Laminating machine remote data control system and method based on PLC control

Abstract

The invention discloses a remote data control system and method for a laminating machine based on PLC control, and belongs to the technical field of industrial automation control. The system comprises a data acquisition and preprocessing module, a simulation model construction and data analysis module, a fault diagnosis and anomaly detection module and an abnormal change trend analysis and feedback module. The data acquisition and preprocessing module is used for acquiring historical sensor data and laminating machine fault logs and marking historical fault event data segments; the simulation model construction and data analysis module is used for constructing a simulation model of the laminating machine by utilizing simulation software, generating simulation data, enabling the simulation data to correspond to historical fault event data segments, and carrying out trend analysis; the fault diagnosis and anomaly detection module analyzes a corresponding relation between a historical fault event data segment and simulation data so as to obtain a fault event trend anomaly point; and the abnormal change trend analysis and feedback module performs trend analysis according to the real-time sensor data and outputs corresponding prompt information.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation control, and in particular to a remote data control system and method for a laminating machine based on PLC control. Background Art

[0002] In the current field of industrial automation, laminating machines, as an important production equipment, are widely used in the pressing and processing of various materials. With the development of technology, the demand for real-time monitoring and remote control of the operating status of laminating machines is increasing. Traditional laminating machine control systems can usually only realize on-site operation and control, and cannot meet the needs of remote monitoring and management. PLC (Programmable Logic Controller), as a powerful industrial automation control device, has high reliability and stability and has been widely used in various industrial control occasions. Using PLC for control can realize accurate monitoring and precise control of the operating status of the laminating machine, thereby improving production efficiency and product quality.

[0003] Although the current PLC-based laminating machine remote data control system can realize the remote control of the laminating machine, there are still some shortcomings. For example, when the existing technology uses PLC to remotely control the laminating machine, the fault detection of the laminating machine often relies on pre-set fault modes and characteristics. This fault detection method has some limitations, especially when dealing with new faults or emergencies, and therefore may not be able to respond to new problems or abnormal situations in a timely manner. Summary of the invention

[0004] The object of the present invention is to provide a laminating machine remote data control system and method based on PLC control to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A remote data control method for a laminating machine based on PLC control, the method comprising the following steps:

[0007] Step S100. Obtain historical sensor data of the laminator operation collected by the PLC from the database, and correspond the historical sensor data with the fault event timestamp according to the fault event timestamp recorded in the fault log of the laminator in the database; based on the correspondence result, for each fault event, obtain the historical sensor data of the time period where the fault event is located, and mark it as a historical fault event data segment;

[0008] Step S200. Acquire simulation data of the PLC-controlled laminating machine, divide the simulation data into a plurality of data segments according to the timestamp of the fault event, and correspond the data segments of the simulation data with the corresponding data segments of the historical fault event; perform trend analysis on the data segments of the simulation data and the data segments of the historical fault event respectively, so as to obtain trend characteristics of the simulation data and trend characteristics of the historical fault event;

[0009] Step S300. Based on the correspondence between the historical fault event data segments and the simulation data data segments, the historical fault event trend characteristics are analyzed with the corresponding simulation data trend characteristics to obtain the fault event trend abnormal points;

[0010] Step S400. Collect real-time sensor data of the laminator through PLC, perform trend analysis on the real-time sensor data every selected time period, and obtain corresponding real-time event trend characteristics; analyze the real-time event trend characteristics, and perform corresponding processing according to the analysis results.

[0011] Furthermore, step S100 includes:

[0012] S101. Obtain historical sensor data and laminator fault log from the database, and mark the fault event timestamp on the timeline corresponding to the historical sensor data according to the fault event timestamp recorded in the laminator fault log, so as to correspond the historical sensor data to the fault event timestamp;

[0013] S102. Based on the fault event timestamp, the time period of a time length t before and after the fault event timestamp is taken as the time period of the corresponding fault event; for each fault event, the historical sensor data of the time period of the fault event is obtained and marked as a historical fault event data segment; the historical fault event data segments of all fault events are aggregated to form a fault event data segment set G, and G = {g1, g2, ..., gn}, where g1 represents the historical fault event data segment corresponding to the first fault event, g2 represents the historical fault event data segment corresponding to the second fault event, and so on, gn represents the historical fault event data segment corresponding to the nth fault event; and n represents the number of fault events, which is a positive integer.

[0014] Further, step S200 includes:

[0015] S201. According to the task requirements of the laminator, a simulation model is constructed using simulation software, a PLC control program is written to control the operation of the laminator in the simulation model, and simulation data of the PLC-controlled laminator is obtained, and the timeline of the simulation data completely overlaps with the timeline of the historical sensor data; according to the time period of the fault event, the simulation data is divided into several data segments, and the data segments of the simulation data corresponding to the same time period are matched with the data segments of the historical fault event;

[0016] S202. Perform trend analysis on the data segment of the simulation data and the data segment of the historical fault event respectively, and the trend analysis process of the data segment of the simulation data and the data segment of the historical fault event is completely consistent. The specific trend analysis process is as follows:

[0017] The data segment x(t) is decomposed by wavelet to obtain the wavelet coefficients W(j,k) of different scales j and positions k, and the calculation formula corresponding to the wavelet coefficients W(j,k) is:

[0018] W(j,k)= <x(t),ψ j,k (t)>=∫x(t)·ψ * j,k (t)dt;

[0019] Among them, ψ * j,k (t) represents the wavelet basis function, j represents the scale parameter, and k represents the location parameter;

[0020] Perform spectral analysis on the wavelet coefficients W(j,k) of each scale j and calculate the corresponding energy spectrum E j (ω), the specific calculation formula is:

[0021] E j (ω)=(1 / N)Σ k |W j,k (ω)|^2;

[0022] Among them, W j,k (ω) represents the amplitude of the wavelet coefficient at frequency ω, and N represents the number of wavelet coefficients;

[0023] According to the energy spectrum E j (ω), extract the characteristics of the data segment at different frequency scales and summarize them, and the data segment refers to the data segment of the simulation data or the data segment of the historical fault event, so as to obtain the trend characteristics of the simulation data and the trend characteristics of the historical fault event.

