A method for perceiving the operating state of an intelligent dust collector driven by data

Abstract

The present invention relates to the technical field of data processing, and particularly relates to an intelligent dust collector operation state perception method driven by data, including: collecting monitoring data of the dust collector in real time; preprocessing the operation parameter monitoring data stream; respectively performing preliminary time window truncation processing on the operation data stream of each dimension to obtain the mother data stream of each dimension; respectively performing secondary time window truncation processing on the mother data stream of each dimension to obtain the sub-data stream of each dimension; calculating the operation fault possibility based on the similarity between the sub-data stream of each dimension and the corresponding mother data stream; and judging the operation state of the dust collector based on the operation fault possibility. In the present invention, the sub-data stream extracted by the sub-time window that flexibly changes according to the data change degree structure and data fluctuation degree of the data stream itself is used to perceive the operation state, which can effectively improve the accuracy of perception.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly relates to a method for perceiving the operating state of an intelligent dust collector driven by data. Background Art

[0002] As an important part of industrial environmental protection facilities, the technology for perceiving the operating state of dust collectors has developed rapidly in recent years. With the continuous improvement of industrial emission standards and the increasingly strict requirements for energy conservation, consumption reduction, and environmental protection, the technology for perceiving the operating state of dust collectors has become a key technology for improving the efficiency of dust collectors, reducing operating costs, and extending the service life of equipment. With the progress of information technology and sensing technology, the technology for perceiving the operating state of dust collectors has been significantly improved. Modern dust collectors generally adopt an online monitoring system to continuously monitor the operating state of the equipment through sensors.

[0003] A large amount of streaming data is generated during the operation of a dust collector. Usually, when the operating state is perceived by analyzing a fixed-size window of the streaming data, this method of a fixed window size cannot effectively adapt to the operating environment, and often there is a problem that the accuracy of the perceived result of the operating state of the dust collector is low due to unreasonable window size division. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a method for perceiving the operating state of an intelligent dust collector driven by data. To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a method for perceiving the operating state of an intelligent dust collector driven by data, the method comprising: continuously collecting monitoring data of the dust collector and converting it into a monitoring data stream of operating parameters, the monitoring data including monitoring data of operating parameters in at least two dimensions; preprocessing the monitoring data stream of operating parameters to obtain an operating data stream corresponding to each dimension; respectively performing a preliminary time window truncation process on the operating data stream of each dimension to obtain a mother data stream of each dimension within a mother time window, the duration of the mother time window being calculated from the degree of data change in each operating data stream, and the end time of the mother time window being the current time; respectively performing a secondary time window truncation process on the mother data stream of each dimension to obtain a sub-data stream of each dimension within a sub-time window, the duration of the sub-time window being calculated from the degree of data fluctuation in each mother data stream, and the end time of the sub-time window being the current time; calculating the possibility of an operating fault based on the similarity between the sub-data stream of each dimension and the corresponding mother data stream; and determining the operating state of the dust collector based on the possibility of the operating fault and a preset judgment threshold.

[0006] In combination with the first aspect, in a possible implementation manner, the method for calculating the duration of the mother time window includes: obtaining the cycle duration of the dust removal cycle; performing data extraction and processing within the running data stream based on the cycle duration, and extracting from the current moment forward by the cycle duration to obtain an initial running data stream, the duration of the initial running data stream being an integer multiple of the cycle duration; respectively calculating the sum of the change degrees of each value in each dust removal cycle within the initial running data stream to obtain a cumulative change degree value, each of the value change degrees being the difference between the data corresponding to one moment within a dust removal cycle and the corresponding data of the previous moment; calculating the duration of the mother time window by calculating the initial running data stream according to the cycle duration, all the cumulative change degree values, and a preset first duration calculation functional formula.

