WO2010041355A1 - 異常検知方法及びシステム - Google Patents
異常検知方法及びシステム Download PDFInfo
- Publication number
- WO2010041355A1 WO2010041355A1 PCT/JP2009/002391 JP2009002391W WO2010041355A1 WO 2010041355 A1 WO2010041355 A1 WO 2010041355A1 JP 2009002391 W JP2009002391 W JP 2009002391W WO 2010041355 A1 WO2010041355 A1 WO 2010041355A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- abnormality
- modeling
- sensors
- abnormality detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
- G05B23/0254—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2137—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on criteria of topology preservation, e.g. multidimensional scaling or self-organising maps
- G06F18/21375—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on criteria of topology preservation, e.g. multidimensional scaling or self-organising maps involving differential geometry, e.g. embedding of pattern manifold
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Definitions
- the present invention relates to a method for detecting an abnormality in a plant or equipment at an early stage.
- Electric power companies use waste heat from gas turbines to supply hot water for district heating and supply high-pressure steam and low-pressure steam to factories.
- Petrochemical companies operate gas turbines and other power sources.
- it is extremely important to detect the abnormality at an early stage because damage to society can be minimized.
- SmartSignal in the United States provides services for abnormality detection mainly for engines.
- the past data is stored as a database (DB)
- the similarity between the observation data and the past learning data is calculated by an original method
- the estimated value is calculated by linear combination of the data with high similarity
- the past learning data stored in the database must comprehensively include various states. If observation data that is not in the learning data is observed, all of these are treated as not included in the learning data, and are judged as outliers. It will drop significantly. For this reason, it is indispensable for the user to store all data in all past states as a DB.
- the object of the present invention is to solve the above-mentioned problems, allow incompleteness of learning data and mixing of abnormalities, reduce user load, and detect abnormalities with high sensitivity at an early stage. It is to provide a method and system for detecting an abnormality.
- the present invention (1) pays attention to temporal data behavior and divides the trajectory into clusters over time.
- the present invention even if the learning data is not complete, it is possible to allow an abnormality to be mixed, and it is possible to detect an abnormality early and highly accurately in a facility such as a plant. In other words, unlike SmartSignal, there is no need for normal part learning and comprehensive data collection according to multiple states.
- FIG. 1 shows an embodiment of the present invention
- FIG. 2 shows an example of a target signal.
- the target signal is a plurality of time series signals as shown in FIG.
- four types of signals of series 1, 2, 3, 4 are shown.
- the number of signals is not limited to four, and may be several hundred to several thousand.
- Each signal corresponds to the output from a plurality of sensors provided in the target plant or facility. For example, the temperature of cylinders, oil, cooling water, etc., oil and cooling water pressure, shaft rotation speed, room temperature, operation time, etc. are observed from various sensors several times a day or at regular intervals. Is done. It may be a control signal (input) for controlling something as well as representing an output or state. In some cases, ON / OFF control is performed, and in other cases, control is performed so that the value is constant. Some of these data are highly correlated with each other. All these signals can be of interest. The presence or absence of abnormality is judged by looking at these data. Here, it is treated as a multidimensional time series signal.
- a multidimensional time series signal is acquired by the multidimensional signal acquisition unit 1.
- the missing value correction / deletion unit 2 corrects / deletes the missing value.
- defect correction is generally performed by replacing previous and subsequent data or moving average. Deletion eliminates abnormalities as data, such as when many data are simultaneously reset to zero.
- correction and deletion of missing values are performed based on the state of the equipment and the knowledge of the engineer stored in the DB of state data / knowledge 3.
- the invalid signal is deleted by the correlation analysis by the invalid signal deleting unit 4 by the correlation analysis.
- the principal component analysis unit 5 performs data dimension reduction.
- the M-dimensional multidimensional time-series signal is linearly converted into an r-dimensional multi-dimensional time-series signal having the number of dimensions r by principal component analysis.
- Principal component analysis generates an axis that maximizes variation. KL conversion may be used.
- the number of dimensions r is determined based on a value that is a cumulative contribution ratio obtained by arranging eigenvalues obtained by principal component analysis in descending order and dividing the eigenvalue added from the larger one by the sum of all eigenvalues.
- FIG. 4 shows the pattern.
- the three-dimensional display (referred to as feature space) in the upper left of FIG. 4 is a three-dimensional display with a high contribution ratio of the r-dimensional multidimensional time-series signal after the principal component analysis. In this state, it can be seen that the state of the target facility is still observed as being complicated.
- the remaining eight three-dimensional displays in FIG. 4 are obtained by tracking a trajectory along time and performing clustering, and represent each cluster. In clustering, if the distance between data exceeds a predetermined threshold value over time, another cluster is used, and if the threshold value is not exceeded, the same cluster is used.
- clusters 1, 3, 9, 10, and 17 are clusters in the operation ON state, and the clusters 6, 14, and 20 are separated into the clusters in the operation OFF state.
- Clusters not shown such as cluster 2 are in the transition period.
- the locus moves linearly when the operation is on, and an unstable locus moves when the operation is off.
- clustering by trajectory division has several advantages. It can be classified into a plurality of states such as an operation ON state and an operation OFF state. As seen in the operation ON state, these clusters can be expressed by a low-dimensional model such as linear.
- These clustering operations may be performed with these strings attached in consideration of the alarm signal of the facility and maintenance information. Specifically, information such as an alarm signal is added to each cluster as an attribute.
- FIG. 5 shows another example of the state of labeling by clustering in the feature space.
- FIG. 6 shows the result of clustering on one time series signal. In this case, 16 clusters can be generated, and it can be seen that the time-series signal is divided into 16 clusters.
- the operation time (cumulative time) was also displayed. The horizontal part is the operation OFF. It can be seen that operation ON and operation OFF are separated with high accuracy.
- modeling in a low-dimensional subspace is performed for each clustered cluster by the modeling unit 8 for each cluster.
- modeling is performed by regression analysis.
- the general formula for regression analysis is as follows. “Y” corresponds to an r-dimensional multidimensional time series signal for each cluster. “X” is a variable that explains y. "Y ⁇ " becomes a model. “E” is a deviation.
- y Objective variable (column r) b: Regression coefficient (1 + p column)
- ⁇ min b (X'X) -1X'y ('represents transpose)
- FIG. 24 illustrates a general CLAFIC method. The case of 2 class, 2D pattern is shown. A subspace of each class, that is, a subspace represented here as a one-dimensional straight line is obtained.
- eigenvalue decomposition is performed on the autocorrelation matrix of each class of data, and an eigenvector is obtained as a basis.
- the eigenvectors corresponding to the upper eigenvalues having a large value are used.
- the unknown pattern q (latest observation pattern) is classified into class A.
- the multidimensional time series signal shown in FIG. 2 since the normal part is basically targeted, it becomes a problem of class 1 identification (illustrated in FIG. 1).
- the distance from the latest observation pattern) to class A is obtained, and this is taken as the deviation. If the deviation is large, it is determined as an outlier.
- the influence is mitigated when the dimension is reduced and the subspace is made. This is an advantage of applying the subspace method.
- the center of gravity of each class is used as the origin.
- the eigenvector obtained by applying KL expansion to the covariance matrix of each class is used as a basis.
