102 results about "Statistical index" patented technology

The invention relates to the technical field of exception prediction, in particular to a storage cabinet exception trend prediction system based on time series data analysis, which comprises a state monitoring module, an interval sensing module, a path reconstruction module, a symptom activation module and an evolution prediction module. According to the method, the state vectors including the temperature, the voltage, the current and the door lock state are constructed and combined with the timestamp information to form the time sequence data sequence, and the dynamic expression mode of state change is established; a jump characteristic is analyzed by using a ratio of a time interval to a state change amplitude, a short-time disturbance path and a trend evolution path are distinguished by combining a jump rate statistical index, and an evolution activation signal is identified based on trend maintenance and non-fallback characteristics. On the basis, a neural network structure with long-time dependent learning ability is introduced to capture an aperiodic thermal anomaly trend in a state sequence, and the accuracy and timeliness of anomaly recognition are improved through multi-dimensional parameter cooperative processing and path construction logic.

The invention relates to the technical field of industrial Internet of Things data processing, in particular to a dual-channel anomaly detection method based on local entropy and isolated forest, and the method comprises the steps: collecting multi-dimensional time sequence data of equipment operation in real time; extracting local data of the multi-dimensional time series data in each sliding window by using a variable step size sliding window technology, calculating four statistical indexes according to the local data to construct a multi-dimensional entropy model, and obtaining a local entropy score according to the multi-dimensional entropy model; utilizing an isolated forest algorithm to calculate an isolated forest abnormal score of the multi-dimensional time series data; and constructing a joint scoring model, dynamically adjusting the weight of the local entropy score and the weight of the isolated forest abnormal score to obtain a joint score, and judging the abnormal level of the equipment according to the joint score and a preset abnormal threshold value. Therefore, the problems of insufficient real-time performance, insufficient feature utilization, low decision collaboration efficiency and the like in an industrial scene are solved.

The invention discloses a perspective analysis method and system based on multi-element space superposition, and belongs to the technical field of geographic information processing.The perspective analysis method comprises the steps that analysis range data input by a user is obtained, and the analysis range data is generated through map drawing or external vector file import; selecting at least two analysis object layers; performing spatial alignment, coordinate system conversion and attribute field standardization processing on the analysis range data and the analysis object layer; a parallel spatial indexing technology is adopted, all analysis object layers and analysis range data are superposed at a time, and a fine-divided result library is generated; storing a space unit ID and an original layer attribute field in the fine crushing result library; calculating a target field value according to the area proportion of the space unit based on a preset adjustment rule, and outputting the target field value to the space data file and the table file; generating a multi-dimensional statistical report according to user-defined dimensions and statistical indexes through a perspective analysis engine on the basis of the fine crushing result library; and outputting the fine space data file and the analysis report according to the filtering condition.

The invention relates to the field of intelligent positioning, and provides an IMU (Inertial Measurement Unit) and UWB (Ultra Wideband) combined positioning method based on extended Kalman filtering, which comprises the following steps: acquiring IMU and UWB data, establishing a combined positioning model, carrying out preliminary distance measurement according to the combined positioning model, carrying out consistency check, and carrying out autonomous detection and gross error identification on UWB measurement. According to the method, measured values which are geometrically inconsistent are suppressed by using information, when detection exceeds a threshold, UWB data at the moment are not adopted, only an IMU prediction result is reserved, and when detection passes, standardized information square of each channel is taken as a statistical index and mapped to a basic probability distribution function in a DS evidence theory; according to the method, the UWB measurement value is subjected to UWB measurement, multi-channel measurement evidences are fused through a DS evidence combination rule, a fusion result is converted into measurement noise covariance for adaptive adjustment, filtering updating of the UWB measurement value in an extended Kalman filtering algorithm is achieved, and the positioning precision and stability of IMU and UWB combined positioning in a complex indoor environment are improved.

