CN117170303A - PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction - Google Patents
PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction Download PDFInfo
- Publication number
- CN117170303A CN117170303A CN202311456602.3A CN202311456602A CN117170303A CN 117170303 A CN117170303 A CN 117170303A CN 202311456602 A CN202311456602 A CN 202311456602A CN 117170303 A CN117170303 A CN 117170303A
- Authority
- CN
- China
- Prior art keywords
- data
- plc
- prediction
- time series
- fault
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 46
- 238000012423 maintenance Methods 0.000 title claims abstract description 41
- 238000001514 detection method Methods 0.000 claims abstract description 59
- 238000000034 method Methods 0.000 claims abstract description 33
- 239000000428 dust Substances 0.000 claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 15
- 238000004891 communication Methods 0.000 claims abstract description 12
- 238000013527 convolutional neural network Methods 0.000 claims description 27
- 238000013528 artificial neural network Methods 0.000 claims description 22
- 230000000306 recurrent effect Effects 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 239000002245 particle Substances 0.000 claims description 12
- 238000002360 preparation method Methods 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000006399 behavior Effects 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims 4
- 238000003909 pattern recognition Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 230000004913 activation Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000009776 industrial production Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of PLC equipment fault diagnosis, and discloses a PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction, which comprises an equipment layer, a database layer and a control layer, wherein the equipment layer comprises an input device and an output device, the input device comprises a temperature sensor, a humidity sensor, a noise measuring instrument and a dust concentration detector, the output device comprises a display device and an equipment fault early warning device, the database layer is used for storing data acquired in the working process of the system, the control layer comprises a CPU, a data prediction module, a fault detection module, a fault early warning module and a communication module, and the fault detection module is used for detecting a fault mode in the predicted data according to the predicted data of the data prediction module; the fault early warning module is used for carrying out early warning on the PLC equipment with faults. The invention can early warn the faulty PLC equipment and improve the reliability of the PLC equipment.
Description
Technical Field
The invention relates to the technical field of PLC equipment fault diagnosis, in particular to a PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction.
Background
PLC devices in an industrial automation control system play a vital role in monitoring and controlling a plurality of process parameters, and ensuring the normal operation of a production line. The stable operation of the PLC device is critical to the normal operation of industrial production. Once the PLC fails, the production line may be stopped, thereby causing industrial production interruption, production efficiency reduction and even serious safety accidents. Currently, conventional PLC fault diagnosis methods generally rely on experienced human operators for fault determination and removal. This approach may lead to inaccuracy in the fault determination. Second, relying on a human operator for fault diagnosis requires a longer time, extending the period of fault diagnosis and maintenance. In addition, most of the current PLC fault detection methods are mainly based on monitoring and analysis of real-time data. These methods only inform the device of the current state and do not predict in advance whether a fault will occur. This makes maintenance personnel often take maintenance measures after a fault has occurred, and cannot be prepared sufficiently in advance. Moreover, the operation states of different PLC devices may be related to each other, and there is a relationship between different operation state indexes of the PLC devices, and the existing PLC fault detection method does not consider the relationship between the two aspects, but performs fault detection on a single PLC device and a certain operation state index of the PLC device.
A fault diagnosis system of a PLC automation device is disclosed in the patent of application publication No. CN113741389a, aiming at detecting the fault of the PLC device from real-time data. The invention comprises a core integrated management module, a real-time drawing driving module, an instantaneous waveform screenshot module and an analysis diagnosis algorithm module, wherein the core integrated management module receives real-time signals; the core comprehensive management module is connected with the real-time drawing driving module, the instantaneous waveform screenshot module and the analysis and diagnosis algorithm module; the real-time drawing driving module is used for dynamically displaying the waveform curve of each signal in real time according to the received real-time signal values, the instantaneous waveform screenshot module is used for intercepting the waveform curve in real time, and the analysis and diagnosis algorithm module is used for analyzing the waveform curve in the real-time drawing driving module and outputting the signals. The invention improves the visualization degree and the automation degree. However, the invention only detects the faults of the PLC equipment based on the real-time data, and the possible faults of the PLC equipment in the future during operation cannot be predicted.
The patent with the application publication number of CN110209110A discloses a PLC fault detection method for remanufacturing a shield machine, and aims to realize the fault detection of a fully-disassembled and fully-detected PLC module of the shield machine. The invention comprises an industrial personal computer, analysis display equipment, a data acquisition card input module, a data acquisition card output module, an input terminal row and an output terminal row; according to the invention, a shield machine PLC fault detection testing device is built, an upper computer data acquisition and analysis system is built, a module fault detection method is provided, and the complete disassembly and complete detection of the shield machine PLC module fault detection is realized. The invention has high integration level, can accurately detect the parameters of each module, can display in real time on an upper computer and perform error analysis with a standard signal curve, has reliable analysis result, high data acquisition stability and low data acquisition cost; the 8 paths of channels of the PLC module can be detected simultaneously, and the detection of faults of various types of PLC modules can be realized. But the invention does not take into account the correlation between different PLC devices/modules, different PLC data.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art.
Disclosure of Invention
Aiming at the defects of the prior art, the main purpose of the invention is to provide a PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction, which can effectively solve the problems in the background art: the existing PLC fault detection method is mainly based on monitoring and analysis of real-time data, whether the PLC equipment can fail or not cannot be predicted in advance, and the existing PLC fault detection method only carries out fault detection on single PLC equipment and a certain running state index of the PLC equipment, and ignores the relevance among the equipment and the indexes. The specific technical scheme of the invention is as follows:
a PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction comprises an equipment layer, a database layer and a control layer; the device layer comprises an input device and an output device which are connected with the PLC device, wherein the input device comprises a temperature sensor for measuring the temperature of the PLC device when the PLC device is in operation, a humidity sensor for measuring the ambient humidity of the PLC device when the PLC device is in operation, a noise measuring instrument for measuring the noise level of the PLC device when the PLC device is in operation and a dust concentration detector for measuring the ambient dust particle concentration of the PLC device when the PLC device is in operation, the output device comprises a display device for displaying the operation state and the operation environment of the PLC device and a device fault early warning device for early warning the PLC device with the abnormal operation state and the abnormal operation environment, the PLC device integrates a plurality of communication ports, supports a plurality of network protocols and can rapidly communicate between different devices; the database layer is used for storing data acquired in the working process of the system.
