Forestry waste crushing device and method

Abstract

The invention relates to the technical field of clustering data processing, in particular to a forestry waste crushing device and method.The forestry waste crushing method comprises the steps that multiple vibration data time periods and power data time periods of a crusher are obtained, and multiple class clusters are obtained through clustering; according to the fault possibility of each vibration data time period in the class cluster, screening to obtain an abnormal time period, and based on the fault possibility of each abnormal time period and the association between each abnormal time period and each non-abnormal time period in the class cluster, obtaining the fault degree of each abnormal time period; according to the method, the fault probability of the pulverizer is obtained based on the fault degree and duration of each abnormal time period, the duration of the adjacent abnormal time period of each abnormal time period and the time interval between the adjacent abnormal time period, fault monitoring is not carried out singly according to vibration data of the pulverizer, and different types of forestry wastes are considered; analysis is carried out from multiple aspects, the accurate fault probability of the pulverizer is obtained, and the accuracy of pulverizer fault monitoring is improved.

Description

Technical Field

[0001] The invention relates to the technical field of cluster data processing, and in particular to a forestry waste crushing device and method. Background Art

[0002] Forestry waste mainly includes a type of lignocellulosic waste composed of forest logging residues, wood processing residues and afforestation pruning residues. It is a potential resource with great potential. By crushing forestry waste, forestry waste can be used for garden covering, and inorganic or organic matter can be regularly spread on the soil, which not only protects the soil but also improves the soil quality. Therefore, crushing is a key step in the treatment of forestry waste. In the process of forestry waste treatment, the crusher is the main equipment for crushing forestry waste. If the crusher fails and is not detected in time, it will cause production interruption. Once production is interrupted, subsequent production links will be affected, causing the entire production line to stagnate.

[0003] When the existing technology performs fault monitoring on the pulverizer used for forestry waste, it usually directly analyzes the initial vibration data of the pulverizer and determines whether the pulverizer has a fault based on the analysis results. However, in the actual forestry waste treatment environment, forestry wastes vary in hardness, toughness, size and other aspects. The vibration modes generated by forestry wastes with different characteristics during the pulverization process vary greatly. If the differences between forestry wastes are not considered and fault monitoring is performed directly based on the vibration data of the pulverizer, the accuracy of the pulverizer fault monitoring will be affected. Summary of the invention

[0004] In order to solve the technical problem that the fault monitoring accuracy of the existing pulverizer used for forestry waste is low, the purpose of the present invention is to provide a forestry waste pulverizing device and method, and the technical solution adopted is as follows: In a first aspect of the present invention, a method for comminuting forestry waste is provided, comprising: Acquire multiple vibration data periods and power data periods of the pulverizer, and cluster the vibration data periods to obtain multiple clusters; The abnormal period is screened out based on the possibility of failure in each vibration data period in the cluster. The possibility of failure is obtained by the clustering effect of the cluster and the correlation between the vibration data period in the cluster and the power data period in the same period. Based on the possibility of failure in each abnormal period and the association between each abnormal period and each non-abnormal period in the cluster, the degree of failure in each abnormal period is obtained; Based on the fault degree and duration of each abnormal period, the duration of each abnormal period adjacent to the abnormal period, and the time interval between adjacent abnormal periods, the failure probability of the crusher is obtained.

[0005] In an exemplary embodiment, the process of obtaining the fault degree includes: According to the time interval and characteristic value difference between the first abnormal period and each non-abnormal period in the cluster, combined with the possibility of failure of the first abnormal period and each non-abnormal period in the cluster, the initial failure degree of the first abnormal period is obtained; the first abnormal period is any abnormal period; The fault degree of the first abnormal period is obtained according to the initial fault degree of the first abnormal period and the duration ratio of the first abnormal period.

[0006] In an exemplary embodiment, the process of obtaining the initial fault degree of the first abnormal period includes: According to the first time interval, the first eigenvalue difference, and the possibility of failure in the first non-abnormal period, a fault influence sub-coefficient of the first non-abnormal period on the first abnormal period is obtained; the fault influence sub-coefficient is inversely proportional to the first time interval, proportional to the first eigenvalue difference, and inversely proportional to the possibility of failure in the first non-abnormal period; the first non-abnormal period is any non-abnormal period in the cluster where the first abnormal period is located; the first time interval is the time interval between the first abnormal period and the first non-abnormal period, and the first eigenvalue difference is the eigenvalue difference between the first abnormal period and the first non-abnormal period; The fault influence sub-coefficients of each non-abnormal period in the cluster where the first abnormal period is located on the first abnormal period are integrated to obtain the fault influence coefficient of the first abnormal period; According to the fault influence coefficient of the first abnormal period and the possibility of the fault occurring in the first abnormal period, the initial fault degree of the first abnormal period is obtained.

[0007] In an exemplary embodiment, the process of obtaining the failure probability of a pulverizer includes: According to the duration of the adjacent abnormal time periods of each abnormal time period, and the time interval between each abnormal time period and the adjacent abnormal time period, the fault probability impact weight related to each abnormal time period is obtained; the fault probability impact weight is proportional to the duration of the adjacent abnormal time period of the abnormal time period, and inversely proportional to the time interval between the abnormal time period and the adjacent abnormal time period; Based on the failure probability impact weights associated with each abnormal period, the failure degree of each abnormal period is weighted and summed, and combined with the mean of the time interval between any two abnormal periods, the failure probability of the crusher is obtained.

