222 results about "Interval data" patented technology

The invention discloses a remote printer monitoring and fault prediction system based on the Internet of Things, and relates to the technical field of the Internet of Things and artificial intelligence, and the system comprises a data collection module which is used for collecting equipment state data of a remote printer in real time and carrying out the preprocessing of the equipment state data; the data processing module is used for integrating the preprocessed equipment state data into a data matrix and a label vector; the model construction module is used for outputting a final feature set by using a random forest model and combining a Gumbel-Sigmoid activation function; and the fault prediction module is used for constructing a fault prediction model by using the bidirectional LSTM to output the predicted fault probability of the current time point and dividing a fault interval data pool. By introducing the combination of a random forest algorithm and a Gumbel-Sigmoid activation function, a feature selection process can be automatically optimized, a feature set which is crucial to fault prediction can be accurately screened out, and a bidirectional LSTM model is adopted to capture time sequence information, so that the fault prediction precision and the real-time response capability are remarkably improved.

The invention provides a performance optimization method and system for cross-data-source paging query, and relates to the technical field of data processing, and the method comprises the steps: calculating a data distribution weight and a data source routing table, determining an optimal data access path, and splitting a paging query request into sub-query tasks; dynamically adjusting the number of parallel query threads based on the load state and distributing sub-query tasks; establishing a local index and adaptively adjusting an index structure; dynamically adjusting the connection number of the connection pool and acquiring data in batches; performing feature analysis according to a query result, determining the size of a data fragment, and selecting an optimal merging strategy to merge the data; constructing a cache access mode graph, and aggregating and storing related data based on the access association degree; and optimizing a performance strategy set through reinforcement learning, adjusting a cache updating interval, a data access path and a feature matrix, and finally obtaining a query result. According to the method, the cross-data-source paging query efficiency can be effectively improved, the query delay is reduced, and the overall performance of the system is improved.

A method for detecting an anomaly, which is performed by a computing device including one or more processors according to some embodiments of the present disclosure may include: collecting a CAN message generated in a controller area network (CAN); generating first traffic time interval data and first payload data based on the CAN message; obtaining a first latent vector generated by inputting the first traffic time interval data into a first neural network model; obtaining a second latent vector generated by inputting the first payload data into a second neural network model; generating merged data in which the first latent vector and the second latent vector are merged; and determining whether the CAN message is anomalous by inputting the merged data into a third neural network model.

The invention relates to the technical field of satellite communication, and discloses a residual frequency offset compensation method, device and equipment, a storage medium and a computer program product, and the method comprises the steps: determining a linear interpolation phase step length according to frame header information in a data frame; calculating a phase offset estimation value and a phase change parameter based on the linear interpolation phase step length and interval data between any two pilots in the data frame; and when it is detected that the phase shift estimation value has a phase cross period, compensating the phase shift estimation value according to the phase change parameter to obtain modulation data phase information. Large residual frequency offset estimation and residual phase linear interpolation compensation can be carried out in each frame, the complexity of carrier fine synchronization and phase recovery is reduced, and the system phase recovery time under the condition of low symbol rate is greatly shortened.

The invention discloses a multi-source heterogeneous alarm log anomaly detection method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining multi-source heterogeneous alarm logs in a to-be-detected time period and time interval data between adjacent logs; analyzing the log to generate a log template; inputting the log template into a semantic model to generate a log semantic feature vector; performing linear transformation on the time interval data to generate a log time feature vector; converting each word in the log template into a numerical vector, and generating a log index feature vector; splicing the log semantic feature vector, the time feature vector and the index feature vector to generate a log fusion feature vector; inputting the log fusion feature vector into an anomaly detection model, randomly shielding a plurality of features and predicting a result; and if the proportion of the prediction result of the shielding feature to the actual inconsistency exceeds a preset threshold value, determining that an abnormal alarm exists in the log. By implementing the method, the accuracy of log anomaly detection can be improved.