[0007] In combination with the first aspect, in a possible implementation manner, the method for calculating the duration of the sub - time window includes: obtaining the cycle duration of the dust removal cycle; calculating an initial coefficient based on the extreme value difference within the mother data stream of the target dimension, the target dimension being any dimension; calculating a similarity change correction coefficient based on the change similarity between all mother data streams within the mother time window of the target dimension; calculating a noise correction coefficient based on the noise intensity within the mother data stream of the target dimension; and correcting the cycle duration based on the noise correction coefficient, the similarity change correction coefficient, and the initial coefficient to obtain the duration of the sub - time window.

[0008] In combination with the first aspect, in a possible implementation manner, calculating an initial coefficient based on the extreme value difference within the mother data stream of the target dimension includes: screening out the maximum value within the mother data stream; calculating the difference by subtracting a preset index threshold from the maximum value; calculating the difference between adjacent extreme values within the mother data stream one by one; and calculating the initial coefficient based on all the differences between adjacent extreme values and the difference.

[0009] In combination with the first aspect, in a possible implementation manner, calculating a similarity change correction coefficient based on the change similarity between all mother data streams within the mother time window of the target dimension includes: processing the mother data streams of all dimensions based on a preset clustering algorithm to obtain a clustering result, the clustering object of the clustering algorithm being the data collected and pre - processed at each moment; extracting the clustering sub - clusters under the target dimension from the clustering result; calculating the Euclidean distance between every two of the clustering sub - clusters one by one and summing to obtain an accumulated Euclidean distance value; calculating the dynamic time warping distance between the mother data stream of the target dimension and the mother data streams of other dimensions respectively; and calculating the similarity change correction coefficient based on the accumulated Euclidean distance value and all the dynamic time warping distances.

[0010] In combination with the first aspect, in a possible implementation manner, the clustering algorithm is the k-means clustering algorithm.

[0011] In combination with the first aspect, in a possible implementation manner, calculating the noise correction coefficient according to the noise intensity in the mother data stream of the target dimension includes: decomposing the mother data stream of the target dimension based on a preset decomposition algorithm to obtain at least one intrinsic mode function component and a residual; respectively counting the maximum peak amplitude in each of the intrinsic mode function components; respectively calculating the variance of each of the intrinsic mode function components; and calculating the noise correction coefficient based on the variance and the maximum peak amplitude of each of the intrinsic mode function components.

[0012] In combination with the first aspect, in a possible implementation manner, the decomposition algorithm is the intrinsic time-scale decomposition algorithm.

[0013] In combination with the first aspect, in a possible implementation manner, the judgment threshold is 0.8.

[0014] In combination with the first aspect, in a possible implementation manner, the preprocessing includes data cleaning, data standardization, and normalization.

[0015] The present invention has the following beneficial effects:

[0016] In the present invention, first, a preliminary interception is performed according to the degree of data change in the running data stream to obtain a mother data stream. Then, a secondary interception is performed according to the data fluctuation state of this mother data stream to obtain a sub-data stream. Through the above method, the sub-data stream extracted by the sub-time window that flexibly changes according to the structural characteristics of the data itself is used to sense the running state, which can effectively improve the accuracy of sensing the running state. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a schematic flowchart of a method for sensing the running state of an intelligent dust collector driven by data provided by an embodiment of the present invention;

[0019] Figure 2 It is a schematic flowchart of step S3 provided by an embodiment of the present invention;

[0020] Figure 3The flowchart of step S4 provided by an embodiment of the present invention. Detailed implementation manners

[0021] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following combines the accompanying drawings and preferred embodiments to detail the specific implementation manners, structures, features and effects of a data-driven intelligent dust collector operation state perception method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0023] The following specifically describes the specific solution of a data-driven intelligent dust collector operation state perception method provided by the present invention with reference to the accompanying drawings.

[0024] Please refer to Figure 1 , which shows the flowchart of a data-driven intelligent dust collector operation state perception method provided by an embodiment of the present invention, specifically including steps S1 - S6.

[0025] S1. Real-time collect the monitoring data of the dust collector and convert it into an operation parameter monitoring data stream. The monitoring data includes operation parameter monitoring data in at least two dimensions.