[0008] In an exemplary embodiment, the process of obtaining the possibility of a fault occurrence includes: Obtaining a correlation coefficient between a first vibration data period in a first cluster and a power data period in the same period; the first cluster is any cluster, and the first vibration data period is any vibration data period in the first cluster; The clustering effect of the first cluster is used as the fault occurrence credibility of the first cluster, and the fault occurrence possibility of the first vibration data period is obtained by combining the correlation coefficient of the first vibration data period and the power data period of the same period.

[0009] In an exemplary embodiment, the process of obtaining the clustering effect of the first cluster includes: obtaining the clustering effect of the first cluster according to the average value of the similarity between any two vibration data time periods in the first cluster and the silhouette coefficient of the first cluster.

[0010] In an exemplary embodiment, before acquiring a plurality of vibration data periods and power data periods of the pulverizer, the forestry waste pulverizing method further includes: Acquire a vibration data sequence and a power data sequence of a pulverizer; The STL algorithm is used to decompose the vibration data sequence and obtain the trend term curve; The APCA segmentation method is used to segment the trend item curve, and based on the segmentation points, the vibration data series and the power data series are segmented to obtain multiple vibration data periods and power data periods respectively.

[0011] In an exemplary embodiment, the vibration data time periods are clustered to obtain a plurality of clusters, including: Obtain the difference of the APCA approximation values ​​of any two trend item periods, where the trend item periods are obtained by segmenting the trend item curve using the APCA segmentation method; The K-means clustering algorithm is used to cluster the trend item periods with the difference of APCA approximation between any two trend item periods as the clustering distance. Based on the relationship between the trend item period and the vibration data period, multiple clusters are obtained, which include multiple vibration data periods.

[0012] In an exemplary embodiment, the forestry waste comminution method further comprises: comparing the failure probability of the pulverizer with a preset failure threshold; If the failure probability of the crusher is greater than the preset failure threshold, a crusher shutdown command is output.

[0013] In a second aspect of the present invention, a forestry waste crushing device is provided, comprising: a memory and a processor; the memory is connected to the processor; the memory is used to store program instructions; the processor is used to implement the above-mentioned forestry waste crushing method when the program instructions are executed.

[0014] The present invention has the following beneficial effects: clustering the vibration data time periods of the pulverizer to obtain multiple clusters, the crushed forestry waste in each cluster is similar, so as to better reflect the similarity between the changes in the vibration data of the pulverizer in different time periods, and improve the accuracy of the pulverizer fault detection; then, focusing on the analysis of the abnormal time periods obtained by screening, since there is a certain correlation between each abnormal time period and each non-abnormal time period in the cluster, this correlation is related to the degree of fault of each abnormal time period, therefore, based on this, the degree of fault of each abnormal time period is obtained, and a single abnormal time period is used as the analysis object to obtain the accurate degree of fault of each abnormal time period, and finally, since there is a certain correlation between the abnormal time period and the adjacent abnormal time period, whether it is the duration of the abnormal time period or the time interval, this correlation affects the failure probability of the pulverizer. Therefore, the technical solution provided by the present invention does not perform fault monitoring based on the vibration data of the pulverizer alone, but takes into account different types of forestry waste, analyzes from multiple aspects, obtains the accurate probability of failure of the pulverizer, and improves the accuracy of monitoring the pulverizer fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a forestry waste crushing method provided by one embodiment of the present invention; Figure 2 A flowchart of the steps of the forestry waste comminution method provided by the present invention is provided in one embodiment of the present invention; Figure 3 It is a curve schematic diagram obtained by decomposing the vibration data fitting curve using the STL algorithm provided by an embodiment of the present invention; Figure 4 is a specific clustering process diagram provided by an embodiment of the present invention; Figure 5 is a flowchart for obtaining the possibility of a fault occurring provided by an embodiment of the present invention; Figure 6 is a flowchart of obtaining the fault degree provided by an embodiment of the present invention; Figure 7 is a flow chart for obtaining the initial fault degree of the first abnormal period provided by an embodiment of the present invention; Figure 8 is a flow chart for obtaining the failure probability of a pulverizer provided by an embodiment of the present invention; Fig. 9 It is a flow chart of the steps included in the forestry waste crushing method provided by the present invention, provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs. The data and information collected in this application have been obtained with full consent and authorization, and the collection, use and processing of relevant information must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0018] The present embodiment provides an application scenario of a forestry waste crushing method: there are many types of forestry wastes, and there are large differences in hardness, toughness, size, etc. During the crushing process, wastes of different properties have different vibration trends. Therefore, the vibration data in several time periods of the same vibration trend are analyzed for abnormality to improve the accuracy of crusher fault monitoring.

[0019] like Figure 1 As shown, the forestry waste comminution method comprises the following steps: Step S1: obtaining multiple vibration data periods and power data periods of a pulverizer, and clustering the vibration data periods to obtain multiple clusters; Step S2: Screening out abnormal periods based on the probability of failure in each vibration data period in the cluster, where the probability of failure is obtained by the clustering effect of the cluster and the correlation between the vibration data period in the cluster and the power data period in the same period; Step S3: based on the possibility of failure in each abnormal period and the association between each abnormal period and each non-abnormal period in the cluster, the degree of failure in each abnormal period is obtained; Step S4: Based on the fault degree and duration of each abnormal period, the duration of the adjacent abnormal period of each abnormal period, and the time interval between the adjacent abnormal periods, the failure probability of the crusher is obtained.