Techniques for processing a large dataset to extract information from the dataset, where the processing is performed by multiple systems and where the entire dataset, due to its size, cannot be loaded into the memory of any one of the multiple systems. In certain implementations, the information extracted from the dataset is in the form of a set of one or more statistical metrics computed for the dataset. For example, the dataset may include datapoints related to a machine-learning (ML) model, and a metric value can be computed for the ML model based upon the dataset datapoints. The metric value may, for example, be a metric that measures the performance of the ML model.

The invention relates to the technical field of fault detection, and discloses a big data-based machine room equipment fault diagnosis method and system, and the method comprises the steps: collecting load information data streams generated by all equipment in a machine room in real time, and carrying out the filtering, and obtaining a filtered information sequence; performing grouping processing on the filtered information sequence by using a K-means clustering algorithm to obtain a dynamic change mode group, calculating a key statistical index, and obtaining an abnormal feature set; if the abnormal feature set exceeds a preset variance threshold, classifying the abnormal feature set and determining a fault mode to obtain a fault mode category set; matching the fault mode category set with a preset fault mode database to obtain a fault mode matching result, and generating a corresponding risk assessment report to obtain a risk assessment report; and adjusting the operation parameters of the machine room equipment to obtain the equipment operation parameters. According to the method, equipment faults can be diagnosed in time.

The invention discloses a differential early warning method for a water turbine speed regulating system based on data driving, and aims to solve the problems of poor real-time performance, low accuracy, weak adaptability and lack of differential early warning of a traditional method. The method comprises the following steps: collecting data such as guide vane opening and unit frequency in real time, and preprocessing the data through median filtering and Kalman filtering; identifying working conditions by using a rule engine and fuzzy logic and processing according to the working conditions; extracting time domain and frequency domain characteristic parameters, and constructing statistical indexes; integrating statistics, nonlinearity and twinborn simulation to build a threshold model, and combining a production rule, a fault tree and a Bayesian network to build a rule model; and three types of early warning models are constructed, early warning levels are divided according to an expert scoring method, information is pushed, and a differentiation strategy is formulated. The method improves the early warning real-time performance and accuracy, adapts to complex working conditions, and guarantees the safe operation of the hydropower station.

The invention discloses an unmanned vending machine point location alarm method and system based on Flink + Dores multi-dimensional statistical analysis, and the method comprises the steps: collecting external data streams of equipment transaction, sensors, weather / holidays and the like in real time through Flink, generating minute-level statistical indexes through window calculation, and associating a dimension table in Dores to construct a multi-dimensional real-time data stream; calculating a dynamic threshold value by adopting a space-time dynamic weight algorithm in combination with historical data pre-aggregated by the Dores star model; multi-dimensional alarm conditions are defined through an SQL-like rule engine, composite alarm triggering of cross-time periods, geographic positions and equipment states is supported, and work orders are automatically distributed to events which are abnormal and last for more than 30 minutes in combination with a network quality perception suppression false alarm and hierarchical upgrading mechanism. The method solves the problems that a traditional scheme is insufficient in real-time performance, single in dimension and rigid in static threshold value.

The invention provides a cross-border trade data statistical method and system based on big data, and relates to the technical field of big data analysis and cross-border trade, and the method comprises the steps: obtaining original cross-border trade data based on a distributed crawler cluster and a streaming batch integrated collection engine, and carrying out the preprocessing of the original cross-border trade data to generate target cross-border trade data; constructing a multi-dimensional node relation network corresponding to the target cross-border trade data, and identifying key cross-border trade nodes and abnormal cross-border trade links based on a graph calculation algorithm; constructing a trade index analysis model and generating a cross-border trade statistical index according to the target cross-border trade data; constructing a trade trend prediction model, and predicting a future cross-border trade trend according to the key cross-border trade nodes, the abnormal cross-border trade links and the cross-border trade statistical indexes; and summarizing all the data to generate a cross-border trade statistical result, so that real-time statistics and secure transmission of the cross-border trade data can be realized, and the processing efficiency, dynamic adaptability and security of the cross-border trade data are remarkably improved.