[0026] In this step, a total of dimensions of operation parameter monitoring data are collected. For example, in this embodiment, the operation parameter monitoring data involved includes the air flow pressure, dust concentration, temperature, and air flow rate collected by different sensors. These parameters together constitute dimensions of data monitoring. That is, in this embodiment, specifically takes the value of 4. However, for those skilled in the art, some dimensions can be selected according to actual needs and specific situations, or other dimensions of operation data can be added. For example, in addition to the above four dimensions, parameters such as the motor running current and motor running voltage can also be added for monitoring, and a total of dimensions of operation parameter monitoring data are obtained. That is, regarding the specific selection of operation parameter monitoring data, those skilled in the art can flexibly decide according to actual application needs and specific situations, and no specific limitation is made in this embodiment.

[0027] S2. Preprocess the operation parameter monitoring data stream to obtain the operation data stream corresponding to each dimension.

[0028] The preprocessing involved in this stage includes but is not limited to data cleaning, data standardization, and normalization. Among them, data cleaning aims to ensure the accuracy and integrity of data by identifying and correcting errors, missing values, and outliers in the data. Data standardization refers to converting data according to established rules to make it conform to a specific format or standard for subsequent processing and analysis. Normalization is to scale the data to a range of 0 - 1 to eliminate the influence of different dimensions and magnitudes, ensuring fairness and consistency in the analysis process. Through these preprocessing steps, the data quality and the reliability of the analysis results can be significantly improved. That is, the operation data obtained after the above processing is a set of one-dimensional time-series data formed by the preprocessed data collected at each moment. It can also be called the time-history curve of each dimension in time series.

[0029] Furthermore, in this embodiment, considering that during the operation of the dust collector, regular ash cleaning operations are carried out, and the accumulation and regular cleaning of dust will affect the air flow pressure, dust concentration, temperature, and air flow rate. For example, during the normal operation of the dust collector, the air flow pressure increases with the increase in the dust accumulation on the filter bag. The accumulated dust will increase the resistance of the air flow, resulting in an increase in pressure. The periodic ash cleaning operation will reduce the dust accumulation on the filter bag, reduce the air flow resistance, and cause the air flow pressure to drop, and the system pressure will return to a lower level. Therefore, correspondingly, when the operation data streams of these dimensions fluctuate normally, they will all be within a certain change cycle.

[0030] Therefore, in this embodiment, the duration of one ash cleaning cycle of the dust collector is denoted as , and the change correlation between the monitoring data of the operation parameters of each dimension and the ash cleaning operation is analyzed one by one. When the correlation is stronger, it indicates that the periodicity of the change of this dimension is stronger. Therefore, in this embodiment, a mother time window containing fewer dust collector ash cleaning cycles can be first selected to ensure that the data stream in the mother time window is data during the normal operation of the dust collector, and then the subsequently intercepted data is compared with the normal operation data according to the corresponding time for similarity comparison, so as to achieve fault identification. The detailed steps can be seen in steps S3 - S5.

[0031] S3. Perform preliminary time window interception processing on the operation data stream of each said dimension to obtain the mother data stream of each said dimension within the mother time window. The duration of the mother time window is calculated from the degree of data change in each said operation data stream, and the end time of the mother time window is the current time.

[0032] Therefore, according to the above analysis, this embodiment believes that when analyzing the correlation between the operation data stream of each dimension and the change of the dust removal cycle, it should be observed whether there is a numerical change in the operation data stream of one dimension accompanied by the dust cleaning operation, and the degree of numerical change in multiple dust cleaning operations is similar. If the above change characteristics are met, it indicates that the data of this dimension has a higher correlation with the periodic change of the dust cleaning operation. Therefore, this feature is used to confirm the duration of the mother time window. See Figure 2 , Figure 2 It is shown in Figure 2 that step S3 in this step further includes steps S31 - S34.

[0033] S31. Obtain the cycle duration of the dust removal cycle .

