Fault monitoring method, device, equipment and medium

Abstract

The present invention relates to the field of fault detection technology and provides a fault monitoring method, device, equipment and medium, comprising: obtaining historical waveform data of a galvanizing unit to obtain a plurality of waveform data blocks; determining the time of occurrence of historical faults in the waveform data blocks and extracting the historical fault data; collecting real-time operation data of the galvanizing unit; and obtaining the fault data according to the operation matrix X. i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j , determine the distance matrix D i，j The optimal path P; extract the operation matrix X i，j and the fault matrix Y i，j The algorithm calculates the corresponding data points on the optimal path P in the image; calculates the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P; and issues a fault warning when the similarity exceeds a threshold. This approach improves the accuracy and real-time performance of fault detection for high-speed galvanizing lines, ensuring stable operation of the lines.

Description

Technical Field

[0001] The present disclosure relates to the field of fault detection technology, and in particular to a fault monitoring method, apparatus, device, and medium. Background Art

[0002] As the core equipment of steel mills, the operating speed of galvanizing units has been increased from the traditional tens of meters per minute to hundreds of meters per minute. While high-speed production improves efficiency, it also poses severe challenges to fault detection technology.

[0003] Existing fault detection techniques typically rely on traditional sensor monitoring and simple data analysis. While these methods may be effective at low speeds or when fault characteristics are obvious, their limitations become particularly pronounced when galvanizing lines operate at high speeds. Specifically, fault characteristics at high speeds often exhibit rapid changes and weak signals, making it difficult for traditional methods to accurately capture these characteristics in real time. Furthermore, existing techniques typically only detect obvious fault signals, but lack the ability to provide early warning of faults.

[0004] Therefore, how to solve the real-time fault identification problem when the galvanizing unit is running at high speed has become a technical problem that needs to be broken through urgently. Summary of the Invention

[0005] The present disclosure provides a fault monitoring method, device, equipment and medium for realizing real-time and accurate detection of faults of a galvanizing unit.

[0006] In a first aspect, the present disclosure provides a fault monitoring method, comprising:

[0007] Acquire historical waveform data of the galvanizing unit, and cut the historical waveform data into blocks to obtain multiple waveform data blocks;

[0008] Determine the time when a historical fault occurs in the waveform data block, and expand the adaptive time length forward and backward in the waveform data block with the time when the historical fault occurs as the center to extract the historical fault data;

[0009] Collect the real-time operation data of the galvanizing unit, and convert the operation data and the fault data into the operation matrix X of the same dimension respectively. i，j and the fault matrix Y i，j , where i and j are the sampling time point and the characteristic parameter value respectively; according to the operation matrix X i，j The difference between the fault matrix Yi,j determines the distance matrix D i，j , from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner is found; the optimal path P is mapped to the operation matrix X i，j and the fault matrix Y i，j, extract the operating matrix X i，j and the fault matrix Y i，j The corresponding data points on the optimal path P in the equation are the corresponding operation matrices X i,j and the fault matrix Y i,j The characteristic parameter value on the optimal path; calculating the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data point;

[0010] When the similarity is greater than the threshold, a fault warning is issued.

[0011] According to the fault monitoring method provided by the present disclosure, extending the adaptive duration forward and backward respectively includes:

[0012] Determine the time span from the initiation of a fault to its occurrence in the waveform data block, and determine the leading oscillation time T_pre and the subsequent decay time T_post of the historical fault based on the time span from the initiation of a fault to its occurrence.

[0013] The mean and standard deviation of the leading oscillation time T_pre and the subsequent decay time T_post are calculated respectively, and the adaptive duration is calculated based on the mean and standard deviation. The adaptive duration includes the forward extension duration T forward and backward extension time T backward .