The present disclosure provides a flow detection method, device, electronic device and storage medium, which relate to the fields of computer technology such as the Internet and big data. The specific implementation scheme is: obtaining the business flow data to be detected, the corresponding detection dimensions and statistical indicators; determining multiple dimension values ​​of the detection dimensions and determining multiple index values ​​of the statistical indicators; based on the multiple dimension values ​​and the multiple index values, aggregating the business flow data to obtain the flow distribution data of each dimension value under the multiple index values; determining the correlation between the flow distribution data of each dimension value under the multiple index values, and determining the target business flow data in the business flow data based on the correlation. In this way, it is achieved that the target business flow data with abnormalities can be detected from the business flow data to be detected, and it has strong versatility, low complexity and high detection efficiency.

The invention discloses a wetting time missing data filling model and a fine characterization method. Environmental factor data of a test station is collected, and a meteorological station environmental factor data set is collected; multi-source data granularity alignment is completed; performing feature analysis on data of the test station and the weather station based on statistical indexes, and evaluating temperature and humidity differences of the test station and the weather station; constructing an LSTM-based environmental factor mapping model and model parameters with optimal curing performance; on the basis of the optimal mapping model, missing data is filled, a complete data set is formed, and by comparing and analyzing characteristics before and after data filling, it is ensured that the data rule is unified; and the wetting time definition is given for different metals and different environments, the features of different time ranges such as annual wetting time and monthly wetting time are generated based on the complete data set, the features of different mechanism meanings are generated, fine representation of the wetting time is completed, and a basis is provided for subsequent metal environment adaptability evaluation.

Techniques for processing a large dataset to extract information from the dataset, where the processing is performed by multiple systems and where the entire dataset, due to its size, cannot be loaded into the memory of any one of the multiple systems. In certain implementations, the information extracted from the dataset is in the form of a set of one or more statistical metrics computed for the dataset. For example, the dataset may include datapoints related to a machine-learning (ML) model, and a metric value can be computed for the ML model based upon the dataset datapoints. The metric value may, for example, be a metric that measures the performance of the ML model.

The application discloses an automatic verification and evaluation method and system for simulation effect of an air quality model, and comprises the following steps: determining meteorological stations and air quality stations in a simulation area range based on the simulation area range, and obtaining corresponding formatted monitoring data; selecting the nearest grid point in a grid file corresponding to the monitoring stations based on the longitude and latitude of each station, batch extracting simulation values in the corresponding grid, and obtaining formatted simulation data; matching and merging the monitoring data and the simulation data, and obtaining a corresponding merged file; calculating the merged file, and obtaining statistical index results of different parameters of each station; and performing statistical analysis, time series analysis and spatial distribution analysis on the statistical index results. The application can realize rapid extraction, analysis and comparison of simulation results, improve the verification efficiency, reduce the need for human interference and subjective judgment, provide reliable quantitative analysis, and improve the objectivity and repeatability of the verification.

System and method for automatic selection of training region selection for deep learning model is disclosed. An identification module 110 identifies equipment shutdowns and excludes buffer region data. A segmentation module 120 estimates operational regions and divides data into predefined operating regions using K-means clustering. A feature analysis module 130 determines top contributing tags and identifies low and high values within operational regions. A domain comparison module 140 compares contributing tags with domain knowledge, excluding regions associated with mapped failures. A data cleaning module 150 ensures data integrity by recalculating statistical measures and eliminating outliers. A filtering module 160 refines data and consolidates regions to reduce training ranges. This integrated approach enhances equipment monitoring and predictive analytics, enabling proactive maintenance strategies and optimizing operational efficiency.

The present invention discloses a data analysis method and device, relating to the field of computer technology. A specific implementation of the method includes: utilizing model configuration information to process raw user data to generate model data for at least one user model, the model data including user identification and statistical indicator values; obtaining stratification rule information, the stratification rule information including stratification identification and filtering rules for statistical indicators; and determining stratification data for at least one user stratum based on the model data and the stratification rule information. This implementation can reduce the time and labor costs required for user operations analysis.