- Various subspace methods have been proposed, but if there is a distance scale, the degree of deviation can be calculated. In the case of the density, the degree of deviation can be determined based on the magnitude.
- the CLAFIC method is a similarity measure because it determines the length of the orthogonal projection.
- Subspace methods such as the projection distance method are discriminators based on distance, and as a learning method when abnormal data can be used, vector quantization that updates dictionary patterns and metric learning that learns distance functions can be used. .
- k multidimensional time series signals close to an unknown pattern q (latest observation pattern), called a local subspace method, are obtained, and a linear manifold is generated such that the nearest neighbor pattern of each class is the origin, A method of classifying an unknown pattern into a class having a minimum projection distance to the linear manifold can also be applied (see the frame of the local subspace method in FIG. 31).
- Local subspace method is also a kind of subspace method.
- the local subspace method will be applied to each cluster after clustering already described.
- k is a parameter.
- the anomaly detection since it becomes a problem of one class identification as before, the class A to which the majority of data belongs is regarded as a normal part, and the distance from the unknown pattern q (latest observation pattern) to the class A is obtained, This is the deviation.
- an orthogonal projection point from an unknown pattern q (latest observation pattern) to a partial space formed using k multidimensional time series signals can be calculated as an estimated value ( (The estimated value data described in the frame of the local subspace method in FIG. 31). It is also possible to rearrange the k multi-dimensional time series signals in the order closer to the unknown pattern q (latest observation pattern) and perform weighting inversely proportional to the distance to calculate the estimated value of each signal.
- the estimated value can be calculated in the same manner by the projection distance method or the like.
- the parameter k is usually set to one type, but if the parameter k is changed and executed several times, the target data will be selected according to the similarity, and it can be judged comprehensively from those results. is there.
- the selected data in the cluster is targeted, even if anomalous values are mixed slightly, the influence is greatly reduced when the local subspace is used.
- a classifier such as a one-class support vector machine is also applicable if it is simply considered as a problem of one-class identification.
- kernelization such as radial ⁇ basis function that maps to higher-order space can be used.
- the side close to the origin is an outlier, that is, an abnormality.
- the support vector machine can cope with a large dimension of the feature amount, there is a drawback that the calculation amount becomes enormous as the number of learning data increases.
- FIG. 8 shows a case where clustering by trajectory division is not performed for reference. In the case of FIG. 7, the deviation is large when the time-series signal behaves in an oscillating manner in the operation OFF section and the operation ON section.
- the outlier detection unit 10 obtains the outlier.
- the magnitude of the threshold is checked. Since the detected abnormal signal is after the principal component analysis, it can be inversely converted to confirm at what ratio the original signal is synthesized and determined as abnormal.
- FIG. 28 shows an application example of the local subspace method.
- the signal is divided into the first half and the second half (according to a verification method called intersection confirmation), and each is used as learning data to determine the distance to the remaining data.
- the parameter k was 10.
- a stable result can be obtained by changing some k and taking a majority vote thereof (based on the same idea as the method of bagging described later).
- This local subspace method has an advantage that N data is automatically removed.
- an irregular behavior during operation OFF is detected.
- the necessity of clustering is alleviated, but the cluster other than the cluster to which the observation data belongs may be used as learning data, and the local subspace method may be applied to this data and the observation data. According to this method, the degree of deviation from another cluster can be evaluated. The same applies to the projection distance method.
- FIG. 29 shows examples thereof. Data other than the cluster to which the observation data belongs was used as learning data. This idea is effective because the most similar data can be excluded from the “local” region when similar data continues like time-series data.
- N data extraction was demonstrated as a feature-value (sensor signal), the data of a time-axis direction may be sufficient.
- FIG. 9 shows some examples.
- the diagram on the left side of FIG. 9 is a two-dimensional display of the r-dimensional time series signal after principal component analysis. This is an example of visualizing data behavior.
- the diagram on the right side of FIG. 9 illustrates clustering by performing clustering by trajectory division. In this example, each cluster is expressed by a simple low-order model (here, a straight line).
- the diagram on the left side of FIG. 10 is an example shown so that the moving speed of data can be understood. By applying the Wavelet analysis described later, it is possible to analyze the speed, that is, the frequency, and handle it as a multivariate.
- the right diagram in FIG. 10 is an example displayed so that the deviation from the model shown in the right diagram in FIG. 9 can be understood.
- the figure on the left side of FIG. 11 is another example. This is an example in which clusters determined to be similar based on a distance criterion or the like are merged (in the figure, adjacent clusters are merged), the model after merging is shown, and the deviation from the model is illustrated.
- the diagram on the right side of FIG. 11 represents the state. Three types of states A, B, and C are separately displayed. When the states are considered separately, changes in the state A and the like can be illustrated as shown in the left diagram of FIG.
- FIG. 12 shows changes from a model (low-order subspace) obtained from past learning data, and state changes can be observed. In this way, better understanding can be promoted by processing the data, showing the processed data to the user, and visualizing the current situation.
- FIG. 13 shows an abnormality detection method.
- a random number of r-dimensional multi-dimensional time series signals are selected for each cluster.
- By random selection There is an advantage that it can be calculated in a shorter time than all combinations except for an ineffective signal in which a characteristic that cannot be seen appears when all signals are used. Further, it is possible to select a number of r-dimensional multi-dimensional time series signals that are randomly determined in the time axis direction.
- a cluster may be used as a unit, the inside of the cluster is divided, and the determined number is randomly selected.
- FIG. 14 shows another embodiment. A portion 12 for processing an alarm signal is added. As shown in FIG. 15A, an alarm signal generation history is acquired. Then, the histogram is displayed. It can be easily imagined that the degree of abnormality is high in the high frequency section. Therefore, as shown in FIG. 15 (b), in consideration of the frequency of the histogram, the abnormality specifying unit 13 shown in FIG. Abnormality judgment is performed.
- FIG. 16 shows another embodiment. This is an example in which Wavelet (conversion) analysis is added.
- the wavelet analysis signal adding unit 14 performs the wavelet analysis shown in FIG. 17 for the M-dimensional multidimensional time series signal, and adds these signals to the M-dimensional multidimensional time series signal. It is also possible to replace it with an M-dimensional multidimensional time series signal. Anomalies are detected by a discriminator such as a local subspace method for such newly added or replaced multidimensional time series signals.
- 17A corresponds to the signal of scale 1 in FIG. 25 described later
- FIG. 17B corresponds to the fluctuation of scale 8 in FIG. 25 described later
- FIG. 25 corresponds to the variation of the scale 4 in FIG. 25
- FIG. 17D corresponds to the variation of the scale 2 in FIG.
- FIG. 25 illustrates the Wavelet analysis.
- the signal of scale 1 is the original signal. This is sequentially added to the adjacent signal to create a scale 2 signal, and the difference from the original signal is calculated to generate a scale 2 fluctuation signal. When this is repeated in sequence, finally, a constant value signal of scale 8 and its fluctuation signal are obtained.
- the original signal is decomposed into each fluctuation signal of scale 2, 4, 8 and DC signal of scale 8. it can. Therefore, each fluctuation signal of such scales 2, 4, and 8 is regarded as a new characteristic signal and is added to the multidimensional time series signal.
