CN113253037B

CN113253037B - Current ripple-based edge cloud cooperative equipment state monitoring method and system and medium

Info

Publication number: CN113253037B
Application number: CN202110693850.4A
Authority: CN
Inventors: 张颖华
Original assignee: Beijing Cyberconsortium Technology Co ltd
Current assignee: Beijing Cyberconsortium Technology Co ltd
Priority date: 2021-06-22
Filing date: 2021-06-22
Publication date: 2021-10-08
Anticipated expiration: 2041-06-22
Also published as: CN113253037A

Abstract

The invention belongs to the technical field of intelligent monitoring equipment, and relates to a side cloud cooperative type equipment state monitoring method and system based on current ripple and a distributed storage medium. The method comprises the following steps: selecting a characteristic current ripple library through a terminal according to the type of the monitored equipment, wherein the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process; the method comprises the steps that current data of a power point of a monitored device are collected in real time by an edge side; preprocessing the acquired current data to obtain current ripples; judging the equipment state or identifying an abnormal event for the monitored equipment based on the current ripple according to the equipment preset state threshold and the characteristic current ripple library; and uploading the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data correspondingly collected by the abnormal event to a cloud for storage, downloading and displaying on the terminal. The current ripple is obtained by acquiring current data of a power point of monitored equipment and preprocessing the current data, so that the real-time monitoring of the equipment state is realized.

Description

Current ripple-based edge cloud cooperative equipment state monitoring method and system and medium

Technical Field

The invention belongs to the technical field of intelligent monitoring equipment, and particularly relates to a side cloud cooperative type equipment state monitoring method based on current ripples, a side cloud cooperative type equipment state monitoring system based on current ripples, and a distributed storage medium.

Background

With the continuous development of science and technology, more and more mechanical equipment and higher value are provided. Generally, mechanical devices have long life and the period of updating is long. Thus, mechanical equipment has been in service for many years. On the one hand, in the manufacturing industry today, production plants that seem to be functioning well do not actually work in the best state, which inevitably causes huge losses for the enterprise. On the other hand, as the degree of integration and automation of the equipment becomes higher and higher, the efficiency of the whole plant and factory is directly related to the operation state of the equipment. The value of the equipment has a great improvement space, and by constructing a digital factory, knowing the real-time running state of the equipment in real time and evaluating the comprehensive efficiency in real time, the abnormal production can be known in time, the loss is avoided, and the execution performance of the production plan is evaluated in time.

The internet of things platform provides the possibility of Equipment state monitoring and Equipment comprehensive efficiency (OEE for short) analysis, however, the mechanical Equipment has many application fields, various types and models, and the following problems exist when the internet of things platform is constructed:

the device itself already contains a management Controller (PLC, for example, Programmable Logic Controller), and the acquisition of the device status can be obtained by communicating with the PLC of the device. However, dozens of PLC communication protocols exist in the market, some communication protocols are in a manufacturer-defined form, and the PLC communication protocols of different manufacturers and different devices may be different; on the other hand, even if different manufacturers and different devices can use the same PLC communication protocol, because the PLC programs for each device to control according to a specific application are different, the PLC program still needs to be programmed specifically to obtain the PLC status. The method for developing the data acquisition system by customizing each device is high in cost and long in period.

The problem is more serious for a large number of earlier devices, which are not even equipped with PLC, or even if PLC is included, the protocol is not open to the outside, and how to identify the device condition of such dummy devices becomes a troublesome problem in the current digital plant construction. The existing proposed solution is that a specific photoelectric sensor, a distance sensor, a camera and other sensors are additionally arranged to acquire specific parameters of equipment so as to facilitate analysis, however, the sensor is high in cost and cannot be standardized in installation, and one piece of equipment often needs a plurality of sensors, so that great economic expenditure is inevitably brought by modification; when the sensor is installed, equipment needs to be stopped, so that the normal production is greatly influenced, and the support of a production client is difficult to obtain.

Disclosure of Invention

The invention aims to solve the technical problem of providing a side cloud cooperative type equipment state monitoring method and system based on current ripple and a distributed storage medium aiming at the defects in the prior art, and equipment state monitoring is realized in a side cloud cooperative mode based on the current ripple of the equipment.