[0014] According to the fault monitoring method provided by the present disclosure, according to the operation matrix X i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j include:

[0015] Calculate the running matrix X i,j With the fault matrix Y i,j The difference value at each sampling time point i and characteristic parameter value j is expressed as: a = |X i,j - Y i,j |;

[0016] For the fault matrix Y i,j At each time point i, a neighborhood window [iw,i+w] is selected with i as the center, and the local variance matrix Var is calculated. i,j , where w is the preset neighborhood window size parameter;

[0017] The difference value D a Multiply the fault matrix Y i，j The local variance matrix Var i,j , calculate the distance matrix D i,j .

[0018] According to the fault monitoring method provided by the present disclosure, the operation data and the fault data are converted into an operation matrix X of the same dimension. i，j and the fault matrix Y i，j include:

[0019] For i, the timestamps of the operating data and the fault data are aligned by interpolation method to make the length of the time series of the two consistent;

[0020] For j, the characteristic parameter values ​​are standardized to ensure that different characteristic parameters are comparable.

[0021] According to the fault monitoring method provided by the present disclosure, from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner includes:

[0022] Starting from the distance matrix D i，j When searching from the lower right corner to the upper left corner, the reverse search path from the upper left corner to the lower right corner is started simultaneously;

[0023] When the search boundaries of the two paths are both within the distance matrix D i，j When the paths are within the central area, the two paths are integrated and duplicate data are removed to obtain the optimal path P.

[0024] According to the fault monitoring method provided by the present disclosure, the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data points includes:

[0025] From the running matrix X i,j and the fault matrix Y i,j In the optimal path P, the corresponding data points correspond to the operation matrix X. i,j and the fault matrix Y i,j The characteristic parameter values ​​on the optimal path;

[0026] For each corresponding data point on the optimal path P, calculate the difference between the operating data and the fault data;

[0027] On the optimal path P, calculate the slope at each corresponding data point;

[0028] The similarity is obtained by multiplying the difference between the corresponding data points and the slope of the optimal path P at that point.

[0029] According to the fault monitoring method provided by the present disclosure, when the similarity is greater than a threshold, performing a fault warning includes:

[0030] Pre-set similarity threshold;

[0031] Calculating the similarity in real time and comparing it with the threshold;

[0032] When the similarity exceeds the threshold, a fault warning mechanism is triggered to send a warning signal to the operator;

[0033] At the same time, the current operation data, similarity value and warning timestamp information are recorded and stored in the system log for subsequent analysis.

[0034] In a second aspect, the present disclosure further provides a fault monitoring device, comprising:

[0035] A slicing module, which obtains historical waveform data of the galvanizing unit and slices the historical waveform data into blocks to obtain multiple waveform data blocks;

[0036] a fault module, determining a historical fault occurrence time in the waveform data block, and extending adaptive time lengths forward and backward with the historical fault occurrence time as the center in the waveform data block to extract historical fault data;

[0037] Data module, collects the real-time operation data of the galvanizing unit, and converts the operation data and the fault data into the operation matrix X of the same dimension respectively. i，j and the fault matrix Y i，j , where i and j are the sampling time point and the characteristic parameter value respectively; according to the operation matrix X i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j , from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner is found; the optimal path P is mapped to the operation matrix X i，j and the fault matrix Y i，j , extract the operating matrix X i，j and the fault matrix Y i，j The corresponding data points on the optimal path P in the equation are the corresponding operation matrices X i,j and the fault matrix Y i,j The characteristic parameter value on the optimal path; calculating the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data point;

[0038] The early warning module issues a fault warning when the similarity is greater than the threshold.