The invention discloses a satellite film law enforcement statistical analysis method based on a data center station, and particularly relates to the field of data processing and statistical analysis. According to the method, firstly, a distributed database cluster is built, space and time sequence data processing capacity is integrated, and unified access and standardized processing of multi-source data are achieved; a distributed computing engine is adopted to process the remote sensing image in parallel, a deep learning model is combined to extract suspected illegal pattern spots, and an accurate boundary is output through multi-stage optimization; secondly, sorting statistical indexes based on mass data, constructing 9 types of analysis models, packaging the analysis models into API interfaces, and supporting real-time generation of annual and trend results; and finally, through multi-dimensional display of a three-dimensional map, a trend chart and a report, providing support for decision making in combination with space-time analysis and an intelligent report engine.

The invention provides a steady state identification method, system and device in a roll package production process, and a medium. The method comprises the following steps: obtaining production process data of a roll package; wherein the production process data comprises production statistical indexes, equipment parameters, process parameters and equipment failure shutdown data; preprocessing the production process data to obtain multivariate time series data; based on the multivariate time sequence data, constructing and using a multivariate time sequence change point detection model, and identifying change points in the multivariate time sequence data; and based on the positions and the number of the change points, identifying the systematic steady state of the rolling and packaging production process, and outputting a systematic steady state identification result. According to the method, the positions and the number of the change points are determined in a self-adaptive manner by using the multivariate production process data of the roll and package, the limitation by priori knowledge is avoided, and the systematic steady state and the non-steady state of the roll and package production process can be accurately identified.

The present invention discloses a method and apparatus for elastic resource allocation based on dynamic perception of operator scopes. The method includes: constructing a directed acyclic graph of a data stream processing application, where the nodes in the directed acyclic graph represent operators for processing data streams, and the edges represent data streams; evaluating the scope of an upstream operator according to the static screening rate index and the dynamic screening rate statistical index of the operator, and adaptively partitioning the directed acyclic graph. Predicting the load of each partition in a future period of time; for each operator in each of the partitions, generating a corresponding number of operator instances based on the load to perform real-time processing on the received data and meet the service quality requirements. The present invention plans the parallelism of operators in units of partitions, thereby ensuring end-to-end processing latency and improving resource utilization rate.

The invention belongs to the technical field of aerobic composting, and provides an artificial intelligence-based aerobic composting intelligent control method, which comprises the following steps of: acquiring environmental data and gas generation conditions in a composting process in real time, uploading the acquired data of a sensor to a monitoring platform, and fusing corresponding gas weights to obtain microbial community activity conditions; preprocessing the collected data, filling missing items, removing abnormal values, arranging the collected data according to a time sequence, selecting representative features from the collected data through correlation analysis, and establishing a dynamic model; selecting a regression decision tree as a machine learning model, segmenting the collected data into subsets of different categories according to the features, calculating the validity of the features in classification, inputting the regression decision tree to obtain a prediction result, and sorting an instruction according to the prediction result; and counting environmental parameters and microbial activity data after instruction execution, calculating statistical indexes, arranging the statistical indexes into a feedback data set to calculate error conditions, and iterating a machine learning model according to errors.

The application discloses a multi-channel track detection data adaptive mileage correction and evaluation method, relates to the track detection field, and comprises the following steps: S1, acquiring reference data X and to-be-repaired data Y of a track; S2, extracting special points in the reference data and the to-be-repaired data as base points; S3, adopting digital curve matching of a local section to obtain optimal correction points; S4, correcting mileage error of to-be-repaired data between adjacent base points according to the optimal correction points; and S5, comprehensively evaluating correction results according to the precision of mileage deviation correction of the to-be-repaired data and the change of waveforms before and after correction. According to the track reference data X, the to-be-repaired data Y is accurately corrected in terms of mileage error, higher-precision mileage information is obtained, original data features of track detection data such as track orientation and height, track gauge, level, twist, curvature of left and right tracks are reserved, and the quality of mileage correction is evaluated by using scientific and faithful statistical indexes, so that the real geometric state of the track can be more accurately mastered.

The invention relates to the technical field of data processing, and particularly provides a statistical method, system and device for a multi-modal evaluation sample and a storage medium, and the method comprises the steps: obtaining evaluation data generated after a multi-modal large model completes reasoning, and the evaluation data comprises a plurality of sample records; using the data source, the prompt word and the sample identifier to form a compound key, using the compound key as a unique identifier of a sample, and aggregating the evaluation value fields with the same compound key into the same statistical group to form a sample-level statistical unit; performing statistical calculation on the plurality of evaluation value fields in each sample-level statistical unit, including calculating a mean value and a standard deviation of the plurality of evaluation value fields, and calculating an optimal boundary index and a worst boundary index of the plurality of evaluation value fields based on replacement sampling to obtain a group of statistical indexes; and distributing and outputting the statistical indexes based on a preset rule. According to the method, sample-level accurate aggregation is realized through the compound key, and statistical deviation caused by the same cue word is avoided.

The invention provides an LDL-C concentration estimation method based on an interpretability enhancement machine, and belongs to the technical field of medical detection and artificial intelligence. The LDL-C concentration estimation method comprises the following steps: selecting a subject with a complete blood fat detection record from a preset database to obtain initial data; performing data cleaning on the initial data, and longitudinally merging the blood fat measurement data under the two visits to obtain a data set; performing regression modeling on the EBM based on the training set and the test set to obtain an EBM model; calculating a contribution value of each input ratio variable by constructing a square view of a lookup table based on an EBM model, adding all the contribution values, and calculating a final LDL-C predicted value through a connection function g; and performing comprehensive evaluation and optimization on the EBM model by adopting a plurality of statistical indexes, and integrating the EBM model into various medical systems. Through high-quality data, an optimized data processing strategy, an accurate EBM prediction model, an interpretable prediction result and efficient system integration, accurate estimation and wide application of LDL-C are realized, and the method is superior to a traditional formula calculation method.

The invention provides a multi-source observation data fusion method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining original L0 data of satellite observation, carrying out the positioning and calibration preprocessing of the original L0 data, so as to generate L1 data, and carrying out the inversion processing according to the generated L1 data, so as to obtain L2 data; performing space-time matching on the L2 data subjected to quality control through a preset space-time interpolation method so as to correct mismatching of different data sources in time and space; and performing weighted average calculation on the data after time-space matching according to a weight distribution algorithm to obtain fused data, verifying the fused data through verification data, and performing data evaluation on the fused data by calculating a statistical index between the fused data and the verification data. According to the method, the problem of space-time mismatching of the multi-source observation data can be effectively solved, so that the fused data can more accurately reflect the real state of the atmosphere, and the availability and value of the data are improved.

The invention discloses an abnormal monitoring data identification and restoration method based on an improved box type method, and the method comprises the steps: S1, collecting a monitoring data segment with a fixed length, building a data label through a box type graph in combination with the analysis features of a scatter diagram, and obtaining the features of the box type graph; step S2, calculating four statistical parameters of each segment of data according to the box diagram characteristics and the maximum value, the minimum value, the median, the lower quartile and the upper quartile of the display data; s3, calculating a relative difference rate by taking a statistical index of fault-free data in the same structure monitoring time sequence as a standard; s4, identifying fault data segments according to the relative difference rate of the data segments, and classifying the fault data segments; s5, adopting different repair strategies according to different fault data segments; and S6, finally, verifying the repair effect of repairing the fault data segment by adopting the repair strategy. According to the invention, accurate identification and classification of various types of abnormal data are realized.

A method for extracting the power consumption of an individual piece of equipment within a set of equipment (E1, E2, E3, …, En), comprising transforming a stream of measurements of overall power consumption into a time series, and dividing this series (S1) into a set of sequences. Then, for each sequence, the following steps are performed: generation (S4) of a positional encoding matrix (PEM); extraction (S5) of features from the sequence to form a feature matrix (FM); projection (S6) of statistical metrics into a statistics vector (SM); and concatenation (S7) of these different elements (PEM, FM, SM) to form an input matrix (IM) provided to a transformation neural network designed to infer (S8) a time sequence corresponding to the individual piece of equipment. Figure for the abstract: Fig. 1