[0034] S32. Based on the cycle duration, perform data extraction processing in the operation data stream. Extract from the current moment forward by the cycle duration to obtain the initial operation data stream, and the duration of the initial operation data stream is an integer multiple of the cycle duration.

[0035] Specifically, in this example, the duration of the initial operation data stream is 5 times the cycle duration. At the same time, those skilled in the art can select other multiples, and this embodiment does not make specific limitations on this.

[0036] S33. Calculate the sum of the degree of change of each value in each dust removal cycle in the initial operation data stream respectively to obtain the cumulative value of the degree of change. Each degree of change of value is the difference between the data corresponding to one moment in a dust removal cycle and the data corresponding to the previous moment.

[0037] S34. Calculate the duration of the mother time window for the initial operation data stream according to the cycle duration, all the cumulative values of the degree of change, and a preset first duration calculation function formula.

[0038] Specifically, the first duration calculation function formula mentioned in this step is as follows:

[0039] ;

[0040] Wherein, represents the duration of the mother time window of the th dimension; represents the cycle duration of the dust removal cycle; represents the multiple of the duration of the initial operation data stream to the dust removal interval duration. In this embodiment, ; represents the maximum - minimum normalization function; represents the total number of dimensions; represents in the initial operation data stream, at the Cumulative value of the degree of change in the nth dimension within the secondary dust removal interval; Represents the variance of the initial operating data stream.

[0041] In the above calculation formula, the variance Can express the degree of data dispersion within a dust removal cycle or between two ash cleaning operations, which can show the degree of data change from one aspect. At the same time, the cumulative value of the degree of change Can express the degree of data change within a dust removal cycle or between two ash cleaning operations, which can show the degree of data change from another aspect. Combining the data characteristic displays of the above two aspects can better reflect whether the operating data stream of a dimension has a numerical change accompanied by the ash cleaning operation, and the degree of numerical change in multiple ash cleaning operations is similar. If the above change characteristics are met, it indicates that the data of this dimension has a higher correlation with the periodic change of the ash cleaning operation, The larger the value of, the smaller the correlation, The smaller the value of.

[0042] At the same time, in this embodiment, it is also considered that during the operation of the dust collector, the faulty data often shows a section of abnormal data in order to perceive the operating state of the dust collector based on the sensor data obtained in real time. Therefore, it is also necessary to further segment the mother data stream of each dimension using a sub-time window. Therefore, see step S4 for details.

[0043] S4. Perform secondary time window truncation processing on the mother data stream of each dimension to obtain the sub-data stream of each dimension within the sub-time window. The duration of the sub-time window is calculated from the degree of data fluctuation within each mother data stream, and the cut-off moment of the sub-time window is the current moment.

[0044] Since, in this embodiment, it is considered that an overly large sub-time window will result in untimely feedback of the operating state, and too much interference data intercepted by the same sub-time window cannot accurately capture the faulty data, but an overly small sub-time window may cause an excessive burden on computing resources and affect the real-time performance of the system. Therefore, in this embodiment, it is necessary to adjust the duration of the sub-time window in real time according to the data change situation. Therefore, in this embodiment, the duration calculation of the sub-time window is demonstrated with one of the dimensions as an example. See Figure 3 Steps S41 - S44 shown in.

[0045] S41. Calculate the initial coefficient according to the extreme value difference within the mother data stream of the target dimension, where the target dimension is any dimension.

[0046] In this embodiment, it is considered that when the data changes relatively smoothly, the possibility that the data is faulty data is small. To save computing resources, a larger sub-time window should be selected; when the data changes greatly, it may indicate that the dust collector has a running fault. At this time, a smaller sub-time window needs to be selected to better obtain the fault status data of the dust collector. Specifically, in this step S4, it also includes the process of calculating the initial duration shown in steps S411 - S414.

[0047] S411. Screen out the maximum value in the mother data stream.

[0048] S412. Calculate the difference by subtracting the preset index threshold from the maximum value.