[0024] The present invention adopts a multidimensional data analysis method, including wavelet analysis, spectrum analysis, etc., to compare and analyze the historical fault event data segments with the simulation data, and comprehensively considers the characteristics at different scales to make the fault detection process more comprehensive and accurate.

[0025] Furthermore, step S300 includes:

[0026] S301. According to the correspondence between the historical fault event data segments and the simulation data data segments, the historical fault event data segments and the simulation data data segments corresponding to the same time period are extracted, so as to obtain the corresponding historical fault event trend characteristics and simulation data trend characteristics, and the historical fault event trend characteristics and the simulation data trend characteristics are format-converted respectively, and converted into data representations of the same format, so as to form the historical fault event trend feature vector L and the simulation data trend feature vector M;

[0027] S302. The historical fault event trend feature vector L and the simulation data trend feature vector M are both represented on the same radar chart; the radar chart is composed of a starting point and a plurality of directed line segments, and the number of directed line segments is equal to the dimension of the historical fault event trend feature vector L and the simulation data trend feature vector M; the eigenvalues ​​corresponding to the historical fault event trend feature vector L and the simulation data trend feature vector M are marked on the radar chart in sequence, and the points on the two adjacent directed line segments are connected in sequence according to the order of the vector eigenvalues, so as to obtain the closed polygon A corresponding to the historical fault event trend feature vector L and the closed polygon B corresponding to the simulation data trend feature vector M;

[0028] S303. For the closed polygons A and B corresponding to the historical fault event trend feature vector L and the simulation data trend feature vector M, the vertex deviation values ​​R on all directed line segments are counted in turn, and the vertex deviation value R is equal to the vertex difference on the same directed line segment of the closed polygon A and the closed polygon B; for the vertex deviation values ​​R on all directed line segments, the corresponding average value μ and standard deviation σ are calculated, the vertex deviation value R is compared with the corresponding deviation value interval Q, and the vertex of the closed polygon A whose first vertex deviation value R does not belong to the deviation value interval Q is taken as the target point, and the interval range of the deviation value interval Q is [μ-eσ,μ+eσ], where e represents a coefficient, which is used to define the range of the deviation value; according to the target point, the characteristic value in the historical fault event trend feature vector L is obtained, and the frequency scale of the corresponding historical fault event data segment is obtained based on the characteristic value, so as to find the abnormal point of the fault event trend in the historical fault event data segment;

[0029] Obtain the fault event trend abnormal points in all historical fault event data segments, and associate the fault codes corresponding to the historical fault events with the fault event trend abnormal points, so as to obtain the association relationship between the historical fault events and the fault event trend abnormal points; obtain the historical fault event data of a time period of a time length t before and after the fault event trend abnormal points, so as to obtain the abnormal change trend curve Y of the historical fault event data, so as to obtain the association relationship between the fault code, the fault event trend abnormal points and the abnormal change trend curve Y.

[0030] The introduction of radar chart and polygon comparison methods can intuitively compare the trend characteristics of historical fault events with the trend characteristics of real-time events, further improving the accuracy and efficiency of fault diagnosis.

[0031] Furthermore, step S400 includes:

[0032] S401. Use PLC to collect real-time sensor data of the laminator, obtain the real-time sensor data segment of the real-time sensor data in the selected time period every selected time period, and obtain the corresponding data segment of the simulation data according to the selected time period; perform trend analysis on the real-time sensor data segment and the simulation data segment respectively, so as to obtain the corresponding real-time event trend characteristics and simulation data trend characteristics, perform format conversion on the real-time event trend characteristics and the simulation data trend characteristics respectively, and obtain the real-time event trend characteristic vector S and the simulation data trend characteristic vector M1 respectively; wherein the simulation data trend characteristic vector M and the simulation data trend characteristic vector M1 have differences in the corresponding time periods;

[0033] S402. For the real-time event trend feature vector S and the simulation data trend feature vector M1, refer to the analysis process of the historical fault event trend feature vector L and the simulation data trend feature vector M in S302, so as to obtain the closed polygon A1 corresponding to the real-time event trend feature vector S, and the closed polygon B1 corresponding to the simulation data trend feature vector M1; refer to the analysis process of S303, if there is no real-time event trend abnormal point in the real-time sensor data segment, no notification is made, and the analysis of the real-time sensor data of the next selected time period is performed;

[0034] If there is a real-time event trend abnormal point in the real-time sensor data segment, the time period of a time length t before and after the real-time event trend abnormal point is taken as the corresponding time period of the real-time event, and the corresponding real-time sensor data is obtained according to the time period of the real-time event, so as to obtain the abnormal change trend curve Y1 of the real-time sensor data, and the real-time event trend abnormal point is associated with the abnormal change trend curve Y1;

[0035] S403. Perform similarity analysis on the abnormal change trend curve Y1 and the abnormal change trend curve Y corresponding to the historical fault event. If there is an abnormal change trend curve Y with a similarity greater than the similarity threshold, the abnormal change trend curve Y with the greatest similarity is selected as the matching result, and the fault code corresponding to the abnormal change trend curve Y is output; if there is an abnormal change trend curve Y with a similarity less than or equal to the similarity threshold, the current abnormal change trend curve Y1 is output, and the event code is marked for the current real-time event, and output to relevant personnel for analysis. If the current real-time event is a new fault, the current event code, the real-time event trend abnormal point and the abnormal change trend curve Y1 are associated, and the above-mentioned association relationship is added to the database.