[0039] Compared with the prior art, the present invention relates to a fault monitoring method, which comprises: obtaining historical waveform data of a galvanizing unit to obtain a plurality of waveform data blocks; determining the time of occurrence of historical faults in the waveform data blocks and extracting the historical fault data; collecting real-time operation data of the galvanizing unit; and i，j and the fault matrix Y i，j The difference between the two determines the distance matrix Di，j , determine the distance matrix D i，j The optimal path P; extract the operation matrix X i，j and the fault matrix Y i，j The algorithm calculates the corresponding data points on the optimal path P in the image; calculates the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P; and issues a fault warning when the similarity exceeds a threshold. This approach improves the accuracy and real-time performance of fault detection for high-speed galvanizing lines, ensuring stable operation of the lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the present disclosure, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0041] Figure 1 A flowchart of a fault monitoring method provided by the present disclosure;

[0042] Figure 2 A schematic diagram of a matrix mapping provided by the present disclosure;

[0043] Figure 3 A schematic diagram of a fault monitoring device provided by the present disclosure;

[0044] Figure 4 A schematic diagram of the electronic device provided in the present disclosure. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of this disclosure more clear, the technical solutions of this disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of this disclosure, not all of them. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of this disclosure without creative effort shall fall within the scope of protection of this disclosure.

[0046] In modern steel production, galvanizing lines are a core component of plate surface treatment, and their operating status directly impacts product quality and production capacity. Since galvanizing lines often operate at high speeds of hundreds of meters per minute and involve multiple processes, including uncoiling, galvanizing, annealing, and coiling, a failure can not only lead to batch coating quality defects (such as missing plating and uneven thickness) but can also trigger a chain reaction, including equipment downtime for maintenance and production line shutdowns. Therefore, their normal operation is crucial to the production efficiency and safety of steel mills.

[0047] Figure 1This is a flow chart of a fault monitoring method provided by the present disclosure, such as Figure 1 As shown, the method includes:

[0048] Step 1: Acquire historical waveform data of the galvanizing unit, and cut the historical waveform data into blocks to obtain multiple waveform data blocks;

[0049] Furthermore, the data acquisition system collects historical operating data through a sensor array (including temperature sensors, current transformers, and speed encoders) deployed at key workstations in the galvanizing line. The sampling frequency is set at 10kHz-50kHz to cover transient characteristics under high-speed operating conditions (100m / min-200m / min). The collected data includes: zinc pot temperature waveform, electroplating current waveform, roller speed waveform, tension sensor waveform, etc.

[0050] The historical waveform data refers to the waveform data recorded when the galvanizing unit is operating normally and when a fault occurs. These data reflect the dynamic characteristics of the unit under different operating conditions and are the basis for subsequent fault detection and early warning.

[0051] Specifically, the historical waveform data is cut into blocks using a sliding window, and the sliding window overlap rate is set to ≥50% to avoid feature omissions; when the fluctuation amplitude of a certain characteristic parameter in the window exceeds a preset threshold, independent cutting of the current window is triggered; the waveform data blocks after cutting are standardized, and the starting point of the unified time axis is 100ms before the fault, which facilitates the subsequent alignment of fault features.

[0052] Step 2: Determine the time when a historical fault occurs in the waveform data block, and expand the adaptive time length forward and backward with the time when the historical fault occurs as the center in the waveform data block to extract the historical fault data;

[0053] Furthermore, when the characteristic parameters of a waveform data block match any fault type in a preset fault signature model, the time of historical fault occurrence in the waveform data block is determined to be the time corresponding to the matched characteristic parameters. For example, if the number of sudden changes and the slope change rate of the electroplating current waveform in a waveform data block match the abnormal characteristics of the electroplating current in the fault signature model, the time corresponding to the sudden change point is determined to be the time of fault occurrence.

[0054] Through the above steps, this embodiment realizes the accurate determination of the time when historical faults occurred in the waveform data block, and extracts complete historical fault data based on adaptive time expansion, which can effectively capture the dynamic characteristics before and after the fault occurs, and provide an accurate data basis for subsequent fault detection and early warning.