The technical scheme adopted for solving the technical problem of the invention is to provide the following scheme:

as an aspect of the present invention, a method for monitoring a state of a current ripple-based edge cloud collaborative device is provided, which includes the steps of:

constructing or updating a characteristic current ripple library according to current data of sample equipment, wherein the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to a process, and obtaining the characteristic current ripple template at least comprises the following steps: preprocessing the current of the sample equipment through an edge side to obtain a current ripple, determining a process section of the current ripple through a cloud end, and calculating the mode of sampling points in the current ripple of the process section;

selecting a characteristic current ripple library through a terminal according to the type of monitored equipment, wherein the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to a process, and the characteristic current ripple library is stored in a cloud database and downloaded to an edge side;

the edge side collects the current data of the power point of the monitored equipment in real time, and the collection frequency and amplitude unit selects the parameter setting which is the same as the reference current ripple in the characteristic current ripple template; and the number of the first and second groups,

preprocessing the acquired current data to obtain current ripples, wherein the preprocessing method and the processing parameter setting are the same as the method and the parameter setting for constructing the characteristic current ripple library; and judging the equipment state or identifying an abnormal event for the monitored equipment based on the current ripple according to the equipment preset state threshold and the characteristic current ripple library, wherein the equipment state comprises: shutdown state, standby state, working state, abnormal state, the abnormal event includes: an abnormal startup event, an unplanned shutdown event, a minor shutdown event, a speed loss event, a processing abnormal event;

uploading the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data correspondingly collected by the abnormal event to the cloud for storage;

and downloading and displaying the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data correspondingly acquired by the abnormal event at a terminal.

Preferably, constructing a library of characteristic current ripples from the current data of the sample device comprises the steps of:

acquiring a plurality of groups of current data at a power point of at least one sample device on the edge side; and the number of the first and second groups,

preprocessing the acquired current data to obtain current ripples, wherein the preprocessing at least comprises filtering the current data;

the cloud end determines an equipment process section according to different characteristics of the current ripple along with different process changes; selecting characteristic points for the data of different process segments to align the characteristic points of the different process segments, wherein the characteristic points comprise extreme values; normalizing the data of at least two groups of same process segments, and aligning time based on the positions of the feature points; and, calculating a mode at each sampled data point in the current ripple; and the number of the first and second groups,

and determining the characteristic current ripple template corresponding to the process section according to the mode of the sampling points in the current ripple, wherein the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process, and the characteristic current ripple template is used as a reference standard for identifying the working condition of the equipment.

Preferably, the device preset state threshold and the characteristic current ripple library are issued to the edge side through the cloud, and the device preset state threshold includes a shutdown current ripple threshold, a standby current ripple threshold and a correlation threshold; on the side of the edge, it is,

if the current ripple is smaller than the shutdown current ripple threshold, judging that the state of the monitored equipment is a shutdown state;

if the current ripple is smaller than the standby current ripple threshold and larger than the shutdown current ripple threshold, the state of the monitored equipment is judged to be a standby state;

if the correlation coefficient of the current ripple and the reference current ripple in the characteristic current ripple template is greater than the correlation threshold, judging the state of the monitored equipment to be a normal state;

and if the current ripple does not belong to any one of the situations, judging the state of the monitored equipment to be an abnormal state.

Preferably, when the monitored device is in a working state, determining the device working condition of the monitored device includes:

calculating Pearson correlation coefficient values of the current ripple of the monitored device and the reference current ripple of the characteristic current ripple template of various processes in the characteristic current ripple library, wherein the Pearson correlation coefficient value formula is as follows:

wherein the content of the first and second substances,

xi is the current ripple of the monitored device at present, i is a data point position serial number, i is 1,2, … Nk, and Xi represents the current ripple position and the past total Nk data points;

nk is the data length of the reference current ripple in the kth characteristic current ripple template in the characteristic current ripple library;

yki is the ith data point of the reference current ripple in the kth characteristic current ripple template in the characteristic current ripple library, k represents the template serial number in the characteristic current ripple library, k is 1,2, … M, i is the data point position serial number, i is 1,2, … Nk, and Yki represents the total Nk data points of the reference current ripple in the kth characteristic current ripple template in the characteristic current ripple library;