- the frequency spectrum obtained by performing Fourier transform spreads over the entire area, and it is difficult to extract features for individual signals.
- the Wavelet transform which can obtain a spectrum localized in time, is convenient when data including a lot of non-stationary signals, such as pulses and impulses, is processed.
- Wavelet analysis edited by the Institute of Electrical Engineers of Japan, is detailed in Asakura Publishing, 2005, “Industrial Application of Wavelet Analysis” written by Seiichi Shin. It is applied to many objects such as chemical plant control system diagnosis, abnormality detection in air conditioning plant control, cement firing process abnormality monitoring, and glass melting furnace control.
- the difference from the prior art is that the wavelet analysis is treated as a multi-resolution expression, and the information of the original multi-dimensional time series signal is revealed by the wavelet transform.
- these as multivariate variables it is possible to detect the abnormality early from the stage where the abnormality is weak. That is, it becomes possible to detect early as a sign.
- ⁇ Classification is possible based on the above formula.
- the above is an example of classification into multi-classes, but if the number of classes K is 2, it becomes a problem of two classifications, that is, an abnormality detection. Then, by applying the result of Wavelet analysis to the observed value, it is possible to detect anomalies that target even anomalies localized in time. As described above, when there is a lot of abnormality data and it can be taught, it is possible to perform abnormality detection with higher accuracy in a statistical sense.
- FIG. 18 shows another embodiment. This is an example in which a scatter diagram / correlation analysis unit 15 is added.
- FIG. 19 shows an example in which a scatter diagram / correlation analysis is performed on an r-dimensional multidimensional time series signal.
- a delay lag is considered.
- the position of the maximum value of the cross correlation function is called a lag.
- the time lag for the two phenomena is equal to the lag of the cross-correlation function.
- the positive / negative of the lag is determined by which of the two phenomena occurs earlier.
- the results of such scatter diagram analysis and cross-correlation analysis represent the correlation between time series signals, but can also be used to characterize each cluster and can be a measure of similarity between clusters. .
- similarity between clusters is determined based on the degree of coincidence of lag amounts. This makes it possible to merge similar clusters shown in FIG. Model using the merged data.
- the merging method may be another method.
- FIG. 20 shows another embodiment. This is an example in which a time / frequency analysis unit 16 is added.
- FIG. 21 shows an example in which time / frequency analysis is performed on an r-dimensional multidimensional time series signal. It is also possible to perform time / frequency analysis or scatter diagram / correlation analysis and add these signals to the M-dimensional multi-dimensional time series signal or replace them with the M-dimensional multi-dimensional time series signal.
- FIG. 22 shows another embodiment. This is an example in which learning data DB 17 and modeling (1) 18 are added.
- FIG. 23 shows the details.
- the learning data is modeled as a plurality of models, the similarity with the observation data is judged, the corresponding model is applied, and the deviation from the observation data is calculated.
- Modeling (2) is the same as in FIG. 1, and from this, the deviation from the model obtained from the observation data is calculated.
- modeling (1) and (2) can be handled equally, but weighting may be performed. That is, if the learning data is considered basic, the weight of the model (1) is increased, and if the observation data is considered basic, the weight of the model (2) is increased.
- the subspace model comprised by the model (1) is compared between clusters, and if they are the clusters of the same state originally, the state change can be known. And if the subspace model of observation data has moved from it, a state change can be read. If the state change is intended such as part replacement, that is, if the design side knows and should allow the change, the weight of the model (1) is decreased and the weight of the model (2) is decreased. Enlarge. If the state change is not intended, the weight of the model (1) is increased. For example, if the parameter ⁇ is used as the weight of the model (1), ⁇ ⁇ model (1) + (1 ⁇ ) ⁇ model (2) Can be formulated as
- the model (1) may have a forgetting type in which the weight of the older model is smaller. In this case, a model based on recent data is emphasized.
- a physical model 22 is a model that simulates a target engine and the like by simulation. If the target knowledge is sufficient, the target engine or the like can be expressed by a discrete-time (non) linear state space model (represented by a state equation or the like), so that an intermediate value or output thereof can be estimated. Therefore, according to this physical model, it is possible to detect an abnormality based on the deviation from this model.
- the learning data model (1) it is also possible to modify the learning data model (1) according to the physical model.
- the physical model can be modified in accordance with the learning data model (1).
- knowledge as a past record it is also possible to incorporate data transitions accompanying the occurrence of alarms and parts replacement into the physical model.
- the learning data (individual data, the position of the center of gravity, etc.) may be moved with the occurrence of an alarm or part replacement.
- the statistical model is mainly used for the physical model when the statistical model is less understood about the process of generating data. By being effective. Distance and similarity can be defined even if the data generation process is unclear. Even in the case of an image, the statistical model is effective when the image generation process is unclear. If the knowledge about the object can be used even a little, the physical model 22 can be used.
- FIG. 26 shows abnormality detection and diagnosis after abnormality detection.
- an abnormality is detected from the time-series signal from the facility by the method described so far.
- incidental information such as maintenance events (alarms, work results, etc.) is captured to detect abnormalities with high sensitivity.
- abnormality diagnosis is performed based on this sign.
- Abnormality diagnosis can be easily divided into phenomenon diagnosis that identifies sensors that contain signs and cause diagnosis that identifies parts that may cause failure.
- the abnormality detection unit outputs information regarding the feature amount in addition to a signal indicating the presence / absence of abnormality to the abnormality diagnosis unit.
- the abnormality diagnosis unit makes a diagnosis based on this information.
- Fig. 27 shows the hardware configuration.
- Sensor data such as a target engine is input to a processor that executes abnormality detection, and missing values are repaired and stored in the database DB.
- the processor performs abnormality detection using the observation data and DB data, and outputs an abnormality detection signal.
- the display unit performs various displays as described in the embodiments. It is also possible to display a trend.
- Database DB can be operated by skilled engineers. In particular, abnormal cases and countermeasure cases can be taught and stored. (1) Learning data (normal), (2) abnormal data, (3) countermeasure contents are stored. By making the database DB a structure that can be manipulated by skilled engineers, a sophisticated and useful database can be created. Data manipulation is performed by moving learning data (individual data, the position of the center of gravity, etc.) with the occurrence of an alarm or part replacement. It is also possible to add acquired data. If there is abnormal data, a method such as generalized vector quantization can be applied to the movement of the data.
- the time series signal is used, it does not matter. It can also be applied to human body measurement data. According to this embodiment, even if the number of states and the number of transitions are large, it can be handled.
- each function described in the embodiment for example, clustering, principal component analysis, wavelet analysis, etc. is not necessarily performed, and may be appropriately performed according to the nature of the target signal.
- Clustering is not limited to time trajectories, and it goes without saying that methods in the data mining field can be used, including the EM (Expectation-Maximization IV) algorithm and k-means clustering for mixed distributions.
- a classifier may be applied to the obtained cluster as a target, but a cluster may be grouped and a classifier may be applied to the cluster.
- the simplest example is to separate the cluster to which the daily observation data belongs and the cluster to which the daily observation data belongs (the current data as the data of interest shown in the feature space on the right side of FIG. Applicable to past data).