[0055] Step 3: Collect the real-time operation data of the galvanizing unit and convert the operation data and the fault data into the operation matrix X of the same dimension.i，j and the fault matrix Y i，j , where i and j are the sampling time point and the characteristic parameter value respectively; according to the operation matrix X i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j , from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner is found; the optimal path P is mapped to the operation matrix X i，j and the fault matrix Y i，j , extract the operating matrix X i，j and the fault matrix Y i，j The corresponding data points on the optimal path P in the equation are the corresponding operation matrices X i,j and the fault matrix Y i,j The characteristic parameter value on the optimal path; calculating the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data point;

[0056] Furthermore, real-time operating data is collected using a sensor array (including temperature sensors, current transformers, and speed encoders) deployed at key workstations in the galvanizing line. This sensor array is identical to the one used to collect historical waveform data in step 1, ensuring data consistency and comparability. The sampling frequency is also set between 10kHz and 50kHz to cover transient characteristics under high-speed operating conditions. Real-time data collected includes the zinc pot temperature waveform, plating current waveform, roller speed waveform, and tension sensor waveform.

[0057] The pre-processed real-time operation data and historical fault data are converted into operation matrices X i,j and the fault matrix Y i,j Where i represents the sampling time point and j represents the characteristic parameter value. Operation matrix X i,j Contains the values ​​of real-time operation data at each time point and characteristic parameters, the fault matrix Y i,j Contains the values ​​of historical fault data at various time points and characteristic parameters.

[0058] For the running matrix X i,j and the fault matrix Y i,j First, timestamp alignment is performed to ensure that the lengths of the two time series are consistent. Specifically, this involves extracting timestamp information from the operating and fault data and identifying the time series range and sampling frequency. If the time series lengths are inconsistent, adjustments are made through data cropping to ensure that the time series of the operating and fault data are fully aligned. During the alignment process, the data's time sequence and sampling intervals are kept consistent to avoid analysis errors caused by time deviations.

[0059] Further, such as Figure 2 As shown, the left figure shows the operation matrix X i,j and the fault matrix Y i,j The distance matrix D calculated by the difference i,j Each element in the matrix represents the difference measure at the corresponding time point and feature parameter value.

[0060] Using the dynamic programming algorithm, we start from the lower right corner of the matrix and find the optimal path P to the upper left corner. The optimal path P is represented by a line segment, which represents the path with the smallest difference metric, that is, the path where the operating data and the fault data are most similar in time series and characteristic parameters. Figure 2 D on the left i,j The optimal path P includes 13 points P t , t is a natural number from 1 to 13.

[0061] Map the optimal path P to the operation matrix X i,j The corresponding data point on the path is extracted. This data point reflects the operating status of the galvanizing unit under the time series and characteristic parameters represented by the optimal path P. Figure 2 X in the middle i,j The optimal mapping path P includes 13 data points X t .

[0062] Similarly, the optimal path P is mapped to the failure matrix Y i,j The corresponding data point on the path is extracted. This data point reflects the fault state of the galvanizing unit under the time series and characteristic parameters represented by the optimal path P, as shown in the attached figure. Figure 2 Y on the right i,j The optimal mapping path P includes 13 data points Y t .

[0063] Calculating the difference of the corresponding data points includes calculating X t With Y t The difference.

[0064] Through the above steps, this embodiment realizes the i,j Extract the optimal path P and map it to the operation matrix X i,j and the fault matrix Y i,j The method can effectively capture the similarity between operating data and fault data, provide a scientific basis for fault early warning, and significantly improve the accuracy and real-time performance of fault detection.

[0065] Step 4: When the similarity is greater than the threshold, a fault warning is issued.

[0066] Furthermore, the relevant data when the fault warning is triggered is stored in the data storage unit, and each warning event is assigned a unique identifier, which contains information such as the warning timestamp, similarity value, and operation data snapshot, which facilitates subsequent data tracing and analysis.

