Industrial key equipment mutation detection method and system

Abstract

The invention discloses an industrial key equipment abrupt change detection method and system, which are used for abrupt change detection of key equipment on a production line, one or more key indexes of monitored industrial key equipment are selected, and a data stream of the key indexes in a complete production line rhythm under the condition that the equipment normally operates is acquired as a reference data stream; real-time data streams of the key indexes are collected in real time, feature value sequences representing time domain features and frequency domain features of the key indexes in each complete production line beat are also extracted, and the feature value sequences of the data streams to be detected and the feature reference value sequences are compared and analyzed; and according to a comparison result, judging whether the key equipment is suddenly changed or not. According to the method and the device, multi-dimensional mutation detection is realized by using relatively low resource cost, the method and the device are very suitable for detecting the operation state of the industrial key equipment in an industrial scene, and the defect that the mutation of the industrial key equipment cannot be economically and effectively detected in time in the prior art is overcome.

Description

Technical Field

[0001] The present application belongs to the field of industrial control digital information, and in particular relates to a method and system for detecting sudden changes in key industrial equipment. Background Art

[0002] Industrial key equipment is an important facility in the process of industrial production activities. Whether industrial key equipment can operate normally and stably will directly affect whether industrial activities can proceed normally. The current traditional monitoring of the health status of industrial key equipment mainly monitors the data generated by the key indicators of the equipment during operation. If the data is observed to exceed the set safety threshold, it is highly likely that the equipment has an abnormality or even a failure.

[0003] However, if the equipment safety threshold is used for detection, when abnormal jitter or mutation occurs and the data values ​​are far less than the upper limit of the equipment safety threshold, it cannot be discovered in time, thus delaying the best fault diagnosis time, which may bury huge safety hazards for future production. The commonly used multi-dimensional detection methods rely on a large amount of historical data due to the lack of in-depth understanding of the specific characteristics of industrial production. They require complicated calculation and deduction processes and data processing, making the system less practical and inefficient. And because it only focuses on a single production status data, its data source and analysis results are not comprehensive and accurate, and problems cannot be discovered in time.

[0004] Based on the above situation, the traditional method of detecting anomalies of key industrial production equipment does not fit the actual situation of industrial scenarios. There is an urgent need for better and more targeted methods to help key industrial equipment detect anomalies and failures in a timely manner. Summary of the invention

[0005] The embodiment of the present application provides a method and system for detecting mutations in key industrial equipment. Due to the in-depth analysis and research of the industrial production process, the application uses the characteristics of task clarity, stability and periodicity of key industrial equipment to complete work, and summarizes the law of mutations in key indicator data streams when potential problems occur in key industrial equipment in industrial production, especially in assembly line production scenarios. This solution uses a relatively small resource cost to achieve multi-dimensional mutation detection, which is very suitable for detecting the operating status of key industrial equipment in industrial scenarios, and overcomes the defect that the existing technology lacks economical and effective timely detection of mutations in key industrial equipment.

[0006] In a first aspect, an embodiment of the present application provides a method for detecting sudden changes in industrial key equipment, which is used for detecting sudden changes in key equipment on a production line. The method comprises: Select one or more key indicators of the monitored industrial key equipment, wherein the key indicators can reflect the operating status and performance of the equipment; Acquire a data stream within a production line beat when the key indicator is operating normally as a reference data stream, extract a feature value sequence representing the time domain features and frequency domain features of the reference data stream as a feature reference, and form a feature reference value sequence; Collect the real-time data stream of the key indicators in real time, and also extract the characteristic value sequence representing the time domain characteristics and frequency domain characteristics within each complete production line beat to form the characteristic value sequence of the data stream to be detected; Compare and analyze the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence; Based on the comparison results, determine whether the key equipment has mutated.

[0007] Furthermore, the time domain features in the feature value sequence are manifested as the arithmetic mean and / or standard deviation of interval data of the data stream in the time domain space.

[0008] Furthermore, the data stream is a time series data stream, and the data of the frequency domain characteristics in the eigenvalue sequence are expressed as n discrete eigenvalues ​​in the frequency domain space of the time series data stream within a production line cycle, where n is the number of samples and is greater than 1.