The invention is further improved in that the control layer comprises a CPU, a data prediction module, a fault detection module, a fault early warning module and a communication module, wherein the CPU is used for managing and controlling the operation of the system; the data prediction module is used for predicting various data of the PLC equipment in operation; the fault detection module is used for detecting a fault mode in the predicted data according to the predicted data of the data prediction module; the fault early warning module is used for carrying out early warning on the PLC equipment with faults through the equipment fault early warning device in the output equipment when the fault detection module detects the fault mode in the prediction data; the communication module is used for constructing a communication network in the system to realize the mutual transmission of data in the system.
A further development of the invention consists in that the temperature sensor and the noise measuring device are mounted on the PLC device for measuring the device temperature and the noise level of the PLC device during operation.
The invention further improves that the humidity sensor and the dust concentration detector are arranged in the running environment of the PLC equipment and are used for measuring the ambient humidity and the ambient dust particle concentration of the PLC equipment when the PLC equipment is running.
The invention further improves that the input device collects device temperature data, noise level data, environment humidity data and environment dust particle concentration data when the PLC device is operated, and the device temperature data, the noise level data, the environment humidity data and the environment dust particle concentration data are all time series data.
A further improvement of the present invention is that,the data prediction module comprises a PLC device data preparation unit, a first prediction unit and a second prediction unit, wherein the PLC device data preparation unit collects data in one operation period of the PLC device through the input device to prepare a multi-element time sequence data set, and the data prediction module is provided withA plurality of PLC devices, each PLC device collecting +_total through the input device>Seed data, let->No. H of the individual PLC devices>The seed data is->，/>，Wherein->，/>，/>The representation dimension is +.>Real space of>Representing the total number of time steps involved in one operating cycle of said PLC device, +.>No. H of personal device>Seed data at->The value over the individual time steps is +.>The PLC device data preparation unit is according to +.>Constructing two sets of multivariate time series data sets, the first set of multivariate time series data sets +.>Is >The sample is marked by->Different types of data of the individual PLC devices, the first set of multivariate time series data sets +.>Is>The samples are represented as a matrix->，/>，/>Wherein->The representation dimension is +.>Is a real space of (2)Middle, wherein->Total number of PLC device data categories collected for said input device,/->Representing the total number of time steps involved in one operating cycle of said PLC device, a second set of multivariate time series data sets +.>Is>The individual samples are defined by the +.>Seed data, said second set of multivariate time series data sets->Is>The samples are represented as a matrix->，/>，/>Wherein->The representation dimension is +.>Of (2) wherein>Total number of PLC devices contained in said system, < >>Representing the total number of time steps involved in one run cycle of the PLC device.
The invention further improves that the first prediction unit and the second prediction unit predict each item of data of the PLC device collected by the input device by using the multi-element time sequence prediction model with the same structure, the first prediction unit trains the multi-element time sequence prediction model of the first prediction unit based on the first multi-element time sequence training data set, the second prediction unit trains the multi-element time sequence prediction model of the second prediction unit based on the second multi-element time sequence training data set, and the input data of the multi-element time sequence prediction model are uniformly expressed as a matrix ，Wherein->The representation dimension is +.>Real space of>Is->The%>Column vector,/->Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series contained in +.>Representing the total number of time steps involved in one operation cycle of the PLC device, the prediction target of the multivariate time series prediction model is based on +.>Generating a predictive value +.>Prediction->I.e. the data of the current operation cycle of the PLC device is used to predict the data of the next operation cycle.
A further improvement of the present invention is that the multivariate time series prediction model comprises a convolutional neural network part and a recurrent neural network part, the convolutional neural network part is composed of a layer of convolutional neural network layers, the convolutional neural network layers use the high of the convolutional kernel and the input dataAre equal in number of lines, i.e. are +.>，/>Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series contained in +.>Representing the total number of time steps contained in one operation period of the PLC device, and setting the convolutional neural network layer to contain +.>A convolution kernel, th->Input data +.>After the convolution operation, input data +. >Conversion into vectors->，/>Wherein，/>The representation dimension is +.>Real space of>Representing the total number of time steps contained in one operation period of the PLC equipment, extracting the interdependence relation of different time sequences in a multi-element time sequence by the conversion process, wherein the output of the convolutional neural network layer is matrix->，/>Wherein->The representation dimension is +.>Real space of>For the total number of convolution kernels>Representing the total number of time steps contained in one operation period of the PLC equipment, wherein the working process of the convolutional neural network layer is as follows: />Wherein->Representing convolutional neural network layer operations,/->Is>Behavior vector->The output matrix of the convolutional neural network layer +.>Is an input to the recurrent neural network portion.
The invention further improves that the recurrent neural network part adopts a GRU network structure, comprises six GRU network layers and is provided withIs>Column->，/>Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series comprised in will +.>In the input GRU network structure, the output is in time step +.>Implicit status on->，/>，/>For the final prediction of the recurrent neural network part, wherein +.>The representation dimension is +.>Real space of >Representation matrix->The working process of the recursive neural network part is as follows: />Wherein->For the final prediction result of said recurrent neural network part,/->Input data for said recurrent neural network portion +.>Is>Column vector,/->Representing a GRU network layer operation.