[0067] Compared with the prior art, the fault monitoring method disclosed in the present invention obtains historical waveform data of the galvanizing unit to obtain multiple waveform data blocks; determines the time when the historical fault occurs in the waveform data block and extracts the historical fault data; collects the real-time operation data of the galvanizing unit; and calculates the fault status of the galvanizing unit according to the operation matrix X. i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j , determine the distance matrix D i，j The optimal path P; extract the operation matrix X i，j and the fault matrix Y i，j The algorithm calculates the corresponding data points on the optimal path P in the image; calculates the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P; and issues a fault warning when the similarity exceeds a threshold. This approach improves the accuracy and real-time performance of fault detection for high-speed galvanizing lines, ensuring stable operation of the lines.

[0068] In one embodiment, extending the adaptive duration forward and backward in step 2 includes:

[0069] Determine the time span from the initiation of a fault to its occurrence in the waveform data block, and determine the leading oscillation time T_pre and the subsequent decay time T_post of the historical fault based on the time span from the initiation of a fault to its occurrence.

[0070] Specifically, based on the determined historical fault occurrence time, the leading oscillation time and subsequent decay time in the waveform data block are further analyzed. The leading oscillation time T_pre refers to the length of time the waveform data exhibits abnormal fluctuations before the fault occurs, while the subsequent decay time T_post refers to the length of time the waveform data gradually returns to stability after the fault occurs.

[0071] The mean and standard deviation of the leading oscillation time T_pre and the subsequent decay time T_post are calculated respectively, and the adaptive duration is calculated based on the mean and standard deviation. The adaptive duration includes the forward extension duration T forward and backward extension time T backward;

[0072] Specifically, after obtaining the leading oscillation time and subsequent decay time in multiple historical fault data blocks, the mean value μ of the leading oscillation time T_pre is calculated. pre and standard deviation σ pre; Calculate the mean μ of the subsequent decay time T_post post and standard deviation σ post ;

[0073] Based on the mean and standard deviation of the leading oscillation time and the subsequent decay time, the adaptive extension duration is calculated. The adaptive extension duration includes the forward extension duration T forward and backward extension time T backward .

[0074] T forward =α×(μ pre +k×σ pre );

[0075] T backward =β×(μ post +k×σ post );

[0076] α and β are preset proportional coefficients, both in the range of [0.5, 2], which can be adjusted according to actual working conditions and fault characteristics; k is used to control the flexibility of the extension time, in the range of [1, 3].

[0077] In the waveform data block, take the time of historical fault occurrence as the center and expand forward T forward Duration, extended backward T backward The waveform data within this time period is extracted as historical fault data. In this way, the extracted historical fault data can be ensured to fully cover the key features before and after the fault occurs, providing high-quality data support for subsequent fault analysis and model training.

[0078] In one embodiment, the operation matrix X in step 3 is used to i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j include:

[0079] Calculate the running matrix X i,j With the fault matrix Y i,j The difference value at each sampling time point i and characteristic parameter value j is expressed as: a = |X i,j - Y i,j |;

[0080] For the fault matrix Y i,j At each time point i, a neighborhood window [iw,i+w] is selected with i as the center, and the local variance matrix Var is calculated. i,j , where w is the preset neighborhood window size parameter;

[0081] The difference value Da With the fault matrix Y i，j The local variance matrix Var i,j Multiply and calculate the distance matrix D i,j .

[0082] In one embodiment, the operation data and the fault data are converted into an operation matrix X of the same dimension. i，j and the fault matrix Y i，j include:

[0083] For i, the timestamps of the operating data and the fault data are aligned by interpolation method to make the length of the time series of the two consistent;

[0084] For j, the characteristic parameter values ​​are standardized to ensure that different characteristic parameters are comparable.

[0085] In one embodiment, from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner includes:

[0086] Starting from the distance matrix D i，j When searching from the lower right corner to the upper left corner, the reverse search path from the upper left corner to the lower right corner is started simultaneously;

[0087] When the search boundaries of the two paths are both within the distance matrix D i，j When the paths are within the central area, the two paths are integrated and duplicate data are removed to obtain the optimal path P.