[0020] Each step is described in detail below with reference to the accompanying drawings.

[0021] Step S1: Acquire multiple vibration data time periods and power data time periods of a pulverizer, and cluster the vibration data time periods to obtain multiple clusters.

[0022] Acquire multiple vibration data periods and power data periods of the pulverizer. The multiple vibration data periods and power data periods of the pulverizer may be processed in advance and directly used in step S1 of the present invention. In an exemplary embodiment, before acquiring multiple vibration data periods and power data periods of the pulverizer, as Figure 2 As shown, the forestry waste crushing method provided by the present invention also includes: Step S1-1: Acquire the vibration data sequence and power data sequence of the pulverizer.

[0023] This embodiment takes a blade crusher as an example for fault detection. A blade crusher usually cuts and crushes forestry waste by a blade installed on a rotating shaft. Therefore, the rotating shaft is the core component of the crusher. By transmitting the power of the motor to the crushing tool, the tool can rotate at high speed to crush forestry waste. In the structure of the forestry waste crusher, the rotating shaft is a relatively easy position to install a vibration sensor. Compared with some other complex components, the surface of the rotating shaft is relatively regular, which is convenient for fixing the sensor and can ensure good contact between the vibration sensor and the rotating shaft, so as to obtain accurate vibration data. Since the vibration of the crusher is caused by the rotation of the rotating shaft, as another embodiment, the vibration sensor can also be fixed on the body of the crusher, and the vibration data of the crusher body obtained by detection is used to characterize the vibration of the rotating shaft. In addition, a power sensor is also provided at the power supply end of the crusher to detect the operating power of the crusher.

[0024] The vibration sensor and power sensor of the pulverizer have the same acquisition frequency and both collect data synchronously. In an exemplary embodiment, the acquisition frequency is once every 5 seconds.

[0025] This embodiment also sets a monitoring time period. In the subsequent processing, the vibration data and power data of the crusher in the monitoring time period are analyzed to monitor the crusher fault. The end time of the monitoring time period is the current time, and the length of the monitoring time period is set according to the actual situation. Then, the monitoring time period includes multiple sampling moments, and each sampling moment corresponds to a vibration data and a power data. The vibration data collected at each moment is arranged in time sequence to obtain the vibration data sequence of the crusher, and the power data collected at each moment is arranged in time sequence to obtain the power data sequence of the crusher.

[0026] Step S1-2: Decompose the vibration data sequence using the STL algorithm to obtain a trend term curve.

[0027] In the scenario of forestry waste shredders, the vibration data of the shredders is affected by many factors, including long-term trend changes caused by changes in the operating status of the equipment itself, as well as some periodic or random interference factors. By extracting trend items through the STL (Seasonal-Trend decomposition using Loess) algorithm, we can focus on the long-term change pattern in the vibration data, remove interference such as short-term fluctuations, seasonality, and random noise, and more clearly understand the basic change trend of rotating shaft vibration.

[0028] Therefore, the STL algorithm is used to decompose the vibration data sequence to obtain the trend term curve. Before the STL algorithm is used to decompose the vibration data sequence, the vibration data sequence is curve fitted to obtain the vibration data fitting curve. The horizontal axis of the vibration data fitting curve is the time axis, and the vertical axis is the amplitude axis of the vibration data. In addition to obtaining the trend term curve, the STL algorithm is used to decompose the vibration data fitting curve. The period term and the residual term are also obtained, such as Figure 3 As shown, the curves from bottom to top are: residual term curve, period term curve, trend term curve and vibration data fitting curve. Figure 3 The horizontal axis is time, and the vertical axis is the amplitude of the corresponding data.

[0029] The vibration trend of the rotating shaft of the pulverizer may have different change rates and characteristics at different stages. For example, when crushing harder branches, the vibration trend of the rotating shaft may be continuous and large-amplitude fluctuations, and the corresponding characteristic value will reflect this higher vibration level; while when crushing softer leaves, the vibration trend is relatively gentle, and the corresponding characteristic value will be lower.

[0030] Step S1-3: Use the APCA segmentation method to segment the trend item curve, and segment the vibration data sequence and the power data sequence based on the segmentation points to obtain multiple vibration data time periods and power data time periods respectively.

[0031] The trend item curve is segmented using the APCA (Adaptive Piecewise Constant Approximation) segmentation method to obtain several trend item time periods and the APCA approximation of each trend item time period. The APCA approximation can be the data mean in the trend item time period. Accordingly, multiple segmentation points are obtained. Then, based on the multiple segmentation points obtained, the vibration data sequence is segmented to obtain multiple vibration data time periods, and the power data sequence is segmented to obtain multiple power data time periods. Therefore, the number of vibration data time periods is the same as that of power data time periods, and each vibration data time period corresponds to each power data time period one by one.

[0032] By segmenting and obtaining the APCA approximate value through the APCA segmentation method, the entire continuous trend change can be divided into multiple representative time periods according to the inherent characteristic differences, which is convenient for capturing the characteristics of vibration trends in different stages in more detail.

[0033] Then, the vibration data time periods are clustered to obtain multiple clusters. In an exemplary embodiment, since the APCA approximation of each trend item time period is obtained as described above, then, Figure 4 As shown, a specific clustering process is given as follows: Step S1-4: Obtain the difference between the APCA approximations of any two trend item periods.

