An exhaust gas purification device abnormal operation state detection method and system

Abstract

The application discloses an exhaust gas purification equipment abnormal operation state detection method and system, and relates to the field of fault detection. The method is applied to detection equipment, and the method comprises the following steps: acquiring multi-source sensor data of the purification equipment; calculating fluctuation indexes of the multi-source sensor data, and updating the fluctuation indexes to a preset fault influence relationship graph; determining a plurality of abnormal equipment nodes according to the fluctuation indexes of a plurality of equipment nodes; taking the plurality of abnormal equipment nodes as starting points, traversing the fault influence relationship graph to obtain a plurality of fault chains; calculating chain responsibility degrees of the plurality of fault chains, and determining a fault chain with the largest chain responsibility degree as a root cause result; and outputting the root cause result to an equipment management interface, so that troubleshooting personnel can quickly troubleshoot abnormal reasons. By implementing the technical scheme, the problem of low troubleshooting efficiency caused by various and complex fault types of the exhaust gas purification equipment is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault detection, and in particular to a method and system for detecting abnormal operation state of exhaust gas purification equipment. BACKGROUND

[0002] Sand blasting process is to use compressed air to spray abrasive materials at high speed to the workpiece surface to achieve cleaning, rust removal, strengthening and other purposes; in this process, compressed air will carry dust on the workpiece surface and a small amount of abrasive particles into the air at high speed, thereby forming exhaust gas.

[0003] For such exhaust gas containing solid particles, mechanical dust collectors or filter dust collectors are generally used for purification; but due to the often harsh working environment, the abnormal frequency of these exhaust gas purification equipment is also relatively high, so frequent detection is required; the current detection method is to use automatic monitoring means to provide part of the equipment operation information, and then the troubleshooting personnel check the abnormality according to the equipment operation information, however, due to the complex and diverse types of exhaust gas purification equipment failures, it is difficult for less experienced troubleshooting personnel to quickly troubleshoot the root cause, thereby reducing the troubleshooting efficiency. SUMMARY

[0004] In view of the problem of low troubleshooting efficiency caused by the complex and diverse types of exhaust gas purification equipment failures, the present application provides a method and system for detecting abnormal operation state of exhaust gas purification equipment.

[0005] In a first aspect, the present application provides a method for detecting abnormal operation state of exhaust gas purification equipment, applied to a detection device, the method comprising:

[0006] Obtaining multi-source sensor data of the purification equipment;

[0007] Calculating the fluctuation index of the multi-source sensor data and updating it to a preset fault influence relationship graph, the fault influence relationship graph comprising a plurality of equipment nodes and influence edges between the plurality of equipment nodes, wherein one equipment node corresponds to one fluctuation index;

[0008] Determining a plurality of abnormal equipment nodes according to the fluctuation indexes of the plurality of equipment nodes;

[0009] Taking the plurality of abnormal equipment nodes as the starting point, traversing the fault influence relationship graph to obtain a plurality of fault chains;

[0010] Calculating the chain responsibility degree of the plurality of fault chains, and determining the fault chain with the largest chain responsibility degree as the root cause result;

[0011] Outputting the root cause result to a device management interface to facilitate the troubleshooting personnel to quickly troubleshoot the abnormal reason.

[0012] Optionally, the fluctuation index of the multi-source sensor data is calculated and updated into a preset fault influence relationship graph, specifically comprising:

[0013] The plurality of device nodes are classified to obtain a plurality of single-sensor device nodes and a plurality of multi-sensor device nodes.

[0014] If the first device node is a single-sensor device node, the sensor data of the first device node is reconstructed in a phase space to obtain a phase space point set.

[0015] The correlation dimension of the phase space point set is calculated, and the fluctuation index of the first device node is determined based on the correlation dimension.

[0016] Optionally, the fluctuation index of the multi-source sensor data is calculated and updated into a preset fault influence relationship graph, specifically further comprising:

[0017] If the second device node is a multi-sensor device node, the plurality of sensor data corresponding to the second device node is constructed into a plurality of slope curves.

[0018] Any one of the plurality of slope curves is selected as a reference curve.

[0019] The fluctuation similarity between the plurality of slope curves and the reference curve is calculated respectively, and the fluctuation similarity includes fluctuation time similarity and fluctuation direction similarity.

[0020] The fluctuation index of the second device node is determined according to the mean value of the plurality of fluctuation similarities.

[0021] Optionally, the fluctuation index of the multi-source sensor data is calculated and updated into a preset fault influence relationship graph, specifically further comprising:

[0022] The abnormal frequency of the plurality of device nodes is determined according to the historical multi-source sensor data of the plurality of device nodes.