[0088] Furthermore, the forward search path includes: i,j Initialize the starting point of the forward search path starting at the bottom right corner of the matrix (i.e., the last time point and the last characteristic parameter point in the matrix). Set the cumulative distance of this point to the distance value of that point. Expand the path step by step from the bottom right corner to the top left corner. At each step, consider three possible paths from the current point: upward, left, or diagonal. Compare the cumulative distances of these three paths and select the path with the smallest cumulative distance as the next movement direction. The cumulative distance is the sum of all distance values ​​along the path from the starting point to the current point.

[0089] Reverse search path: At the same time, from the distance matrix D i,j Initialize the starting point of the reverse search path starting from the top left corner of the matrix (i.e., the first time point and the first characteristic parameter point). Set the cumulative distance of this point to the distance value of that point. Gradually expand the path from the top left corner to the bottom right corner. At each step, consider three possible paths from the current point: downward, right, or diagonal. Again, compare the cumulative distances of these three paths and select the path with the smallest cumulative distance as the next direction of movement.

[0090] When the search boundaries of both the forward search path and the reverse search path enter the distance matrix D i,j When the two paths are within the central area of ​​the distance matrix, it is determined whether the two paths meet. The central area range can be defined by a preset threshold. For example, when the path enters the middle 1 / 3 area of ​​the distance matrix, the path is considered to have entered the central area.

[0091] Check whether the current nodes of the forward search path and the reverse search path coincide or are adjacent. If so, the two paths are considered to have met.

[0092] When two paths meet, the forward search path and the reverse search path are integrated:

[0093] Remove duplicate nodes: If two paths have duplicate nodes at their intersection, remove the duplicate nodes to ensure path uniqueness. The forward and reverse search paths are joined at the intersection to form the complete optimal path P. This joined path is optimized to ensure smoothness and continuity. This optimization may include adjusting the path slope to avoid large jumps or sudden changes.

[0094] In one embodiment, the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data points includes:

[0095] From the running matrix X i,j and the fault matrix Y i,j In the optimal path P, the corresponding data points correspond to the operation matrix X. i,j and the fault matrix Y i,j The characteristic parameter values ​​on the optimal path;

[0096] For each corresponding data point on the optimal path P, calculate the difference between the operating data and the fault data;

[0097] On the optimal path P, calculate the slope at each corresponding data point;

[0098] The similarity is obtained by multiplying the difference between the corresponding data points and the slope of the optimal path P at that point.

[0099] In one embodiment, when the similarity is greater than a threshold, performing a fault warning includes:

[0100] Pre-set similarity threshold;

[0101] Calculating the similarity in real time and comparing it with the threshold;

[0102] When the similarity exceeds the threshold, a fault warning mechanism is triggered to send a warning signal to the operator;

[0103] At the same time, the current operation data, similarity value and warning timestamp information are recorded and stored in the system log for subsequent analysis.

[0104] A fault monitoring device provided by the present disclosure is described below. The detection system described below and the detection method described above can be referenced to each other.

[0105] like Figure 3 As shown, a fault monitoring device includes:

[0106] A slicing module, which obtains historical waveform data of the galvanizing unit and slices the historical waveform data into blocks to obtain multiple waveform data blocks;

[0107] a fault module, determining a historical fault occurrence time in the waveform data block, and extending adaptive time lengths forward and backward with the historical fault occurrence time as the center in the waveform data block to extract historical fault data;

[0108] Data module, collects the real-time operation data of the galvanizing unit, and converts the operation data and the fault data into the operation matrix X of the same dimension respectively. i，j and the fault matrix Y i，j , where i and j are the sampling time point and the characteristic parameter value respectively; according to the operation matrix X i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j , from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner is found; the optimal path P is mapped to the operation matrix X i，j and the fault matrix Y i，j , extract the operating matrix X i，j and the fault matrix Y i，j The corresponding data points on the optimal path P in the equation are the corresponding operation matrices X i,j and the fault matrix Y i,j The characteristic parameter value on the optimal path; calculating the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data point;

[0109] The early warning module issues a fault warning when the similarity is greater than the threshold.