[0009] Furthermore, the discrete eigenvalues ​​are obtained by performing continuous wavelet transform on the time series data stream, and the wavelet basis function used in the wavelet transform is Complex Morlet wavelet.

[0010] Furthermore, the comparison analysis uses a Euclidean distance algorithm to compare and analyze the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence to evaluate the similarity between the two. If the Euclidean distance between the characteristic value sequence of the data stream to be detected and the characteristic reference value sequence is less than a preset threshold, it is determined that the key indicator is normal and the key device is normal; if the Euclidean distance is greater than or equal to the preset threshold, it is determined that the key indicator has mutated and the key device has mutated.

[0011] Furthermore, the method also includes dynamically updating the characteristic reference value sequence, that is, adjusting and optimizing the characteristic reference value sequence regularly or according to preset conditions based on the historical data of the long-term operation of the key equipment.

[0012] Furthermore, the production line rhythm includes the complete production line rhythm including preparation time, production time and waiting time; the key indicators include: power, and / or temperature, and / or voltage, and / or current, and / or pressure, and / or impedance, and / or robot arm load rate, and / or energy consumption rate, and / or coordinate positioning deviation.

[0013] In a second aspect, an embodiment of the present application provides a system for detecting sudden changes in key industrial equipment, the system comprising: Selection module: selects one or more key indicators of the monitored industrial key equipment, wherein the key indicators can reflect the operating status and performance of the equipment; Feature benchmark module: obtains the data stream of a production line beat under the condition of normal operation of the key indicator as the benchmark data stream, extracts the feature value sequence representing the time domain characteristics and frequency domain characteristics of the benchmark data stream as the feature benchmark, and forms a feature benchmark value sequence; Data acquisition module: collects the real-time data stream of the key indicators in real time, and also extracts the characteristic value sequence representing the time domain characteristics and frequency domain characteristics within each complete production line beat to form the characteristic value sequence of the data stream to be detected; Data comparison module: compares and analyzes the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence; Abnormal judgment module: judge whether the key indicator has a sudden change according to the comparison result of the data comparison module.

[0014] In a third aspect, an embodiment of the present application provides a target device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for detecting mutations in critical industrial equipment described in any one of the first aspects when executing the computer program.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for detecting mutations in critical industrial equipment as described in any one of the first aspects above is implemented.

[0016] It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0017] Compared with the prior art, the beneficial effects of the embodiments of the present application are as follows: the present application aims at the characteristics that industrial production equipment is in steady-state operation most of the time, and the fault signs usually occur under steady-state conditions. By extracting the time domain and frequency domain characteristics of the data stream within a production line beat under normal operation of the key indicators of the monitored equipment as the feature benchmark, it is possible to meet the requirements of mutation detection, without the need to establish a complex benchmark model or long-term historical data accumulation, and to achieve rapid deployment and implementation of the system. At the same time, the present application independently extracts the time domain and frequency domain characteristics of the data stream, and comprehensively considers the deviation of the data stream from the feature benchmark in these two different dimensions, which can solve the problem that the previous detection means cannot accurately identify the mutation of the frequency of data changes under the normal threshold, making the judgment result more accurate and comprehensive. In addition, in terms of the evaluation method, the Euclidean distance algorithm, which is mature and consumes less system computing resources, is adopted, which can meet the requirements of rapid and smooth deployment of the system under the current industrial production status with old computing equipment and poor computing performance. Finally, this application integrates the equipment with the specific business it performs. The extraction of data streams no longer focuses only on single production status data, but on the entire process data of unit products including preparation time, production time and waiting time, that is, the data within the rhythm of a complete production line. It is closely integrated with the actual scenario of industrial production, and the data sampling range and judgment criteria are more comprehensive. It can not only timely discover whether the equipment itself is in a normal state, but also discover whether the equipment is in a normal operating state when processing specific business. It can timely discover and solve problems, help managers to issue early warnings and prevent risks, and ensure the safe and efficient operation of industrial production. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a schematic diagram of a process flow of a method for detecting mutations in key industrial equipment provided by an embodiment of the present application; Figure 2 It is a key device-key indicator-data flow-feature set relationship diagram provided in an embodiment of the present application; Figure 3 This is a schematic diagram of sudden change detection of key industrial equipment provided by an embodiment of the present application; Figure 4 This is a schematic diagram of the periodic regularity of a power index data stream of a certain power distribution equipment provided by an embodiment of the present application; Figure 5This is a schematic diagram of converting a time domain signal into a frequency signal provided by an embodiment of the present application; Figure 6 This is a schematic diagram of a mutation detection system for key industrial equipment provided by an embodiment of the present application; Figure 7 It is a schematic diagram of an industrial control host using a method for detecting sudden changes in key industrial equipment provided by an embodiment of the present application. DETAILED DESCRIPTION