m is the number of the characteristic current ripple templates of a certain process in the characteristic current ripple library;

or, further according to the Pearson correlation coefficient values of the current ripple of the monitored device and the reference current ripple of the characteristic current ripple template, determining the current process of the monitored device:

if the Pearson correlation coefficient value of the current ripple of the monitored equipment and the reference current ripple of the characteristic current ripple template of a certain process section is between the preset state threshold value of the equipment and 1.0, the monitored equipment carries out the process currently;

otherwise, the monitored equipment is not currently performing such a process.

Preferably, the method further comprises the following steps: calculating the comprehensive efficiency of the equipment according to the production arrangement, the equipment state and the abnormal events, wherein the formula for calculating the comprehensive efficiency of the equipment is as follows:

wherein the content of the first and second substances,

OEE is the equipment comprehensive efficiency of the monitored equipment;

i is the first workpiece produced by the monitored equipment in the analysis period;

l is the total number of workpieces produced by the monitored equipment during an analysis cycle;

p is the number of process steps to produce the first workpiece;

tlp is the time length of the pth process for producing the pth workpiece;

t is the planned production time;

and uploading the comprehensive efficiency and the analysis period of the equipment to the cloud for storage;

and downloading and displaying the comprehensive efficiency and the analysis period of the equipment on a terminal.

As another aspect of the present invention, a current ripple-based edge cloud collaborative device status monitoring system is provided, which includes an edge side, a cloud side and a terminal, wherein the cloud side includes a library modeling unit, the edge side includes a data acquisition unit, a data processing unit and a determination unit, the terminal includes a library selection unit and a display unit, and the edge side, the cloud side and the terminal are communicatively connected to each other, wherein:

the library modeling unit is configured to construct or update a characteristic current ripple library according to current data of sample equipment, the characteristic current ripple library includes an equipment preset state threshold and at least one characteristic current ripple template corresponding to a process, and obtaining the characteristic current ripple template at least includes: preprocessing the current of the sample equipment through an edge side to obtain a current ripple, determining a process section of the current ripple through a cloud end, and calculating the mode of sampling points in the current ripple of the process section;

the library selection unit is configured to select a characteristic current ripple library through a terminal according to the type of the monitored equipment, wherein the characteristic current ripple library is stored in a cloud database and downloaded to the edge side;

the data acquisition unit is configured to acquire current data of a power point of the monitored equipment in real time, and the acquisition frequency and amplitude unit selects parameter setting which is the same as the reference current ripple in the characteristic current ripple template;

the data processing unit is configured to preprocess the acquired current data to obtain current ripples, and the preprocessing method and the processing parameter setting are the same as those of the characteristic current ripple library;

the judging unit is configured to judge a device state or identify an abnormal event for the monitored device based on the current ripple according to the device preset state threshold and the characteristic current ripple library, where the device state includes: shutdown state, standby state, working state, abnormal state, the abnormal event includes: an abnormal startup event, an unplanned shutdown event, a minor shutdown event, a speed loss event, a processing abnormal event;

the display unit is configured to display the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data collected correspondingly.

Preferably, a plurality of sets of current data are acquired at the power point of at least one sample device through the data acquisition unit;

preprocessing the acquired current data to obtain current ripples through the data processing unit, wherein the preprocessing at least comprises filtering the current data;

the library modeling unit is further configured to:

determining equipment process sections according to different characteristics of the current ripple changing along with different processes; and the number of the first and second groups,

selecting characteristic points for the data of different process segments to align the characteristic points of the different process segments, wherein the characteristic points comprise extreme values; normalizing the data of at least two groups of same process segments, and aligning time based on the positions of the feature points; and, calculating a mode at each sampled data point in the current ripple; and the number of the first and second groups,

Preferably, the device preset state threshold and the characteristic current ripple library are issued to the determination unit on the edge side through a cloud, the device preset state threshold includes a shutdown current ripple threshold, a standby current ripple threshold and a correlation threshold, and in the determination unit:

Preferably, the edge side further includes a computing unit connected to the determining unit, and the computing unit is configured to, when determining whether the monitored device is in an operating state:

calculating Pearson correlation coefficient values of the current ripple of the monitored device and the reference current ripple of the characteristic current ripple template of various processes in the characteristic current ripple library; calculating the comprehensive efficiency of the equipment according to the production arrangement, the equipment state and the abnormal events;

the judging unit is further configured to judge a current process of the monitored device according to a Pearson correlation coefficient value of the current ripple of the monitored device and the reference current ripple of the characteristic current ripple template:

otherwise, the monitored equipment is not currently performing such a process.