[0110] An electronic device provided in an embodiment of the present application is Figure 4 As shown, the electronic device includes a processor 401 and a memory 403, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the steps of the method provided in the above embodiment are implemented.

[0111] See also Figure 4 The electronic device further includes: a bus 404 and a communication interface 402, a processor 401, a communication interface 402 and a memory 403 connected via the bus 404; the processor 401 is used to execute executable modules stored in the memory 403, such as computer programs.

[0112] Memory 403 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk drive. Communication between the system network element and at least one other network element is achieved via at least one communication interface 402 (which may be wired or wireless), and may utilize the Internet, a wide area network, a local area network, a metropolitan area network, or the like.

[0113] The bus 404 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0114] Among them, the memory 403 is used to store programs, and the processor 401 executes the program after receiving the execution instruction. The method executed by the device defined by the process disclosed in any embodiment of the present application can be applied to the processor 401 or implemented by the processor 401.

[0115] The processor 401 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by an integrated logic circuit of hardware in the processor 401 or by instructions in the form of software. The above-mentioned processor 401 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in memory 403, and processor 401 reads the information in memory 403 and performs the steps of the above method in conjunction with its hardware.

[0116] Corresponding to the above-mentioned fault monitoring method, an embodiment of the present application also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer-readable storage medium stores computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are called and executed by the processor, the computer-executable instructions prompt the processor to execute the steps of the above-mentioned fault monitoring method.

[0117] The fault monitoring device provided in the embodiment of the present application can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in the embodiment of the present application are the same as those of the aforementioned method embodiment. For the sake of brief description, for any part not mentioned in the device embodiment, reference can be made to the corresponding content in the aforementioned method embodiment. Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0118] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0119] For another example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0120] 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, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0121] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0122] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the fault monitoring method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0123] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0124] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The scope of protection of the present application is not limited thereto. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-mentioned embodiments within the technical scope disclosed in the present application, or perform equivalent replacements for some of the technical features thereof. However, these modifications, changes, or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application. They should all be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A fault monitoring method, characterized in that: include: Acquire historical waveform data of the galvanizing unit, and cut the historical waveform data into blocks to obtain multiple waveform data blocks; Determine the time when a historical fault occurs in the waveform data block, and expand the adaptive time length forward and backward in the waveform data block with the time when the historical fault occurs as the center to extract the historical fault data; Collect the real-time operation data of the galvanizing unit, and convert the operation data and the fault data into the operation matrix X of the same dimension respectively. i，j and the fault matrix Y i，j , where i and j are the sampling time point and the characteristic parameter value respectively; according to the operation matrix X i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j , from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner is found; the optimal path P is mapped to the operation matrix X i，j and the fault matrix Y i，j , extract the operating matrix X i，j and the fault matrix Y i，j The corresponding data points on the optimal path P in the equation are the corresponding operation matrices X i,j and the fault matrix Y i,j The characteristic parameter value on the optimal path; calculating the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data point; When the similarity is greater than the threshold, a fault warning is issued.

2. The fault monitoring method according to claim 1, characterized in that: The forward and backward extension of the adaptive duration includes: Determine the time span from the initiation of a fault to its occurrence in the waveform data block, and determine the leading oscillation time T_pre and the subsequent decay time T_post of the historical fault based on the time span from the initiation of a fault to its occurrence. The mean and standard deviation of the leading oscillation time T_pre and the subsequent decay time T_post are calculated respectively, and the adaptive duration is calculated based on the mean and standard deviation. The adaptive duration includes the forward extension duration T forward and backward extension time T backward .