[0023] The influence coefficient of the plurality of influence edges in the fault influence relationship graph is calculated according to the abnormal frequency of the plurality of device nodes.

[0024] Optionally, the chain responsibility degree of the plurality of fault chains is calculated, specifically:

[0025] The fault frequency of the plurality of fault chains is obtained.

[0026] The path correction coefficient of the plurality of fault chains is determined according to the fault frequency of the plurality of fault chains.

[0027] According to the fluctuation indexes corresponding to the plurality of device nodes in the first fault chain and the influence edges, a node responsibility degree of the plurality of device nodes in the first fault chain is calculated, the first fault chain being any one of the plurality of fault chains;

[0028] Based on the node responsibilities of the plurality of device nodes in the first fault chain and the path correction coefficient corresponding to the first fault chain, a chain responsibility degree of the first fault chain is calculated.

[0029] Optionally, after the root cause result is output to the device management interface, the method further includes:

[0030] A target fault device node input by the troubleshooting personnel is obtained.

[0031] When the target fault device node belongs to the root cause result, the root cause result and the target fault device node are constructed into a corresponding relationship and stored in a preset fault library.

[0032] In a second aspect, the application provides an abnormal operation state detection system of a waste gas purification device. The system is a detection device, and the detection device includes an acquisition module, a processing module, and a display module. The acquisition module is configured to acquire multi-source sensor data of the purification device. The processing module is configured to calculate fluctuation indexes of the multi-source sensor data and update the fluctuation indexes in a preset fault influence relationship graph. The fault influence relationship graph includes a plurality of device nodes and influence edges between the plurality of device nodes. One device node corresponds to one fluctuation index. The processing module is further configured to determine a plurality of abnormal device nodes according to the fluctuation indexes of the plurality of device nodes, traverse the fault influence relationship graph starting from the plurality of abnormal device nodes, obtain a plurality of fault chains, calculate chain responsibility degrees of the plurality of fault chains, and determine a fault chain with the largest chain responsibility degree as a root cause result. The display module is configured to output the root cause result to a device management interface to facilitate a troubleshooting personnel to quickly troubleshoot an abnormal cause.

[0033] The acquisition module is configured to acquire multi-source sensor data of the purification device.

[0034] The processing module is configured to calculate fluctuation indexes of the multi-source sensor data and update the fluctuation indexes in a preset fault influence relationship graph. The fault influence relationship graph includes a plurality of device nodes and influence edges between the plurality of device nodes. One device node corresponds to one fluctuation index. The processing module is further configured to determine a plurality of abnormal device nodes according to the fluctuation indexes of the plurality of device nodes, traverse the fault influence relationship graph starting from the plurality of abnormal device nodes, obtain a plurality of fault chains, calculate chain responsibility degrees of the plurality of fault chains, and determine a fault chain with the largest chain responsibility degree as a root cause result.

[0035] The display module is configured to output the root cause result to a device management interface to facilitate a troubleshooting personnel to quickly troubleshoot an abnormal cause.

[0036] In a third aspect, the application provides an electronic device. The electronic device includes a processor, a memory, a user interface, and a network interface. The memory is configured to store instructions. The user interface and the network interface are configured to communicate with other devices. The processor is configured to execute the instructions stored in the memory to enable the electronic device to perform the method according to any one of the first aspect.

[0037] In a fourth aspect, the present application provides a computer readable storage medium storing instructions which, when executed, perform the method of any one of the first aspect.

[0038] In summary, the one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0039] For the case that the exhaust gas purification equipment has various and complex fault types, according to the multi-source sensor data of the purification equipment, the purification equipment is disassembled into a plurality of equipment nodes in the present application, wherein the performance parameters of each equipment node are detected by at least one type of sensor, and then a fault influence relationship graph is constructed according to the influence relationship of each equipment node, so as to facilitate accurate mining of the fault equipment node of the exhaust gas purification equipment. Then, the real-time detected multi-source sensor data is converted into a fluctuation index and updated to the fault influence relationship graph, at this time each equipment node in the fault influence relationship graph corresponds to a fluctuation index. It needs to be explained that when the fluctuation index of a certain equipment node is abnormal, it is not necessarily caused by the equipment node itself, but also possible that the associated equipment node is faulty, resulting in abnormal detection data of the equipment node. Therefore, after determining a plurality of abnormal equipment nodes from the fluctuation indexes of a plurality of equipment nodes, the present application takes the plurality of abnormal equipment nodes as the starting point to traverse the fault influence relationship graph, mines the fault chain of each abnormal equipment node, and finally calculates the chain responsibility degree of a plurality of fault chains to narrow the fault troubleshooting range and determine the accurate position of the root cause result. At this time, the root cause result is displayed to the troubleshooting personnel, and the troubleshooting personnel can quickly find the fault cause by preferentially troubleshooting the equipment node corresponding to the root cause result, thereby improving the troubleshooting efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 is a flowchart of an exhaust gas purification equipment abnormal running state detection method provided by the embodiments of the present application.