[0049] S413. Calculate the difference between adjacent extreme values in the mother data stream one by one.

[0050] S414. Calculate the initial coefficient based on all the differences between adjacent extreme values and the difference.

[0051] Specifically, the calculation functional formula of the initial coefficient in this step is as follows:

[0052] ;

[0053] Among them, represents the initial coefficient of the sub-time window in the th dimension; represents the number of extreme values in the mother data stream in the th dimension; represents the difference between the th extreme value and the th extreme value in the mother data stream in the th dimension; represents the absolute value calculation function; represents the difference between the maximum value in the mother data stream in the th dimension and the preset index threshold; represents the natural constant.

[0054] In this embodiment, it is considered that for the determination of the sub-time window, when in a certain dimension, the greater the difference between the data at each moment represented, the greater the change in the data of this dimension, and the larger the sub-time window corresponding to this data dimension should be, so as to avoid the selected sub-window being too small to capture important features during fault identification. At the same time, in this embodiment, it is also considered that when a certain value in a mother data stream is closer to the preset index threshold of this dimension, the possibility of the dust collector having a fault is greater. It should also be noted that the preset index threshold of a dimension mentioned in this embodiment can usually be set to ten times greater than the average value of this dimension.

[0055] S42. Calculate a similarity change correction coefficient based on the change similarity between all the mother data streams in the mother time window of the target dimension.

[0056] In this embodiment, when the data change curves of multiple dimensions are relatively similar, the information repetition of these dimensions may be high, and the sub-window corresponding to the corresponding data dimension should be smaller to avoid considering too much repeated information during fault identification. Therefore, the similarity between multiple dimensions is also considered in this application. Therefore, in order to clarify the calculation process of the similarity change correction coefficient, see steps S421-S425 for details.

[0057] S421. Processing the parent data streams of all the dimensions based on a preset clustering algorithm to obtain clustering results, wherein the clustering object of the clustering algorithm is the data collected and pre-processed at each moment.

[0058] Specifically, the clustering algorithm used in this embodiment is the k-means clustering algorithm, which is an algorithm widely used in the field of data mining and statistical analysis. It distributes data points to k clusters through an iterative process so that the sum of the squares of the distances between each data point and the center point of the cluster to which it belongs is minimized. However, those skilled in the art should understand that in addition to the k-means clustering algorithm, there are many other types of clustering algorithms to choose from, such as hierarchical clustering, DBSCAN, spectral clustering, etc. Those skilled in the art may also use other clustering algorithms, which are not specifically limited in this embodiment.

[0059] S422: extracting cluster subclusters under the target dimension from the clustering results.

[0060] S423, calculating the Euclidean distance between every two of the cluster sub-clusters one by one, and summing them up to obtain an accumulated Euclidean distance value.

[0061] S424, respectively calculating the dynamic time warping distances of the mother data streams of the target dimension and other dimensions.

[0062] In this step, the dynamic time warping distance is calculated by a dynamic time warping algorithm (DTW for short), which is an algorithm for measuring the similarity between two time series (data series with a time order). Since it is a prior art, the specific calculation process is not described in detail in this embodiment.

[0063] S425. Calculate a similarity change correction coefficient based on the Euclidean distance accumulation value and all the dynamic time warping distances.

[0064] Specifically, the calculation functional formula of the similarity change correction coefficient in this step is as follows:

[0065] ;

[0066] Among them, represents the similarity change correction coefficient of the th dimension; represents the accumulated value of the Euclidean distance; represents the total number of dimensions; represents the dynamic time warping distance between the mother data stream of the th dimension and the mother data stream of the th dimension.