[0036] According to the processing results of real-time events, relevant information is added to the database, and the fault diagnosis model is continuously optimized to adapt to new fault conditions that may arise in the future, thereby improving the intelligence and adaptability of the overall system.

[0037] A remote data control system for a laminating machine based on PLC control, the system comprises: a data acquisition and preprocessing module, a simulation model building and data analysis module, a fault diagnosis and anomaly detection module, and an abnormal change trend analysis and feedback module;

[0038] The data acquisition and preprocessing module is responsible for obtaining historical sensor data and laminator fault logs from the database, matching the historical sensor data with the fault event timestamp, and marking the historical fault event data segments; obtaining the historical fault event data segments, and summarizing them to form a set of fault event data segments;

[0039] The simulation model building and data analysis module uses simulation software to build a simulation model of the laminating machine, writes a PLC control program according to task requirements, generates simulation data and corresponds it with the historical fault event data segment, and performs trend analysis to obtain the trend characteristics of the simulation data and the trend characteristics of the historical fault events;

[0040] The fault diagnosis and anomaly detection module analyzes the correspondence between the historical fault event data segments and the simulation data, extracts the corresponding simulation data trend feature vectors and the historical fault event trend feature vectors, and presents them in radar charts, calculates the vertex deviation value, and thus obtains the fault event trend anomaly points;

[0041] The abnormal change trend analysis and feedback module performs trend analysis based on real-time sensor data, compares the real-time sensor data corresponding to the abnormal point of the real-time event trend with the data segment corresponding to the historical fault event for the similarity of the corresponding abnormal change trend curve, outputs the matching result or marks the new fault event, and provides it to relevant personnel for further analysis and processing.

[0042] The data acquisition and preprocessing module includes a database data acquisition unit, a fault event timestamp marking unit, and a historical fault event data segment extraction unit;

[0043] The database data acquisition unit acquires the historical sensor data collected by the PLC and the laminator fault log from the database; the fault event timestamp marking unit matches the fault event timestamp with the historical sensor data timeline, and marks the fault event timestamp on the historical sensor data timeline; the historical fault event data segment extraction unit extracts the historical sensor data of the time period where the fault event is located according to the fault event timestamp, and forms a historical fault event data segment set.

[0044] Further, the simulation model construction and data analysis module includes a simulation model establishment unit, a simulation data generation unit, and a data segment division and correspondence unit;

[0045] The simulation model building unit uses simulation software to build a simulation model of the laminating machine and writes a PLC control program to simulate the operation of the laminating machine; the simulation data generation unit obtains the simulation data of the simulation model so that the time line of the simulation data completely corresponds to the time line of the historical sensor data; the data segment division and correspondence unit divides the simulation data into several data segments according to the fault event timestamp, and corresponds them to the historical fault event data segments.

[0046] Further, the fault diagnosis and anomaly detection module includes a trend analysis unit and a fault event trend anomaly point detection unit;

[0047] The trend analysis unit performs wavelet decomposition and spectrum analysis on the simulation data segment and the historical fault event data segment to obtain corresponding simulation data trend characteristics and historical fault event trend characteristics.

[0048] The abnormal change trend analysis and feedback module includes a real-time sensor data acquisition unit, a real-time event trend feature extraction unit, an abnormal change trend comparison unit and an abnormal change trend feedback unit;

[0049] Furthermore, the real-time sensor data acquisition unit periodically acquires the real-time sensor data of the laminator; the real-time event trend feature extraction unit performs trend analysis on the real-time sensor data to obtain the trend features of the real-time events; the abnormal change trend comparison unit compares the trend features of the real-time events with the trend features of the simulation data to detect whether there are abnormal points in the real-time event trends; the abnormal change trend feedback unit does not process the real-time sensor data without abnormal points in the real-time event trends, and only records the normal state; the real-time sensor data with abnormal points in the real-time event trends are analyzed for abnormal change trend curves, the real-time sensor data and the historical fault event data are analyzed for corresponding trend curves, and the trend curves of the two are matched and analyzed, and the corresponding notification information is output.

[0050] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention collects sensor data of the laminator in real time, and combines historical fault event data to perform trend analysis on the real-time data, so as to timely discover and identify abnormal points in the fault event trend; this method not only relies on pre-set fault modes and characteristics, but also can effectively deal with new faults or emergencies, thereby improving the timeliness and accuracy of fault detection; by comparing the trend characteristics of simulation data with the trend characteristics of historical fault events, the present invention can identify potential abnormal points in the fault trend, thereby realizing early prediction of laminator faults; this prediction capability helps maintenance personnel to take preventive measures in advance to reduce production interruptions and equipment damage caused by faults; the present invention associates fault codes, abnormal points in the fault event trend, and abnormal change trend curves to form a complete fault diagnosis system; when the real-time sensor data is abnormal, the system can quickly match the historical fault data, quickly locate the cause of the fault, simplify the fault diagnosis process, and improve the efficiency of fault diagnosis; the fault detection method of the present invention does not rely on fixed fault modes and characteristics, but identifies faults by analyzing data trends; this method enables the system to flexibly deal with various complex situations, and with the accumulation of data and the advancement of analysis technology, the performance of the system will continue to improve. In addition, when new fault types emerge, the system is able to expand its fault diagnosis capabilities by learning new data features. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0052] Figure 1 It is a schematic diagram of a remote data control system module of a laminating machine based on PLC control of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] See also Figure 1 , the present invention provides a technical solution:

[0055] A remote data control system for a laminating machine based on PLC control, the system comprises: a data acquisition and preprocessing module, a simulation model building and data analysis module, a fault diagnosis and anomaly detection module, and an abnormal change trend analysis and feedback module;

[0056] The data acquisition and preprocessing module is responsible for obtaining historical sensor data and laminator fault logs from the database, matching the historical sensor data with the fault event timestamp, and marking the historical fault event data segments; obtaining the historical fault event data segments, and summarizing them to form a set of fault event data segments;

[0057] The simulation model building and data analysis module uses simulation software to build a simulation model of the laminating machine, writes a PLC control program according to task requirements, generates simulation data and corresponds it with the historical fault event data segment, and performs trend analysis to obtain the trend characteristics of the simulation data and the trend characteristics of the historical fault events;

[0058] The fault diagnosis and anomaly detection module analyzes the correspondence between the historical fault event data segments and the simulation data, extracts the corresponding simulation data trend feature vectors and the historical fault event trend feature vectors, and presents them in radar charts, calculates the vertex deviation value, and thus obtains the fault event trend anomaly points;

[0059] The abnormal change trend analysis and feedback module performs trend analysis based on real-time sensor data, compares the real-time sensor data corresponding to the abnormal point of the real-time event trend with the data segment corresponding to the historical fault event for the similarity of the corresponding abnormal change trend curve, outputs the matching result or marks the new fault event, and provides it to relevant personnel for further analysis and processing.

[0060] The data acquisition and preprocessing module includes a database data acquisition unit, a fault event timestamp marking unit, and a historical fault event data segment extraction unit;

[0061] The database data acquisition unit acquires the historical sensor data collected by the PLC and the laminator fault log from the database; the fault event timestamp marking unit matches the fault event timestamp with the historical sensor data timeline, and marks the fault event timestamp on the historical sensor data timeline; the historical fault event data segment extraction unit extracts the historical sensor data of the time period where the fault event is located according to the fault event timestamp, and forms a historical fault event data segment set.

[0062] The simulation model building and data analysis module includes a simulation model building unit, a simulation data generating unit, and a data segment division and corresponding unit;

[0063] The simulation model building unit uses simulation software to build a simulation model of the laminating machine and writes a PLC control program to simulate the operation of the laminating machine; the simulation data generation unit obtains the simulation data of the simulation model so that the time line of the simulation data completely corresponds to the time line of the historical sensor data; the data segment division and correspondence unit divides the simulation data into several data segments according to the fault event timestamp, and corresponds them to the historical fault event data segments.

[0064] The fault diagnosis and anomaly detection module includes a trend analysis unit and a fault event trend anomaly point detection unit;

[0065] The trend analysis unit performs wavelet decomposition and spectrum analysis on the simulation data segment and the historical fault event data segment to obtain corresponding simulation data trend characteristics and historical fault event trend characteristics.

[0066] The abnormal change trend analysis and feedback module includes a real-time sensor data acquisition unit, a real-time event trend feature extraction unit, an abnormal change trend comparison unit and an abnormal change trend feedback unit;

[0067] The real-time sensor data acquisition unit periodically acquires the real-time sensor data of the laminating machine; the real-time event trend feature extraction unit performs trend analysis on the real-time sensor data to obtain the trend features of the real-time events; the abnormal change trend comparison unit compares the trend features of the real-time events with the trend features of the simulation data to detect whether there are abnormal points in the real-time event trends; the abnormal change trend feedback unit does not process the real-time sensor data without abnormal points in the real-time event trends, and only records the normal state; the real-time sensor data with abnormal points in the real-time event trends are analyzed for abnormal change trend curves, the real-time sensor data and the historical fault event data are analyzed for the corresponding trend curves, and the trend curves of the two are matched and analyzed, and the corresponding notification information is output.

[0068] A remote data control method for a laminating machine based on PLC control, the method comprising the following steps:

[0069] Step S100. Obtain historical sensor data of the laminator operation collected by the PLC from the database, and correspond the historical sensor data with the fault event timestamp according to the fault event timestamp recorded in the fault log of the laminator in the database; based on the correspondence result, for each fault event, obtain the historical sensor data of the time period where the fault event is located, and mark it as a historical fault event data segment;

[0070] Step S200. Acquire simulation data of the PLC-controlled laminating machine, divide the simulation data into a plurality of data segments according to the timestamp of the fault event, and correspond the data segments of the simulation data with the corresponding data segments of the historical fault event; perform trend analysis on the data segments of the simulation data and the data segments of the historical fault event respectively, so as to obtain trend characteristics of the simulation data and trend characteristics of the historical fault event;

[0071] Step S300. Based on the correspondence between the historical fault event data segments and the simulation data data segments, the historical fault event trend characteristics are analyzed with the corresponding simulation data trend characteristics to obtain the fault event trend abnormal points;

[0072] Step S400. Collect real-time sensor data of the laminator through PLC, perform trend analysis on the real-time sensor data every selected time period, and obtain corresponding real-time event trend characteristics; analyze the real-time event trend characteristics, and perform corresponding processing according to the analysis results.

[0073] Step S100 includes:

[0074] S101. Obtain historical sensor data and laminator fault log from the database, and mark the fault event timestamp on the timeline corresponding to the historical sensor data according to the fault event timestamp recorded in the laminator fault log, so as to correspond the historical sensor data to the fault event timestamp;

[0075] S102. Based on the fault event timestamp, the time period of a time length t before and after the fault event timestamp is taken as the time period of the corresponding fault event; for each fault event, the historical sensor data of the time period of the fault event is obtained and marked as a historical fault event data segment; the historical fault event data segments of all fault events are aggregated to form a fault event data segment set G, and G = {g1, g2, ..., gn}, where g1 represents the historical fault event data segment corresponding to the first fault event, g2 represents the historical fault event data segment corresponding to the second fault event, and so on, gn represents the historical fault event data segment corresponding to the nth fault event; and n represents the number of fault events, which is a positive integer.

