for data-driven fault diagnosis under small samples and unbalanced working conditions.

The application relates to the technical field of perovskite photovoltaic material design, and discloses a perovskite material band gap regulation method and system based on machine learning, which comprises the following steps: S1, constructing a perovskite material composition database to form a multi-component material combination space; S2, performing feature extraction and fusion processing on the material composition data to generate a comprehensive material feature vector; S3, inputting the comprehensive feature vector into a pre-trained band gap prediction model to obtain a band gap prediction value; S4, based on the deviation of the band gap prediction result and a target range, completing material screening through multi-constraint optimization; and S5, performing performance evaluation on candidate materials and executing iterative optimization. The perovskite material band gap regulation method and system based on machine learning improve the band gap prediction accuracy and material screening efficiency of the multi-component system, effectively represent the complex coupling relationship between components, realize the collaborative optimization of structural stability and component feasibility in the target band gap interval, and greatly shorten the new material development cycle.

The invention relates to the technical field of machine learning, in particular to a multi-view machine learning assisted alcohol reforming catalyst screening method which comprises the following steps: S1, acquiring data to construct a first data set which comprises composition of an alcohol reforming catalyst, an alcohol reforming condition and a catalyst performance evaluation result; s2, weighting the composition data of the alcohol reforming catalyst in the first data set by adopting a multi-stage weighting element descriptor to construct a second data set; s3, inputting the second data set and the first data set into a parameter-free multi-view clustering model for training to obtain a catalyst performance prediction model; and S4, inputting the catalyst composition and the alcohol reforming condition which are acquired in real time into the catalyst performance prediction model to obtain a catalyst performance evaluation prediction index. The screening method can predict the performance of the catalyst, so that the screening process of the catalyst is accelerated.

The application relates to a rendering scene division and storage optimization method and system for a large model, which comprises collecting geometric data of all components in a scene, obtaining vertex information and calculating bounding box information of the components and graphics primitives, counting material information and reuse conditions, and obtaining component preprocessing information; based on the component preprocessing information, initializing a BVH root node and setting a global bounding box, and performing BVH division on the components based on the space diagonal length of the bounding box, so as to obtain a scene division result; according to the scene division result, organizing geometric data files according to BVH node numbers, classifying vertex, index, normal and uv data into Block blocks, and recording geometric body position information in the blocks; and based on the BVH tree, constructing a streaming rendering Node structure directory, recording component transformation information, material markers and independent rendering block composition data in the node.

The invention discloses an aggregate source optimization method and system based on aggregate interface performance data modeling. The method comprises the following steps: acquiring three-dimensional surface texture data, chemical composition data and surface adhesion data of multi-source aggregates with the same lithology; the method comprises the following steps: screening core parameters strongly related to interface adhesion strength in three-dimensional surface texture data, chemical composition data and surface adhesion data of the same-lithology multi-source aggregate through a Pearson correlation analysis method, and carrying out dimensionality reduction on the parameters by adopting a principal component analysis method so as to obtain principal component characteristic parameters; and constructing a multiple regression adhesion performance prediction model, obtaining an interface adhesion strength prediction value of the candidate aggregate by using the multiple regression adhesion performance prediction model and the candidate aggregate, and selecting the candidate aggregate source corresponding to the maximum value as an optimal aggregate source. According to the method, the technical problems that aggregate selection depends on experience, the adhesion performance prediction capability is weak, and a cross-scale data fusion and mapping mechanism is lacked in the prior art are solved.

A computer-implemented method for automatically preserving data in a data preservation system. The method comprises receiving, by the data preservation system, first data to be preserved, the first data comprising one or more first constituent data. The method further comprises processing, by the data preservation system, each one of the one or more first constituent data to generate second data representing the received first data to be preserved. The second data comprises structured data adhering to a pre-defined data model and one or more second constituent data representing the one or more first constituent data. The method further comprises storing, by the data preservation system, the second data in the data preservation system for later retrieval.

The application discloses a fuel property prediction method and device, equipment and a storage medium, and the method comprises the following steps: analyzing a to-be-tested fuel sample by means of a two-dimensional gas chromatograph to obtain hydrocarbon composition data of the to-be-tested fuel sample, inducing a plurality of key chemical component variables and constructing a component feature vector; performing standardization processing and principal component analysis on the component feature vector to obtain principal components meeting preset conditions and calculating a principal component score vector; inputting the principal component score vector into a trained nonlinear regression prediction model to obtain a predicted value of a physical and chemical property of the to-be-tested fuel sample; wherein the nonlinear regression prediction model comprises linear terms, square terms and interaction terms of the principal component score vector; wherein the square terms and the interaction terms are used to represent the nonlinear effect of the components on the physical and chemical property and the synergistic effect between the components. Thus, the problem of large fuel sample consumption and long test period can be solved, and the fuel property can be accurately predicted.

The application provides a kind of based on nonparametric density ratio's component data classification method and system, it is related to data processing technical field, method includes: obtaining component data;Through L2 norm normalization, each variable in component data is mapped to non-negative sphere;Through finite reflection group, non-negative spherical data is expanded to entire sphere;Based on kernel density function, estimate the class conditional probability density of spherical data;Using pullback operation, obtain the class conditional probability density of component data;Through log marginal density ratio transformation, obtain the feature enhancement data set of each component type variable;SVM model is used to train feature enhancement data set and corresponding class label;Target feature enhancement data set of target component data is input into the SVM model after training, and output component data classification result.

A compositional visualization system comprises a sensor to collect contextual information, a particle generator to generate a first stream of one or more types of particles, and a detector to receive a second stream of one or more detectable products. The second stream is generated by interaction of the first stream with the environment. The system further comprises computer-executable instructions to cause the system to transform the received second stream into compositional data, and merge the compositional data with the contextual information to generate a merged digital representation. The merged digital representation can be displayed at one or more devices and can also be used directly to drive autonomous robotic systems.

The present application relates to a kind of based on the improved YOLOv7 clean energy field key equipment defect method, comprising: obtaining preprocessed image and composition data set;YOLOv7 model is improved, and the improved YOLOv7 model is obtained;Get trained model, i.e. The target detection model;Substation equipment to be detected image is input into target detection model, and the position of substation equipment to be detected image and defect information are output.The present application effectively suppresses the noise interference in complex background, so that the network can accurately focus on the defect features of small targets, while reducing the computational redundancy, significantly improving the detection accuracy and feature extraction ability of the model;By introducing auxiliary frame to fine constraint on original bounding box, the regression process is optimized, and a dynamic adjustment factor is designed to balance the contribution weight of high-quality and low-quality samples to the loss function, which greatly enhances the recognition ability of small target defects with low resolution, insufficient information and susceptible to noise interference.