[0067] In the above calculation formula, when the similarity of data changes between the target dimension and other dimensions, when the data change curves of the target dimension and other dimensions are relatively similar, the information redundancy shown by the two dimensions may be relatively high, and the corresponding sub-time window of the data dimension should be smaller. Therefore, it can better reflect the higher the similarity, the smaller the similarity change correction coefficient, and the smaller the corresponding sub-time window. represents the influence degree of the target dimension under other dimensions. When the information redundancy shown by the target dimension and other dimensions is relatively high, the Euclidean distance between the corresponding clustering sub-clusters should be smaller, the corresponding similarity change correction coefficient is smaller, and the corresponding sub-window of the data dimension is smaller.

[0068] S43. Calculate the noise correction coefficient according to the noise intensity in the mother data stream of the target dimension.

[0069] At the same time, considering that the data collected by the sensor may generate some noise signals due to the unique mode in the operation process of the dust collector. For example, when the fan starts and stops, the air flow pressure may fluctuate short-term, resulting in noise in the data. Therefore, the degree of noise signals contained in the data of a certain dimension should also be monitored in real time. When a certain data contains more noise signals, the corresponding sub-window of the data dimension should be smaller. Therefore, to clarify the calculation process of the noise correction coefficient, see steps S431 - S434 for details.

[0070] S431. Decompose the mother data stream of the target dimension based on a preset decomposition algorithm to obtain at least one intrinsic mode function component and a residual.

[0071] In this embodiment, the decomposition algorithm adopted is the Intrinsic Time-Scale Decomposition (ITD) algorithm. In this embodiment, the ITD algorithm can adaptively decompose the mother data stream into a series of Intrinsic Mode Functions (IMFs) and a residual (Trend). These IMF components can more intuitively reflect the data changes of the mother data stream at a certain frequency.

[0072] S432. Statistically calculate the maximum peak amplitude in each of the said intrinsic mode function components respectively.

[0073] S433. Calculate the variance of each of the said intrinsic mode function components respectively.

[0074] S434. Calculate the noise correction coefficient based on the variance and the maximum peak amplitude of each of the said intrinsic mode function components.

[0075] Specifically, the calculation functional formula of the noise correction coefficient in this step is as follows:

[0076] ;

[0077] Wherein, represents the noise correction coefficient of the th dimension; total represents the total number of intrinsic mode function components; represents the maximum peak amplitude of the intrinsic mode function component of the th dimension; represents the average value of the variances of all intrinsic mode function components.

[0078] In this embodiment, considering that noise usually appears as high-frequency and random fluctuations. Therefore, higher-order intrinsic mode function components are more likely to contain noise. Noise usually increases the volatility of the signal, resulting in an increase in the amplitude or an irregular frequency of these intrinsic mode function components. At the same time, noise usually causes an increase in the variance of the signal. When the variance of the intrinsic mode function component is larger, the intensity of the noise is higher. Therefore, in the above calculation formula can better describe the variance change of the intrinsic mode function component. At the same time, assigns a larger weight to higher-order intrinsic mode function components, and combined with the maximum peak amplitude of this intrinsic mode function component, it can better reflect the noise intensity of the mother data stream.

[0079] S44. Modify the cycle duration based on the noise correction coefficient, the similarity change correction coefficient, and the initial coefficient to obtain the duration of the sub-time window.

[0080] Specifically, the calculation function of the duration of the sub-time window in this step is as follows:

[0081] ;

[0082] Among them, represents the duration of the sub-time window corresponding to the th dimension; represents the cycle duration of the dust removal cycle; represents the maximum-minimum normalization function; represents the initial coefficient of the sub-time window of the th dimension; represents the similarity change correction coefficient of the th dimension; represents the noise correction coefficient of the th dimension.

[0083] By combining the coefficients in terms of noise, similarity between different dimensions, and characteristics of the data itself, a sub-data stream length with appropriate amount of information can be better intercepted.

[0084] S5. Calculate the likelihood of operating failure based on the similarity between the sub-data stream of each dimension and the corresponding mother data stream.