[0041] Figure 2 is a structural schematic diagram of an exhaust gas purification equipment abnormal running state detection system provided by the embodiments of the present application.

[0042] Figure 3 is a structural schematic diagram of an electronic device provided by the embodiments of the present application.

[0043] Explanation of reference signs: 1, acquisition module; 2, processing module; 3, display module; 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0044] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0045] There has been a purification problem for waste gas containing solid particles, mainly because the particle size range of the solid particles can be from several nanometers to several hundred microns, which has strong diffusion. In addition, these solid particles can include metal oxides, carbides, silicates and other substances, which can cause the purification equipment to be worn and corroded when the solid particles come into contact with the surface of the purification equipment, thereby causing the purification equipment for treating waste gas containing solid particles to have a higher failure frequency than the purification equipment for treating pure chemical waste gas. At this time, frequent detection is required. The current detection method is to use automatic monitoring means to provide part of the equipment operation information, and then the troubleshooting personnel check the abnormality according to the equipment operation information. However, this detection method is difficult to cope with the complex working conditions of the waste gas purification equipment with various types of faults, and it is difficult for less experienced troubleshooting personnel to quickly identify the root cause, thereby reducing the troubleshooting efficiency.

[0046] Therefore, in order to solve this problem, the present application provides a waste gas purification equipment abnormal operation state detection method, which is applied to a detection device, as shown in the figure, the method comprises steps S101 to S106, and the steps are as follows: Figure 1

[0047] S101, obtaining multi-source sensor data of the purification equipment.

[0048] In the above steps, the multi-source sensor data includes data composed of multiple different types of sensors, which represents the performance parameters of different equipment nodes. For example, for a mechanical dust collector, it includes equipment nodes such as a fan and a cyclone separator. The fan needs air volume sensor data to detect the ventilation condition and vibration sensor data to detect the ventilation stability, and the cyclone separator needs pressure sensor data and particle concentration sensor data to detect the separation effect of solid particles in the waste gas.

[0049] S102, calculating the fluctuation index of the multi-source sensor data and updating it to the preset fault influence relationship graph, the fault influence relationship graph includes multiple equipment nodes and the influence edges between the multiple equipment nodes, wherein one equipment node corresponds to one fluctuation index.

[0050] ​In the above steps, for the waste gas purification device, there is a mutual influence relationship between multiple device nodes, for example, if the air volume of the fan is insufficient, it will lead to the decrease of the waste gas flow entering the cyclone separator, the decrease of the airflow velocity in the cyclone separator, the decrease of the centrifugal force, thereby reducing the separation efficiency of the cyclone separator for solid particles, and the decrease of the separation efficiency of the cyclone separator will increase the content of solid particles in the waste gas entering the bag-type dust collector, thereby causing the filter bag of the bag-type dust collector to be more easily blocked, the filter resistance to increase, and the purification efficiency of the bag-type dust collector to decrease. Therefore, according to the influence relationship between each purification device, the application pre-constructs a fault influence relationship graph, which contains multiple device nodes and influence edges between multiple device nodes. The influence edge can be understood as the influence degree of one device node on another device node, for example, the influence edge of the fan on the cyclone separator can be preset as 0.8, and the influence edge of the cyclone separator on the bag-type dust collector can be preset as 0.9.

[0051] For each device node, the performance parameter is monitored by the sensor data, so when the sensor data is abnormal, it often indicates that the running state of the device node is abnormal; the application can calculate the fluctuation index of the sensor data of each device node to determine the abnormality degree, which can be calculated by calculating the variance, mean or standard deviation to obtain the fluctuation index, and then updating the fluctuation index of each device node to the fault influence relationship graph, thereby facilitating subsequent accurate mining of the root cause of the abnormality according to the fault influence relationship graph.