The invention further improves that the multivariate time sequence prediction model uses the data of at least three running periods of the PLC equipment collected by the input equipment in training, a training set is constructed by the PLC equipment data preparation unit, and the multivariate time sequence prediction model is trained by the following loss function:
；
wherein the method comprises the steps ofRepresenting the total number of time steps involved in one operating cycle of said PLC device,/>Representing training data at +.>True value on time step, +.>Is indicated at->Predicted value on time step,/->Representing the F-norm.
The invention further improves that the first prediction unit and the second prediction unit respectively predict two groups of prediction data by using the multi-element time sequence prediction model, wherein each group of prediction data comprises all kinds of prediction data of all PLC devices, and the length is the total number of time steps contained in one operation period of the PLC devices 。
The invention further improves that the fault detection module comprises a hard index detection unit and a fault mode identification unit, wherein the hard index detection unit sets a section range for each type of data of the PLC equipment collected by the input equipment, the section range comprises a temperature section range, a humidity section range, a noise section range and a dust concentration section range, the hard index detection unit comprises a first hard index detection subunit and a second hard index detection subunit, the first hard index detection subunit detects whether the predicted data obtained by the first prediction unit belongs to the corresponding section range or not, if the predicted data does not belong to the corresponding section range, the fault number of the PLC equipment corresponding to the predicted data is recorded as 1 time, if the predicted data belongs to the corresponding section range, the fault number of the PLC equipment corresponding to the predicted data is recorded as 0 time, the second hard index detection subunit detects whether the predicted data obtained by the second prediction unit belongs to the corresponding section range or not, if the predicted data does not belong to the corresponding section range, the fault number of the PLC equipment corresponding to the predicted data is recorded as 1 time, and if the predicted data does not belong to the corresponding section range, the predicted data is recorded as 0 time.
In the present invention, the failure mode identifying unit uses a pre-trained two-classification SVM model to perform two classifications on the input time-series data, that is, when classifying the input time-series data, the classification class includes two classes of failure and no failure, the failure mode identifying unit includes a first failure mode identifying subunit and a second failure mode identifying subunit, the first failure mode identifying subunit classifies the predicted data obtained by the first prediction unit using the pre-trained two-classification SVM model, if the class is failure, the failure number of the PLC device corresponding to the predicted data is recorded as 1, if the class is no failure, the failure number of the PLC device corresponding to the predicted data is recorded as 0, the second failure mode identifying subunit classifies the predicted data obtained by the second prediction unit using the pre-trained two-classification SVM model, if the class is failure, the failure number of the PLC device corresponding to the predicted data is recorded as 1, and if the class is no failure number of the PLC device corresponding to the predicted data is recorded as 0.
The invention is further improved in that the fault early warning module counts the times of faults detected by the first hard index detection subunit, the second hard index detection subunit, the first fault mode identification subunit and the second fault mode identification subunit on the PLC equipment, and the times of faults counted by the fault early warning module on the PLC equipment are set as follows Secondary, whereinIf->Early warning is not performed if->The PLC equipment is pre-warned through the equipment fault pre-warning device,and the early warning level is set as +.>A stage.
A PLC fault intelligent diagnosis maintenance method based on multivariate time sequence prediction comprises the following specific steps:
a1: collecting data in one operation period of the PLC equipment through a sensor;
a2: constructing a first group of multi-element time sequence data set and a second group of multi-element time sequence data set according to the data in the A1;
a3: according to the first group of multi-element time sequence data sets and the second group of multi-element time sequence data sets in the A2, respectively predicting data in the next operation period of the PLC equipment by using a multi-element time sequence prediction model, and respectively predicting to obtain two groups of prediction data;
a4: detecting whether data in the two groups of predicted data do not belong to a specified interval range according to the two groups of predicted data in the A3;
a5: detecting whether the two groups of prediction data have faults or not based on a pre-trained two-class SVM model according to the two groups of prediction data in the A3;
a6: according to the detection results of A4 and A5, counting the times of faults on the PLC equipment;
a7: and (3) carrying out hierarchical early warning on the PLC equipment with faults according to the times of the faults on the PLC equipment in the step A6.
A computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the intelligent diagnosis and maintenance method for PLC faults based on multivariate time series prediction.
An apparatus, comprising:
a memory for storing instructions;
and the processor is used for executing the instructions to enable the equipment to execute the operation of realizing the intelligent diagnosis and maintenance method for the PLC fault based on the multivariate time sequence prediction.
Compared with the prior art, the invention has the following beneficial effects:
when diagnosing the faults of the PLC equipment, the invention predicts the data of the next operation period based on the data of the current operation period of the PLC equipment, and detects the faults of the PLC equipment based on the predicted data, and the early warning mechanism can effectively improve the operation safety of the PLC equipment and discover the faults in advance;
c2, when the operation data of the PLC equipment is predicted, a multi-element time sequence prediction model is adopted for prediction, correlations between the same kind of data of different PLC equipment and different kinds of data of the same PLC equipment are respectively extracted, and a better effect is achieved when multi-element time sequence data formed by the operation data of the PLC equipment is predicted;
And C3, the invention respectively carries out fault detection on the PLC equipment based on the hard index detection unit and the fault mode identification unit, and can carry out grading early warning according to the detection result, thereby reasonably guiding operators to detect the faults of the PLC equipment and maintain the PLC equipment.
Drawings
Fig. 1 is a schematic diagram of a framework of a PLC fault intelligent diagnosis maintenance system based on multivariate time series prediction.
Fig. 2 is a schematic diagram of steps of a PLC fault intelligent diagnosis maintenance system based on multivariate time series prediction according to the present invention.
Fig. 3 is a diagram illustrating an example of a fault detection module and a fault early warning module of the intelligent diagnosis and maintenance system for PLC fault based on multivariate time series prediction according to the present invention.
Detailed Description
The following detailed description of the present invention is made with reference to the accompanying drawings and specific embodiments, and it is to be understood that the specific features of the embodiments and the embodiments of the present invention are detailed description of the technical solutions of the present invention, and not limited to the technical solutions of the present invention, and that the embodiments and the technical features of the embodiments of the present invention may be combined with each other without conflict.