[0085] Furthermore, in this embodiment, in order to determine the similarity between the sub-data stream of a dimension and the mother data stream. In this embodiment, the Dynamic Time Warping (DTW) algorithm is used to compare their similarity. And in this embodiment, the duration of the sub-time window of each dimension is used as the weight of that dimension, and when the sub-time window of that dimension is longer, the more useful information it contains, and higher reference should be provided in the similarity comparison. At the same time, in this embodiment, it is also considered that when comparing the sub-data stream with the mother data stream, the position of the sub-data stream in a dust cleaning cycle should be found, and the data at the corresponding position of the sub-data stream should be compared with the data at the corresponding position of the mother data stream. When the similarity is higher, it indicates that the likelihood of operating failure of the dust collector at this time is lower. That is, first, the mother data stream should be divided into several data stream segments according to the duration of the dust removal cycle, and then, the similarity between the sub-data stream and each other comparison data stream is compared respectively, and the sum of their similarities is accumulated to obtain the similarity between the sub-data stream and the mother data stream, where each comparison data stream is extracted according to the time period of the sub-data stream in a dust removal cycle in other data stream segments.

[0086] Therefore, based on the above analysis, the calculation function of the likelihood of operating failure in this step is as follows:

[0087] ;

[0088] Among them, represents the possibility of operation failure; represents the maximum-minimum normalization function; represents the total number of dimensions; represents the duration of the sub-time window corresponding to the th dimension; represents the duration of the parent time window of the th dimension; represents the dust removal interval duration of the dust collector; represents the DTW distance between the sub-data stream of the th dimension and the th comparison data stream in the parent data stream. Among them, the th comparison data stream is obtained by intercepting the th dust removal cycle in the parent data stream according to the duration of the sub-data stream. The cut-off moment of the comparison data stream is equal to the cut-off moment of the th dust removal cycle in the parent data stream.

[0089] S6. Determine the operation state of the dust collector based on the determined operation failure possibility.

[0090] Therefore, in this embodiment, the operation state of the dust collector can be determined whether it is faulty by the value of the operation failure possibility and a preset judgment threshold. For example, the judgment threshold in this embodiment is 0.8. Then, when the operation failure possibility is greater than 0.8. It is considered that the dust collector is diagnosed as faulty. Once a fault is diagnosed, a fault tolerance control strategy needs to be adopted to maintain the normal operation of the dust collector.

[0091] In this embodiment, for the large amount of streaming data generated during the operation of the dust collector, the prior art usually adopts the method of a fixed time window size, and analyzes the streaming data to realize the perception of the operation state. However, this method of a fixed window size has an adverse effect on the accuracy of the operation perception result. In each judgment, the rationality of the window size cannot be guaranteed, which leads to a decrease in the accuracy of the perception result of the operation state of the dust collector. Therefore, in this embodiment, first, the parent data stream is initially intercepted according to the data change degree of the operation data stream. Subsequently, secondary interception is performed based on the data fluctuation state of the parent data stream to obtain the sub-data stream. During the interception process of the sub-data stream, the characteristics of the operation data stream of each dimension, the similarity between different dimensions, and the internal noise intensity of each dimension are also fully considered, and the window size is adjusted accordingly. Finally, the operation state is perceived through the sub-data stream extracted by the above-mentioned flexible and variable sub-time window, which can significantly improve the accuracy of perception.

[0092] It should be noted that the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