[0020] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0021] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0022] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0023] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0024] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0025] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0026] After analyzing a large amount of data on key industrial equipment at industrial sites, we found that the work performed by most key industrial equipment is clear and stable. Especially on the production line, the corresponding equipment repeats its own business and has a periodicity. Figure 2 As shown, if the equipment is in good operating condition, the data stream generated by the key indicators of the equipment should also be in a stable and regular state. Once the equipment is abnormal or fails, the equipment operating state will inevitably change suddenly, and the data stream generated by it will also change suddenly. These data stream changes not only appear in the production state, but also in the production preparation and production waiting states in many cases. Therefore, the embodiment of the present application provides a method and system for detecting sudden changes in industrial key equipment, which uses the sudden changes in the data stream in the rhythm of the entire production line as a basis for judging the equipment operating status detection, and can more effectively monitor whether the industrial key equipment is in a normal working state. The embodiment of the present application provides a method and system for detecting mutations in industrial key equipment. In this example, a feature benchmark is formed by extracting the time domain and frequency domain characteristics of the data stream of the key indicators of the monitored equipment within a complete production line beat under normal operation. The benchmark is used to compare with the subsequent periodic real-time data stream characteristics to be detected. Based on the comparison results, it is determined whether the key indicator data stream has mutated, and then it is inferred whether the key equipment has a sudden change in its operating status, and whether an abnormality or failure has occurred in the equipment operation. This allows timely alarms to remind relevant personnel to detect, check and repair as soon as possible, so as to discover and eliminate abnormalities or failures as soon as possible and ensure stable and continuous production.

[0027] In one example, the industrial key equipment mutation detection method of the present application is used for mutation detection of key equipment on a production line, and includes the following main steps: S101, selecting one or more key indicators of the monitored industrial key equipment, wherein the key indicators can reflect the operating status and performance of the equipment; S102, obtaining a data stream within a production line beat when the key indicator is operating normally as a reference data stream, extracting a feature value sequence representing time domain features and frequency domain features of the reference data stream as a feature reference, and forming a feature reference value sequence; S103, collecting the real-time data stream of the key indicators in real time, and similarly extracting the feature value sequence representing the time domain characteristics and frequency domain characteristics of each production line beat, to form a feature value sequence of the data stream to be detected; S104, comparing and analyzing the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence; S105. Determine whether a mutation occurs to the key device based on the comparison result.

[0028] At present, the production equipment detection method is usually based on the construction of production models, which requires the use of a large amount of historical data to train the reference model. The benchmark extraction process is complicated, and in the detection process, the essence of the actual industrial production data flow changes is not grasped. The detection results are often not satisfactory. The main steps of the industrial key equipment mutation detection method of this application are explained in combination with the embodiments to illustrate the detection method of the present invention.

[0029] S101. Select one or more key indicators of the monitored industrial key equipment, where the key indicators can reflect the operating status and performance of the equipment: In production practice, if the key industrial equipment on the industrial site is in good operating condition, the data stream generated by the key indicators of the equipment should also be in a stable and regular state. Once the equipment is abnormal or fails, the equipment operating state will inevitably change suddenly, and the data stream it generates will also change suddenly. If there is a sudden change in the data stream, it is considered that there is a sudden change in the equipment state. Since key indicators can reflect the operating status and performance of the equipment, it is necessary to purposefully select one or more key indicators of the monitored industrial key equipment as the detection object, and the standard for these indicators to be used as key indicators of the detection object also includes that they must be periodic.