- the selection of sensor signals (features) can be done by using existing methods such as the wrapper method (for example, using the backward ⁇ ⁇ stepwise selection to remove the most unnecessary features one by one from the state where all feature quantities are present). is there.
- the discriminator can prepare several discriminators and take the majority of them. That is, ensemble (group) learning using different classifier groups can be applied.
- FIG. 30 shows a configuration example thereof.
- the classifier 1 is a projection distance method
- the classifier 2 is a local subspace method
- the classifier 3 is a linear regression method. Any classifier can be applied as long as it is based on case data.
- the reason for using multiple discriminators is that the discriminators require different criteria and different target data ranges (depending on segmentation and their integration), so that there is a slight difference in the results. For this reason, a majority decision is taken to stabilize, or an OR (outlier value itself, that is, maximum value detection in the case of multiple values) logic is used, and if an abnormality is detected by one of the discriminators, it is output that an abnormality has occurred. , Try to detect without abnormality, or with AND (minimum value detection in the case of multiple values) logic, if any abnormality is detected at the same time in any discriminator, it will output that an abnormality has occurred and minimize false detection
- the discriminator is configured based on a higher standard such as limiting. Of course, it is needless to say that the integration can be performed in consideration of information such as maintenance information such as alarm signals and parts replacement.
- bagging takes K pieces of data by allowing duplication from N data (restoration extraction), creates the first discriminator h1 based on this K pieces, and duplicates from N pieces of data.
- a second learner h2 based on this K pieces (the contents of which are different from the first one), and collect several discriminators from different data. This is the method of taking a majority vote when actually using it as a discriminator.
- an equal weight 1 / N is first assigned to N pieces of data, and the first discriminator h1 learns using all N pieces of data.
- the accuracy rate is checked for each piece, and the reliability ⁇ 1 (> 0) is obtained based on the accuracy rate.
- the weight of the data correctly answered by the first discriminator is reduced by exp ( ⁇ 1) to reduce the weight, and the weight of the data that cannot be correctly answered is increased by exp ( ⁇ 1).
- the second discriminator h2 performs weighted learning using all N pieces of data, obtains a reliability ⁇ 2 (> 0), and updates the data weight.
- the weight of the correct data for both is light, and the weight of the wrong data for both is heavy. Thereafter, this is repeated to create M discriminators, and when actually used as discriminators, a majority vote with reliability is taken. By applying these methods to the cluster group, performance improvement can be expected.
- FIG. 31 shows an example of a configuration example of the entire abnormality detection including the discriminator shown in FIG.
- ensemble learning is performed to achieve a high identification rate.
- the method of integrating the discriminator outputs is as described above, but there are some combinations of which discriminator is applied to which cluster.
- the local subspace method is applied to clusters different from the observed data to grasp the degree of deviation from the different clusters (calculate the estimated value), and the regression analysis method is applied to the same cluster as the observed data. Apply and grasp the degree of deviation from its own cluster.
- the degree of deviation from other clusters can also be performed by the projection distance method or the regression analysis method.
- the degree of deviation from the own cluster can also be performed by the projection distance method. If the alarm signal can be utilized, the cluster can be a cluster to which no severe alarm signal is added depending on the severity level of the alarm signal.
- the integration of the discriminator outputs may be scalar conversion processing such as addition of outliers, maximum / minimum, OR / AND, or the output of the discriminator can be treated as a vector and multidimensional.
- the scales of the discriminator outputs are matched as much as possible.
- the first report of abnormality is detected for other clusters, and when the data of the own cluster is collected, the abnormality of the second report is targeted for the own cluster. Detection may be performed. In this way, it is possible to urge customers to call attention.
- the present embodiment can be said to be an embodiment that pays more attention to signal behavior and behavior in relation to the target cluster group.
- Each cluster trajectory scatter diagram / correlation analysis unit 16 ⁇ ⁇ ⁇ Time / frequency analysis section for each cluster, 17 ⁇ ⁇ ⁇ Learning data, 18 Modeling part (1), 19 ... Processor, 20 ... Display, 21 ... Database 22: Physical model.
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
Description
運転ONの状態、運転OFFの状態など、複数の状態に分類できる。
運転ONの状態に見られるように、これらのクラスタは、例えば線形など、低次元のモデルで表現できる。
これらのクラスタリングは、設備のアラーム信号や保全情報を加味して、これらのひも付きとして、実施してもよい。具体的には、各クラスタに、アラーム信号などの情報が属性として付加される。
y:目的変数(r列)
b:回帰係数(1+p列)
X:説明変数行列(r行、1+p列)
||y-Xb||⇒min
b=(X'X)-1X'y('は転置を表す)
y~=Xb=X(X'X) -1X'y(説明変数の影響を表す部分)
e=y-y~(y~で近似できない部分。説明変数の影響を取り除いた部分)
ただし、rank X=p+1
ここでは、各クラスタのr次元の多次元時系列信号に対し、データのN個抜き(N=0,1,2,・・・)の回帰分析を行う。例えばN=1の場合は、異常信号が1種混入していると考え、これを除いた信号を「X」としてモデル化する。N=0の場合は、全r次元多次元時系列信号を扱うことになる。
全ての信号を使う場合に見えなかった特性が現れる
有効ではない信号を除く
全ての組み合わせより、短時間で計算できる
という利点がある。
さらに、時間軸方向にr次元多次元時系列信号をランダムに定めた個数分選択するという選択も考えられる。ここでは、クラスタを単位とすることもあるが、クラスタ内を区分し、これらを定めた個数分、ランダムに選択する。
α×モデル(1)+(1-α)×モデル(2)
として定式化できる。
正常データから、異常を検知できる、
データ収集が不完全でも精度の高い異常検知が可能となる、
異常データが包含されていても、この影響を許容できる、
といった直接的効果に加え、
ユーザにとって、現象を理解しやすい、
エンジニアの知識を活用できる
物理モデルも併用できる、
と言った副次的な効果がある。
2・・・欠損値の修正・削除部、
3・・・状態データ・知識データベース、
4・・・相関解析による無効信号の削除部、
6・・・軌跡分割クラスタリング、
7・・・アラーム信号/保全情報、
8・・・各クラスタを対象にモデル化部、
9・・・モデルからの偏差算出部、
10・・・はずれ値検出部、
11・・・各クラスタを特徴選択でモデル化部、
12・・・アラーム信号などの一定区間累積ヒストグラム、
13・・・異常特定部、
14・・・Wavelet(変換)解析部、
15・・・各クラスタ軌跡散布図・相関解析部、
16・・・各クラスタ毎時間・周波数解析部、
17・・・学習データ、
18・・・モデル化(1)部、
19・・・プロセッサ、
20・・・表示器、
21・・・データベース
22・・・物理モデル。
Claims (14)
- プラントまたは設備の異常を早期に検知する異常検知方法であって、
複数のセンサからデータを取得し、前記データの時間的な変化に基づいて、データ空間の軌跡を複数のクラスタに分割し、
前記分割した前記クラスタ群に対して、部分空間法でモデル化し、
はずれ値を異常候補として算出することを特徴とする異常検知方法 - 前記モデル化は、前記複数のセンサから過去に得られたデータからなる学習データを対象にモデル化したリファレンスモデルを併用して、はずれ値を異常候補として算出することを特徴とする請求項1に記載の異常検知方法。
- 前記モデル化は、データのN個抜き(N=0,1,2,・・・)の回帰分析法、または、部分空間法を用いて行うことを特徴とする請求項1記載の異常検知方法。
- 前記部分空間法は、投影距離法、CLAFIC法、または、観測データの近傍を対象とする局所部分空間法であることを特徴とする請求項3記載の異常検知方法。
- 前記はずれ値の検出は、前記回帰分析法、前記投影距離法、前記CLAFIC法、または、観測データの近傍を対象とする前記局所部分空間法の出力を統合して行うものであることを特徴とする請求項4記載の異常検知方法。
- プラントまたは設備の異常を早期に検知する方法において、
複数のセンサからデータを取得し、
時間的なデータの時間的変化に基づいて、データ空間の軌跡を複数のクラスタに分割し、
分割したクラスタが分かるように、データを表示することにより、状態を可視化することを特徴とする異常検知方法。 - 前記複数のセンサからの過去のデータに基づくクラスタ群に対する異常検知により、第1報を出力し、前記複数のセンサの現在のデータに基づくクラスタ群に対する異常検知により第2報を出力することを特徴とする請求項1記載の異常検知方法。
- プラントまたは設備の異常を早期に検知する異常検知システムにおいて、
複数のセンサからデータを取得するデータ取得部と、
前記データの時間的な変化に基づいて、データ空間の軌跡を複数のクラスタ群に分割する軌跡分割部と、
分割した前記クラスタ群に対して、部分空間でモデル化するモデル化部と、
はずれ値を異常候補として算出する異常検出部を有し、
前記軌跡分割部においては前記分割したクラスタ群に対して、部分空間でモデル化し、はずれ値を異常候補として算出することを特徴とする異常検知システム - 前記データ取得部において取得した前記複数のセンサからの過去のデータを蓄積し、かつ、モデル化してリファレンスモデルを作成する学習データベース部を有し、前記学習データベース部における前記リファレンスモデルを併用して、はずれ値を異常候補として算出することを特徴とする請求項8記載の異常検知システム。
- 前記モデル化は、データのN個抜き(N=0,1,2,・・・)の回帰分析法、または、部分空間法を用いて行うことを特徴とする請求項8記載の異常検知システム。
- 前記部分空間法は、投影距離法、CLAFIS法、または、観測データの近傍を対象とする局所部分空間法であることを特徴とする請求項8記載の異常検知システム。
- 前記はずれ値の検出は、前記回帰分析法、前記投影距離法、前記CLAFIS法、または、観測データの近傍を対象とする前記局所部分空間法の出力を統合して行うものであることを特徴とする請求項8記載の異常検知システム。
- プラントまたは設備の異常を早期に検知する異常システムにおいて、
複数のセンサからデータを取得するデータ取得部と、
前記データの時間的な変化に基づいて、データ空間の軌跡を複数のクラスタに分割する軌跡分割部と、
前記分割したクラスタ群に対して、部分空間でモデル化するモデル化部と、
はずれ値を異常候補として算出する異常検出部と、
データ表示部とを有し、
前記分割した複数のクラスタが分かるように、データを表示することにより、
状態を可視化することを特徴とする異常検知システム。 - 前記複数のセンサからの過去のデータに基づくクラスタ群に対する異常検知により、第1報を出力し、前記複数のセンサの現在のデータに基づくクラスタ群に対する異常検知により第2法を出力することを特徴とする請求項13記載の異常検知システム。
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200980130579.0A CN102112933B (zh) | 2008-10-09 | 2009-05-29 | 异常检测方法及系统 |
| KR1020117002716A KR101316486B1 (ko) | 2008-10-09 | 2009-05-29 | 이상 검지 방법 및 시스템 |
| US13/057,831 US8630962B2 (en) | 2008-10-09 | 2009-05-29 | Error detection method and its system for early detection of errors in a planar or facilities |
| EP09818896A EP2333629A4 (en) | 2008-10-09 | 2009-05-29 | ERROR DETECTION METHOD AND SYSTEM |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2008263030A JP5048625B2 (ja) | 2008-10-09 | 2008-10-09 | 異常検知方法及びシステム |
| JP2008-263030 | 2008-10-09 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2010041355A1 true WO2010041355A1 (ja) | 2010-04-15 |
Family
ID=42100317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2009/002391 Ceased WO2010041355A1 (ja) | 2008-10-09 | 2009-05-29 | 異常検知方法及びシステム |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US8630962B2 (ja) |
| EP (1) | EP2333629A4 (ja) |
| JP (1) | JP5048625B2 (ja) |
| KR (1) | KR101316486B1 (ja) |
| CN (1) | CN102112933B (ja) |
| WO (1) | WO2010041355A1 (ja) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101888309A (zh) * | 2010-06-30 | 2010-11-17 | 中国科学院计算技术研究所 | 在线日志分析方法 |
| JP2019057056A (ja) * | 2017-09-20 | 2019-04-11 | 株式会社東芝 | 事故原因分類装置 |
| CN112256750A (zh) * | 2020-09-27 | 2021-01-22 | 上海楚腾信息技术有限公司 | 一种数据纠错预警方法、系统及存储介质 |
| US11215363B2 (en) * | 2017-04-24 | 2022-01-04 | Honeywell International Inc. | Apparatus and method for two-stage detection of furnace flooding or other conditions |
| CN121658871A (zh) * | 2026-02-06 | 2026-03-13 | 西安现代控制技术研究所 | 基于注意力表征集成聚类的低成本飞行器运行支持方法 |
Families Citing this family (92)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5431235B2 (ja) | 2009-08-28 | 2014-03-05 | 株式会社日立製作所 | 設備状態監視方法およびその装置 |
| EP2568433A4 (en) | 2010-05-06 | 2017-10-25 | Nec Corporation | Malfunction analysis apparatus, malfunction analysis method, and recording medium |
| JP5598082B2 (ja) * | 2010-05-17 | 2014-10-01 | 日本電気株式会社 | 異常判定装置、異常判定プログラムおよび異常判定方法 |
| JP5132746B2 (ja) * | 2010-09-30 | 2013-01-30 | 中国電力株式会社 | 動作異常検出装置、動作異常検出方法 |
| US20120259792A1 (en) * | 2011-04-06 | 2012-10-11 | International Business Machines Corporation | Automatic detection of different types of changes in a business process |
| CA2741202A1 (en) * | 2011-05-27 | 2012-11-27 | Hydro-Quebec | Dynamic clustering of transient signals |