[0034] The difference of the APCA approximate values ​​between any two trend item time periods is obtained, wherein the difference of the APCA approximate values ​​is specifically the absolute value of the difference of the APCA approximate values.

[0035] Step S1-5: Using the K-means clustering algorithm, the difference in the APCA approximation between any two trend item periods is used as the clustering distance to cluster the trend item periods, and based on the relationship between the trend item periods and the vibration data periods, multiple clusters are obtained, and the clusters include multiple vibration data periods.

[0036] This embodiment adopts K-means clustering algorithm for clustering, wherein the K value is set according to the actual clustering needs. The difference of the APCA approximation values ​​of any two trend item time periods is used as the clustering distance to cluster the trend item time periods, thereby obtaining multiple clusters. Since there is a one-to-one correspondence between each trend item time period and each vibration data time period, after obtaining each cluster corresponding to the trend item time period, the trend item time period in each cluster is replaced with the vibration data time period of the same period, and multiple clusters corresponding to the vibration data time period are obtained, each cluster includes multiple vibration data time periods, and clustering of the vibration data time periods is realized.

[0037] Therefore, the K-means clustering algorithm is used to cluster the trend item time periods, which is essentially clustering the time periods corresponding to the trend item time periods, and the obtained clusters include multiple time periods. Thus, according to the temporal correspondence between the trend item time periods and the vibration data time periods, a cluster including multiple vibration data time periods is obtained.

[0038] Correspondingly, according to the one-to-one correspondence between the vibration data time periods and the power data time periods, a plurality of power data time periods included in each cluster are obtained.

[0039] Since the vibration trend is determined by the nature of the forestry waste being shredded, the shredded forestry waste in each cluster has similarities in hardness, toughness, size, etc.

[0040] Step S2: Screening out abnormal time periods based on the possibility of failure in each vibration data time period in the cluster, where the possibility of failure is obtained by the clustering effect of the cluster and the correlation between the vibration data time period in the cluster and the power data time period in the same period.

[0041] The clustering process in step S1 only considers the change of the trend item. This step determines whether the clustering division is appropriate based on the fluctuation difference between the vibration data. Whether the clustering division is appropriate is characterized by the clustering effect of each cluster.

[0042] The first cluster is set as any cluster, and based on the clustering process in step S1, the silhouette coefficient of the first cluster is obtained. The larger the silhouette coefficient, the better the clustering effect of the first cluster. Moreover, since the first cluster includes multiple vibration data time periods, the similarity of any two vibration data time periods in the first cluster is obtained. The specific implementation method of the similarity is determined by actual needs. In an exemplary embodiment, the DTW (Dynamic Time Warping Distance) distance of any two vibration data time periods in the first cluster is obtained, and then the DTW distance is negatively correlated and normalized to obtain the similarity of any two vibration data time periods in the first cluster. The negative correlation normalization method here can be: , where exp is an exponential function with the natural constant e as the base, and x is the object that needs negative correlation normalization.

[0043] Then, the average value of the similarity between any two vibration data periods in the first cluster is calculated. Then, the higher the average value of the similarity is, the better the clustering effect of the first cluster is.

[0044] Therefore, according to the average value of the similarity of any two vibration data time periods in the first cluster and the silhouette coefficient of the first cluster, the clustering effect of the first cluster is obtained. The clustering effect is proportional to the silhouette coefficient and proportional to the average value of the similarity. In an exemplary embodiment, in order to facilitate subsequent calculations, the silhouette coefficient of the first cluster is normalized. Since the numerical range of the silhouette coefficient of the first cluster is [-1, 1], the silhouette coefficient of the first cluster is added with a value of 1 and then divided by 2, so that the numerical range of the result is [0, 1], and normalization is achieved.

[0045] The product of the average value of the similarities between any two vibration data periods in the first cluster and the normalized silhouette coefficient of the first cluster is taken as the clustering effect of the first cluster.

[0046] The hardness characteristics of forestry waste will affect both the vibration of the rotating shaft and the power of the crusher. Hard materials will increase power consumption and vibration amplitude. The power is also affected by voltage fluctuations and motor failures, that is, the voltage fluctuations of the power supply system will affect the operating efficiency of the motor, thereby affecting the power stability of the crusher. Motor failure or aging may also lead to unstable power output. Therefore, the smaller the correlation between the vibration of the rotating shaft and the power of the crusher, the more likely it is to fail. Therefore, based on the clustering effect of the first cluster and the correlation between the vibration data period in the first cluster and the power data period in the same period, the possibility of failure of the first cluster is obtained.

[0047] In an exemplary embodiment, Figure 5 As shown in FIG. 1 , the process of obtaining the possibility of a failure includes: Step S2 - 1 : Obtaining the correlation coefficient between the first vibration data period in the first cluster and the power data period in the same period.

[0048] Each vibration data period in the first cluster has a power data period in the same period. For ease of description, the first vibration data period is set to be any vibration data period in the first cluster.

[0049] Obtain a correlation coefficient between the first vibration data period and the power data period of the same period. In an exemplary embodiment, the correlation coefficient between the first vibration data period and the power data period of the same period is specifically a Pearson correlation coefficient between the first vibration data period and the power data period of the same period. The smaller the Pearson correlation coefficient, the more likely an abnormality is.