[0030] In one embodiment, taking a production line business scenario of a factory as an example, the production line requires power distribution equipment to provide a power source. Since the production line business is clear and stable, when the power distribution equipment is operating normally, Figure 4 As shown, its power data has a clear periodic pattern. This key indicator that can reflect the operating status and performance of equipment in industrial production and has a periodic pattern is the analysis object of the sudden change detection method for industrial key equipment in this embodiment.

[0031] In other embodiments, the key indicators include: power, and / or temperature, and / or voltage, and / or current, and / or pressure, and / or impedance. In the field of robotics, the key indicators may also include: robot arm load rate, and / or energy consumption rate, and / or coordinate positioning deviation, etc.

[0032] S102, obtaining a data stream within a production line beat when the key indicator is operating normally as a reference data stream, extracting a feature value sequence representing the time domain characteristics and frequency domain characteristics of the reference data stream as a feature reference, and forming a feature reference value sequence: After a large amount of data analysis of key industrial equipment on industrial sites, we found that the work performed by most of the key industrial equipment is clear and stable, especially on the production line, the corresponding equipment repeats the business it shares, with strict periodicity. If the equipment is in good operating condition, the data stream generated by the key indicators of the equipment should also be in a stable and regular state. This state characteristic is not only reflected when the equipment is in production state, but also when the equipment is in preparation state and waiting state. Therefore, in one embodiment of the present application, only the data stream within a complete production line beat when the key indicators are operating normally is obtained as the reference data stream, which avoids the need for a large amount of historical collection work and greatly improves the extraction efficiency of the reference data.

[0033] Further, in one embodiment, first Figure 1 The offline processing process shown obtains a complete cycle data stream of the power index of the distribution equipment under normal operation (in this embodiment, it is a complete production line rhythm of the production line, including preparation time, production time and waiting time) as a benchmark data stream, and then extracts the characteristic value sequence characterizing the time domain characteristics and frequency domain characteristics of the benchmark data stream as a characteristic benchmark to form a characteristic benchmark value sequence (i.e., the characteristic value sequence of the benchmark feature in the figure).

[0034] In other words, this embodiment obtains the time domain characteristic value sequence and frequency domain characteristic value sequence of the periodic data stream by extracting the characteristics of the complete periodic data stream of the power indicator of the distribution equipment under normal operation, thereby forming a characteristic reference value sequence of the power indicator data stream under normal operation of the distribution equipment as a characteristic reference of the indicator.

[0035] In one embodiment, the time-domain characteristic value sequence of the periodic data stream is the arithmetic mean and / or standard deviation, which are the two most important measurement values ​​for describing the central tendency and dispersion degree of data: The arithmetic mean is the sum of all the data in a set of data divided by the number of data. The sum of n numbers divided by n is called the arithmetic mean of these n numbers. It is an indicator of the trend of the data set. The formula for the arithmetic mean is:

[0036] Standard Deviation, a mathematical term, is the square root of the square of the deviation from the mean (i.e., variance), represented by σ. Standard deviation is also called standard deviation, or experimental standard deviation, and is most commonly used in probability statistics as a measure of the degree of statistical distribution. Standard deviation is the square root of variance. Standard deviation can reflect the degree of dispersion of a data set. Two sets of data with the same mean may not have the same standard deviation. The formula for the overall standard deviation is:

[0037] The arithmetic mean and standard deviation are statistical calculation methods commonly used by technical personnel in this field. When seeing the application of these two values ​​as the characteristic value sequence used by the benchmark data stream in the time domain disclosed in this embodiment, the technical personnel should be able to understand the implementation method of the characteristic value sequence, which will not be elaborated here.

[0038] In one embodiment, the frequency domain feature value sequence of the data stream is n discrete feature values ​​in the frequency domain space of the time series data stream within a complete production line beat, where n is the number of samples and is greater than 1. The discrete feature values ​​are extracted by performing a continuous wavelet transform (CWT) algorithm on the time series data stream.