The term "and/or" is merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone. The character "/", generally indicates that the front and rear associated objects are an or relationship.
Example 1
The embodiment provides a PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction, which is used for solving the problems that whether a PLC device can fail or not can not be predicted in advance in the prior art, and only a single PLC device and a certain operation state index of the PLC device are subjected to fault detection, and the relevance of the operation states among different PLC devices and the relevance among different operation state indexes of the PLC device are ignored. 1-3, the intelligent diagnosis and maintenance system for PLC faults based on multivariate time sequence prediction comprises an equipment layer, a database layer and a control layer. In terms of a device layer, the system comprises an input device and an output device which are connected with the PLC device. The input device comprises a temperature sensor for measuring the temperature of the PLC equipment in operation, a humidity sensor for measuring the ambient humidity of the PLC equipment in operation, a noise measuring instrument for measuring the noise level of the PLC equipment in operation and a dust concentration detector for measuring the concentration of ambient dust particles in the operation of the PLC equipment. The output device comprises a display device for displaying whether the running state and the running environment of the PLC device are abnormal or not and a device fault early warning device for early warning the PLC device with abnormal running state and running environment. The PLC equipment integrates various communication ports, supports various network protocols and can rapidly communicate among different equipment. The database layer is used for storing data collected in the working process of the system.
In this embodiment, the control layer includes a CPU, a data prediction module, a fault detection module, a fault early warning module, and a communication module, where the CPU is configured to manage and control operation of the system; the data prediction module is used for predicting various data of the PLC equipment in operation; the fault detection module is used for detecting a fault mode in the predicted data according to the predicted data of the data prediction module; the fault early warning module is used for carrying out early warning on the PLC equipment with faults through the equipment fault early warning device in the output equipment when the fault detection module detects the fault mode in the prediction data; the communication module is used for constructing a communication network in the system to realize the mutual transmission of data in the system.
In this embodiment, the temperature sensor and noise meter are mounted on the PLC device for measuring the device temperature and noise level of the PLC device when in operation.
In this embodiment, the humidity sensor and the dust concentration detector are installed in an operation environment of the PLC apparatus, and are used to measure the ambient humidity and the ambient dust particle concentration of the PLC apparatus when the PLC apparatus is operated.
In this embodiment, the input device collects device temperature data, noise level data, ambient humidity data, and ambient dust particle concentration data when the PLC device is operating, where the device temperature data, noise level data, ambient humidity data, and ambient dust particle concentration data are all time-series data.
In this embodiment, the data prediction module includes a PLC device data preparation unit, a first prediction unit, and a second prediction unit, where the PLC device data preparation unit collects data in one operation cycle of the PLC device through the input device to prepare a multiple time series data set, where the multiple time series data set is provided withA plurality of PLC devices, each PLC device collecting +_total through the input device>Seed data, let->No. H of the individual PLC devices>The seed data is->，/>，/>Wherein，/>，/>The representation dimension is +.>Real space of>Representing the total number of time steps involved in one operating cycle of said PLC device, +.>No. H of personal device>Seed data at->The value over the individual time steps is +.>The PLC device data preparation unit is according to +.>Constructing two sets of multivariate time series data sets, a first set of multivariate time series data setsIs>The sample is marked by->PLC deviceDifferent classes of data are prepared, the first set of multivariate time series data setsIs>The samples are represented as a matrix->，/>，/>Wherein->The representation dimension is +.>Of (2) wherein>Total number of PLC device data categories collected for said input device,/->Representing the total number of time steps involved in one operating cycle of said PLC device, a second set of multivariate time series data sets +. >Is>The individual samples are defined by the +.>Seed data, said second set of multivariate time series data sets->Is>The samples are represented as a matrix->，/>，/>Wherein->The representation dimension is +.>Of (2) wherein>Total number of PLC devices contained in said system, < >>Representing the total number of time steps involved in one run cycle of the PLC device.
In this embodiment, the first prediction unit and the second prediction unit predict each item of data of the PLC device collected by the input device using a multi-component time series prediction model of the same structure, the first prediction unit trains the multi-component time series prediction model of the first prediction unit based on the first set of multi-component time series training data sets, the second prediction unit trains the multi-component time series prediction model of the second prediction unit based on the second set of multi-component time series training data sets, and the input data of the multi-component time series prediction models are uniformly represented as a matrix，/>WhereinThe representation dimension is +.>Real space of>Is->The%>Column vector,/->Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series contained in +.>Representing a total number of time steps involved in one operating cycle of the PLC device, the predictive objective of the multivariate time series predictive model being based on Generating a predictive value +.>Prediction->I.e. the data of the current operation period of the PLC equipment is used for predicting the data of the next operation period, in particular, the multiplex timeSequence prediction model is based on->Prediction->Based on->Prediction->Wherein->And finally, the prediction target is achieved for the predicted value of the multi-element time sequence prediction model and so on.
In this embodiment, the multivariate time series prediction model includes a convolutional neural network portion and a recurrent neural network portion, where the convolutional neural network portion is formed by a layer of convolutional neural network layer, and the convolutional neural network layer is activated by using a RELU activation function, where a RELU activation function formula is as follows:wherein->Representing the output of the RELU activation function, +.>Input representing RELU activation function, +.>To take the maximum function, the convolution kernel used by the convolution neural network layer is high and input data +.>Are equal in number of lines, i.e. are +.>，/>Wherein->Representing dimensions asReal space of>Representation matrix->The number of lines of the time series contained in +.>Representing the total number of time steps contained in one operation period of the PLC device, and setting the convolutional neural network layer to contain +.>A convolution kernel, th- >Input data +.>After the convolution operation, input data +.>Conversion into vectors->，/>Wherein，/>The representation dimension is +.>Real space of>Representing the total number of time steps contained in one operation period of the PLC equipment, extracting the interdependence relation of different time sequences in a multi-element time sequence by the conversion process, wherein the output of the convolutional neural network layer is matrix->，/>Wherein->The representation dimension is +.>Real space of>For the total number of convolution kernels>Representing the total number of time steps contained in one operation period of the PLC equipment, wherein the working process of the convolutional neural network layer is as follows: />Wherein->Representing the convolutional neural network layer operation,is>Behavior vector->The convolutional neural network layerOutput matrix->Is an input to the recurrent neural network portion.