[0052] In a possible implementation, due to the difference in the operating conditions of each device node, the types of sensor data required for detection of each device node may not be consistent, for example, some simple structure device nodes only need one type of sensor to represent the performance parameters, and some complex structures may need multiple types of sensors to represent the performance parameters, which causes the update efficiency of the fluctuation index of different device nodes in the fault influence relationship graph to be inconsistent, thereby reducing the timeliness of the sensor data of each device node. In addition, for the sensor data of the simple structure device node, although the data analysis efficiency is relatively fast, the data amount is single and small, thereby reducing the accuracy of the fluctuation index; and for the sensor data of the complex device node, although the data amount is rich, which can improve the calculation accuracy of the fluctuation index, but the data is complex and large, thereby reducing the data analysis efficiency. Therefore, in order to further solve the above problems, the multiple device nodes of the exhaust gas purification device are divided into single-sensor device nodes and multi-sensor device nodes according to the number of sensor types, wherein the single-sensor device node is a device node that only needs one type of sensor to represent the performance parameters, and the multi-sensor node is a device node that needs two or more types of sensors to represent the performance parameters; then the single-sensor node adopts a fluctuation index algorithm with high accuracy, and the multi-sensor device node adopts a fluctuation index algorithm with high efficiency, thereby unifying the analysis efficiency and calculation accuracy of the two, and further improving the detection accuracy and efficiency of the subsequent abnormal operating state. Specifically:

[0053] For the single-sensor device node, since the data amount is single and simple, the fluctuation of the data is random and not obvious, therefore, the application adopts a chaotic algorithm to mine the hidden rules in the data, thereby improving the calculation accuracy of the fluctuation index; specifically, first, according to the preset delay time and embedding dimension, the sensor data is reconstructed in phase space, the one-dimensional sensor data is expanded to multi-dimensional space, so that the linear relationship hidden in the one-dimensional space is clearly displayed in the high-dimensional space, thereby obtaining a phase space point set; for example, for a one-dimensional space sensor data set {x t}, t = n, n is the total number of sensor data, at this time the whole is in a nonlinear relationship, but under the observation angle of delay time 2 and embedding dimension 2, the phase space point set obtained is {x i = (x i , x i+2}, i = 1, 2, 3,..., n-2, presents a certain linear law, at this time the deviation degree between the phase space point set and the presented linear law is the fluctuation index of the single sensor device node, and in order to quantify the deviation degree, the present application calculates the correlation dimension of the phase space point set, and then calculates the difference between the correlation dimension and the embedding dimension, which is the fluctuation index of the single sensor device node. For example, for a phase space data set with an embedding dimension of 2, if its correlation dimension is 2.3, then the dimension deviation degree is 2.3-2=0.3, which further indicates that the fluctuation index of the single sensor device node is 0.3. The correlation dimension calculation method is as follows:

[0054] First, for the phase space point set {x i}, the pair distance of each point is calculated, and a pair data set is obtained. Then, according to the preset distance radius r, the number of point pairs C(r) in the pair data set whose distance is less than the preset distance radius is counted. Then, the preset distance radius r is changed, and the above steps are repeated to obtain a series of point pair numbers. At this time, the fractal structure of the point pair number under different preset distance radius scales in the phase space is calculated to determine the space filling rule of the phase space point set under different scales. Specifically, In(r) is taken as the horizontal coordinate, InC(r) is taken as the vertical coordinate, a curve is drawn, and finally the slope of the fitted curve is calculated to obtain the fractal dimension (representing the fractal structure). The closer the fractal dimension is to the embedding dimension of the current phase space, the stronger the linear law of the phase space point set, and the stronger the correlation. Therefore, the fractal dimension at this time is the correlation dimension of the phase space point set. The lower limit of the value of the preset distance radius r is determined by the minimum value of the pair distance in the pair data set, and the upper limit is determined by the maximum value of the pair distance in the pair data set.

[0055] It should be noted that the delay time is the time interval for sampling sensor data. If the delay time is too small, the difference between adjacent data points may not be obvious, and the linear change cannot be effectively presented. If the delay time is too large, some short-term dynamic information may be lost. The embedding dimension is the dimension number of the reconstructed phase space. If the embedding dimension is too low, the phase space cannot fully describe the linear law of the system. If the embedding dimension is too high, too much noise or redundant information may be introduced. Therefore, in order to select appropriate delay time and embedding dimension, the present application uses autocorrelation function method to determine the delay time and false nearest neighbor method to determine the embedding dimension. Both the autocorrelation function method and the false nearest neighbor method are conventional technical means for those skilled in the art, and will not be described in detail here.