| US8620853B2 (en) | 2011-07-19 | 2013-12-31 | Smartsignal Corporation | Monitoring method using kernel regression modeling with pattern sequences |
| US8660980B2 (en) * | 2011-07-19 | 2014-02-25 | Smartsignal Corporation | Monitoring system using kernel regression modeling with pattern sequences |
| WO2013030984A1 (ja) | 2011-08-31 | 2013-03-07 | 株式会社日立エンジニアリング・アンド・サービス | 設備状態監視方法およびその装置 |
| CN102509001B (zh) * | 2011-10-18 | 2014-06-04 | 西安交通大学 | 一种自动去除时序数据野值点的方法 |
| JP5648157B2 (ja) * | 2011-12-28 | 2015-01-07 | 株式会社日立ハイテクノロジーズ | 半導体製造装置 |
| US20140032506A1 (en) * | 2012-06-12 | 2014-01-30 | Quality Attributes Software, Inc. | System and methods for real-time detection, correction, and transformation of time series data |
| US9336302B1 (en) | 2012-07-20 | 2016-05-10 | Zuci Realty Llc | Insight and algorithmic clustering for automated synthesis |
| KR101978018B1 (ko) * | 2012-12-18 | 2019-05-13 | 아우디 아게 | 예후 및 건전성 관리를 위한 한도 기반 임계치 추정 |
| JP5643372B2 (ja) * | 2013-03-28 | 2014-12-17 | 株式会社トーエネック | 回転機の良否診断システム |
| CA2815161A1 (en) * | 2013-05-06 | 2014-11-06 | Hydro-Quebec | Quantitative analysis of signal related measurements for trending and pattern recognition |
| US10069691B2 (en) * | 2013-11-26 | 2018-09-04 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and apparatus for anomaly detection in a network |
| EP2881822A1 (de) | 2013-12-05 | 2015-06-10 | Bayer Technology Services GmbH | Computer-implementiertes Verfahren und System zur automatischen Überwachung und Statusermittlung ganzer Prozessabschnitte in einer Process Unit |
| JP6299299B2 (ja) * | 2014-03-14 | 2018-03-28 | オムロン株式会社 | 事象検出装置および事象検出方法 |
| CN105338541B (zh) * | 2014-06-30 | 2018-11-13 | 华为技术有限公司 | 一种基于移动无线网络数据的异常轨迹检测方法及装置 |
| CN104166718B (zh) * | 2014-08-18 | 2017-11-03 | 国家电网公司 | 一种适用于大电网的不良数据检测与辨识方法 |
| JP6301791B2 (ja) * | 2014-09-17 | 2018-03-28 | 株式会社東芝 | 配電網の故障予兆診断システムおよびその方法 |
| FR3032273B1 (fr) * | 2015-01-30 | 2019-06-21 | Safran Aircraft Engines | Procede, systeme et programme d'ordinateur pour phase d'apprentissage d'une analyse acoustique ou vibratoire d'une machine |
| US9915942B2 (en) | 2015-03-20 | 2018-03-13 | International Business Machines Corporation | System and method for identifying significant and consumable-insensitive trace features |
| US9864823B2 (en) | 2015-03-30 | 2018-01-09 | Uop Llc | Cleansing system for a feed composition based on environmental factors |
| CN106156913B (zh) * | 2015-04-02 | 2019-02-12 | 中国商用飞机有限责任公司 | 用于飞机部附件的健康管理方法 |
| CN105467174B (zh) * | 2015-11-23 | 2018-11-09 | 上海华岭集成电路技术股份有限公司 | 一种获取悬臂式探针系统维护周期的方法 |
| US20190011506A1 (en) * | 2016-01-20 | 2019-01-10 | Mitsubishi Electric Corporation | Malfunction detection apparatus capable of detecting actual malfunctioning device not due to abnormal input values |
| KR102444442B1 (ko) | 2016-03-07 | 2022-09-21 | 한국전자통신연구원 | 설비 고장 진단 장치 및 방법 |
| JP6686593B2 (ja) | 2016-03-23 | 2020-04-22 | 日本電気株式会社 | データ処理装置、データ処理システム、データ処理方法及びプログラム |
| CN105973595B (zh) * | 2016-04-27 | 2019-03-19 | 清华大学深圳研究生院 | 一种滚动轴承故障的诊断方法 |
| CN105910806A (zh) * | 2016-05-30 | 2016-08-31 | 重庆大学 | 加注泵早期健康状态监测方法 |
| EP3258333A1 (en) * | 2016-06-17 | 2017-12-20 | Siemens Aktiengesellschaft | Method and system for monitoring sensor data of rotating equipment |
| CN109690641B (zh) * | 2016-08-29 | 2022-11-22 | 韩国水力原子力株式会社 | 包括确定设备重要度和警报有效性的处理程序的用于预检核电站设备异常迹象的方法及系统 |
| JP6798825B2 (ja) * | 2016-08-29 | 2020-12-09 | 公立大学法人大阪 | データ解析装置、制御装置、データ解析装置の制御方法、制御プログラム、および記録媒体 |
| US10545487B2 (en) * | 2016-09-16 | 2020-01-28 | Uop Llc | Interactive diagnostic system and method for managing process model analysis |
| JP2018077757A (ja) * | 2016-11-11 | 2018-05-17 | 横河電機株式会社 | 情報処理装置、情報処理方法、情報処理プログラム及び記録媒体 |
| US11205103B2 (en) | 2016-12-09 | 2021-12-21 | The Research Foundation for the State University | Semisupervised autoencoder for sentiment analysis |
| FR3061324B1 (fr) * | 2016-12-22 | 2019-05-31 | Electricite De France | Procede de caracterisation d'une ou plusieurs defaillances d'un systeme |
| JP6778132B2 (ja) * | 2017-03-13 | 2020-10-28 | 株式会社日立製作所 | 設備機器の異常診断システム |
| CN107103039B (zh) * | 2017-03-24 | 2021-03-30 | 北明软件有限公司 | 针对海量数据的异常数据检测与处理方法 |
| US10754359B2 (en) | 2017-03-27 | 2020-08-25 | Uop Llc | Operating slide valves in petrochemical plants or refineries |
| US10678272B2 (en) | 2017-03-27 | 2020-06-09 | Uop Llc | Early prediction and detection of slide valve sticking in petrochemical plants or refineries |
| US10670353B2 (en) | 2017-03-28 | 2020-06-02 | Uop Llc | Detecting and correcting cross-leakage in heat exchangers in a petrochemical plant or refinery |
| US10962302B2 (en) | 2017-03-28 | 2021-03-30 | Uop Llc | Heat exchangers in a petrochemical plant or refinery |
| US11130111B2 (en) | 2017-03-28 | 2021-09-28 | Uop Llc | Air-cooled heat exchangers |
| US11396002B2 (en) | 2017-03-28 | 2022-07-26 | Uop Llc | Detecting and correcting problems in liquid lifting in heat exchangers |
| US10794644B2 (en) | 2017-03-28 | 2020-10-06 | Uop Llc | Detecting and correcting thermal stresses in heat exchangers in a petrochemical plant or refinery |
| US10752845B2 (en) | 2017-03-28 | 2020-08-25 | Uop Llc | Using molecular weight and invariant mapping to determine performance of rotating equipment in a petrochemical plant or refinery |