[0050] Step S2-2: taking the clustering effect of the first cluster as the fault occurrence credibility of the first cluster, and combining the correlation coefficient between the first vibration data period and the power data period of the same period to obtain the fault occurrence possibility of the first vibration data period.

[0051] The clustering effect of the first cluster is taken as the reliability of the fault occurrence of the first cluster. The greater the reliability of the fault occurrence, the more similar the forestry wastes in different time periods within the first cluster are, and the more reliable the possibility of the fault occurrence is.

[0052] According to the reliability of the fault occurrence of the first cluster and the correlation coefficient between the first vibration data period and the power data period of the same period, the possibility of the fault occurrence in the first vibration data period is obtained. In an exemplary embodiment, the correlation coefficient between the first vibration data period and the power data period of the same period is normalized to the negative correlation, which is convenient for subsequent calculations. Since the numerical range of the Pearson correlation coefficient is [-1, 1], the value 1 is subtracted from the Pearson correlation coefficient and then divided by 2, so that the numerical range of the result is [0, 1], and negative correlation normalization is achieved.

[0053] Then, the correlation coefficient between the normalized first vibration data period and the power data period of the same period is multiplied by the clustering effect of the first cluster, and the result is used as the failure possibility of the first vibration data period. In this way, the failure possibility of each vibration data period in each cluster is obtained.

[0054] The possibility of failure of each vibration data time period in the first cluster is used to screen out the abnormal time period in the first cluster, and the abnormal time period is specifically the abnormal vibration data time period. In an exemplary embodiment, an abnormal threshold is preset, and the numerical range of the preset abnormal threshold is 0-1, and the specific numerical value of the preset abnormal threshold is set by the actual situation. If a safer monitoring logic is required, the preset abnormal threshold can be set slightly smaller, so that it is easier to have abnormal time periods that meet the monitoring requirements. The preset abnormal threshold in this embodiment takes 0.6 as an example. Then, the possibility of failure of each vibration data time period in the first cluster is compared with the preset abnormal threshold, and the vibration data time period corresponding to the possibility of failure greater than the preset abnormal threshold is determined as an abnormal time period. Then, through screening, the first cluster includes abnormal time periods and non-abnormal time periods (i.e., normal time periods).

[0055] By adopting the above process, the abnormal time periods in each cluster are obtained. Then, as a whole, multiple abnormal time periods are obtained.

[0056] Step S3: Based on the possibility of failure in each abnormal period and the association between each abnormal period and each non-abnormal period in the cluster, the failure degree of each abnormal period is obtained.

[0057] The abnormal period represents the period when the crusher may fail. Due to the complex working environment of the forestry waste crusher and the wide variety of forestry waste, which differ greatly in hardness, water content, etc., different types of failures may occur under long-term high-load operation. However, the impact of different types of failures is different. It is necessary to further determine the type of failure and the degree of failure of the crusher based on the distribution characteristics of the abnormal period in time series.

[0058] In an exemplary embodiment, Figure 6 As shown in FIG. 1 , the process of obtaining the fault degree includes: Step S3-1: According to the time interval and characteristic value difference between the first abnormal period and each non-abnormal period in the cluster, combined with the possibility of failure of the first abnormal period and each non-abnormal period in the cluster, the initial failure degree of the first abnormal period is obtained.

[0059] For each abnormal time period obtained, each abnormal time period is sorted according to the time sequence of each abnormal time period, so as to obtain the time sequence of each abnormal time period.

[0060] For the sake of convenience, the first abnormal period is set to any abnormal period, and the cluster in which the first abnormal period is located and the non-abnormal periods in the cluster in which the first abnormal period is located can be obtained.

[0061] According to the time interval and characteristic value difference between the first abnormal period and each non-abnormal period in the cluster, combined with the failure possibility of the first abnormal period and each non-abnormal period in the cluster, the initial failure degree of the first abnormal period is obtained.

[0062] In an exemplary embodiment, Figure 7 As shown, the process of obtaining the initial fault degree of the first abnormal period is given as follows, including: Step S3-1-1: Obtain a fault influence sub-coefficient of the first non-abnormal period on the first abnormal period according to the first time interval, the first characteristic value difference, and the possibility of a fault occurring in the first non-abnormal period.

[0063] For the convenience of explanation, the first non-abnormal time period is set to be any non-abnormal time period in the cluster where the first abnormal time period is located.

[0064] The time interval between the first abnormal period and the first non-abnormal period is obtained. Specifically, if the first abnormal period is earlier than the first non-abnormal period, the end moment of the first abnormal period and the start moment of the first non-abnormal period are obtained, and the time interval between the end moment of the first abnormal period and the start moment of the first non-abnormal period is used as the time interval between the first abnormal period and the first non-abnormal period; if the first non-abnormal period is earlier than the first abnormal period, the end moment of the first non-abnormal period and the start moment of the first abnormal period are obtained, and the time interval between the end moment of the first non-abnormal period and the start moment of the first abnormal period is used as the time interval between the first abnormal period and the first non-abnormal period. For the sake of convenience, the first time interval is defined as the time interval between the first abnormal period and the first non-abnormal period.

[0065] The characteristic value of the first abnormal period and the characteristic value of the first non-abnormal period are obtained. The characteristic value represents the characteristic situation of the corresponding period. In an exemplary embodiment, based on the above APCA segmentation method, the characteristic value is specifically the APCA approximation of the corresponding period, so as to obtain the APCA approximation of the first abnormal period and the APCA approximation of the first non-abnormal period. The characteristic value difference between the first abnormal period and the first non-abnormal period is obtained. The characteristic value difference is specifically the absolute value of the difference between the APCA approximation of the first abnormal period and the APCA approximation of the first non-abnormal period, which is defined as the first characteristic value difference.