[0056] For the multi-sensor device node, due to the large and complex data quantity, the fluctuation index calculation efficiency is low; therefore, the application utilizes the characteristics that multiple sensor data can jointly represent performance parameters, analyzes the change trend of multiple sensor data, to determine the fluctuation index, it can be understood that under normal circumstances, the change trend of multiple sensor data is consistent, and in the abnormal state, due to the chaotic state of the working condition, the change trend of each sensor data is uncertainly deviated, so as to present a chaotic state as a whole; based on the characteristics, the application first constructs multiple sensor data into multiple slope curves, then selects any one of the multiple slope curves as a reference curve, it can be understood that the slope curve can represent the change trend of the sensor data, at this time, the fluctuation similarity between the remaining multiple slope curves and the reference curve is calculated respectively, wherein the fluctuation similarity includes fluctuation time similarity and fluctuation direction similarity, the fluctuation time similarity can be understood as whether two data points fluctuate at the same time, the fluctuation direction similarity can be understood as whether the fluctuation direction (increase or decrease) of two data points is consistent at the same time, if the fluctuation direction of two data points is consistent at the same time, the fluctuation similarity of the data point is 1, otherwise it is 0, then the distribution of 0 and 1 in the curve is counted, so that the fluctuation similarity can be calculated, for example, a certain slope curve and the reference curve both contain 100 data points, and the data points with fluctuation similarity of 1 contain 80, and the data points with fluctuation similarity of 0 contain 20, then the fluctuation similarity of the two curves is 0.5. Then the fluctuation similarity of multiple slope curves and the reference curve is averaged, at this time, if the average is small, it means that the change trend of multiple slope curves is different from the change trend of the reference curve, which further indicates that the abnormal degree of the current multi-sensor device node is high, and the fluctuation index of the current multi-sensor device node can be represented by 1 minus the average of the fluctuation similarity. In the above process, compared with considering the correlation between each sensor data, directly converting multiple sensor data into a unified slope curve representing the change trend can greatly reduce the calculation logic of the algorithm, thereby simplifying the calculation process, and further improving the calculation efficiency of the fluctuation index.

[0057] In a possible implementation, in the fault relationship influence graph, as the use time increases, the device nodes will age, and the influence degree between the device nodes will change, so that the preset influence edges cannot accurately characterize the influence degree of one device node on another device node. Therefore, the application also needs to update and adjust the influence edges. First, the historical multi-source sensor data of the plurality of device nodes are acquired, and then the abnormal frequency of the plurality of device nodes is counted. Then, a plurality of high-frequency abnormal device nodes with high abnormal frequency are screened from the plurality of device nodes. It can be understood that the aging rate of the device node with high abnormal frequency is also relatively fast, and the higher the aging degree, the greater the influence degree on the downstream device node. Therefore, the difference between the abnormal frequency of the high-frequency abnormal device node and the abnormal frequency of the downstream node is calculated, and then the difference is divided by the abnormal frequency of the high-frequency abnormal device node, so as to obtain an adjustment coefficient. Then, the adjustment coefficient is multiplied by the influence edge between the original high-frequency abnormal device node and the downstream node, so as to correct the influence coefficient, so that the corrected influence edge can accurately describe the influence degree of the high-frequency abnormal device node on the downstream device node.

[0058] S103, determine a plurality of abnormal device nodes according to the fluctuation indexes of the plurality of device nodes.

[0059] S104, take the plurality of abnormal device nodes as starting points to traverse the fault influence relationship graph to obtain a plurality of fault chains.

[0060] S105, calculate the chain responsibility degrees of the plurality of fault chains, and determine the fault chain with the largest chain responsibility degree as the root cause result.

[0061] In the above steps S103 to S105, after updating the fault influence relationship graph, the fluctuation indexes of the plurality of device nodes are compared with the fluctuation index threshold one by one. For the device node with the fluctuation index greater than or equal to the fluctuation index threshold, it can be determined as an abnormal device node, so as to obtain a plurality of abnormal device nodes. Then, the plurality of abnormal device nodes are taken as starting points to traverse the fault influence relationship graph to obtain a plurality of fault chains.

[0062] Then, the chain responsibility degrees of the plurality of fault chains are calculated. The chain responsibility degree can be understood as the influence degree of the fault chain on the overall operation of the system. The greater the chain responsibility degree, the greater the influence degree, and it is more likely to exist abnormity. The chain responsibility degree calculation method is:

[0063] Taking the first fault chain as an example, if the first fault chain is A-B-C, the fluctuation index of A is a, the fluctuation index of B is b, the fluctuation index of C is c, the influence edge between A and B is t1, and the influence edge between B and C is t2, first, the node responsibility degree of each device node is calculated, which can be understood as the influence degree of the device node on the overall operation of the system, at this time, the node responsibility degree of A node is a, the node responsibility degree of B node is b*t1, and the node responsibility degree of C node is c*t2; then the node responsibility degrees of A, B and C are added, and the chain responsibility degree of the first fault chain is obtained.