The invention provides a silent upgrading method for an air conditioner, and relates to the technical field of air conditioner upgrading, the method comprises the following steps: receiving an incremental upgrading package to obtain upgrading resource demand estimation information, and monitoring key environment parameters; judging whether to trigger a refrigeration interval period identification mechanism of the air conditioner or not according to the key environment parameters; after a refrigeration interval period identification mechanism is triggered, multiple refrigeration interval periods are identified to obtain historical interval data, and the prediction duration of the refrigeration interval periods under multiple different temperature difference ranges is predicted according to the historical interval data; and when the air conditioner enters the current refrigeration interval period, the duration of the current refrigeration interval is determined according to the key environment parameters, and the duration of the current refrigeration interval period, the key environment parameters and the upgrading resource demand estimation information are compared to judge whether silent upgrading operation is executed or not in the current refrigeration interval period. According to the method, whether the air conditioner has enough resources to complete silent upgrading or not is judged by utilizing the pre-cooling release time of the refrigeration interval period of the air conditioner, and non-stop silent upgrading of the air conditioner is achieved.

The invention discloses a training method, a recognition method and a system of an unmanned aerial vehicle behavior pattern recognition model, and belongs to the technical field of unmanned aerial vehicle safety. Then, sensing data are extracted according to the flight parameters, a time interval of behavior mode execution is traversed in combination with an adaptive sliding window method, a time window of a key action corresponding to the behavior model is determined, and a key action data set is constructed according to time sequence interval data of the time window; and finally, training the behavior recognition mode by adopting the key action data set to obtain a trained behavior recognition model. According to the method, the crash risk of the unmanned aerial vehicle due to improper behavior mode execution can be remarkably reduced.

The invention discloses a heart tension index dynamic calculation and risk assessment method based on HRV, and the method comprises the steps: carrying out the filtering and denoising of an electrocardiosignal through a Butterworth filter, and obtaining a filtered electrocardiosignal; calculating interval data of adjacent wave crests; processing the interval data according to fast Fourier transform to obtain frequency domain features, and combining the interval data to obtain time domain features; performing normalization processing on the time domain and frequency domain features based on a maximum and minimum normalization method; calculating a cardiac tension index according to the normalized features and a pre-trained multiple linear regression model; matching a pre-constructed reference database according to the individual age, gender and body mass index to obtain a reference cardiac tension index; calculating a cardiovascular risk score in combination with the normalized frequency domain features and a reference cardiac tension index; and repeating the above steps regularly, smoothing through a moving average method based on continuous multiple cardiovascular risk scores, and analyzing the trend to send out cardiovascular risk early warning information.

The invention belongs to the technical field of gas pipelines, and particularly relates to a gas engineering construction quality evaluation system based on big data, which comprises the following steps of: firstly, constructing a defect feature tensor, introducing a space-time grid analysis technology to analyze defect disposal aging interval data and location defect reproduction frequency data, and constructing a defect feature tensor on the basis of a preset quality-efficiency reliability quantification rule; defect features are converted into quantifiable quality-effect scores, then deep fusion of multi-source heterogeneous data is achieved by establishing a space-time incidence matrix of welding defects, welding process parameters and environment monitoring parameters, and on this basis, a fuzzy correlation analysis technology and a preset process compliance quantification rule are adopted to achieve quality-effect evaluation of the welding defects. Complex process parameters and environmental conditions are converted into quantifiable process scores, the process scores and quality-effect scores are integrated to generate a gas pipeline welding quality comprehensive evaluation value of a target contractor, and a quantifiable intelligent decision support system is provided for gas pipeline construction quality management and control of the contractor.

The invention relates to the technical field of urban rail transit operation optimization, and provides an energy consumption peak value optimization method and system for a peak period of an urban rail train group, and the method comprises the steps: obtaining a historical train operation timetable, real-time position data and an energy consumption curve, and generating a training sample set; classifying the operation intervals in the sample set based on the operation interval difference, and determining the peak type of each interval; training a time sequence classification model by utilizing interval data and peak types to obtain an energy consumption feature classification model, and training a multi-objective optimization neural network model to obtain an energy consumption peak value optimization model; and inputting real-time operation data into the classification model to identify a current peak type, and inputting the data into the optimization model according to the type to generate a dynamic timetable adjustment strategy so as to smooth an energy consumption peak value. According to the method, the energy consumption peak value of the urban rail train group can be reduced, the energy utilization efficiency is improved, departure interval fluctuation and delay are reduced, and the operation stability is improved.