| US10663238B2 (en) | 2017-03-28 | 2020-05-26 | Uop Llc | Detecting and correcting maldistribution in heat exchangers in a petrochemical plant or refinery |
| US10695711B2 (en) | 2017-04-28 | 2020-06-30 | Uop Llc | Remote monitoring of adsorber process units |
| KR101926257B1 (ko) | 2017-05-15 | 2018-12-06 | 두산중공업 주식회사 | 이상 신호 복원 시스템 및 방법 |
| US10913905B2 (en) | 2017-06-19 | 2021-02-09 | Uop Llc | Catalyst cycle length prediction using eigen analysis |
| US11365886B2 (en) | 2017-06-19 | 2022-06-21 | Uop Llc | Remote monitoring of fired heaters |
| US10739798B2 (en) | 2017-06-20 | 2020-08-11 | Uop Llc | Incipient temperature excursion mitigation and control |
| US11130692B2 (en) | 2017-06-28 | 2021-09-28 | Uop Llc | Process and apparatus for dosing nutrients to a bioreactor |
| JP6691082B2 (ja) * | 2017-08-28 | 2020-04-28 | 株式会社日立製作所 | 指標選択装置及びその方法 |
| JP6815954B2 (ja) * | 2017-08-29 | 2021-01-20 | 株式会社神戸製鋼所 | プラントの操業状態推定装置及び操業状態推定方法 |
| CN107818135B (zh) * | 2017-09-26 | 2020-02-04 | 广东电网有限责任公司电力调度控制中心 | 一种基于灰色关联法的维诺图电力大数据异常检测方法 |
| US11194317B2 (en) | 2017-10-02 | 2021-12-07 | Uop Llc | Remote monitoring of chloride treaters using a process simulator based chloride distribution estimate |
| US11105787B2 (en) | 2017-10-20 | 2021-08-31 | Honeywell International Inc. | System and method to optimize crude oil distillation or other processing by inline analysis of crude oil properties |
| WO2019087787A1 (ja) * | 2017-10-30 | 2019-05-09 | マクセル株式会社 | 異常データ処理システムおよび異常データ処理方法 |
| CN108197254B (zh) * | 2017-12-29 | 2018-12-28 | 清华大学 | 一种基于近邻的数据修复方法 |
| US10445401B2 (en) | 2018-02-08 | 2019-10-15 | Deep Labs Inc. | Systems and methods for converting discrete wavelets to tensor fields and using neural networks to process tensor fields |
| US10901403B2 (en) | 2018-02-20 | 2021-01-26 | Uop Llc | Developing linear process models using reactor kinetic equations |
| US10734098B2 (en) | 2018-03-30 | 2020-08-04 | Uop Llc | Catalytic dehydrogenation catalyst health index |
| WO2019229977A1 (ja) * | 2018-06-01 | 2019-12-05 | 株式会社 東芝 | 推定システム、推定方法及び推定プログラム |
| CN108897640B (zh) * | 2018-06-11 | 2021-08-27 | 上海交通大学 | 群智感知中错误位置数据的检测系统及方法 |
| CN112272763B (zh) | 2018-06-14 | 2023-05-19 | 三菱电机株式会社 | 异常探测装置、异常探测方法以及计算机可读取的存储介质 |
| JP7029362B2 (ja) * | 2018-08-16 | 2022-03-03 | 三菱重工業株式会社 | 異常検出装置、異常検出方法、及びプログラム |
| CN109145996B (zh) * | 2018-08-31 | 2021-11-26 | 创新先进技术有限公司 | 异常环境下的指标数据生成方法、装置及电子设备 |
| CN109710636B (zh) * | 2018-11-13 | 2022-10-21 | 广东工业大学 | 一种基于深度迁移学习的无监督工业系统异常检测方法 |
| US10953377B2 (en) | 2018-12-10 | 2021-03-23 | Uop Llc | Delta temperature control of catalytic dehydrogenation process reactors |
| JP7093603B2 (ja) * | 2019-01-23 | 2022-06-30 | 東芝三菱電機産業システム株式会社 | ドライブ応答監視装置 |
| JP7225984B2 (ja) * | 2019-03-20 | 2023-02-21 | 株式会社リコー | システム、演算装置、及びプログラム |
| KR102191843B1 (ko) * | 2019-05-20 | 2020-12-17 | 서울대학교산학협력단 | 계층 정보를 이용한 지식 베이스 제공을 위한 방법 및 장치 |
| ES2961321T3 (es) * | 2019-09-30 | 2024-03-11 | Siemens Mobility GmbH | Diagnóstico de sistemas técnicos mediante el análisis de componentes principales (ACP) |
| WO2021079472A1 (ja) * | 2019-10-24 | 2021-04-29 | 富士通株式会社 | 異常検出方法、異常検出プログラム及び異常検出装置 |
| KR102226050B1 (ko) * | 2019-11-06 | 2021-03-11 | 한국전력공사 | 보일러 배관 모니터링 시스템 및 방법 |
| CN110794797A (zh) * | 2019-11-27 | 2020-02-14 | 江南大学 | 一种基于互信息和多块信息提取的pca故障监测方法 |
| JP2021144415A (ja) | 2020-03-11 | 2021-09-24 | 本田技研工業株式会社 | 情報処理方法、情報処理装置、異常判定システム、異常判定装置、異常判定方法、及びプログラム |
| US11698628B2 (en) | 2020-03-16 | 2023-07-11 | Vitro Flat Glass Llc | System, method, and computer program product for optimizing a manufacturing process |
| WO2022026079A1 (en) * | 2020-07-31 | 2022-02-03 | Northrop Grumman Systems Corporation | Fleet level prognostics for improved maintenance of vehicles |
| JP7010343B1 (ja) * | 2020-08-20 | 2022-01-26 | トヨタ自動車株式会社 | 機械学習装置 |
| JP7419207B2 (ja) * | 2020-09-29 | 2024-01-22 | 株式会社日立ハイテクソリューションズ | 異常予測装置及び異常予測方法 |
| CN112215436B (zh) * | 2020-11-03 | 2024-04-05 | 广州铭诚计算机科技有限公司 | 一种基于机器学习的学业异常预测方法 |
| CN116601650B (zh) * | 2020-12-08 | 2025-08-29 | 三菱电机株式会社 | 学习装置、异状检测装置及异状检测方法 |
| JP7342848B2 (ja) * | 2020-12-08 | 2023-09-12 | Jfeスチール株式会社 | 異常判定モデル生成装置、異常判定装置、異常判定モデル生成方法および異常判定方法 |
| KR102438189B1 (ko) | 2021-06-14 | 2022-08-30 | 주식회사 마키나락스 | 비정상 데이터 검출 방법 및 장치 |
| CN120188003A (zh) * | 2022-11-01 | 2025-06-20 | 杰富意钢铁株式会社 | 异常诊断装置及异常诊断方法 |
| JP7592230B1 (ja) * | 2023-02-15 | 2024-11-29 | 三菱電機株式会社 | 学習装置、状態推論装置、状態監視システム、及び学習方法 |
| US12468593B2 (en) | 2023-03-30 | 2025-11-11 | International Business Machines Corporation | Root cause identification in hybrid applications via probing |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0728766A (ja) * | 1993-06-24 | 1995-01-31 | Toshiba Corp | パターン認識システム及び監視システム |
| JP2000184593A (ja) * | 1998-12-17 | 2000-06-30 | Kyushu Electric Power Co Inc | 送電線事故原因判別方法 |
| US6216066B1 (en) | 1998-07-01 | 2001-04-10 | General Electric Company | System and method for generating alerts through multi-variate data assessment |
| US6952662B2 (en) | 2000-03-30 | 2005-10-04 | Smartsignal Corporation | Signal differentiation system using improved non-linear operator |
| US6975962B2 (en) | 2001-06-11 | 2005-12-13 | Smartsignal Corporation | Residual signal alert generation for condition monitoring using approximated SPRT distribution |
| JP2006107179A (ja) * | 2004-10-06 | 2006-04-20 | Ntt Data Corp | 異常値検出装置、変化点検出装置及び異常値検出方法、変化点検出方法 |