[0064] Finally, the chain responsibility degrees of multiple fault chains are compared, and for the fault chain with the largest chain responsibility degree, the influence degree on the overall operation of the system is greater, at this time, it can be determined as the root cause result.

[0065] In a possible implementation, in the actual operation of the system, the higher the occurrence frequency of the fault chain, the greater the possibility of the fault, therefore, in order to more accurately describe the occurrence possibility of each fault chain, the application also obtains the fault frequencies of multiple fault chains, then normalizes the fault frequencies of multiple fault chains to obtain the path correction coefficients of multiple fault chains, and then multiplies the chain responsibility degrees of multiple fault chains by the respective path correction coefficients, so as to be more in line with the actual operation.

[0066] S106, output the root cause result to the device management interface, so that the troubleshooting personnel can quickly troubleshoot the abnormal reason.

[0067] In the above steps, after the root cause result is determined, the root cause result is displayed on the device management interface, at this time, the troubleshooting personnel can determine which device nodes to prioritize troubleshooting from the device management interface, at this time, the troubleshooting personnel can more easily understand the complex device operation logic relationship, greatly reducing the experience requirement of the troubleshooting personnel, and also being able to quickly troubleshoot the cause, thereby improving the troubleshooting efficiency.

[0068] In a possible implementation, for the analyzed root cause result, it is not 100% accurate, therefore, when the troubleshooting personnel cannot troubleshoot the fault device from the fault chain corresponding to the root cause result, the chain responsibility degrees of multiple fault chains are displayed in the device management interface in descending order, to help the troubleshooting personnel more quickly troubleshoot the fault reason. When the troubleshooting personnel troubleshoots the fault device node, the troubleshooting personnel will input the target fault device node, at this time, the detection device will judge whether the target fault device node belongs to the root cause result, if it belongs, it means that the current analysis result is reliable, therefore, the root cause result and the target fault device node can be constructed as a corresponding relationship, and stored in the preset fault library, when the same type of fault occurs next time, not only the fault chain will be displayed, but also the more accurate fault device node will be displayed, thereby further improving the troubleshooting efficiency.

[0069] Reference Figure 2 The application also provides an abnormal operation state detection system of an exhaust gas purification device. The system is a detection device, which comprises an acquisition module 1, a processing module 2 and a display module 3, wherein:

[0070] The acquisition module 1 is configured to acquire multi-source sensor data of the purification device.

[0071] The processing module 2 is configured to calculate a fluctuation index of the multi-source sensor data and update the fluctuation index into a preset fault influence relationship graph, wherein the fault influence relationship graph comprises a plurality of device nodes and influence edges between the plurality of device nodes, one device node corresponds to one fluctuation index, a plurality of abnormal device nodes are determined according to the fluctuation indexes of the plurality of device nodes, a plurality of fault chains are obtained by traversing the fault influence relationship graph from the plurality of abnormal device nodes as starting points, a chain responsibility degree of the plurality of fault chains is calculated, and a fault chain with the largest chain responsibility degree is determined as a root cause result.

[0072] The display module 3 is configured to output the root cause result to a device management interface, so as to facilitate a troubleshooting personnel to quickly troubleshoot an abnormal cause.

[0073] In a possible implementation, the fluctuation index of the multi-source sensor data is calculated and updated into the preset fault influence relationship graph, and specifically includes the following steps.

[0074] The plurality of device nodes are classified to obtain a plurality of single-sensor device nodes and a plurality of multi-sensor device nodes.

[0075] If the first device node is a single-sensor device node, the sensor data of the first device node is reconstructed in a phase space to obtain a phase space point set.

[0076] The correlation dimension of the phase space point set is calculated, and the fluctuation index of the first device node is determined based on the correlation dimension.

[0077] In a possible implementation, the fluctuation index of the multi-source sensor data is calculated and updated into the preset fault influence relationship graph, and specifically further includes the following steps.

[0078] If the second device node is a multi-sensor device node, a plurality of sensor data corresponding to the second device node are constructed into a plurality of slope curves.

[0079] Any one of the plurality of slope curves is selected as a reference curve.

[0080] The fluctuation similarity between the plurality of slope curves and the reference curve is calculated respectively, and the fluctuation similarity includes a fluctuation time similarity and a fluctuation direction similarity.

[0081] The fluctuation index of the second device node is determined according to the mean value of the plurality of fluctuation similarities.

[0082] In a possible implementation, the fluctuation index of the multi-source sensor data is calculated, and is updated to the preset fault influence relationship graph, and specifically further comprising:

[0083] According to the historical multi-source sensor data of the plurality of device nodes, the abnormal frequency of the plurality of device nodes is determined.