In this embodiment, the recurrent neural network part adopts a GRU network structure, and includes six GRU network layers, and is provided withIs>Column->，/>Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series comprised in will +.>In the input GRU network structure, the output is in time step +.>Implicit status on->，，/>For the final prediction of the recurrent neural network part, wherein +. >The representation dimension is +.>Real space of>Representation matrix->The working process of the recursive neural network part is as follows: />Wherein->For the final prediction result of said recurrent neural network part,/->Input data for said recurrent neural network portion +.>Is>Column vector,/->Representing a GRU network layer operation.
In this embodiment, the multivariate time series prediction model uses the data of at least three operation periods of the PLC device collected by the input device during training, constructs a training set by the PLC device data preparation unit, and trains the multivariate time series prediction model by using a random gradient descent method by using the following loss function:
；
wherein the method comprises the steps ofRepresenting the total number of time steps involved in one operating cycle of said PLC device,/>Representing training data at +.>True value on time step, +.>Is indicated at->Predicted value on time step,/->Representing the F-norm.
In this embodiment, the first prediction unit and the second prediction unit respectively predict two sets of prediction data by using the multivariate time series prediction model, where each set of prediction data includes all kinds of prediction data of all PLC devices, and the length is the total number of time steps included in one operation cycle of the PLC devices 。
In this embodiment, the fault detection module includes a hard index detection unit and a fault mode identification unit, where the hard index detection unit sets a range for each type of data of the PLC device collected by the input device, where the range includes a temperature range, a humidity range, a noise range, and a dust concentration range, the hard index detection unit includes a first hard index detection subunit and a second hard index detection subunit, where the first hard index detection subunit detects whether the predicted data obtained by the first prediction unit belongs to the corresponding range, if the predicted data does not belong to the corresponding range, the number of times of faults of the PLC device corresponding to the predicted data is recorded as 1, if the predicted data belongs to the corresponding range, the number of times of faults of the PLC device corresponding to the predicted data is recorded as 0, and the second hard index detection subunit detects whether the predicted data obtained by the second prediction unit belongs to the corresponding range, if the predicted data does not belong to the corresponding range, the number of times of faults of the PLC device corresponding to the predicted data is recorded as 1, and if the predicted data does not belong to the corresponding range, the number of times of faults of the PLC device corresponding to the predicted data is recorded as 0.
In this embodiment, the failure mode identifying unit uses a pre-trained two-classification SVM model to perform two classifications on the input time-series data, that is, when classifying the input time-series data, the classification class includes two classes including failure and no failure, the failure mode identifying unit includes a first failure mode identifying subunit and a second failure mode identifying subunit, the two classes SVM model uses a linear kernel function, the first failure mode identifying subunit classifies the predicted data obtained by the first prediction unit using the pre-trained two-classification SVM model, if the class is failure, the failure number of the PLC device corresponding to the predicted data is recorded as 1, if the class is no failure, the failure number of the PLC device corresponding to the predicted data is recorded as 0, the second failure mode identifying subunit classifies the predicted data obtained by the second prediction unit using the pre-trained two-classification SVM model, if the class is failure, the failure number of the PLC device corresponding to the predicted data is recorded as 1, and if the class is no failure number of the PLC device corresponding to the predicted data is recorded as 0.
In this embodiment, the failure early-warning module counts the number of times that the first hard index detection subunit, the second hard index detection subunit, the first failure mode identification subunit and the second failure mode identification subunit detect the failure on the PLC device, and sets the number of times that the failure early-warning module counts the failure on the PLC device asIn a second time, the first time,wherein->If (if)Early warning is not performed if->The PLC equipment is pre-warned through the equipment fault pre-warning device, and the pre-warning grade is set to be +.>A stage.
Example 2
The embodiment provides a PLC fault intelligent diagnosis and maintenance method based on multivariate time sequence prediction, which comprises the following specific steps:
a1: collecting data in one operation period of the PLC equipment through a sensor;
a2: constructing a first group of multi-element time sequence data set and a second group of multi-element time sequence data set according to the data in the A1;
a3: according to the first group of multi-element time sequence data sets and the second group of multi-element time sequence data sets in the A2, respectively predicting data in the next operation period of the PLC equipment by using a multi-element time sequence prediction model, and respectively predicting to obtain two groups of prediction data;
a4: detecting whether data in the two groups of predicted data do not belong to a specified interval range according to the two groups of predicted data in the A3;
A5: detecting whether the two groups of prediction data have faults or not based on a pre-trained two-class SVM model according to the two groups of prediction data in the A3;
a6: according to the detection results of A4 and A5, counting the times of faults on the PLC equipment;
a7: and (3) carrying out hierarchical early warning on the PLC equipment with faults according to the times of the faults on the PLC equipment in the step A6.
Example 3
The embodiment provides a computer readable storage medium, which uses a special storage server, a hard disk array or cloud service to store computer programs and data required by a PLC fault intelligent diagnosis and maintenance system, and the computer programs realize the PLC fault intelligent diagnosis and maintenance method based on multivariate time sequence prediction when being executed by a processor.