[0039] Continuous Wavelet Transform (CWT) is an algorithm used to analyze the time-frequency characteristics of signals. By selecting appropriate wavelet basis functions, CWT can decompose complex signals into wavelet coefficients of different scales and positions, thereby converting time domain signals into a set of specific frequency domain coefficients, revealing the hidden features and patterns in the signal. The formula is as follows:

[0040] Among them, C(a,b) is the wavelet coefficient, f(t) is the original signal, and ψ(t) is the wavelet basis function (a is the scale parameter and b is the translation parameter).

[0041] Among them, the ψ(t) wavelet basis function includes many types. In this case, the Complex Morlet wavelet basis function is used. The formula is as follows (where B represents the bandwidth and C represents the center frequency. These two parameters determine the wavelet waveform):

[0042] The above describes the basic process of wavelet transform, that is, after sampling the power index time series data stream of the distribution equipment, continuous wavelet transform is performed based on Complex Morlet wavelet to extract frequency domain features.

[0043] After continuous wavelet transform, the time domain signal is transformed into a frequency domain signal, as shown in the schematic diagram. Figure 5 , and then sample and extract its frequency domain eigenvalues. These discrete frequency domain eigenvalues ​​constitute the frequency domain eigenvalue sequence of the data stream.

[0044] In one embodiment, by extracting the features of the data stream of a complete cycle of power indicators (including a complete production line beat including preparation time, production time and waiting time) of the power distribution equipment under normal working conditions, the time domain feature value (mean and standard deviation) and frequency domain feature value sequence of the data stream of this period are obtained, forming a feature set of the power indicator data stream of the power distribution equipment under normal working conditions, i.e., a feature benchmark, as a feature benchmark value sequence of the indicator. The feature benchmark value sequence is: <<mean, standard deviation>,<C1…Ci…Cn> >, where C represents the frequency domain eigenvalue sequence and n is the number of sampling times.

[0045] In another embodiment, the frequency domain eigenvalue sequence of the data stream that also uses continuous wavelet transform can also be n vectors of the time series data stream in the frequency domain space. Compared with the eigenvalue, the vector is not a simple numerical value, but also has a quantity with time or space scale. The specific scale depends on which data transformation is performed on the time domain data stream to obtain its frequency domain characteristics. In one of the embodiments, the time series data stream of the complete cycle of the power indicator (such as a production line beat) of the power distribution equipment under normal operation is subjected to continuous wavelet transform to obtain the corresponding n frequency domain vectors with time dimension, which can also be used as the characteristic reference value sequence of the indicator. The obtained characteristic reference value sequence is: <<mean, standard deviation>,<C1…Ci…Cn> >, where C represents the frequency domain feature vector sequence and n is the number of sampling times. The specific operation process will not be elaborated here.

[0046] In another embodiment, in order to ensure the accuracy of mutation detection results, taking into account equipment aging and other reasons, these feature baseline value sequences are also dynamically updated, that is, based on the historical data of the long-term operation of the key equipment, the feature baseline value sequences are adjusted and optimized regularly or according to preset conditions.

[0047] S103, collecting the real-time data stream of the key indicators in real time, and similarly extracting the feature value sequence representing the time domain characteristics and frequency domain characteristics of each complete production line beat, to form a feature value sequence of the data stream to be detected; After analyzing a large amount of data on key industrial equipment at industrial sites, we found that the work performed by most key industrial equipment is clear and stable. Especially on the production line, the corresponding equipment repeats the business it is responsible for and is cyclical.

[0048] This embodiment collects the real-time data stream of the key indicators in real time, and based on the same feature value sequence extraction method as S102, obtains the feature value sequence of the time domain characteristics and frequency domain characteristics of the data stream in each complete production line beat, and forms the feature value sequence of the data stream to be detected as: <<mean, standard deviation>,<C1…Ci…Cn> >, where C represents the frequency domain eigenvalue and / or vector sequence, and n is the number of sampling times. The specific operation process will not be elaborated here.