[0076] In this embodiment, taking fault event 1 as an example, assuming that the fault event timestamp corresponding to fault event 1 is t0, then the time period of a time length t before and after the fault event timestamp is taken as the corresponding time period of the fault event, that is, the time interval corresponding to the time period of the fault event is: [t0-t, t0+t].

[0077] Step S200 includes:

[0078] S201. According to the task requirements of the laminator, a simulation model is constructed using simulation software, a PLC control program is written to control the operation of the laminator in the simulation model, and simulation data of the PLC-controlled laminator is obtained, and the timeline of the simulation data completely overlaps with the timeline of the historical sensor data; according to the time period of the fault event, the simulation data is divided into several data segments, and the data segments of the simulation data corresponding to the same time period are matched with the data segments of the historical fault event;

[0079] S202. Perform trend analysis on the data segment of the simulation data and the data segment of the historical fault event respectively, and the trend analysis process of the data segment of the simulation data and the data segment of the historical fault event is completely consistent. The specific trend analysis process is as follows:

[0080] The data segment x(t) is decomposed by wavelet to obtain the wavelet coefficients W(j,k) of different scales j and positions k, and the calculation formula corresponding to the wavelet coefficients W(j,k) is:

[0081] W(j,k)= <x(t),ψ j,k (t)>=∫x(t)·ψ * j,k (t)dt;

[0082] Among them, ψ * j,k (t) represents the wavelet basis function, j represents the scale parameter, and k represents the location parameter;

[0083] Perform spectral analysis on the wavelet coefficients W(j,k) of each scale j and calculate the corresponding energy spectrum E j (ω), the specific calculation formula is:

[0084] E j (ω)=(1 / N)Σ k |W j,k (ω)|^2;

[0085] Among them, W j,k (ω) represents the amplitude of the wavelet coefficient at frequency ω, and N represents the number of wavelet coefficients;

[0086] According to the energy spectrum E j (ω), extract the characteristics of the data segment at different frequency scales and summarize them, and the data segment refers to the data segment of the simulation data or the data segment of the historical fault event, so as to obtain the trend characteristics of the simulation data and the trend characteristics of the historical fault event.

[0087] The present invention adopts a multidimensional data analysis method, including wavelet analysis, spectrum analysis, etc., to compare and analyze the historical fault event data segments with the simulation data, and comprehensively considers the characteristics at different scales to make the fault detection process more comprehensive and accurate.

[0088] Step S300 includes:

[0089] S301. According to the correspondence between the historical fault event data segments and the simulation data data segments, the historical fault event data segments and the simulation data data segments corresponding to the same time period are extracted, so as to obtain the corresponding historical fault event trend characteristics and simulation data trend characteristics, and the historical fault event trend characteristics and the simulation data trend characteristics are format-converted respectively, and converted into data representations of the same format, so as to form the historical fault event trend feature vector L and the simulation data trend feature vector M;

[0090] S302. The historical fault event trend feature vector L and the simulation data trend feature vector M are both represented on the same radar chart; the radar chart is composed of a starting point and a plurality of directed line segments, and the number of directed line segments is equal to the dimension of the historical fault event trend feature vector L and the simulation data trend feature vector M; the eigenvalues ​​corresponding to the historical fault event trend feature vector L and the simulation data trend feature vector M are marked on the radar chart in sequence, and the points on the two adjacent directed line segments are connected in sequence according to the order of the vector eigenvalues, so as to obtain the closed polygon A corresponding to the historical fault event trend feature vector L and the closed polygon B corresponding to the simulation data trend feature vector M;

[0091] S303. For the closed polygons A and B corresponding to the historical fault event trend feature vector L and the simulation data trend feature vector M, the vertex deviation values ​​R on all directed line segments are counted in turn, and the vertex deviation value R is equal to the vertex difference on the same directed line segment of the closed polygon A and the closed polygon B; for the vertex deviation values ​​R on all directed line segments, the corresponding average value μ and standard deviation σ are calculated, the vertex deviation value R is compared with the corresponding deviation value interval Q, and the vertex of the closed polygon A whose first vertex deviation value R does not belong to the deviation value interval Q is taken as the target point, and the interval range of the deviation value interval Q is [μ-eσ,μ+eσ], where e represents a coefficient, which is used to define the range of the deviation value; according to the target point, the characteristic value in the historical fault event trend feature vector L is obtained, and the frequency scale of the corresponding historical fault event data segment is obtained based on the characteristic value, so as to find the abnormal point of the fault event trend in the historical fault event data segment;

[0092] Obtain the fault event trend abnormal points in all historical fault event data segments, and associate the fault codes corresponding to the historical fault events with the fault event trend abnormal points, so as to obtain the association relationship between the historical fault events and the fault event trend abnormal points; obtain the historical fault event data of a time period of a time length t before and after the fault event trend abnormal points, so as to obtain the abnormal change trend curve Y of the historical fault event data, so as to obtain the association relationship between the fault code, the fault event trend abnormal points and the abnormal change trend curve Y.

[0093] The introduction of radar chart and polygon comparison methods can intuitively compare the trend characteristics of historical fault events with the trend characteristics of real-time events, further improving the accuracy and efficiency of fault diagnosis.

[0094] Step S400 includes:

[0095] S401. Use PLC to collect real-time sensor data of the laminator, obtain the real-time sensor data segment of the real-time sensor data in the selected time period every selected time period, and obtain the corresponding data segment of the simulation data according to the selected time period; perform trend analysis on the real-time sensor data segment and the simulation data segment respectively, so as to obtain the corresponding real-time event trend characteristics and simulation data trend characteristics, perform format conversion on the real-time event trend characteristics and the simulation data trend characteristics respectively, and obtain the real-time event trend characteristic vector S and the simulation data trend characteristic vector M1 respectively; wherein the simulation data trend characteristic vector M and the simulation data trend characteristic vector M1 have differences in the corresponding time periods;

[0096] S402. For the real-time event trend feature vector S and the simulation data trend feature vector M1, refer to the analysis process of the historical fault event trend feature vector L and the simulation data trend feature vector M in S302, so as to obtain the closed polygon A1 corresponding to the real-time event trend feature vector S, and the closed polygon B1 corresponding to the simulation data trend feature vector M1; refer to the analysis process of S303, if there is no real-time event trend abnormal point in the real-time sensor data segment, no notification is made, and the analysis of the real-time sensor data of the next selected time period is performed;

[0097] If there is a real-time event trend abnormal point in the real-time sensor data segment, the time period of a time length t before and after the real-time event trend abnormal point is taken as the corresponding time period of the real-time event, and the corresponding real-time sensor data is obtained according to the time period of the real-time event, so as to obtain the abnormal change trend curve Y1 of the real-time sensor data, and the real-time event trend abnormal point is associated with the abnormal change trend curve Y1;

[0098] S403. Perform similarity analysis on the abnormal change trend curve Y1 and the abnormal change trend curve Y corresponding to the historical fault event. If there is an abnormal change trend curve Y with a similarity greater than the similarity threshold, the abnormal change trend curve Y with the greatest similarity is selected as the matching result, and the fault code corresponding to the abnormal change trend curve Y is output; if there is an abnormal change trend curve Y with a similarity less than or equal to the similarity threshold, the current abnormal change trend curve Y1 is output, and the event code is marked for the current real-time event, and output to relevant personnel for analysis. If the current real-time event is a new fault, the current event code, the real-time event trend abnormal point and the abnormal change trend curve Y1 are associated, and the above-mentioned association relationship is added to the database.

[0099] According to the processing results of real-time events, relevant information is added to the database, and the fault diagnosis model is continuously optimized to adapt to new fault conditions that may arise in the future, thereby improving the intelligence and adaptability of the overall system.

[0100] In this embodiment, it is assumed that there is a laminating machine, which uses a PLC (programmable logic controller) to collect real-time sensor data, including pressure, temperature, speed and other information, and performs data collection and analysis every 10 minutes; trend analysis is performed on the real-time sensor data segment and the simulation data segment, for example, the pressure change trend over time, the temperature change trend, etc. are analyzed, and it is assumed that a real-time event trend feature vector S and a simulation data trend feature vector M1 are obtained; according to the real-time event trend feature vector S and the simulation data trend feature vector M1, corresponding closed polygons A1 and B1 are generated. If a real-time event trend abnormal point is found in the real-time sensor data segment, the time period of a time length t before and after the real-time event trend abnormal point is defined as the time period of the real-time event, and the real-time event trend abnormal point is obtained according to this time period. Take detailed real-time sensor data to obtain the abnormal change trend curve Y1; perform similarity analysis on the abnormal change trend curve Y1 and the abnormal change trend curve Y corresponding to the historical fault event. Assuming that the similarity threshold is set to 0.7, if an abnormal change trend curve Y with a similarity greater than 0.7 is found, the fault event corresponding to the abnormal change trend curve Y with the largest similarity is selected as the real-time fault event, and its corresponding fault code is output; if the similarity is less than or equal to 0.7, the current abnormal change trend curve Y1 is output and the corresponding event code is marked. After analysis by relevant personnel, if it is determined that the current event is a new fault event, the event code, the real-time event trend abnormal point and the associated information of the abnormal change trend curve Y1 are recorded in the database, and the current real-time event is marked as a new fault.

[0101] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0102] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A remote data control method for a laminating machine based on PLC control, characterized in that: The method comprises the following steps: Step S100. Obtain historical sensor data of the laminator operation collected by the PLC from the database, and correspond the historical sensor data with the fault event timestamp according to the fault event timestamp recorded in the fault log of the laminator in the database; based on the correspondence result, for each fault event, obtain the historical sensor data of the time period where the fault event is located, and mark it as a historical fault event data segment; Step S200. Acquire simulation data of the PLC-controlled laminating machine, divide the simulation data into a plurality of data segments according to the timestamp of the fault event, and correspond the data segments of the simulation data with the corresponding data segments of the historical fault event; perform trend analysis on the data segments of the simulation data and the data segments of the historical fault event respectively, so as to obtain trend characteristics of the simulation data and trend characteristics of the historical fault event; Step S300. Based on the correspondence between the historical fault event data segments and the simulation data data segments, the historical fault event trend characteristics are analyzed with the corresponding simulation data trend characteristics to obtain the fault event trend abnormal points; Step S400. Collect real-time sensor data of the laminator through PLC, perform trend analysis on the real-time sensor data every selected time period, and obtain corresponding real-time event trend characteristics; analyze the real-time event trend characteristics, and perform corresponding processing according to the analysis results.