[0084] According to the abnormal frequency of the plurality of device nodes, the influence coefficient of the plurality of influence edges in the fault influence relationship graph is calculated.

[0085] In a possible implementation, the chain responsibility degree of the plurality of fault chains is calculated, and specifically:

[0086] The fault frequency of the plurality of fault chains is obtained.

[0087] According to the fault frequency of the plurality of fault chains, the path correction coefficient of the plurality of fault chains is determined.

[0088] According to the fluctuation index corresponding to each of the plurality of device nodes in the first fault chain and the influence edge, the node responsibility degree of the plurality of device nodes in the first fault chain is calculated, the first fault chain being any one of the plurality of fault chains.

[0089] Based on the node responsibility of the plurality of device nodes in the first fault chain and the path correction coefficient corresponding to the first fault chain, the chain responsibility degree of the first fault chain is calculated.

[0090] In a possible implementation, after the root cause result is output to the device management interface, specifically further comprising:

[0091] The target fault device node input by the troubleshooting personnel is obtained.

[0092] When the target fault device node belongs to the root cause result, the root cause result and the target fault device node are constructed as a corresponding relationship, and are stored in the preset fault library.

[0093] It should be noted that, when the apparatuses provided in the above embodiments implement their functions, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above described functions. In addition, the apparatuses and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0094] The application also discloses an electronic device. Referring to Figure 3 , Figure 3is a structural schematic diagram of an electronic device disclosed by an embodiment of the present application. The electronic device 300 can include at least one processor 301, at least one network interface 304, a user interface 303, a memory 305, and at least one communication bus 302.

[0095] The communication bus 302 is configured to realize connection and communication between the components.

[0096] The user interface 303 can include a display and a camera. Optionally, the user interface 303 can further include a standard wired interface and a wireless interface.

[0097] The network interface 304 can optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0098] The processor 301 can include one or more processing cores. The processor 301 is connected to various parts of the server through various interfaces and lines, and performs various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one of a hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 301 can be integrated with a combination of one or more of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU is mainly used to process an operating system, a user interface, and an application program. The GPU is used to render and draw the content to be displayed on the display. The modem is used to process wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 301, but can be implemented by a separate chip.

[0099] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include a non-transitory computer-readable storage medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. (Refer to...) Figure 3 The memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for detecting abnormal operating conditions of waste gas purification equipment.

[0100] exist Figure 3 In the illustrated electronic device 300, the user interface 303 is mainly used to provide an input interface for the user and to acquire user input data; while the processor 301 can be used to call an application program stored in the memory 305 for detecting abnormal operating status of a waste gas purification device. When executed by one or more processors 301, the electronic device 300 performs one or more of the methods described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0101] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0102] In several embodiments provided in the present application, it should be understood that the disclosed apparatus can be implemented in other manners. For example, the division of the apparatus embodiments is merely illustrative, and the division of units can be changed according to actual conditions, such as a combination or integration of some units, or a deletion of some features, or an addition of some features. In addition, the coupling or direct coupling or communication connection between the shown or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0103] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0104] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0105] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable memory. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a memory and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned memory includes: a U disk, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0106] The above is only exemplary embodiments of the present disclosure, which cannot limit the scope of the present disclosure. Any equivalent changes and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. Other embodiments of the present disclosure will be readily apparent to those skilled in the art upon considering the specification and practicing the true principles of the present disclosure.

[0107] The present application is intended to cover any variations, uses or adaptive changes of the present disclosure that follow the general principles of the present disclosure and include common knowledge or conventional technical means in the technical field not described in the present disclosure. The specification and examples are only considered as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for detecting abnormal operating conditions of a waste gas purification device, characterized in that, The method is applied to a detection device, and comprises the following steps: Obtaining multi-source sensor data of a purification device; Calculating a fluctuation index of the multi-source sensor data and updating the fluctuation index to a preset fault influence relationship graph, the fault influence relationship graph comprising a plurality of device nodes and a plurality of influence edges between the device nodes, wherein one device node corresponds to one fluctuation index, and the fluctuation index specifically comprises the following steps: Classifying the plurality of device nodes to obtain a plurality of single-sensor device nodes and a plurality of multi-sensor device nodes; If a first device node is a single-sensor device node, reconstructing sensor data of the first device node in a phase space to obtain a phase space point set; Calculating a correlation dimension of the phase space point set and determining a fluctuation index of the first device node based on the correlation dimension; If a second device node is a multi-sensor device node, constructing a plurality of types of sensor data corresponding to the second device node into a plurality of slope curves; Selecting any one of the plurality of slope curves as a reference curve; Calculating fluctuation similarities between the plurality of slope curves and the reference curve, the fluctuation similarities comprising a fluctuation time similarity and a fluctuation direction similarity; Determining a fluctuation index of the second device node according to an average of the plurality of fluctuation similarities; Determining a plurality of abnormal device nodes according to the fluctuation indices of the plurality of device nodes; Traversing the fault influence relationship graph from the plurality of abnormal device nodes as starting points to obtain a plurality of fault chains; Calculating chain responsibility degrees of the plurality of fault chains and determining a fault chain with the largest chain responsibility degree as a root cause result; Outputting the root cause result to a device management interface to facilitate a troubleshooter to quickly troubleshoot abnormal causes.