Example 4
The present embodiment provides an apparatus comprising:
c1, a hard disk memory for storing an instruction set, a module, a model and an algorithm of the intelligent diagnosis and maintenance method for PLC fault based on multivariate time sequence prediction;
and c2, a high-performance image processor is used for executing the instruction, so that the equipment executes the operation of realizing the intelligent diagnosis and maintenance method for PLC fault based on multivariate time sequence prediction, has parallel computing capability, and is suitable for rapidly processing image data.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, systems, and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present invention and the scope of the claims, which are all within the protection of the present invention.
Claims (17)
1. The PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction is characterized by comprising an equipment layer, a database layer and a control layer;
the device layer comprises an input device and an output device which are connected with the PLC device, wherein the input device comprises a temperature sensor for measuring the temperature of the PLC device when the PLC device is in operation, a humidity sensor for measuring the ambient humidity of the PLC device when the PLC device is in operation, a noise measuring instrument for measuring the noise level of the PLC device when the PLC device is in operation and a dust concentration detector for measuring the concentration of ambient dust particles when the PLC device is in operation, and the output device comprises a display device for displaying the operation state and the operation environment of the PLC device and a device fault early warning device for early warning the operation state and the operation environment of the PLC device;
the database layer is used for storing data acquired in the working process of the system.
2. The intelligent diagnosis and maintenance system for PLC failure based on multivariate time series prediction according to claim 1, wherein the control layer comprises a CPU, a data prediction module, a failure detection module, a failure early warning module and a communication module, and the CPU is used for managing and controlling the operation of the system; the data prediction module is used for predicting various data of the PLC equipment in operation; the fault detection module is used for detecting a fault mode in the predicted data according to the predicted data of the data prediction module; the fault early warning module is used for carrying out early warning on the PLC equipment with faults through the equipment fault early warning device in the output equipment when the fault detection module detects the fault mode in the prediction data; the communication module is used for constructing a communication network in the system.
3. The intelligent diagnosis and maintenance system for PLC fault based on multivariate time series prediction according to claim 2, wherein the temperature sensor and the noise measuring instrument are installed on the PLC apparatus for measuring the apparatus temperature and the noise level of the PLC apparatus when in operation.
4. The intelligent diagnosis and maintenance system for PLC failure based on multivariate time series prediction according to claim 3, wherein the humidity sensor and the dust concentration detector are installed in the operation environment of the PLC equipment and are used for measuring the environment humidity and the environment dust particle concentration of the PLC equipment when the PLC equipment is operated.
5. The intelligent diagnosis and maintenance system for PLC failure based on multivariate time series prediction according to claim 4, wherein the input device collects device temperature data, noise level data, environmental humidity data and environmental dust particle concentration data of the PLC device during operation, and the device temperature data, the noise level data, the environmental humidity data and the environmental dust particle concentration data are all time series data.
6. The intelligent diagnosis and maintenance system for PLC failure based on multiple time series prediction according to claim 5, wherein the data prediction module comprises a PLC device data preparation unit, a first prediction unit and a second prediction unit, the PLC device data preparation unit collects data in one operation period of the PLC device through the input device to prepare multiple time series data sets, and the intelligent diagnosis and maintenance system is provided with A plurality of PLC devices, each PLC device collecting +_total through the input device>Seed data, let->No. H of the individual PLC devices>The seed data is->，/>，/>Wherein->，/>，The representation dimension is +.>Real space of>Representing the total number of time steps involved in one operating cycle of said PLC device, +.>No. H of personal device>Seed data at->The value over the individual time steps is +.>The PLC equipment data preparation unit is used for preparing the data according to the following steps ofConstructing two sets of multivariate time series data sets, the first set of multivariate time series data sets +.>Is>The sample is marked by->Different types of data of the individual PLC devices, the first set of multivariate time series data sets +.>Is>The samples are represented as a matrix->，/>，/>Wherein->The representation dimension is +.>Of (2) wherein>Total number of PLC device data categories collected for said input device,/->Representing the total number of time steps involved in one operating cycle of said PLC device, a second set of multivariate time series data sets +.>Is>The individual samples are defined by the +.>Seed data, said second set of multivariate time series data sets->Is>The samples are represented as a matrix->，/>，/>WhereinThe representation dimension is +.>Of (2) wherein>Total number of PLC devices contained in said system, < > >Representing the total number of time steps involved in one run cycle of the PLC device.
7. The intelligent diagnosis and maintenance system for PLC failure based on multiple time series prediction according to claim 6, wherein the first prediction unit and the second prediction unit predict each item of data of the PLC device collected by the input device using multiple time series prediction models of the same structure, the first prediction unit trains the multiple time series prediction model of the first prediction unit based on the first multiple time series training data set, the second prediction unit trains the multiple time series prediction model of the second prediction unit based on the second multiple time series training data set, and the input data of the multiple time series prediction models are collectively expressed as a matrix，/>Wherein->Representing dimensions asReal space of>Is->The%>Column vector,/->Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series contained in +.>Representing the total number of time steps involved in one operation cycle of the PLC device, the prediction target of the multivariate time series prediction model is based on +.>Generating a predicted value Prediction->I.e. the data of the current operation cycle of the PLC device is used to predict the data of the next operation cycle.
8. The intelligent diagnosis and maintenance system for PLC failure based on multi-element time series prediction according to claim 7, wherein the multi-element time series prediction model comprises a convolutional neural network part and a recurrent neural network part, the convolutional neural network part is composed of a layer of convolutional neural network layers, and the convolutional neural network layers use the high of the convolutional kernel and input dataAre equal in number of lines, i.e. are +.>，/>Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series contained in +.>Representing the total time steps involved in one operating cycle of the PLC deviceThe number is set to be that the convolutional neural network layer comprises +.>A convolution kernel, th->Input data +.>After the convolution operation, input data +.>Conversion into vectors->，/>Wherein->，/>Representing dimensions asReal space of>Representing the total number of time steps contained in one operation period of the PLC equipment, extracting the interdependence relation of different time sequences in a multi-element time sequence in the conversion process, wherein the output of the convolutional neural network layer is a matrix ，/>Wherein->The representation dimension is +.>Real space of>For the total number of convolution kernels>Representing the total number of time steps contained in one operation period of the PLC equipment, wherein the working process of the convolutional neural network layer is as follows:wherein->Representing convolutional neural network layer operations,/->Is>Behavior vector->The output matrix of the convolutional neural network layer +.>Is an input to the recurrent neural network portion.