2. According to claim 1, a laminating machine remote data control method based on PLC control is characterized in that: The step S100 includes: S101. Obtain historical sensor data and laminator fault log from the database, and mark the fault event timestamp on the timeline corresponding to the historical sensor data according to the fault event timestamp recorded in the laminator fault log, so as to correspond the historical sensor data to the fault event timestamp; S102. Based on the fault event timestamp, the time period of a time length t before and after the fault event timestamp is taken as the time period of the corresponding fault event; for each fault event, the historical sensor data of the time period of the fault event is obtained and marked as a historical fault event data segment; the historical fault event data segments of all fault events are aggregated to form a fault event data segment set G, and G = {g1, g2, ..., gn}, where g1 represents the historical fault event data segment corresponding to the first fault event, g2 represents the historical fault event data segment corresponding to the second fault event, and so on, gn represents the historical fault event data segment corresponding to the nth fault event; and n represents the number of fault events, which is a positive integer.

3. A laminating machine remote data control method based on PLC control according to claim 2, characterized in that: The step S200 includes: S201. According to the task requirements of the laminator, a simulation model is constructed using simulation software, a PLC control program is written to control the operation of the laminator in the simulation model, and simulation data of the PLC-controlled laminator is obtained, and the timeline of the simulation data completely overlaps with the timeline of the historical sensor data; according to the time period of the fault event, the simulation data is divided into several data segments, and the data segments of the simulation data corresponding to the same time period are matched with the data segments of the historical fault event; S202. Perform trend analysis on the data segment of the simulation data and the data segment of the historical fault event respectively, and the trend analysis process of the data segment of the simulation data and the data segment of the historical fault event is completely consistent. The specific trend analysis process is as follows: The data segment x(t) is decomposed by wavelet to obtain the wavelet coefficients W(j,k) of different scales j and positions k, and the calculation formula corresponding to the wavelet coefficients W(j,k) is: W(j,k)=<x(t),ψ j,k (t)>=∫x(t)·ψ * j,k (t)dt; Among them, ψ * j,k (t) represents the wavelet basis function, j represents the scale parameter, and k represents the location parameter; Perform spectral analysis on the wavelet coefficients W(j,k) of each scale j and calculate the corresponding energy spectrum E j (ω), the specific calculation formula is: E j (ω)=(1 / N)Σ k |W j,k (ω)|^2; Among them, W j,k (ω) represents the amplitude of the wavelet coefficient at frequency ω, and N represents the number of wavelet coefficients; According to the energy spectrum E j (ω), extract the characteristics of the data segment at different frequency scales and summarize them, and the data segment refers to the data segment of the simulation data or the data segment of the historical fault event, so as to obtain the trend characteristics of the simulation data and the trend characteristics of the historical fault event.

4. A PLC-based remote data control method for a laminating machine according to claim 3, characterized in that: The step S300 includes: S301. According to the correspondence between the historical fault event data segments and the simulation data data segments, the historical fault event data segments and the simulation data data segments corresponding to the same time period are extracted, so as to obtain the corresponding historical fault event trend characteristics and simulation data trend characteristics, and the historical fault event trend characteristics and the simulation data trend characteristics are format-converted respectively, and converted into data representations of the same format, so as to form the historical fault event trend feature vector L and the simulation data trend feature vector M; S302. The historical fault event trend feature vector L and the simulation data trend feature vector M are both represented on the same radar chart; the radar chart is composed of a starting point and a plurality of directed line segments, and the number of directed line segments is equal to the dimension of the historical fault event trend feature vector L and the simulation data trend feature vector M; the eigenvalues ​​corresponding to the historical fault event trend feature vector L and the simulation data trend feature vector M are marked on the radar chart in sequence, and the points on the two adjacent directed line segments are connected in sequence according to the order of the vector eigenvalues, so as to obtain the closed polygon A corresponding to the historical fault event trend feature vector L and the closed polygon B corresponding to the simulation data trend feature vector M; S303. For the closed polygons A and B corresponding to the historical fault event trend feature vector L and the simulation data trend feature vector M, the vertex deviation values ​​R on all directed line segments are counted in turn, and the vertex deviation value R is equal to the vertex difference on the same directed line segment of the closed polygon A and the closed polygon B; for the vertex deviation values ​​R on all directed line segments, the corresponding average value μ and standard deviation σ are calculated, the vertex deviation value R is compared with the corresponding deviation value interval Q, and the vertex of the closed polygon A whose first vertex deviation value R does not belong to the deviation value interval Q is taken as the target point, and the interval range of the deviation value interval Q is [μ-eσ,μ+eσ], where e represents a coefficient, which is used to define the range of the deviation value; according to the target point, the characteristic value in the historical fault event trend feature vector L is obtained, and the frequency scale of the corresponding historical fault event data segment is obtained based on the characteristic value, so as to find the abnormal point of the fault event trend in the historical fault event data segment; Obtain the fault event trend abnormal points in all historical fault event data segments, and associate the fault codes corresponding to the historical fault events with the fault event trend abnormal points, so as to obtain the association relationship between the historical fault events and the fault event trend abnormal points; obtain the historical fault event data of a time period of a time length t before and after the fault event trend abnormal points, so as to obtain the abnormal change trend curve Y of the historical fault event data, so as to obtain the association relationship between the fault code, the fault event trend abnormal points and the abnormal change trend curve Y.