[0066] According to the first time interval, the first eigenvalue difference, and the possibility of failure in the first non-abnormal period, the fault influence sub-coefficient of the first non-abnormal period on the first abnormal period is obtained. The longer the first time interval, the lower the credibility of the fault influence sub-coefficient, and the shorter the first time interval, the higher the credibility of the fault influence sub-coefficient. The larger the first eigenvalue difference, the greater the difference in vibration performance between the first abnormal period and the first non-abnormal period, and the greater the possibility of failure. The smaller the possibility of failure in the first non-abnormal period, the more credible the first eigenvalue difference. Therefore, the fault influence sub-coefficient is inversely proportional to the first time interval, directly proportional to the first eigenvalue difference, and inversely proportional to the possibility of failure in the first non-abnormal period.

[0067] In an exemplary embodiment, the calculation formula of the fault impact sub-coefficient is as follows: ; in, It represents the fault influence sub-coefficient of the zth non-abnormal period in the cluster where the yth abnormal period is located on the yth abnormal period, represents the probability of failure in the zth non-abnormal period of the cluster where the yth abnormal period is located, represents the time interval between the yth abnormal period and the zth non-abnormal period in the cluster where it is located, represents the APCA approximation of the yth abnormal period, It represents the APCA approximation of the zth non-abnormal period in the cluster where the yth abnormal period belongs.

[0068] Express The normalization method here can be: .

[0069] It represents the normalization of the product of three parts. The normalization method here can be: obtain the maximum and minimum values ​​of the product of the three parts of the yth abnormal time period for each non-abnormal time period in the cluster where the yth abnormal time period is located, and then normalize the product of the three parts based on the maximum and minimum value normalization method.

[0070] The larger the value is, the greater the difference in vibration performance between the yth abnormal period and the zth non-abnormal period in the cluster where the yth abnormal period is located, and the greater the possibility of failure. The smaller it is, the higher the credibility of the fault impact sub-coefficient; the possibility of failure The smaller the corresponding The more credible.

[0071] Step S3-1-2: The fault impact sub-coefficients of each non-abnormal period in the cluster to which the first abnormal period belongs are integrated to obtain the fault impact coefficient of the first abnormal period.

[0072] Step S3-1-1 obtains the fault impact sub-coefficients of each non-abnormal period in the cluster where the y-th abnormal period is located on the y-th abnormal period, and fuses the fault impact sub-coefficients of each non-abnormal period in the cluster where the y-th abnormal period is located on the y-th abnormal period to obtain the fault impact coefficient of the y-th abnormal period. In an exemplary embodiment, the average value of the fault impact sub-coefficients of each non-abnormal period in the cluster where the y-th abnormal period is located on the y-th abnormal period is calculated as the fault impact coefficient of the y-th abnormal period.

[0073] Step S3-1-3: Obtain the initial fault degree of the first abnormal period according to the fault influence coefficient of the first abnormal period and the possibility of the fault occurring in the first abnormal period.

[0074] The product of the fault influence coefficient of the yth abnormal period and the possibility of the fault in the yth abnormal period is calculated as the initial fault degree of the yth abnormal period.

[0075] Step S3-2: Obtain the fault degree of the first abnormal period according to the initial fault degree of the first abnormal period and the duration ratio of the first abnormal period.

[0076] Get the duration of the yth abnormal period, and get the total duration of all abnormal periods, calculate the ratio of the duration of the yth abnormal period to the total duration, and use it as the duration ratio of the yth abnormal period. The larger the duration ratio of the yth abnormal period, the more serious the fault in the yth abnormal period, and the greater the degree of the fault.

[0077] The product of the initial fault degree of the y-th abnormal period and the duration ratio of the y-th abnormal period is calculated as the fault degree of the y-th abnormal period.

[0078] By adopting the above process, the fault degree of each abnormal period is obtained.

[0079] Step S4: Based on the fault degree and duration of each abnormal period, the duration of the adjacent abnormal period of each abnormal period, and the time interval between the adjacent abnormal periods, the failure probability of the crusher is obtained.

[0080] The more dispersed the abnormal time periods are in the timing, the more likely it is that they are caused by unstable power supply voltage. For example, a short-term increase or decrease in voltage will cause changes in parameters such as the motor speed and torque, thereby causing abnormal vibration of the crusher. And the abnormality will return to normal immediately after it occurs. The impact on the crusher is relatively short, and the degree of failure is relatively mild. However, when the abnormal time periods are relatively continuous in the timing, it means that the crusher is affected by continuous interference, and the crusher may have more serious failures, such as damage to the crushing tools (such as hammers, blades, etc.), bearing failure, etc., which will not only affect the waste crushing effect, but also the degree of failure may gradually increase as the crusher runs. Therefore, the more continuous the abnormal time periods are in the timing, the greater the probability of the crusher failure and the higher the degree of failure.

[0081] Therefore, based on the fault degree and duration of each abnormal period, the duration of the adjacent abnormal period of each abnormal period, and the time interval between the adjacent abnormal periods, the failure probability of the crusher is obtained. Figure 8 As shown in FIG. 1 , the process of obtaining the failure probability of the crusher includes: Step S4-1: Obtain the fault probability impact weight associated with each abnormal period according to the duration of the adjacent abnormal period of each abnormal period and the time interval between each abnormal period and the adjacent abnormal period.