9. The intelligent diagnosis and maintenance system for PLC failure based on multivariate time series prediction according to claim 8, wherein the recurrent neural network part adopts a GRU network structure, comprises six GRU network layers, and is provided withIs>Is arranged as，/>Wherein->The representation dimension is +.>Real space of>Representation matrix->The number of lines of the time series comprised in will +.>In the input GRU network structure, the output is in time step +.>Implicit status on->，/>，For the final prediction of the recurrent neural network part, wherein +.>The representation dimension is +.>Real space of>Representation matrix->The working process of the recurrent neural network part is as follows:wherein->For the final prediction result of said recurrent neural network part,/- >Input data for said recurrent neural network portion +.>Is>Column vector,/->Representing a GRU network layer operation.
10. The intelligent diagnosis and maintenance system for PLC failure based on multivariate time series prediction according to claim 9, wherein the multivariate time series prediction model is trained by using the data of not less than three operation cycles of the PLC device collected by the input device during training, constructing a training set by the PLC device data preparation unit, and training the multivariate time series prediction model by the following loss function:
；
wherein the method comprises the steps ofRepresenting the total number of time steps involved in one operating cycle of said PLC device,/>Representing training data inTrue value on time step, +.>Is indicated at->Predicted value on time step,/->Representing the F-norm.
11. The intelligent diagnosis and maintenance system for PLC failure based on multiple time series prediction according to claim 12, wherein the first prediction unit and the second prediction unit respectively predict two sets of prediction data by using the multiple time series prediction model, each set of prediction data contains all kinds of prediction data of all PLC devices, and the length is the total number of time steps contained in one operation period of the PLC devices 。
12. The intelligent diagnosis and maintenance system for PLC fault based on multivariate time series prediction according to claim 11, wherein the fault detection module comprises a hard index detection unit and a fault pattern recognition unit, the hard index detection unit sets a range for each type of PLC device data collected by the input device, the range includes a temperature range, a humidity range, a noise range and a dust concentration range, the hard index detection unit comprises a first hard index detection subunit and a second hard index detection subunit, the first hard index detection subunit detects whether the predicted data obtained by the first prediction unit belongs to the corresponding range, if not, the number of times of the fault of the PLC device corresponding to the predicted data is counted as 1, if not, the number of times of the fault of the PLC device corresponding to the predicted data is counted as 0, if not, the predicted data obtained by the second prediction unit belongs to the corresponding range, if not, the number of times of the fault of the PLC device corresponding to the predicted data is counted as 1, and if not, the number of times of the fault of the PLC device corresponding to the predicted data is counted as 0.
13. The intelligent diagnosis and maintenance system for PLC failure based on multivariate time series prediction according to claim 12, wherein the failure mode recognition unit uses a pre-trained bi-classification SVM model to bi-classify the input time series data, i.e., when classifying the input time series data, the classification class includes two classes of failure and no failure, the failure mode recognition unit includes a first failure mode recognition subunit and a second failure mode recognition subunit, the first failure mode recognition subunit classifies the predicted data predicted by the first prediction unit using the pre-trained bi-classification SVM model, if the class is failure, the failure count of the PLC device corresponding to the predicted data is 1, if the class is no failure, the failure count of the PLC device corresponding to the predicted data is 0, the second failure mode recognition subunit uses the pre-trained bi-classification SVM to classify the predicted data predicted by the second prediction unit, if the class is failure, the failure count of the PLC device corresponding to the predicted data is 1, and if the class is no failure count of the PLC device corresponding to the predicted data is 0.
14. The intelligent diagnosis and maintenance system for PLC according to claim 13, wherein the failure pre-warning module counts the number of failures detected by the first hard index detecting subunit, the second hard index detecting subunit, the first failure mode identifying subunit, and the second failure mode identifying subunit, and sets the number of failures counted by the failure pre-warning module as the number of failures counted by the PLC asSecondary, wherein->If->Early warning is not performed if->The PLC equipment is pre-warned through the equipment fault pre-warning device, and the pre-warning grade is set to be +.>A stage.
15. A method for intelligent diagnosis and maintenance of PLC faults based on multivariate time series prediction, which is implemented based on the intelligent diagnosis and maintenance system of PLC faults based on multivariate time series prediction as claimed in claims 1 to 14, characterized in that the method comprises the following specific steps:
a1: collecting data in one operation period of the PLC equipment through a sensor;
a2: constructing a first group of multi-element time sequence data set and a second group of multi-element time sequence data set according to the data in the A1;
a3: according to the first group of multi-element time sequence data sets and the second group of multi-element time sequence data sets in the A2, respectively predicting data in the next operation period of the PLC equipment by using a multi-element time sequence prediction model, and respectively predicting to obtain two groups of prediction data;
A4: detecting whether data in the two groups of predicted data do not belong to a specified interval range according to the two groups of predicted data in the A3;
a5: detecting whether the two groups of prediction data have faults or not based on a pre-trained two-class SVM model according to the two groups of prediction data in the A3;
a6: according to the detection results of A4 and A5, counting the times of faults on the PLC equipment;
a7: and (3) carrying out hierarchical early warning on the PLC equipment with faults according to the times of the faults on the PLC equipment in the step A6.
16. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements a method for intelligent diagnosis and maintenance of PLC faults based on multivariate time series prediction as set forth in claim 15.