[0049] S104: Compare and analyze the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence: In one embodiment, the frequency domain feature data in the feature value sequence is represented by n discrete feature values ​​of the time series data stream in the frequency domain space, that is, the feature reference value sequence is: <<mean, standard deviation>,<C1…Ci…Cn> >, where C represents the frequency domain feature value sequence, and n is the number of sampling times. In this embodiment, the comparison analysis uses the Euclidean distance algorithm to compare and analyze the feature value sequence of the data stream to be detected with the feature reference value sequence to evaluate the similarity between the two.

[0050] Because the eigenvalue is actually a set of data containing mean, standard deviation, and frequency domain sampling values, comparing features is to compare two sets of sequence values. In n-dimensional space, let Xn and Yn be points in n-dimensional space, then their Euclidean distance is defined as:

[0051] Specifically here, Xn and Yn can represent the mean, standard deviation, and frequency domain eigenvalues ​​in the characteristic value sequence of the data stream to be detected and the characteristic reference value sequence, respectively. The closer the result calculated according to the above formula is to 0, that is, the smaller the Euclidean distance of the two sets of data is, the more similar they are considered to be. If the result is 0, it is considered that the two sets of data overlap in space and there is no difference between the two.

[0052] S105. Determine whether a mutation occurs to the key device based on the comparison result.

[0053] If the Euclidean distance between the characteristic value sequence of the data stream to be detected and the characteristic reference value sequence is less than a preset threshold, the key indicator is judged to be normal and the key device is judged to be normal; if the Euclidean distance is greater than or equal to the preset threshold, the key indicator is judged to have a mutation and the key device is judged to have a mutation.

[0054] In one embodiment, a tolerance value K is defined, that is, if the difference between the two is less than or equal to K, they are considered to be the same, that is, the characteristics of the two cycles are consistent and no data mutation occurs.

[0055] As described above, in one embodiment, features are extracted from the data to be detected in units of cycles, and the extracted feature values ​​are compared with standard features. If the comparison result is less than or equal to the tolerance value, it is considered that the power index of the power distribution equipment operates well within the cycle; if the comparison result exceeds the tolerance value, it is considered that the data has mutated within the cycle, thereby inferring that the operating status of the equipment within the cycle has mutated, and therefore an alarm is issued in time to remind relevant personnel to check, verify and repair, so as to promptly discover and eliminate abnormalities or faults, ensure the continuity and stability of the production process, and reduce losses.

[0056] like Figure 3 As shown in the figure, the mutation detection method based on key indicator data stream is mainly suitable for abnormal detection of key industrial equipment with stable working environment, clear business content and periodicity. This method uses data cycle as a unit, locks the data characteristics of the equipment indicator within a normal working cycle, and uses this as a standard to compare with the subsequent cycle data to find equipment anomalies and failures. This method can make up for the missed detection caused by single threshold detection, and at the same time integrate the equipment with the specific business it performs. It can not only timely find out whether the equipment itself is in a normal state, but also find out whether the equipment is in a normal operating state when processing specific business. It can timely find and solve problems, help managers to warn early, prevent risks, and ensure the safe and effective operation of industrial production.

[0057] Based on the same technical concept, such as Figure 6 As shown, the embodiment of the present application also provides a mutation detection system for key industrial equipment. For the sake of convenience, only the parts related to the embodiment of the present application are shown. Figure 6 The system comprises a selection module 301, a feature reference module 302, a data acquisition module 303, a data comparison module 304, and an abnormality judgment module 305.

[0058] Selection module 301: selecting one or more key indicators of the monitored industrial key equipment, wherein the key indicators can reflect the operating status and performance of the equipment; Feature benchmark module 302: obtaining a data stream of a complete production line beat under the condition of normal operation of the key indicator as a benchmark data stream, extracting a feature value sequence representing the time domain features and frequency domain features of the benchmark data stream as a feature benchmark, and forming a feature benchmark value sequence; Data acquisition module 303: collects the real-time data stream of the key indicators in real time, and also extracts the feature value sequence representing the time domain characteristics and frequency domain characteristics within each complete production line beat to form a feature value sequence of the data stream to be detected; Data comparison module 304: compares and analyzes the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence; Abnormality judgment module 305: judging whether a mutation occurs to the key equipment according to the comparison result of the data comparison module.