| JP2006252259A (ja) * | 2005-03-11 | 2006-09-21 | Ntt Data Corp | データ分析装置及び方法 |
| JP2007198918A (ja) * | 2006-01-26 | 2007-08-09 | Matsushita Electric Works Ltd | 異常監視方法、および異常監視装置 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2783620B1 (fr) * | 1998-09-22 | 2002-03-29 | De Micheaux Daniel Lafaye | Procede et systeme multidimensionnel de maitrise statistique des processus |
| JP4509860B2 (ja) | 2005-05-25 | 2010-07-21 | 株式会社東芝 | データ分割装置、データ分割方法およびプログラム |
| US20080068593A1 (en) * | 2006-09-13 | 2008-03-20 | Hiroyuki Nakano | Method and apparatus for detecting defects |
| US20080154473A1 (en) * | 2006-12-22 | 2008-06-26 | United Technologies Corporation | Gas turbine engine performance data validation |
| US8005771B2 (en) * | 2007-10-04 | 2011-08-23 | Siemens Corporation | Segment-based change detection method in multivariate data stream |
-
2008
- 2008-10-09 JP JP2008263030A patent/JP5048625B2/ja not_active Expired - Fee Related
-
2009
- 2009-05-29 EP EP09818896A patent/EP2333629A4/en not_active Withdrawn
- 2009-05-29 KR KR1020117002716A patent/KR101316486B1/ko not_active Expired - Fee Related
- 2009-05-29 WO PCT/JP2009/002391 patent/WO2010041355A1/ja not_active Ceased
- 2009-05-29 CN CN200980130579.0A patent/CN102112933B/zh not_active Expired - Fee Related
- 2009-05-29 US US13/057,831 patent/US8630962B2/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0728766A (ja) * | 1993-06-24 | 1995-01-31 | Toshiba Corp | パターン認識システム及び監視システム |
| US6216066B1 (en) | 1998-07-01 | 2001-04-10 | General Electric Company | System and method for generating alerts through multi-variate data assessment |
| JP2000184593A (ja) * | 1998-12-17 | 2000-06-30 | Kyushu Electric Power Co Inc | 送電線事故原因判別方法 |
| US6952662B2 (en) | 2000-03-30 | 2005-10-04 | Smartsignal Corporation | Signal differentiation system using improved non-linear operator |
| US6975962B2 (en) | 2001-06-11 | 2005-12-13 | Smartsignal Corporation | Residual signal alert generation for condition monitoring using approximated SPRT distribution |
| JP2006107179A (ja) * | 2004-10-06 | 2006-04-20 | Ntt Data Corp | 異常値検出装置、変化点検出装置及び異常値検出方法、変化点検出方法 |
| JP2006252259A (ja) * | 2005-03-11 | 2006-09-21 | Ntt Data Corp | データ分析装置及び方法 |
| JP2007198918A (ja) * | 2006-01-26 | 2007-08-09 | Matsushita Electric Works Ltd | 異常監視方法、および異常監視装置 |
Non-Patent Citations (3)
| Title |
|---|
| S. SHIN: "Wavelet Kaiseki no Sangyo Oyo (Industry Application of Wavelet Analysis)", 2005, ASAKURA PUBLISHING CO,. LTD. |
| See also references of EP2333629A4 |
| STEPHAN W.: "Wegerich; Nonparametric modeling of vibration signal features for equipment health monitoring, Aerospace Conference", PROCEEDINGS. 2003 IEEE, vol. 7, 2003, pages 3113 - 3121 |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101888309A (zh) * | 2010-06-30 | 2010-11-17 | 中国科学院计算技术研究所 | 在线日志分析方法 |
| CN101888309B (zh) * | 2010-06-30 | 2012-07-04 | 中国科学院计算技术研究所 | 在线日志分析方法 |
| US11215363B2 (en) * | 2017-04-24 | 2022-01-04 | Honeywell International Inc. | Apparatus and method for two-stage detection of furnace flooding or other conditions |
| JP2019057056A (ja) * | 2017-09-20 | 2019-04-11 | 株式会社東芝 | 事故原因分類装置 |
| CN112256750A (zh) * | 2020-09-27 | 2021-01-22 | 上海楚腾信息技术有限公司 | 一种数据纠错预警方法、系统及存储介质 |
| CN121658871A (zh) * | 2026-02-06 | 2026-03-13 | 西安现代控制技术研究所 | 基于注意力表征集成聚类的低成本飞行器运行支持方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2010092355A (ja) | 2010-04-22 |
| CN102112933A (zh) | 2011-06-29 |
| US20110191076A1 (en) | 2011-08-04 |
| EP2333629A4 (en) | 2013-01-02 |
| KR20110026515A (ko) | 2011-03-15 |
| JP5048625B2 (ja) | 2012-10-17 |
| EP2333629A1 (en) | 2011-06-15 |
| KR101316486B1 (ko) | 2013-10-08 |
| US8630962B2 (en) | 2014-01-14 |
| CN102112933B (zh) | 2014-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5048625B2 (ja) | 異常検知方法及びシステム | |
| JP5498540B2 (ja) | 異常検知方法及びシステム | |
| JP5538597B2 (ja) | 異常検知方法及び異常検知システム | |
| JP5301310B2 (ja) | 異常検知方法及び異常検知システム | |
| JP5363927B2 (ja) | 異常検知・診断方法、異常検知・診断システム、及び異常検知・診断プログラム | |
| JP5501903B2 (ja) | 異常検知方法及びそのシステム | |
| JP5431235B2 (ja) | 設備状態監視方法およびその装置 | |
| JP5740459B2 (ja) | 設備状態監視方法 | |
| JP5778305B2 (ja) | 異常検知方法及びそのシステム | |
| JP5364530B2 (ja) | 設備状態監視方法、監視システム及び監視プログラム | |
| JP5439265B2 (ja) | 異常検知・診断方法、異常検知・診断システム、及び異常検知・診断プログラム | |
| JP2011145846A (ja) | 異常検知方法、異常検知システム、及び異常検知プログラム | |
| JP5331774B2 (ja) | 設備状態監視方法およびその装置並びに設備状態監視用プログラム | |
| WO2013030984A1 (ja) | 設備状態監視方法およびその装置 | |
| JP2014056598A (ja) | 異常検知方法及びそのシステム | |
| JP2015181072A (ja) | 設備状態監視方法およびその装置 | |
| Shar et al. | A novel energy performance-based diagnostic model for centrifugal compressor using hybrid ML model | |
| JPWO2013030984A1 (ja) | 設備状態監視方法およびその装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| WWE | Wipo information: entry into national phase |
Ref document number: 200980130579.0 Country of ref document: CN |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09818896 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 20117002716 Country of ref document: KR Kind code of ref document: A |
|
| REEP | Request for entry into the european phase |
Ref document number: 2009818896 Country of ref document: EP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2009818896 Country of ref document: EP |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 13057831 Country of ref document: US |