Claims

1. A data-driven intelligent dust collector operating status perception method, characterized in that: The method comprises: Collect monitoring data of the dust collector in real time and convert it into an operating parameter monitoring data stream, wherein the monitoring data includes operating parameter monitoring data of at least two dimensions; Preprocessing the operating parameter monitoring data stream to obtain an operating data stream corresponding to each dimension; Performing preliminary time window interception processing on the running data stream of each dimension respectively, to obtain the mother data stream of each dimension in the mother time window, the duration of the mother time window is calculated by the degree of data change in each running data stream, and the end time of the mother time window is the current time; Among them, the duration calculation method of the mother time window is as follows: obtain the cycle duration of the dust removal cycle; perform data extraction and processing in the operation data stream based on the cycle duration, extract from the current moment forward through the cycle duration, and obtain the initial operation data stream, and the duration of the initial operation data stream is an integer multiple of the cycle duration; respectively calculate the sum of the degree of change of each numerical value in each dust removal cycle in the initial operation data stream to obtain the cumulative value of the degree of change, and the degree of change of each numerical value is the difference between the data corresponding to one moment in a dust removal cycle and the corresponding data at the previous moment; calculate the initial operation data stream according to the cycle duration, all cumulative values ​​of the degree of change and the preset first duration calculation function to obtain the duration of the mother time window; Performing a secondary time window interception process on the mother data stream of each dimension respectively, to obtain a sub-data stream in a sub-time window of each dimension, wherein the duration of the sub-time window is calculated by the degree of data fluctuation in each mother data stream, and the end time of the sub-time window is the current time; The duration calculation method of the sub-time window is as follows: the initial coefficient is calculated based on the extreme value difference in the mother data stream of the target dimension, and the target dimension is any dimension; the similarity change correction coefficient is calculated based on the change similarity between all mother data streams in the mother time window of the target dimension; the noise correction coefficient is calculated based on the noise intensity in the mother data stream of the target dimension; the period duration is corrected based on the noise correction coefficient, the similarity change correction coefficient, and the initial coefficient to obtain the duration of the sub-time window; Obtaining the possibility of operation failure based on similarity calculation between the sub-data streams of each dimension and the corresponding parent data stream; The operating state of the dust collector is determined based on the operating failure possibility and a preset judgment threshold.

2. The data-driven intelligent dust collector operating status perception method according to claim 1 is characterized in that: The initial coefficients are calculated based on the extreme value differences in the mother data stream of the target dimension, including: Screening the mother data stream to obtain a maximum value; A difference is calculated based on the maximum value and a preset indicator threshold to obtain a difference; Calculating the differences between adjacent extreme values ​​in the mother data stream one by one; The initial coefficients are calculated based on the differences between all adjacent extreme values ​​and the difference.

3. The data-driven intelligent dust collector operating status perception method according to claim 1 is characterized in that: The similarity change correction coefficient is calculated based on the change similarity between all the mother data streams in the mother time window of the target dimension, including: Processing the parent data streams of all the dimensions based on a preset clustering algorithm to obtain clustering results, wherein the clustering object of the clustering algorithm is the data collected and preprocessed at each moment; Extracting cluster subclusters under the target dimension from the clustering results; Calculating the Euclidean distance between each two of the cluster subclusters one by one, and summing them up to obtain the accumulated Euclidean distance value; Calculate the dynamic time warping distance of the target dimension and the mother data stream of the other dimensions respectively; A similarity change correction coefficient is calculated based on the Euclidean distance accumulation value and all the dynamic time warping distances.

4. The data-driven intelligent dust collector operating status perception method according to claim 3 is characterized in that: The clustering algorithm is a k-means clustering algorithm.

5. The data-driven intelligent dust collector operating status perception method according to claim 1 is characterized in that: The noise correction coefficient is calculated according to the noise intensity in the mother data stream of the target dimension, including: Performing data decomposition on the mother data stream of the target dimension based on a preset decomposition algorithm to obtain at least one intrinsic mode function component and a residual; Counting the maximum peak amplitude of each intrinsic mode function component respectively; Calculating the variance of each of the intrinsic mode function components respectively; The noise correction coefficient is calculated based on the variance and the maximum peak amplitude of each of the intrinsic mode function components.

6. The data-driven intelligent dust collector operating status perception method according to claim 5 is characterized in that: The decomposition algorithm is an intrinsic time scale decomposition algorithm.

7. The data-driven intelligent dust collector operating status perception method according to claim 1 is characterized in that: The judgment threshold is 0.

8.

8. The data-driven intelligent dust collector operating status perception method according to claim 1 is characterized in that: The preprocessing includes data cleaning, data standardization and normalization.

Images