5. A PLC-based remote data control method for a laminating machine according to claim 4, characterized in that: The step S400 includes: S401. Use PLC to collect real-time sensor data of the laminator, obtain the real-time sensor data segment of the real-time sensor data in the selected time period every selected time period, and obtain the corresponding data segment of the simulation data according to the selected time period; perform trend analysis on the real-time sensor data segment and the simulation data segment respectively, so as to obtain the corresponding real-time event trend characteristics and simulation data trend characteristics, perform format conversion on the real-time event trend characteristics and the simulation data trend characteristics respectively, and obtain the real-time event trend characteristic vector S and the simulation data trend characteristic vector M1 respectively; S402. For the real-time event trend feature vector S and the simulation data trend feature vector M1, refer to the analysis process of the historical fault event trend feature vector L and the simulation data trend feature vector M in S302, so as to obtain the closed polygon A1 corresponding to the real-time event trend feature vector S, and the closed polygon B1 corresponding to the simulation data trend feature vector M1; refer to the analysis process of S303, if there is no real-time event trend abnormal point in the real-time sensor data segment, no notification is made, and the analysis of the real-time sensor data of the next selected time period is performed; If there is a real-time event trend abnormal point in the real-time sensor data segment, the time period of a time length t before and after the real-time event trend abnormal point is taken as the corresponding time period of the real-time event, and the corresponding real-time sensor data is obtained according to the time period of the real-time event, so as to obtain the abnormal change trend curve Y1 of the real-time sensor data, and the real-time event trend abnormal point is associated with the abnormal change trend curve Y1; S403. Perform similarity analysis on the abnormal change trend curve Y1 and the abnormal change trend curve Y corresponding to the historical fault event. If there is an abnormal change trend curve Y with a similarity greater than the similarity threshold, the abnormal change trend curve Y with the greatest similarity is selected as the matching result, and the fault code corresponding to the abnormal change trend curve Y is output; if there is an abnormal change trend curve Y with a similarity less than or equal to the similarity threshold, the current abnormal change trend curve Y1 is output, and the event code is marked for the current real-time event, and output to relevant personnel for analysis. If the current real-time event is a new fault, the current event code, the real-time event trend abnormal point and the abnormal change trend curve Y1 are associated, and the above-mentioned association relationship is added to the database.

6. A laminating machine remote data control system based on PLC control, applied to a laminating machine remote data control method based on PLC control as claimed in any one of claims 1 to 5, characterized in that: The system includes: a data acquisition and preprocessing module, a simulation model construction and data analysis module, a fault diagnosis and anomaly detection module, and an abnormal change trend analysis and feedback module; The data acquisition and preprocessing module is responsible for obtaining historical sensor data and laminator fault logs from the database, matching the historical sensor data with the fault event timestamp, and marking the historical fault event data segments; obtaining the historical fault event data segments, and aggregating them to form a set of fault event data segments; The simulation model building and data analysis module uses simulation software to build a simulation model of the laminating machine, writes a PLC control program according to task requirements, generates simulation data and corresponds it with historical fault event data segments, and performs trend analysis to obtain simulation data trend characteristics and historical fault event trend characteristics; The fault diagnosis and anomaly detection module analyzes the correspondence between the historical fault event data segments and the simulation data, extracts the corresponding simulation data trend feature vectors and the historical fault event trend feature vectors, and performs radar chart representation, calculates the vertex deviation value, and thus obtains the fault event trend anomaly point; The abnormal change trend analysis and feedback module performs trend analysis based on real-time sensor data, compares the real-time sensor data corresponding to the abnormal point of the real-time event trend with the data segment corresponding to the historical fault event for the similarity of the corresponding abnormal change trend curve, outputs the matching result or marks the new fault event, and provides it to relevant personnel for further analysis and processing.

7. A laminating machine remote data control system based on PLC control according to claim 6, characterized in that: The data acquisition and preprocessing module includes a database data acquisition unit, a fault event timestamp marking unit, and a historical fault event data segment extraction unit; The database data acquisition unit acquires the historical sensor data collected by the PLC and the laminator fault log from the database; the fault event timestamp marking unit matches the fault event timestamp with the historical sensor data timeline, and marks the fault event timestamp on the historical sensor data timeline; the historical fault event data segment extraction unit extracts the historical sensor data of the time period where the fault event is located according to the fault event timestamp, to form a historical fault event data segment set.

8. A laminating machine remote data control system based on PLC control according to claim 6, characterized in that: The simulation model construction and data analysis module includes a simulation model establishment unit, a simulation data generation unit, and a data segment division and correspondence unit; The simulation model building unit uses simulation software to build a simulation model of the laminating machine and writes a PLC control program to simulate the operation of the laminating machine; the simulation data generating unit obtains simulation data of the simulation model so that the time line of the simulation data completely corresponds to the time line of the historical sensor data; The data segment division and correspondence unit divides the simulation data into a plurality of data segments according to the fault event timestamp, and corresponds the data segments with the historical fault event data segments.

9. A laminating machine remote data control system based on PLC control according to claim 6, characterized in that: The fault diagnosis and anomaly detection module includes a trend analysis unit and a fault event trend anomaly point detection unit; The trend analysis unit performs wavelet decomposition and spectrum analysis on the simulation data segment and the historical fault event data segment to obtain corresponding simulation data trend characteristics and historical fault event trend characteristics.

10. A laminating machine remote data control system based on PLC control according to claim 6, characterized in that: The abnormal change trend analysis and feedback module includes a real-time sensor data acquisition unit, a real-time event trend feature extraction unit, an abnormal change trend comparison unit and an abnormal change trend feedback unit; The real-time sensor data acquisition unit periodically acquires real-time sensor data of the laminator; The real-time event trend feature extraction unit performs trend analysis on the real-time sensor data to obtain trend features of the real-time event; The abnormal change trend comparison unit compares the trend characteristics of the real-time event with the trend characteristics of the simulation data to detect whether there are abnormal points in the trend of the real-time event; The abnormal change trend feedback unit does not process the real-time sensor data without abnormal points of real-time event trend, and only records the normal state; For real-time sensor data with abnormal points in real-time event trends, abnormal change trend curves are analyzed, and the corresponding trend curves of real-time sensor data and historical fault event data are analyzed. The trend curves of the two are matched and analyzed, and the corresponding notification information is output.

Images