17. An apparatus, comprising: a memory for storing instructions; a processor for executing the instructions to cause the apparatus to perform operations implementing a method for intelligent diagnostic maintenance of PLC failure based on multivariate time series prediction as set forth in claim 15.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311456602.3A CN117170303B (en) | 2023-11-03 | 2023-11-03 | PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311456602.3A CN117170303B (en) | 2023-11-03 | 2023-11-03 | PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117170303A true CN117170303A (en) | 2023-12-05 |
| CN117170303B CN117170303B (en) | 2024-01-26 |
Family
ID=88930329
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311456602.3A Active CN117170303B (en) | 2023-11-03 | 2023-11-03 | PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117170303B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117170304A (en) * | 2023-11-03 | 2023-12-05 | 傲拓科技股份有限公司 | PLC remote monitoring control method and system based on industrial Internet of things |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115146842A (en) * | 2022-06-24 | 2022-10-04 | 沈阳建筑大学 | Multivariate time series trend prediction method and system based on deep learning |
| CN115688871A (en) * | 2022-09-20 | 2023-02-03 | 中国科学院计算技术研究所 | Multi-element time sequence prediction method and system based on pre-training enhancement |
| CN115860243A (en) * | 2022-12-14 | 2023-03-28 | 四川大学 | Fault prediction method and system based on industrial Internet of things data |
| WO2023071217A1 (en) * | 2021-10-27 | 2023-05-04 | 中国华能集团清洁能源技术研究院有限公司 | Multi-working-condition process industrial fault detection and diagnosis method based on deep transfer learning |
| CN116292247A (en) * | 2023-03-03 | 2023-06-23 | 徐州徐工矿业机械有限公司 | Hydraulic shovel main pump early fault diagnosis method and device based on LSTM flow prediction model |
| CN116893654A (en) * | 2023-08-04 | 2023-10-17 | 山东益健药业有限公司 | Operation state monitoring system of glucosamine sulfate capsule preparation equipment |
-
2023
- 2023-11-03 CN CN202311456602.3A patent/CN117170303B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023071217A1 (en) * | 2021-10-27 | 2023-05-04 | 中国华能集团清洁能源技术研究院有限公司 | Multi-working-condition process industrial fault detection and diagnosis method based on deep transfer learning |
| CN115146842A (en) * | 2022-06-24 | 2022-10-04 | 沈阳建筑大学 | Multivariate time series trend prediction method and system based on deep learning |
| CN115688871A (en) * | 2022-09-20 | 2023-02-03 | 中国科学院计算技术研究所 | Multi-element time sequence prediction method and system based on pre-training enhancement |
| CN115860243A (en) * | 2022-12-14 | 2023-03-28 | 四川大学 | Fault prediction method and system based on industrial Internet of things data |
| CN116292247A (en) * | 2023-03-03 | 2023-06-23 | 徐州徐工矿业机械有限公司 | Hydraulic shovel main pump early fault diagnosis method and device based on LSTM flow prediction model |
| CN116893654A (en) * | 2023-08-04 | 2023-10-17 | 山东益健药业有限公司 | Operation state monitoring system of glucosamine sulfate capsule preparation equipment |
Non-Patent Citations (1)
| Title |
|---|
| 徐圆;卢玉帅;才轶;: "基于多元时序驱动的ELM全流程故障预测及其应用", 化工学报, no. 01 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117170304A (en) * | 2023-11-03 | 2023-12-05 | 傲拓科技股份有限公司 | PLC remote monitoring control method and system based on industrial Internet of things |
| CN117170304B (en) * | 2023-11-03 | 2024-01-05 | 傲拓科技股份有限公司 | PLC remote monitoring control method and system based on industrial Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117170303B (en) | 2024-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109186813A (en) | A kind of temperature sensor self-checking unit and method | |
| CN111866031B (en) | Equipment monitoring and analyzing method and device, server and storage medium | |
| CN102789545B (en) | Based on the Forecasting Methodology of the turbine engine residual life of degradation model coupling | |
| US10373065B2 (en) | Generating database cluster health alerts using machine learning | |
| CN117170303B (en) | PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction | |
| CN107885642A (en) | Business monitoring method and system based on machine learning | |
| KR101948604B1 (en) | Method and device for equipment health monitoring based on sensor clustering | |
| CN102282552A (en) | System, method and computer program for pattern based intelligent control, monitoring and automation | |
| KR102427205B1 (en) | Apparatus and method for generating training data of artificial intelligence model | |
| CN111459700A (en) | Method and apparatus for diagnosing device failure, diagnostic device, and storage medium | |
| Dorgo et al. | Decision trees for informative process alarm definition and alarm-based fault classification | |
| CN102375452A (en) | Event-driven data mining method for improving fault code settings and isolating faults | |
| CN117196066A (en) | Intelligent operation and maintenance information analysis model | |
| CN104460654B (en) | A kind of gear imperfection failure diagnosis Rules extraction method based on quantization characteristic relation | |
| CN107248039A (en) | Real-time quality control method and device based on medical specimen detection project result | |
| CN104596780A (en) | Diagnosis method for sensor faults of motor train unit braking system | |
| CN112581719B (en) | Semiconductor packaging process early warning method and device based on time sequence generation countermeasure network | |
| JPH08234832A (en) | Device and method for monitoring and diagnostic plant | |
| CN116450399A (en) | Fault diagnosis and root cause positioning method for micro service system | |
| CN108415810A (en) | Hard disk state monitoring method and device | |
| CN108153987A (en) | A kind of hydraulic pump Multiple faults diagnosis approach based on the learning machine that transfinites | |
| CN112395684A (en) | Intelligent fault diagnosis method for high-speed train running part system | |
| CN106355314A (en) | A Vehicle Engine Air Valve Quality Streaming Data Processing Method and Device | |
| Hajgató et al. | PredMaX: Predictive maintenance with explainable deep convolutional autoencoders | |
| KR20230125116A (en) | The method for fault detection without training data or diagnosis with prediction of remaining time until breakdown using clustering algorithm and statistical methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |