60 results about "Local outlier factor" patented technology

The invention discloses a power distribution switch terminal communication state analysis method and system, and belongs to the technical field of power distribution network communication. The method comprises the following steps: acquiring dual-source parameters of a terminal and a communication link, and preprocessing to form a time sequence data sequence; an improved local outlier factor algorithm is combined with spatio-temporal clustering to identify abnormal communication events; extracting an analysis sequence containing event context based on the dynamic window; respectively performing communication state evaluation on a terminal side and a link side through a multi-task deep learning model, and outputting state classification and severity score; and when the evaluation results on the two sides conflict, intelligent decision making is performed by adopting a multi-level arbitration mechanism based on an evidence theory and integrating factors such as confidence difference, a historical trend and topological association, so that accurate responsibility determination of a fault source is realized. The method solves the problems that the traditional threshold alarm is high in false alarm rate and the fault source cannot be positioned, and remarkably improves the accuracy and operation and maintenance efficiency of communication state diagnosis.

The invention relates to the technical field of concrete structure monitoring, in particular to a concrete structure service entity durability key index monitoring and analyzing method. Comprising the following steps: performing multi-source data acquisition on a concrete structure through a sensor network, including temperature, humidity, resistivity, steel bar electrochemical parameters, cracks, porosity and the like; acquired data is subjected to synchronous preprocessing and multivariable empirical mode decomposition, and key evolution factors such as temperature and humidity collaboration and electrochemical activity are extracted. Based on the factors, a space-time graph neural network model is constructed, and multi-dimensional features and evolution laws among nodes of the structure are deeply analyzed. And then, a local outlier factor algorithm is adopted to detect structural abnormal nodes and distribution thereof, and finally, historical and environmental data are combined to perform adaptive classification and classification abnormity, and a comprehensive monitoring analysis report is output. According to the invention, the intelligent, precise and dynamic level of concrete structure key index monitoring is comprehensively improved, and the safety guarantee and operation and maintenance efficiency of the concrete structure are improved.

The invention discloses a power grid operation abnormal area identification method and system, and the method comprises the steps: calculating new energy output data based on meteorological data, constructing a unit combination model through combining a generator set parameter and a power grid structure parameter, carrying out the whole-year time sequence production simulation considering a power flow network frame, and generating power grid operation time sequence data; performing feature extraction on the power grid operation time sequence data by using a convolutional neural network to obtain a deep feature vector, extracting a time-space correlation feature in the deep feature vector, outputting an abnormal probability, and screening an area of which the abnormal probability exceeds a preset threshold value as a candidate abnormal area; for candidate abnormal regions, constructing an electrical association neighborhood for each candidate region, calculating local reachable density and a local outlier factor value based on a deep feature vector, and performing abnormal degree quantitative sorting; and performing electrical logic verification, equipment state association verification and time trend verification on the sorted abnormal regions, and performing risk grade division based on a local outlier factor value and an influence range.

The present application relates to the technical field of mine machinery control, and particularly relates to a mine hoist intelligent control method and system. The method comprises: obtaining vibration acceleration and depth and speed data of a hoisting container; constructing a speed adaptive penalty factor negatively correlated with real-time speed, and performing variational modal decomposition on the vibration signal; constructing a dynamic reference potential energy positively correlated with real-time speed square and depth logarithm; combining envelope amplitude relative over-scale and inverse speed weighted difference mean to calculate an impact risk index; and performing fault determination through local outlier factor analysis. The present application realizes relative quantity monitoring of the whole stroke by introducing kinetic energy theorem and cantilever beam stiffness characteristics, avoids false alarms in deep wells, and significantly improves fault sensitivity of low-speed inspection by using inverse speed weighting.

The present application relates to the technical field of electric energy metering, and particularly relates to an insulation performance evaluation system and method for insulation voltage transformers, comprising an insulation monitoring and sensing optimization module, a central storage unit, a data processing and abnormality analysis module, and a device performance evaluation module; the insulation monitoring and sensing optimization module builds an intelligent sensor network and a data acquisition and processing terminal, and realizes real-time monitoring of the insulation voltage transformer; the data processing and abnormality analysis module extracts key features, and uses a local outlier factor algorithm to perform real-time analysis on the key features, and can timely discover abnormal changes in insulation performance; the device performance evaluation module calculates a comprehensive gain value index to perform detailed performance evaluation on the device with problems, which is helpful for operation and maintenance personnel to understand the device state and formulate reasonable maintenance, replacement or optimization strategies. The present application is used to solve the technical problems of incomplete insulation performance monitoring and insufficient evaluation for insulation voltage transformers.

The present invention provides a machine-learning method for detecting data change points in a dynamic social network. The method comprises: capturing a sequence of graph snapshots of a graph representing the dynamic social network; extracting data features from each graph snapshot; applying a sliding-window statistical analysis on the extracted data features of the sequence of graph snapshots to detect a first set of change points; applying a local outlier factor algorithm on the extracted data features of the sequence of graph snapshots to detect a second set of change points; combining the first and second sets of change points to form an output set of change points for the sequence of graph snapshots. The provided method can effectively handle the high-dimensional complex network data, particularly in dynamic environments, to provide more accurate and comprehensive information.

The invention discloses a bridge stay cable monitoring method and system, and the method comprises the steps: determining the size of a sliding window according to the sampling frequency and sampling time of stay cable state data, and determining a target data flow according to the sliding window; performing abnormal data point detection on the target data stream based on a local outlier factor algorithm and a preset weight factor; and preferentially transmitting the abnormal data points to the data center, adjusting the size of the sliding window according to the number of the abnormal data points, re-determining the target data stream according to the adjusted sliding window, and continuing to detect the abnormal data points. Before the stay cable state data is transmitted to the data center, the abnormal data point is preferentially sent to the data center in time, the data transmission efficiency can be effectively improved, and the response delay of the data center is reduced. According to the number of the abnormal data points and the size of the sliding window, the problem of density imbalance of the abnormal data points in the sliding window is avoided to a certain extent, and the accuracy of abnormal data point detection can be improved.

The invention discloses an energy storage battery inconsistency screening method based on an anomaly detection model, and the method comprises the steps: collecting the SOC time series data of a plurality of battery clusters, constructing the multi-dimensional feature data of the SOC between the battery clusters, quantifying the SOC coordination relation between the plurality of battery clusters from different angles, and more comprehensively reflecting the real state of the consistency between the clusters. The anomaly detection model fusing the isolated forest algorithm and the local anomaly factor algorithm is constructed, and the anomaly detection model is trained and tested by using the standardized multi-dimensional feature data, so that the accuracy of a model output result is improved. And inputting standardized multi-dimensional feature data obtained by processing a to-be-detected battery cluster into the anomaly detection model to obtain an anomaly score of the to-be-detected battery cluster, performing contrast screening on the anomaly score of the to-be-detected battery cluster according to a set SOC consistency state grade between the battery clusters, and generating a visual report for a screening result. Therefore, related personnel can visually know the detection condition of the to-be-detected battery cluster.

The application provides a power system secondary circuit fault diagnosis method and system, and belongs to the technical field of power equipment state monitoring and intelligent diagnosis. The method comprises the following steps: acquiring first characteristic data of a target monitoring point in a secondary circuit of a power distribution device and storing historical characteristic data; performing outlier analysis based on the first characteristic data of the target monitoring point; comparing the first characteristic data of the target monitoring point with a reference set, first performing global anomaly detection by using a robust Z-score method, and then performing local anomaly detection on the first monitoring point by using a local outlier factor algorithm; and based on the historical characteristic data of a second monitoring point, performing trend analysis by using a state space model method to identify an abnormal change trend, so as to realize fault diagnosis of the secondary circuit. The application can quickly identify significant global individual anomalies, effectively detect early weak anomalies, can identify a slow development of a persistent degradation trend, and improves the accuracy and reliability of the secondary circuit fault diagnosis.

A kind of observation data anomaly identification method based on SIP-LOF, comprising the following steps: first, the observation data to be detected is extracted by PLR-SIP method All data points, then two data points of starting and ending are used as segmentation point, the point of maximum y direction distance between two segmentation points is calculated, then the point of maximum y direction distance between adjacent two segmentation points is calculated in turn, obtain multiple segmentation points, and sub-sequence is formed between adjacent two segmentation points;Extract multiple characteristic values of sub-sequence, construct multidimensional space according to all characteristic values, then all characteristic values are mapped in multidimensional space, and the distance between each characteristic value is calculated;After arranging distance value according to size, the k distance of each sample and the first k neighbors are found, then the reachable distance of each sample to k neighbors, local reachable density are calculated, then local outlier factor LOF is calculated, and LOF value is obtained.The present application can quickly and automatically locate abnormal data in mass data.

The application discloses a code pre-configuration method fusing calculation power constraint and outlier analysis, comprising the following steps: performing calculation power signature marking on a function function set according to a hardware fingerprint at a management end to obtain a function set recording maximum instantaneous calculation power occupation; performing combination explosion pruning under a target scene calculation power threshold to generate a candidate sequence; inputting the candidate sequence into a space-time joint distribution model based on historical normal output to obtain a predicted output; calculating a local outlier factor of the predicted output and an expected output to form an outlier factor set; obtaining a reordered sequence according to a weighted ranking of the outlier factor and calculation power occupation; and finally selecting a sequence with the minimum outlier factor and lower calculation power occupation than a threshold to realize code pre-configuration. The application can automatically select a function function number sequence most suitable for a target scene before deployment, reduces deployment risks, and improves the predictability and reliability of the whole system.

The invention relates to the technical field of mining machinery control, in particular to an intelligent control method and system for a mine hoist. The method comprises the following steps: acquiring vibration acceleration, depth and speed data of a hoisting container; constructing a speed adaptive penalty factor in negative correlation with the real-time speed, and performing variational mode decomposition on the vibration signal; constructing dynamic reference potential energy in positive correlation with the real-time speed square and the depth logarithm; calculating an impact risk index in combination with a difference mean value of the envelope amplitude relative standard exceeding degree and inverse speed weighting; and carrying out fault judgment through local outlier factor analysis. By introducing the kinetic energy theorem and the rigidity characteristic of the cantilever beam, relative quantity monitoring of the whole stroke is achieved, false alarm of the deep well is avoided, and the fault sensitivity of low-speed inspection is remarkably improved through inverse speed weighting.

The invention discloses an LOF-based machine tool machining abnormality diagnosis method, system and equipment and a medium, and relates to the technical field of machine tool machining monitoring. The method comprises the steps that position information, program instruction information and vibration signals of all shafts of a machine tool are collected; constructing and forming a mapping unit sequence; constructing a local outlier factor algorithm to construct a density distribution model; calculating a local outlier factor of the to-be-measured data point; and selecting a to-be-tested data point based on the local outlier factor, and obtaining corresponding position information and program instruction information. The method does not need to manually set a fixed threshold value, does not need to depend on a fault sample, achieves the time-space fusion of multi-source data and the precise recognition of the machining abnormality, can precisely position the machining position and program instruction corresponding to the abnormality, solves the problems that an existing diagnosis method is poor in adaptability, is difficult to land, and is difficult to trace the abnormality, guarantees the precision machining quality, and improves the precision machining efficiency. And the machining requirements of the high-end manufacturing field are met.

The invention discloses an evolutionary multitask-based sonar data noise robust anomaly detection method, which comprises the following steps of: 1, preprocessing acquired sonar data, and constructing a noise-containing data set in a label overturning manner; 2, screening out noise-free samples by using a local outlier factor algorithm; 3, the number of iterations is set, and populations are initialized on the noise-containing data and the noise-free data respectively; 4, taking a true positive rate TPR and a false positive rate FPR as objective functions; 5, performing iterative optimization on the two populations by adopting an evolutionary multi-task algorithm; and 6, taking an optimal noise individual in the noise sonar population as a sonar noise robust classification model to realize classification of target sonar data. According to the method, the abnormal sample can be identified in a noise scene, and a reliable classification result is obtained, so that the robustness of sonar data anomaly detection can be improved.

This invention discloses a non-contact laser-ultrasound method and system for detecting internal defects in high-entropy alloys. The method first pre-treats the surface of the high-entropy alloy sample, then adaptively adjusts the laser pulse parameters according to the thermal conductivity characteristics of the high-entropy alloy to excite internal ultrasonic waves via thermoelastic effects. After non-contact signal reception using a laser interferometer, signal pre-processing is performed using a wavelet packet transform and principal component analysis fusion algorithm. Defect features are then extracted using a local outlier factor and support vector machine fusion algorithm to distinguish between defects and lattice distortion interference signals. Defect parameters are analyzed using an improved ultrasonic propagation model, combining mechanical parameters and lattice distortion coefficients. Finally, cross-validation and error correction ensure detection accuracy. The system comprises seven modules, including sample pre-processing, laser excitation, and ultrasonic reception, with a main control module coordinating the collaborative work of each module. This invention achieves non-contact detection, improving detection adaptability and accuracy, and enabling automated batch online detection.

This invention provides a data calculation system for electronic current transformer operation, relating to the field of data processing technology. The system includes: a system for scanning the device-dependent adjacency matrix using a sliding time window; calculating the local reachability density under dynamic nearest neighbor topological distance using a local outlier factor algorithm; and extracting a multidimensional anomaly feature vector set when the total harmonic distortion rate gradient enters the statistical hypothesis test rejection region or the winding temperature rise rate exceeds the material's thermal aging critical threshold. The system then uses this multidimensional anomaly feature vector set to generate a risk quantification index set through a pre-trained fault propagation probability model. Based on the risk quantification index set, a multi-objective optimization function is established, and the NSGA-II genetic algorithm is used to perform non-dominated sorting evolution on the disposal instruction chromosomes. A Pareto front solution set is used to screen dynamically optimized instruction sequences that match the real-time load fluctuation coefficient, generating a structured disposal instruction set. This invention improves the reliability of power system operation.

The application discloses a machine tool machining anomaly diagnosis method, system, device and medium based on LOF, and relates to the technical field of machine tool machining monitoring. The method comprises the following steps: collecting position information, program instruction information and vibration signals of each axis of a machine tool; constructing a mapping unit sequence; constructing a density distribution model based on a local outlier factor algorithm; calculating a local outlier factor of a to-be-tested data point; selecting the to-be-tested data point based on the local outlier factor, and obtaining corresponding position information and program instruction information. The application does not need to manually set a fixed threshold value, does not need to rely on fault samples, realizes spatiotemporal fusion of multi-source data and accurate identification of machining anomalies, can accurately locate the machining position and program instruction corresponding to the anomaly, solves the problems that existing diagnosis methods have poor adaptability, are difficult to land, and anomalies are difficult to trace, guarantees precision machining quality, and adapts to machining requirements in the high-end manufacturing field.

The invention provides an intelligent identification method and system for abnormal auditing data of a hospital. The intelligent identification method comprises the steps of obtaining a multi-dimensional data set of the hospital, calculating a local outlier factor LOF of each data record, distributing a differential privacy budget and generating a privacy data set; constructing a weighted K nearest neighbor graph based on the privacy data set and the LOF value, and performing community division on the graph; calculating weighted intermediary centrality, community connection strength and average path length of each node, fusing the weighted intermediary centrality, the community connection strength and the average path length into a structural anomaly score, and screening data records of which the scores are higher than a first threshold value as first-level candidate anomaly data; for each piece of first-level candidate abnormal data, mapping the data structure abnormal score into a time window, extracting a time sequence track, and calculating an average time sequence track of the community as a prototype track; outputting a time sequence similarity distance; constructing a confirmation threshold value, wherein the threshold value is adjusted based on community structure characteristics, data differential privacy budget and a structure anomaly score; and when the time sequence similarity distance is greater than the threshold value, identifying the data as abnormal data.

The application provides a kind of tunnel boring machine's measured data outlier detection and correction method, belongs to the field of outlier data detection, and the measured data of tunnel boring machine under different working conditions is detected by the method of sliding window, and the outlier point is corrected and filled up.The method first divides the original time series into multiple sub-time series by sliding window, and extracts the confidence interval radius of sub-time series slope by fast calculation and identifies abnormal sub-time series, then further determines the outliers using local outlier factor algorithm, and finally uses regression technique to reasonably fill the outliers removed.The application can effectively identify outliers in tunnel boring machine measured data, and reasonably correct and fill the outliers, ensuring the engineering usability of tunnel boring machine measured data and providing good conditions for further data analysis.

This application provides a method, system, electronic device, and storage medium for identifying abnormal batteries in an energy storage system. First, the parameter data of the battery cells in the energy storage system are denoised. Based on the denoised data, the dynamic warping time distance of each battery cell is determined and standardized. The standardized dynamic warping time distance is processed using a local outlier factor algorithm to obtain a set of local outliers. Battery cells corresponding to local outliers that meet a preset threshold condition are identified as abnormal battery cells. This method, combining the dynamic warping time distance algorithm and the local outlier factor algorithm, has a strong filtering capability for both sudden and gradual outliers in battery cells, achieving zero false alarms and zero false negatives, thus improving the robustness and reliability of the detection.

The present application belongs to the technical field of fault prediction and health management of charging facilities, and particularly relates to a sensor-based abnormal fault prediction method and system for a charging port, comprising: acquiring multi-source sensor data during the charging process and performing alignment preprocessing; establishing a thermal-electric coupling physical model, inverting to obtain a contact resistance representing the physical wear level of the contact interface after stripping external thermal interference; extracting multi-dimensional features based on the contact resistance in the initial running stage of the vehicle, and collecting the initial non-damage period feature set as a personalized normal mode reference system; extracting the features of the contact resistance at the current time to form a current three-dimensional feature vector, calculating the local outlier factor in the initial non-damage period feature set to quantify the degree of deviation; and finally, automatically warning and automatically generating a maintenance work order according to the local outlier factor combined with the preset grading threshold. The present application overcomes the hysteresis and false alarm stubbornness of single threshold determination, and realizes early abnormal trend precise warning before numerical out-of-range.

The application discloses a time series network traffic integrated anomaly point and anomaly cluster detection method and device, relates to the field of network security, and comprises the following steps: marking target network traffic data to obtain a network traffic data set; an initial network traffic anomaly detection model is constructed and trained; isolated forest algorithm and local outlier factor algorithm of the trained anomaly detection model are used to detect global anomaly points and local anomaly points respectively; density clustering algorithm and K-means clustering algorithm are used to detect non-spherical cluster anomaly clusters and spherical cluster anomaly clusters respectively; the confidence weight of the initial model parameters is determined according to the anomaly points and the anomaly clusters, and the initial model parameters are adjusted to obtain target model parameters, so that a target network traffic anomaly detection model is constructed based on the target model parameters to detect the network traffic to be detected. Through the detection of network traffic anomaly points and anomaly clusters by various detection algorithms, the robustness and generalization of time series network traffic anomaly detection in complex network attack scenarios are improved.

The invention discloses a code pre-configuration method integrating computing power constraint and outlier analysis, which comprises the following steps of: executing computing power signature labeling on a performance function set according to a hardware fingerprint at a management end to obtain a function set for recording maximum instantaneous computing power occupation; performing combined explosion pruning under a target scene computing power threshold to generate a candidate sequence; inputting the model into a spatio-temporal joint distribution model constructed based on historical normal output to obtain prediction output; calculating local outlier factors of prediction output and expected output to form an outlier factor set; performing weighted reordering according to the outlier factor and the computing power occupancy to obtain a reordering sequence; and finally, selecting a sequence with the minimum outlier factor and the calculation power occupation lower than a threshold value to realize code pre-configuration. According to the method, the performance function number sequence most suitable for the target scene can be automatically selected before deployment, the deployment risk is reduced, and the overall predictability and reliability of the system are improved.

The present application relates to the field of traction machine monitoring technology, and more particularly to a gearless traction machine fault early warning method based on multi-source data fusion. The method comprises: obtaining a braking current sequence, a driving voltage sequence and a micro switch state sequence and extracting a measured action time; using a dynamic time warping algorithm to calculate the waveform distortion index between the braking current sequence and the preset standard fingerprint; calculating the voltage correction factor, the resistance correction factor, and correcting the measured action time to obtain the normalized action time; inputting the waveform distortion index and the normalized action time into a local outlier factor algorithm to calculate the equipment health degree score, and combining the logical consistency to calculate the comprehensive risk early warning index. That is, the scheme of the present application considers mechanical form and electrical physical characteristics when evaluating the state of the brake by fusing multi-source data, thereby achieving more accurate fault identification.

The invention provides a machine learning method for detecting data change points in a dynamic social network. The method includes capturing a series of graphic snapshots representing a graph of the dynamic social network; extracting data features from each graphic snapshot; applying a sliding window statistical analysis on the extracted data features of the series of graphic snapshots to detect a first set of change points; applying a local outlier factor algorithm to the extracted data features of the series of graphic snapshots to detect a second set of change points; the first and second sets of change points are combined to form a set of output change points for the series of graphics snapshots. The method provided by the invention can effectively process high-dimensional complex network data, especially in a dynamic environment, so as to provide more accurate and more comprehensive information.

The present application relates to the technical field of electric digital data processing, in particular to an airborne navigation data cleaning method based on multi-source sensing data, comprising: collecting parameter data of airborne navigation at multiple historical time nodes and constructing as parameter sequence, and constructing a multi-scale window with any parameter data in the parameter sequence as the center; using an improved local outlier factor algorithm to calculate the abnormal score of the any parameter data under any scale window, and determining the parameter data with the mean value of the abnormal scores of the any parameter data under all scale windows greater than a set threshold as abnormal data, and cleaning the abnormal data. The present application solves the problem of misjudgment or omission in the data cleaning process.

The application discloses a kind of based on digital twinning robot operation behavior simulation and optimization method, comprising: collection multi-source data, constructs the basic data set of simulation model;Improved multiscale digital twinning model is constructed, and the initial simulation result is generated multiscale twin state data flow, and output;Data fusion and structured arrangement are carried out, and continuous twin state data flow is obtained;Based on local outlier factor clustering algorithm, behavior anomaly detection and cause and effect tracing analysis are carried out;According to abnormal result adjustment control parameter, and again simulation generates optimization simulation result;Re-execute task, collect real-time data and compare with simulation result;Comprehensive comparison result and deviation information, update control parameter and behavior rule.The application realizes high-precision simulation and adaptive optimization control of robot operation behavior by improved multiscale digital twinning modeling and local outlier factor clustering analysis.

The application discloses a lithium battery fault diagnosis method based on a multi-scale flow improved local outlier factor, and belongs to the field of battery management technology fault diagnosis. The method first collects battery operation data from a big data platform, extracts voltage, voltage change rate, voltage kurtosis and voltage residual after cleaning to form a four-dimensional feature matrix; then a multi-scale sliding window flow processing framework is constructed, and an improved local outlier factor abnormal score is calculated in parallel on multiple time scales; based on the super threshold value distribution fitting, a global threshold value is determined, and a dynamic joint threshold value is formed in combination with a time guardrail mechanism to realize online fault judgment; a dynamic time warping (DTW) distance is introduced to measure the voltage trajectory difference between the fault monomer and the reference monomer, and the fault degree is quantitatively evaluated. The application can realize real-time detection with high precision under complex working conditions and multi-fault concurrent scenes, reduce fault exposure time, improve identification accuracy, and provide reliable support for power battery system safety warning and intelligent operation and maintenance.

In order to facilitate artificial intelligence-based anomaly detection in electronic records, systems and methods include establishing a computer-implemented clustering process based on record attributes (e.g., using a k-means algorithm) such that records are grouped into clusters and pairs of records within each cluster are subsequently analyzed; generating feature vectors by normalizing and concatenating selected attributes and filtering out vectors exhibiting low variance based on first predetermined parameters; applying an ensemble of at least three anomaly detection models (e.g., Local Outlier Factor, DBSCAN and another model) to cast votes on whether each pair is anomalous and flagging pairs that satisfy a second predetermined consensus threshold; and performing actions in response to flagged anomalies, including generating alerts or storing results with associated anomaly scores, whereby the system enhances detection accuracy and scalability for applications such as financial market analysis and other domains requiring robust data evaluation.

The application discloses a drug design method based on two-stage evolutionary multi-task optimization, and is characterized in that the method comprises the following steps: determining a target function of each sub-task, entering an evolution early stage, performing in-task population evolution on a target task to obtain a child population, generating a migration solution better than the target task by using an affine change strategy, monitoring whether the population reaches a stage division point, if not, reselecting the target task and returning to the evolution early stage, if yes, performing an evolution late stage, transferring the solution by using a local outlier factor detection model strategy, merging a parent population, the child population, a mapping solution and the transferred solution, and selecting an optimal solution according to a fitness function, determining whether evolution is completed according to a function evaluation time, if yes, outputting the optimal solution, and if not, returning to the evolution early stage. The method solves the problem that the drug design cannot be completely solved due to uncertain problem properties under the condition of a black box problem.