[0082] Get the adjacent abnormal time periods of the yth abnormal time period in time sequence, where if the yth abnormal time period is the first abnormal time period in time sequence, then the abnormal time period on the right (that is, the second abnormal time period in time sequence) is taken as its adjacent abnormal time period; if the yth abnormal time period is the last abnormal time period in time sequence, then the abnormal time period on the left (that is, the second to last abnormal time period in time sequence) is taken as its adjacent abnormal time period; if the yth abnormal time period is any abnormal time period from the second abnormal time period to the second to last abnormal time period in time sequence, then one abnormal time period on each of its left and right sides is taken as its adjacent abnormal time period.

[0083] The duration of the yth abnormal period and the duration of its adjacent abnormal period are obtained. It should be understood that if the yth abnormal period has adjacent abnormal periods on the left and right sides, the average of the duration of the left adjacent abnormal period and the duration of the right adjacent abnormal period of the yth abnormal period is obtained as the duration of the adjacent abnormal period of the yth abnormal period.

[0084] Get the time interval between the yth abnormal period and the adjacent abnormal period. It should be understood that if the yth abnormal period has adjacent abnormal periods on the left and right sides, then the average of the time interval between the yth abnormal period and the left adjacent abnormal period and the time interval between the yth abnormal period and the right adjacent abnormal period is obtained as the time interval between the yth abnormal period and the adjacent abnormal period.

[0085] According to the length of the adjacent abnormal time period of the yth abnormal time period and the time interval between the yth abnormal time period and the adjacent abnormal time period, the fault probability impact weight related to the yth abnormal time period is obtained. Among them, the shorter the time interval between the yth abnormal time period and the adjacent abnormal time period, the more continuous the yth abnormal time period is in the timing, and the higher the fault probability is; the longer the length of the adjacent abnormal time period of the yth abnormal time period, the more continuous the yth abnormal time period is in the timing, and the higher the fault probability is. Therefore, the fault probability impact weight corresponding to the yth abnormal time period is proportional to the length of the adjacent abnormal time period of the yth abnormal time period, and inversely proportional to the time interval between the yth abnormal time period and the adjacent abnormal time period.

[0086] In an exemplary embodiment, the calculation formula of the fault probability impact weight corresponding to the yth abnormal period is as follows: ; in, represents the fault probability impact weight corresponding to the y-th abnormal period, represents the length of the adjacent abnormal period of the yth abnormal period, To show the time interval between the yth abnormal period and the adjacent abnormal period, Y represents the number of abnormal periods. The denominator is added with 0.01 to avoid the denominator being 0, thus ensuring that the denominator is meaningful.

[0087] pass The calculation formula makes the sum of the failure probability impact weights corresponding to all abnormal time periods equal to 1.

[0088] Step S4-2: Based on the failure probability impact weights associated with each abnormal period, weighted sum is performed on the failure degree of each abnormal period, and combined with the mean of the time interval between any two abnormal periods, the failure probability of the crusher is obtained.

[0089] According to the obtained fault probability impact weights corresponding to each abnormal period and the fault degree of each abnormal period, the fault probability impact weights corresponding to each abnormal period are used as the weight coefficients of the fault degree of the corresponding abnormal period, and the fault degree of each abnormal period is weighted and summed. The calculation formula is: ; in, It represents the result of weighted summation of the fault degree of each abnormal period as the weighted fault degree of the crusher. The higher the fault degree, the higher the fault probability of the crusher. Indicates the fault degree of the yth abnormal period.

[0090] Get the time interval between any two abnormal periods, and calculate the average value of the time interval between any two abnormal periods as the overall interval of the abnormal periods. The smaller the average value of the time interval, the more continuous the abnormal periods are, and accordingly, the higher the probability of failure of the crusher.

[0091] The failure probability of the crusher is obtained by weighted summing the failure degree of each abnormal period and the average value of the time interval between any two abnormal periods. In an exemplary embodiment, the calculation formula of the failure probability of the crusher is given as follows: ; in, represents the failure probability of the crusher, It represents the average value of the time interval between any two abnormal periods.

[0092] Thus, the failure probability of the crusher is obtained. The higher the failure probability, the more likely the crusher is to fail.

[0093] In an exemplary embodiment, Fig. 9 As shown, the forestry waste comminution method also includes: Step S5: comparing the failure probability of the crusher with a preset failure threshold; Step S6: If the failure probability of the grinder is greater than a preset failure threshold, a grinder shutdown instruction is output.

[0094] The value range of the preset fault threshold is 0-1, and the specific value of the preset fault threshold is set according to the actual situation. If a safer monitoring logic is required, the preset fault threshold can be set slightly smaller, so that it is easier to produce a judgment result that meets the monitoring requirements. The preset abnormal threshold in this embodiment takes 0.6 as an example. The pulverizer stops running under the action of the pulverizer stop command to avoid causing more serious accidents.

[0095] This embodiment also provides a forestry waste crushing device, including: a memory and a processor; the memory is connected to the processor, and the memory is used to store program instructions; the processor is used to implement the steps in the above-mentioned forestry waste crushing method embodiment when the program instructions are executed.

[0096] In an exemplary embodiment, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the above-mentioned forestry waste comminution method embodiment.