3. The fault monitoring method according to claim 1, characterized in that: According to the operation matrix X i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j include: Calculate the running matrix X i,j With the fault matrix Y i,j The difference value at each sampling time point i and characteristic parameter value j is expressed as: a = |X i,j - Y i,j |; For the fault matrix Y i,j At each time point i, a neighborhood window [iw,i+w] is selected with i as the center, and the local variance matrix Var is calculated. i,j , where w is the preset neighborhood window size parameter; The difference value D a With the fault matrix Y i，j The local variance matrix Var i,j Multiply and calculate the distance matrix D i,j .

4. The fault monitoring method according to claim 1, characterized in that: The operation data and the fault data are converted into an operation matrix X of the same dimension i，j and the fault matrix Y i，j include: For i, the timestamps of the operating data and the fault data are aligned by interpolation method to make the length of the time series of the two consistent; For j, the characteristic parameter values ​​are standardized to ensure that different characteristic parameters are comparable.

5. The fault monitoring method according to claim 1, characterized in that: The distance matrix D i，j The optimal path P from the lower right corner to the upper left corner includes: Starting from the distance matrix D i，j When searching from the lower right corner to the upper left corner, the reverse search path from the upper left corner to the lower right corner is started simultaneously; When the search boundaries of the two paths are both within the distance matrix D i，j When the paths are within the central area, the two paths are integrated and duplicate data are removed to obtain the optimal path P.

6. The fault monitoring method according to claim 1, characterized in that: The product of the difference based on the corresponding data point and the slope of the optimal path P at the corresponding data point includes: From the running matrix X i,j and the fault matrix Y i,j In the optimal path P, the corresponding data points correspond to the operation matrix X. i,j and the fault matrix Y i,j The characteristic parameter values ​​on the optimal path; For each corresponding data point on the optimal path P, calculate the difference between the operating data and the fault data; On the optimal path P, calculate the slope at each corresponding data point; The similarity is obtained by multiplying the difference between the corresponding data points and the slope of the optimal path P at that point.

7. The fault monitoring method according to claim 1, characterized in that: When the similarity is greater than the threshold, performing a fault warning includes: Pre-set similarity threshold; Calculating the similarity in real time and comparing it with the threshold; When the similarity exceeds the threshold, a fault warning mechanism is triggered to send a warning signal to the operator; At the same time, the current operation data, similarity value and warning timestamp information are recorded and stored in the system log for subsequent analysis.

8. A fault monitoring device, characterized in that: include: A slicing module, which obtains historical waveform data of the galvanizing unit and slices the historical waveform data into blocks to obtain multiple waveform data blocks; a fault module, determining a historical fault occurrence time in the waveform data block, and extending adaptive time lengths forward and backward with the historical fault occurrence time as the center in the waveform data block to extract historical fault data; Data module, collects the real-time operation data of the galvanizing unit, and converts the operation data and the fault data into the operation matrix X of the same dimension respectively. i，j and the fault matrix Y i，j , where i and j are the sampling time point and the characteristic parameter value respectively; according to the operation matrix X i，j and the fault matrix Y i，j The difference between the two determines the distance matrix D i，j , from the distance matrix D i，j The optimal path P from the lower right corner to the upper left corner is found; the optimal path P is mapped to the operation matrix X i，j and the fault matrix Y i，j , extract the operating matrix X i，j and the fault matrix Y i，j The corresponding data points on the optimal path P in the equation are the corresponding operation matrices X i,j and the fault matrix Y i,j The characteristic parameter value on the optimal path; calculating the similarity based on the product of the difference between the corresponding data points and the slope of the optimal path P at the corresponding data point; The early warning module issues a fault warning when the similarity is greater than the threshold.

9. An electronic device, characterized in that: include: processor; A memory storing a program, wherein the program comprises instructions which, when executed by the processor, cause the processor to perform the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer-readable storage medium stores instructions or a computer program, and when the instructions or the computer program are executed on a device, the device is caused to execute the method according to any one of claims 1 to 7.

Images