2. The method of claim 1, wherein, The calculation of the fluctuation index of the multi-source sensor data and the updating of the fluctuation index to the preset fault influence relationship graph further comprise the following steps: Determining abnormal frequencies of the plurality of device nodes according to historical multi-source sensor data of the plurality of device nodes; Calculating influence coefficients of a plurality of influence edges in the fault influence relationship graph according to the abnormal frequencies of the plurality of device nodes.

3. The method of claim 1, wherein, The calculation of the chain responsibility degrees of the plurality of fault chains specifically comprises the following steps: Obtaining fault frequencies of the plurality of fault chains; Determining path correction coefficients of the plurality of fault chains according to the fault frequencies of the plurality of fault chains; Calculating node responsibility degrees of a plurality of device nodes in a first fault chain according to fluctuation indices corresponding to the plurality of device nodes in the first fault chain and influence edges, the first fault chain being any one of the plurality of fault chains; Calculating a chain responsibility degree of the first fault chain based on the node responsibilities of the plurality of device nodes in the first fault chain and a path correction coefficient corresponding to the first fault chain.

4. The method of claim 3, wherein, After the output of the root cause result to the device management interface, the method further comprises the following steps: Obtaining a target fault device node input by the troubleshooter; When the target fault device node belongs to the root cause result, constructing a corresponding relationship between the root cause result and the target fault device node and storing the corresponding relationship in a preset fault library.

5. A system for detecting abnormal operating conditions of waste gas purification equipment, characterized in that, The system is a detection device, and the detection device comprises an obtaining module (1), a processing module (2), and a display module (3), wherein: The acquisition module (1) is configured to acquire multi-source sensor data of a purification device. The processing module (2) is configured to calculate a fluctuation index of the multi-source sensor data and update the fluctuation index to a preset fault influence relationship graph, the fault influence relationship graph comprising a plurality of device nodes and influence edges between the plurality of device nodes, wherein one device node corresponds to one fluctuation index, and the fluctuation index specifically comprises: The plurality of device nodes are classified to obtain a plurality of single-sensor device nodes and a plurality of multi-sensor device nodes. If a first device node is a single-sensor device node, phase space reconstruction is performed on sensor data of the first device node to obtain a phase space point set. The correlation dimension of the phase space point set is calculated, and the fluctuation index of the first device node is determined based on the correlation dimension. If a second device node is a multi-sensor device node, a plurality of sensor data corresponding to the second device node are constructed into a plurality of slope curves. Any one of the plurality of slope curves is selected as a reference curve. The fluctuation similarity between the plurality of slope curves and the reference curve is calculated, the fluctuation similarity comprising fluctuation time similarity and fluctuation direction similarity. The fluctuation index of the second device node is determined according to the average of the plurality of fluctuation similarities. A plurality of abnormal device nodes are determined according to the fluctuation indices of the plurality of device nodes. A plurality of fault chains are obtained by traversing the fault influence relationship graph from the plurality of abnormal device nodes as starting points. The chain responsibility degree of the plurality of fault chains is calculated, and a fault chain with the largest chain responsibility degree is determined as a root cause result. The display module (3) is configured to output the root cause result to a device management interface, so as to facilitate a troubleshooting personnel to quickly troubleshoot an abnormal cause.

6. An electronic device, comprising: The electronic device (300) comprises a processor (301), a memory (305), a user interface (303), and a network interface (304), the memory (305) is configured to store instructions, the user interface (303) and the network interface (304) are configured to communicate with other devices, and the processor (301) is configured to execute the instructions stored in the memory (305) to enable the electronic device (300) to perform the method of any one of claims 1 to 4.

7. A computer readable storage medium characterized in that, The computer-readable storage medium stores instructions, and when the instructions are executed, the method of any one of claims 1 to 4 is performed.

Images