[0097] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages 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.

[0098] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for comminuting forestry waste, characterized in that: include: Acquire multiple vibration data periods and power data periods of the pulverizer, and cluster the vibration data periods to obtain multiple clusters; The abnormal period is screened out based on the possibility of failure in each vibration data period in the cluster. The possibility of failure is obtained by the clustering effect of the cluster and the correlation between the vibration data period in the cluster and the power data period in the same period. Based on the possibility of failure in each abnormal period and the association between each abnormal period and each non-abnormal period in the cluster, the degree of failure in each abnormal period is obtained; Based on the fault degree and duration of each abnormal period, the duration of each abnormal period adjacent to the abnormal period, and the time interval between adjacent abnormal periods, the failure probability of the crusher is obtained.

2. A method for comminuting forestry waste as claimed in claim 1, characterized in that: The process of obtaining the fault extent includes: According to the time interval and characteristic value difference between the first abnormal period and each non-abnormal period in the cluster, combined with the possibility of failure of the first abnormal period and each non-abnormal period in the cluster, the initial failure degree of the first abnormal period is obtained; the first abnormal period is any abnormal period; The fault degree of the first abnormal period is obtained according to the initial fault degree of the first abnormal period and the duration ratio of the first abnormal period.

3. A method for comminuting forestry waste as claimed in claim 2, characterized in that: The process of obtaining the initial fault degree of the first abnormal period includes: According to the first time interval, the first eigenvalue difference, and the possibility of failure in the first non-abnormal period, a fault influence sub-coefficient of the first non-abnormal period on the first abnormal period is obtained; the fault influence sub-coefficient is inversely proportional to the first time interval, proportional to the first eigenvalue difference, and inversely proportional to the possibility of failure in the first non-abnormal period; the first non-abnormal period is any non-abnormal period in the cluster where the first abnormal period is located; the first time interval is the time interval between the first abnormal period and the first non-abnormal period, and the first eigenvalue difference is the eigenvalue difference between the first abnormal period and the first non-abnormal period; The fault influence sub-coefficients of each non-abnormal period in the cluster where the first abnormal period is located on the first abnormal period are integrated to obtain the fault influence coefficient of the first abnormal period; According to the fault influence coefficient of the first abnormal period and the possibility of the fault occurring in the first abnormal period, the initial fault degree of the first abnormal period is obtained.

4. A method for comminuting forestry waste as claimed in claim 1, characterized in that: The process of obtaining the failure probability of the crusher includes: According to the duration of the adjacent abnormal time periods of each abnormal time period, and the time interval between each abnormal time period and the adjacent abnormal time period, the fault probability impact weight related to each abnormal time period is obtained; the fault probability impact weight is proportional to the duration of the adjacent abnormal time period of the abnormal time period, and inversely proportional to the time interval between the abnormal time period and the adjacent abnormal time period; Based on the failure probability impact weights associated with each abnormal period, the failure degree of each abnormal period is weighted and summed, and combined with the mean of the time interval between any two abnormal periods, the failure probability of the crusher is obtained.

5. A method for comminuting forestry waste as claimed in claim 1, characterized in that: The process of obtaining the possibility of failure includes: Obtaining a correlation coefficient between a first vibration data period in a first cluster and a power data period in the same period; the first cluster is any cluster, and the first vibration data period is any vibration data period in the first cluster; The clustering effect of the first cluster is used as the fault occurrence credibility of the first cluster, and the possibility of fault occurrence in the first vibration data period is obtained by combining the correlation coefficient between the first vibration data period and the power data period in the same period.

6. A method for comminuting forestry waste as claimed in claim 5, characterized in that: The process of obtaining the clustering effect of the first cluster includes: obtaining the clustering effect of the first cluster according to the average value of the similarity between any two vibration data time periods in the first cluster and the silhouette coefficient of the first cluster.

7. A method for comminuting forestry waste as claimed in claim 1, characterized in that: Before obtaining a plurality of vibration data periods and power data periods of the pulverizer, the forestry waste pulverizing method further includes: Acquire a vibration data sequence and a power data sequence of a pulverizer; The STL algorithm is used to decompose the vibration data sequence and obtain the trend term curve; The APCA segmentation method is used to segment the trend item curve, and based on the segmentation points, the vibration data series and the power data series are segmented to obtain multiple vibration data periods and power data periods respectively.

8. A method for comminuting forestry waste as claimed in claim 7, characterized in that: Clustering of vibration data time periods yields multiple clusters, including: Obtain the difference of the APCA approximation values ​​of any two trend item periods, where the trend item periods are obtained by segmenting the trend item curve using the APCA segmentation method; The K-means clustering algorithm is used to cluster the trend item periods with the difference of APCA approximation between any two trend item periods as the clustering distance. Based on the relationship between the trend item period and the vibration data period, multiple clusters are obtained, which include multiple vibration data periods.

9. A method for comminuting forestry waste as claimed in claim 1, characterized in that: The forestry waste comminution method also includes: comparing the failure probability of the pulverizer with a preset failure threshold; If the failure probability of the crusher is greater than the preset failure threshold, a crusher shutdown command is output.

10. A forestry waste crushing device, characterized in that it comprises: Memory and processor; The memory is connected to the processor; The memory is used to store program instructions; The processor is used to implement the forestry waste crushing method according to any one of claims 1 to 9 when the program instructions are executed.

Images