The invention provides a two-stage fan equipment quality abnormity tracing method based on hierarchical information characterization, and relates to the technical field of fan operation and maintenance quality management and control, and the method comprises the steps: obtaining SCADA monitoring data, operation logs and operation and maintenance records of fan equipment, carrying out data preprocessing through overlapping sampling, and constructing a time sequence database; calculating the maximum information coefficient, related to the fault, of each feature in the time sequence database, and screening key features to form a feature subset; constructing a double-end Transform model, taking the feature subset as the input of the model, introducing a classification mark, and outputting a feature prediction value and fault category probability distribution; the difference between the fault sample and the normal sample is calculated based on the mahalanobis distance, the tracing key features are screened according to the contribution degree of the difference, and multi-dimensional time domain statistical indexes are extracted from the tracing key features to construct a high-dimensional feature matrix; and inputting the high-dimensional feature matrix into a tracing model represented by hierarchical information, judging an abnormal reason through a feature matrix threshold value, and outputting a tracing result.

The application discloses a content distribution network-based operation and maintenance method and system. In the method, a pre-warning request sent by a management end is received; target data corresponding to a pre-warning type indicated by the pre-warning request is acquired from a central storage cluster according to the pre-warning type; the central storage cluster comprises a service log index and a state index data index; target service logs are stored under the service log index, and target state index data are stored under the state index data index; the target service logs and the target state index data are obtained through edge processing; a data analysis node is called to analyze and count the target data to obtain statistical indexes; the statistical indexes are compared with preset pre-warning rules to obtain comparison results; and an alarm prompt is sent when the comparison results represent that the statistical indexes exceed a threshold. In this way, the target service logs and the target state index data obtained through edge processing are collected in real time, the query range is reduced, and the efficiency of troubleshooting and the operation and maintenance efficiency of the content distribution network are improved.

The application provides a method and system for classifying spatial heterogeneity extreme precipitation events, and the method comprises the following steps: obtaining original precipitation data in a preset time scale and spatial range, calculating a standardized precipitation anomaly index representing the abnormal degree of precipitation based on the original precipitation data; for the standardized precipitation anomaly index spatial field corresponding to each time point, calculating a statistical index for representing the overall spatial heterogeneity degree of the spatial field; based on the time series of the statistical index, screening out spatial heterogeneity extreme events according to a preset high quantile threshold; taking the spatial distribution of the standardized precipitation anomaly index corresponding to the spatial heterogeneity extreme events as an input sample, and performing unsupervised training based on a network model with a preset topological structure to cluster and generate typical spatial distribution modes of the spatial heterogeneity extreme events. Through the application, the extreme events with significant spatial differences can be identified, and the classification results are objective and stable, which provides a scientific basis for refined risk assessment.

The invention relates to a data processing method specially suitable for the purpose of business prediction, in particular to an electric charge pre-collection balance prediction method based on a combined model. The prediction method comprises the following steps: acquiring historical electric charge pre-collection balance time sequence data, meteorological monitoring data, economic statistical index data and user classification data of a power user; clustering and grouping are carried out based on the power utilization characteristic difference of each province through a province clustering module, and a differentiated modeling strategy is established; dividing the historical data into training data and test data through a sliding window according to a time sequence; an independent SARIMA model and an independent LSTM (Long Short Term Memory) model are respectively constructed for each province; and fusing the predicted value of the combined model and outputting a prediction result. According to the method, the advantages of the SARIMA model and the LSTM model are fused, and the defect of insufficient fitting of sudden fluctuation and complex nonlinear modes is effectively overcome.

The invention is suitable for the technical field of production data analysis, and provides a production line test data analysis method, system and device and a program product, and the method comprises the steps: obtaining a plurality of test result value sets corresponding to a plurality of test step identifiers; for each test step identifier, filling a test result value set corresponding to the test step identifier into a template file preset with a calculation formula for calculating a statistical index representing the characteristics of the production process, and triggering a calculation function of the template file to obtain the statistical index corresponding to the test step identifier; and on the basis of the template file which is filled with the test result value set and completes calculation, a single-step analysis file containing the statistical index is generated for each test step identifier, so that the flexibility of the system and the adaptability to service changes are improved, and the accuracy of a statistical analysis result is ensured.