The invention relates to the technical field of single-photon detection, in particular to a single-photon detector performance parameter measurement method, device and equipment and a storage medium, and the method comprises the steps: building a statistical histogram of time domain interval data, and building the statistical distribution characteristics of the photon counting rate and dark counting of a single-photon detector; the probability of no photon detection event in the duration of the optical pulse and the ideal probability of no dark pulse in a preset time window are defined based on the statistical distribution characteristics, and then the probability of generating a single photon signal by the periodic optical pulse is calculated based on the obtained parameters; and finally, determining a periodic statistical peak count value based on the statistical histogram and the probability of generating the single photon signal by the periodic light pulse, and calculating an average detection photon number and a dark count rate so as to obtain the dead zone time, so that information such as the average detection photon number, the dark count rate and the dead zone time of the single photon detector is simultaneously represented through one-time data acquisition, and the detection accuracy of the single photon detector is improved. The problems that in a traditional method, the data storage amount is large, and cost is high are solved.

The invention belongs to the field of sequence recommendation, and particularly relates to a contrast learning sequence recommendation method and system based on time interval data enhancement. The method comprises the following steps: acquiring original sequence recommendation data, wherein the original sequence recommendation data is a user behavior sequence of a plurality of users; dividing each user behavior sequence into a plurality of subsequences, and determining the longest uniform subsequence according to the time standard deviation and the sequence length of each subsequence; replacing and inserting the longest uniform subsequence in each user behavior sequence, and replacing the longest uniform subsequence in different user behavior sequences to obtain enhanced sequence recommendation data; wherein the enhanced sequence recommendation data is used for enhancing training of a contrast learning sequence recommendation model. According to the method, the uniformity of the global sequence and the local sequence is considered. By inserting new items between related items, it can be ensured that the time difference of local sequences is reduced, which helps to better capture user interest changes in a short term. And selecting the longest uniform sub-sequence to ensure that the longest uniform sub-sequence has enough similarity with the global sequence. And the inserted items are more similar to surrounding items, so that the consistency and coherence of recommendation can be kept. Therefore, the accuracy of the contrast learning sequence recommendation model is further ensured.

The invention provides a ceramic card production data state control method and system. The method comprises the following steps: obtaining production data in a ceramic card production process; by adopting the data acquisition module with a distributed architecture, sensor self-diagnosis and self-adaptive adjustment functions and dynamic self-adaptive time interval data acquisition, and combining with the preprocessing and standard storage mode of the data processing and storage module, the problems of inaccurate production data recording and transmission delay errors are solved; the data state updating module monitors in real time, updates the state based on a machine learning model, and realizes data state real-time tracking and anomaly discovery by matching with a visual platform of the data state monitoring and early warning module, dynamic threshold early warning and multi-mode early warning; a multi-dimensional analysis result is generated through a data analysis module by applying a professional tool and an algorithm, and an intelligent decision-making module constructs an intelligent decision-making model and provides an effective data state control mechanism, so that reasonable production scheduling and decision-making are performed according to a production data state.

The invention relates to the technical field of image processing, in particular to an automatic legend adjustment method based on BIM forward design, comprising the following steps: acquiring legend data of all legends in a legend container; performing classification based on the legend data of each legend to obtain classification data; calculating an interval value based on the classification data, and solving an interval mean value based on the interval data; setting a positioning point based on the interval mean value; and carrying out movement adjustment on each legend based on the positioning points, and carrying out adaptation degree detection. According to the method, the typesetting of the legend is automatically adjusted, so that the legend can still keep a clear and ordered layout after the user performs data interaction with the chart, the user experience and the readability of the chart are improved, the self-adaptive capability of the chart is improved, the deep data interaction is promoted, and the user experience is improved. Therefore, the problem that typesetting of the legend is disordered due to data interaction between the user and the chart is solved.

The application discloses a kind of intelligent anchor rod system early warning method and threshold determination method, belong to anchor rod intelligent early warning field, including S1: monitoring data preprocessing, S2: data trend regression analysis and S3: early warning method and index;Step S1 includes substep, S11: detection and correction of singular point, S12: obtain equidistant data and S13: data denoising smoothing;Step S2 includes substep S21: regression model establishment and S22: determine regression coefficient;Step S3 includes substep, S31: the determination method of limit stress static threshold of each monitoring point, S32: the determination method of stress change rate static threshold, S33: sharp end mutation model, the determination method of section stability threshold and S34: multi-factor joint early warning method.The application is by monitoring data preprocessing, data trend regression analysis and division early warning method and index. According to this, the overall trend and law of stress change of roadway surrounding rock can be accurately analyzed, and the anchor rod early warning system early warning threshold is accurately determined.

The invention relates to the technical field of construction early warning, and discloses a visual early warning system based on highway construction environment monitoring. Comprising a monitoring node marking module used for marking monitoring nodes in a construction period, an interval data acquisition module used for acquiring comprehensive construction data of a standard construction interval and a diffusion construction interval, an environment diagram construction module used for constructing a node environment diagram, and an early warning prompt judgment module used for judging whether to send out an environment early warning prompt. According to the invention, an environment pollution area directly or indirectly caused by highway construction on a crossing area can be accurately represented, multi-source construction data of the highway can be comprehensively collected, and pollutant generation and diffusion rule information can be accurately analyzed in combination with a combination mode of field monitoring and experimental simulation, so that the method is simple and convenient to operate. The monitoring and early warning depth of the highway construction on the environmental pollution and harm of the crossing area is improved, and the healthy and stable progress of the highway construction progress is ensured.

The invention relates to the technical field of state monitoring of numerical control machine tools, in particular to an embedded monitoring method for cutter wear of a numerical control machine tool. The method comprises the following steps: performing embedded monitoring and synchronous multi-modal data acquisition and preprocessing on a tool in a machining process to form a real-time working condition data frame queue; cutting effective interval data interception is carried out on the real-time working condition data frame queue to obtain an effective cutting data sequence; performing current and cutter feeding rate discretization processing on the effective cutting data sequence to form an energy-effect reference fingerprint; extracting an expected healthy feed rate from the energy-effect reference fingerprint; and performing performance deviation absolute amount calculation according to the expected healthy feed rate, and performing cutting performance attenuation analysis to obtain the cutting performance attenuation degree. According to the method, a complete process from detection to parameter adaptive closed-loop control is realized on a resource-limited embedded system, so that the service life of a cutter is prolonged, the machining quality is guaranteed, and the production efficiency is improved.

The invention discloses a processing method for a terrain model invading a bridge deck range, and belongs to the technical field of bridge engineering BIM modeling, and the method comprises the steps: searching a bridge deck space range based on a BIM model; a plurality of independent invasion intervals are positioned through intersection of a terrain model and a bridge floor, the invasion side direction of each invasion interval is independently judged, and only interval data of the invasion side are reserved; in the reserved invasion side interval, taking the central line of the route as a reference, iteratively generating a terrain model cutting surface according to a step length: determining a space point by vertically extending downwards to set a length and combining with a bridge floor section edge lateral level, iteratively intersecting with the terrain model on the basis of the space point to obtain an intersecting polyline, connecting all adjacent intersecting polylines in the invasion segmentation interval to form a cutting surface; and performing Boolean operation on the terrain model by using a cutting surface to eliminate an intrusive part. According to the method, rapid, accurate and automatic processing of the terrain model within the invasion bridge floor range is realized, and the accuracy and design efficiency of the bridge BIM model are remarkably improved.

The invention discloses a remote secure connection method and system for an industrial internet of things, and belongs to the technical field of remote secure connection, and the method comprises the steps: obtaining the polling interval data of target industrial equipment, analyzing a polling mode characterization factor, judging a polling interval execution scheme of the remote connection of the target industrial equipment, and carrying out the remote connection of the target industrial equipment; synchronously analyzing a polling mode characterization factor deviation value, executing dynamic polling interval adjustment of remote connection of the target industrial equipment, analyzing dynamic polling interval adjustment effect information, analyzing a network state characterization factor of remote connection, and executing dynamic protocol timeout adjustment of remote connection of the target industrial equipment. And analyzing the dynamic protocol overtime adjustment effect information to obtain a protocol overtime execution scheme of remote connection of the target industrial equipment. According to the invention, the dynamic polling interval is dynamically changed, the inherent rule is broken, the protocol timeout time is dynamically adjusted, the security and real-time performance of remote secure connection are improved, and the compatibility and standardization of the system protocol are enhanced.

A method for identifying a potential supercooled water zone through a satellite-borne multi-parameter ensemble includes: acquiring reference data on potential supercooled water zone identification, collecting parameter data, and pre-processing the reference data and the parameter data; grouping based on distribution features of the parameter data to obtain interval data, and performing feature selection of the parameter data based on the reference data and the interval data according to a cumulative frequency crossover method to obtain an influencing parameter; voting based on the influencing parameter to identify a potential supercooled water zone, and calculating a probability of supercooled water to obtain identification data; establishing a potential supercooled water zone identification model based on the identification data, and optimizing the potential supercooled water zone identification model using the reference data; and inputting data into the potential supercooled water zone identification model to obtain identification results.

The embodiment provides a wind turbine yaw-to-wind error calculation method and system based on goodness of fit, a terminal and a medium, and belongs to the technical field of wind power generation. The embodiment first preprocesses the original data to exclude the interference of variable pitch, shutdown, power limitation and the like on power generation; then performs interval data set division based on wind speed intervals and yaw angles; then uses a hypothesis testing method to check the power distribution of each wind speed interval sub-set, selects a corresponding fitting method through the actual distribution of each sub-set, and determines the yaw wind direction of the maximum power in the interval; and finally comprehensively considers the calculated yaw error angles in each wind speed interval. If the yaw error angles obtained in each wind speed interval have high consistency, and the goodness of the fitting result meets the preset requirement, it is considered that there is a significant yaw error angle in the wind farm, and the yaw error angle is calculated. The present application improves the accuracy of yaw error calculation.

The invention relates to the technical field of plastic particle cutting control, and discloses a high-precision plastic particle cutting equipment control method and system capable of automatically rectifying deviation, and the method comprises the following steps: obtaining a to-be-cut material image, a projection line image and the cutting amount of the current cutting stage; calculating the distance between adjacent projection lines according to the projection line image to obtain spacing data; performing image preprocessing on the to-be-cut material image to obtain an initial sectional area of the to-be-cut material and a corresponding sectional image coordinate; calculating the optimal section coordinate of each to-be-cut material section according to the initial section area and the section image coordinate, and adjusting the initial section area according to the optimal section coordinate to obtain the section area of the to-be-cut material; performing least square fitting on the sectional area of the to-be-cut material and the coordinates of the sectional image to obtain an optimal sectional area; the optimal sectional area is substituted into a preset cutting speed formula to calculate cutting speed data, and the optimal cutting speed is obtained; and the cutting device is controlled to conduct cutting work according to the interval data, the optimal cutting speed and the optimal cutting amount.

The invention discloses an ultra-harmonic detection method based on Nuttall window compression sampling and reconstruction estimation, which comprises the following steps of: firstly, on the basis of a 32-equal-time-interval data window detection method, constructing a window sparse measurement matrix at a sampling end by combining a Gaussian random measurement matrix and a four-item 3-order Nuttall window function to perform windowing compression sampling on an ultra-harmonic signal; and then, reconstructing and estimating a sparse vector s (n) by using an orthogonal matching pursuit reconstruction algorithm (OMP) on the data subjected to windowing compressed sampling at a reconstruction end, and finally correcting a peak frequency point of the estimated sparse vector s (n) by using a three-spectral-line interpolation technology to realize accurate detection of each parameter of the ultra-harmonic signal. According to the invention, on the premise of ensuring certain detection precision, the signal data volume required by super-harmonic detection is greatly reduced; and a basis and conditions are provided for analyzing the generation mechanism and propagation characteristics of the ultraharmonics and subsequent treatment.

A method for performing time constrained queries for event sequence visualization may include receiving an event sequence dataset including event sequences. Each event sequence includes at least two sequentially occurring events. The method further includes generating a forward index having event pattern sets that include pattern occurrence data for each event sequence positioned according to a fixed positional order. The method includes generating a time constraint search structure, such as a B−tree, B+tree, or table, that includes keys and pointer values. The method also includes inserting, into the search structure, a sequence identifier corresponding to each event sequence based on the fixed positional order and time interval data for the at least two sequentially occurring events of each event sequence. The method may also include searching the search structure to identify key values satisfying a received time interval constraint, and outputting the sequence identifiers corresponding to the identified key values.

The invention discloses a traction transformer ratio differential protection performance testing method, system and device, and relates to the technical field of traction transformer performance testing, and the method comprises the steps: obtaining the specification information of a detection object, and collecting the real-time characteristic data information of the high and low voltage sides of the detection object; according to the invention, the action state interval data of the differential action and the braking action in the detected object are determined according to the model specification of the transformer of the detected object and the characteristic expression data of the high-voltage side and the low-voltage side; and then real-time data of the detected object is collected and matched with action state interval data of differential action and braking action in the detected object, the real-time feedback action condition of the detected object is judged, whether the detected object has a fault or not is determined, and the state of the detected object is judged. The defect that in the prior art, a traction transformer ratio differential protection test of a high-speed rail maintenance pipe is not high in reliability is effectively overcome.

The invention relates to the technical field of power systems, and discloses a cable intermediate joint insulation state evaluation method, system, equipment and medium, and the method comprises the steps: collecting a partial discharge signal of a cable intermediate joint, and extracting a plurality of partial discharge characteristic quantities which better represent the insulation state of the cable intermediate joint; sample data of each partial discharge characteristic quantity of the cable intermediate joint in an insulation state under a plurality of different continuous acquisition intervals is represented by constructing a ternary interval data set, so that the uncertainty of information is better processed, the objectivity and credibility of an evaluation result are improved, and by determining the characteristic weight of each partial discharge characteristic quantity, the reliability of the evaluation result is improved. And based on a ternary interval number grey target decision, determining the relative target degree of the cable intermediate joint by using the ternary interval data set and the feature weight of each partial discharge feature quantity, and determining the insulation state of the cable intermediate joint through the relative target degree, thereby improving the evaluation accuracy of the insulation state of the cable intermediate joint.

The present application relates to the technical field of data processing technology and discloses a method for analyzing the reliability of avionics products using multi-source interval data, comprising: summarizing and classifying the multi-source interval data of a specified avionics product to obtain multiple groups of interval data, substituting the multiple groups into an exponential distribution model, and constructing corresponding quality functions according to a unified identification framework; obtaining conflict coefficients and credibility coefficients between the quality functions corresponding to different groups of the interval data, and determining fusion weight coefficients of the respective quality functions; obtaining a weighted quality function according to the fusion weight coefficients, fusing the weighted quality functions based on the D-S evidence theory to obtain a fusion quality function, and transforming the fusion quality function according to the Pignistic probability transformation method; obtaining the reliability of the avionics product at any time according to the transformation result, fitting and obtaining a reliability curve with respect to time changes to perform reliability analysis.

The invention discloses a wide area measurement system abnormal data identification method, system and related device, and the method comprises the steps: carrying out the single electrical interval data abnormity identification of a single PMU at a plant station end, and obtaining a first abnormal data identification result; according to the first abnormal data identification result, when a certain measurement data in a single electrical interval is abnormal, comparing and identifying data anomalies of a plurality of PMUs at a plant-station end to obtain a second abnormal data identification result; if the same PMU is accessed to a plurality of electrical intervals on the same bus, performing wide-area level correlation comparison analysis on wide-area measurement system data at a master station end, and performing wide-area measurement system abnormal data identification through a power grid topology connection relation to obtain a third abnormal data identification result; and integrating the obtained abnormal data identification result, and obtaining an abnormal data identification result of the wide-area measurement system through collaborative analysis of a main station and a plant station. According to the method, the identification accuracy of the WAMS abnormal data is improved, and the work of diagnostic analysis of a master station is reduced.

The invention discloses an intelligent anchor rod system early warning method and a threshold determination method, and belongs to the field of anchor rod intelligent early warning, and the method comprises the steps: S1, monitoring data preprocessing, S2, data trend regression analysis, and S3, early warning method and index. The step S1 comprises the sub-steps of S11, singular point detection and correction, S12, equal interval data acquisition and S13, data noise reduction and smoothing; the step S2 comprises the substeps of S21, establishing a regression model and S22, determining a regression coefficient; the step S3 comprises the sub-steps of S31, determining a limit stress static threshold value of each monitoring point, S32, determining a stress change rate static threshold value, S33, determining a tip sudden change model and a section stability threshold value, and S34, performing a multi-factor combined early warning method. According to the method, the monitoring data is preprocessed, the data trend is subjected to regression analysis, and early warning methods and indexes are divided. Therefore, the overall trend and rule of the stress change of the surrounding rock of the roadway can be accurately analyzed, and the early warning threshold value of the anchor rod early warning system is accurately determined.

A model generation unit (17) generates, on the basis of change interval data obtained from log data stored in a log data storage unit (12), a change interval model obtained by modeling change interval data representing an interval between a start signal that causes operation of a device and a completion signal that appears as a result of the operation of the device. An abnormality degree calculation unit (18) calculates the degree of abnormality of the operation of the device from an actual measurement value, which is an interval between the start signal and completion signal obtained from the device, and a predicted value, which is an interval predicted using the change interval model generated by the model generation unit (17).