[0059] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0060] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0061] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0062] like Figure 7 As shown, an embodiment of the present application also provides an industrial control host, wherein the industrial control host device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the industrial control host program, the steps in any of the above-mentioned method embodiments are implemented.

[0063] An embodiment of the present application provides a computer program product. When the computer program product runs on a terminal, the terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0064] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / target device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0065] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0066] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0067] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as 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 interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0068] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0069] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for detecting sudden changes in key industrial equipment, used for detecting sudden changes in key equipment on a production line, characterized in that: The method comprises: Select one or more key indicators of the monitored industrial key equipment, wherein the key indicators can reflect the operating status and performance of the equipment; Acquire a data stream within a production line beat when the key indicator is operating normally as a reference data stream, extract a feature value sequence representing the time domain features and frequency domain features of the reference data stream as a feature reference, and form a feature reference value sequence; Collect the real-time data stream of the key indicators in real time, and also extract the characteristic value sequence representing the time domain characteristics and frequency domain characteristics of each production line beat to form the characteristic value sequence of the data stream to be detected; Compare and analyze the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence; Based on the comparison results, determine whether the key equipment has mutated.

2. The method for detecting sudden changes in key industrial equipment according to claim 1, characterized in that: The time domain features in the feature value sequence are manifested as the arithmetic mean and / or standard deviation of interval data of the data stream in the time domain space.

3. The method for detecting sudden changes in key industrial equipment according to claim 1, characterized in that: The data stream is a time series data stream, and the data of the frequency domain characteristics in the eigenvalue sequence are expressed as n discrete eigenvalues ​​in the frequency domain space of the time series data stream within a production line cycle, where n is the number of samples and is greater than 1.

4. The method for detecting sudden changes in key industrial equipment according to claim 3, characterized in that: The discrete eigenvalues ​​are obtained by performing continuous wavelet transform on the time series data stream, and the wavelet basis function used in the wavelet transform is Complex Morlet wavelet.

5. The method for detecting sudden changes in key industrial equipment according to claim 1, characterized in that: The comparison analysis uses the Euclidean distance algorithm to compare and analyze the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence to evaluate the similarity between the two. If the Euclidean distance between the characteristic value sequence of the data stream to be detected and the characteristic reference value sequence is less than a preset threshold, it is determined that the key indicator is normal and the key device is normal; if the Euclidean distance is greater than or equal to the preset threshold, it is determined that the key indicator has mutated and the key device has mutated.

6. The method for detecting sudden changes in key industrial equipment according to claim 1, characterized in that: The method further includes dynamically updating the characteristic reference value sequence, that is, adjusting and optimizing the characteristic reference value sequence regularly or according to preset conditions based on the historical data of the long-term operation of the key equipment.

7. The method for detecting sudden changes in key industrial equipment according to any one of claims 1 to 6, characterized in that: The production line beat includes the complete production line beat including preparation time, production time and waiting time; The key indicators include: power, and / or temperature, and / or voltage, and / or current, and / or pressure, and / or impedance, and / or robot arm load rate, and / or energy consumption rate, and / or coordinate positioning deviation.

8. A mutation detection system for key industrial equipment, characterized in that: The system comprises: Selection module: selects one or more key indicators of the monitored industrial key equipment, wherein the key indicators can reflect the operating status and performance of the equipment; Feature benchmark module: obtains the data stream of a production line beat under the condition of normal operation of the key indicator as the benchmark data stream, extracts the feature value sequence representing the time domain characteristics and frequency domain characteristics of the benchmark data stream as the feature benchmark, and forms a feature benchmark value sequence; Data acquisition module: collects the real-time data stream of the key indicators in real time, and also extracts the characteristic value sequence representing the time domain characteristics and frequency domain characteristics of each production line beat to form the characteristic value sequence of the data stream to be detected; Data comparison module: compares and analyzes the characteristic value sequence of the data stream to be detected with the characteristic reference value sequence; Abnormal judgment module: judge whether the key equipment has a sudden change according to the comparison result of the data comparison module.

9. A target device, characterized in that: The target device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for detecting mutations in critical industrial equipment as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for detecting mutations in critical industrial equipment according to any one of claims 1 to 7 is implemented.

Images