As another aspect of the present invention, there is provided a distributed storage medium having a plurality of instructions stored therein, which is provided at a terminal and adapted to be loaded by a processor and to execute:

selecting a characteristic current ripple library according to the type of the monitored equipment, and displaying the relevant time, the relevant time of the abnormal event and the current data acquired correspondingly; the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to a process;

set up in the edge side, is suitable for being loaded and carried out by the processor:

preprocessing the current of the sample equipment to obtain current ripple when the characteristic current ripple template is obtained; and the number of the first and second groups,

acquiring current data of a power point of the monitored equipment in real time, wherein acquisition frequency and amplitude units select parameter settings which are the same as reference current ripples in the characteristic current ripple template; and the number of the first and second groups,

set up in the high in the clouds, be applicable to by the processor loading and carry out:

constructing or updating a characteristic current ripple library according to the current data of the sample equipment; and the number of the first and second groups,

preprocessing the current of the sample equipment to obtain current ripple when the characteristic current ripple template is obtained, determining a process section of the current ripple and calculating the mode of sampling points in the current ripple of the process section; and the number of the first and second groups,

and storing the characteristic current ripple library, the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data acquired correspondingly in a cloud database.

The invention has the beneficial effects that:

the edge cloud cooperative equipment state monitoring method and system based on the current ripple are based on the fact that current data of power points of monitored equipment are collected and preprocessed to obtain the current ripple, and real-time monitoring of the equipment state is achieved through cooperative processing of the edge side and the cloud side.

Drawings

Fig. 1 is a flowchart of a current ripple-based edge cloud collaborative device state monitoring method according to embodiment 1 of the present invention;

fig. 2 is a flowchart of constructing a characteristic current ripple library in the edge cloud collaborative device state monitoring method based on current ripple according to embodiment 1 of the present invention;

FIG. 3 is a schematic view showing current ripples generated when a workpiece is machined by the boring machine according to embodiment 1 of the present invention;

FIG. 4 is a schematic diagram of the current ripple of the process stages of FIG. 3;

FIG. 5 is a schematic diagram of selected characteristic points of the current ripple of the process stages of FIG. 4;

FIG. 6 is a schematic illustration of the mode calculation on the sampled data points of FIG. 2;

fig. 7 is a flowchart of device state determination in the edge cloud collaborative device state monitoring method based on current ripple in embodiment 1 of the present invention;

fig. 8 is a schematic view of a current ripple-based edge cloud cooperative device status monitoring system according to embodiment 2 of the present invention;

in the drawings, wherein:

1-equipment end; 10-a monitored device;

2-edge side; 20-a current collection processor; 21-a data acquisition unit; 22-a data processing unit; 23-a judgment unit; 24-a computing unit;

3-cloud end; 30-a cloud server; 31-library modeling unit; 32-a storage unit;

4-a terminal; 40-a smart device; 41-library selection unit; 42-display unit.

Detailed Description

In order to make those skilled in the art better understand the technical solution of the present invention, the following describes in detail a current ripple-based edge cloud cooperative device status monitoring method, a current ripple-based edge cloud cooperative device status monitoring system, and a distributed storage medium according to the present invention with reference to the accompanying drawings and the detailed description. It should be understood that the detailed description and specific examples, while indicating the preferred embodiment of the invention, are intended for purposes of illustration and explanation only and are not intended to limit the scope of the invention.

It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention. The technical features mentioned in the different embodiments of the invention described below can be combined with each other as long as they do not conflict with each other.

Example 1:

in view of the above problems, the present embodiment provides a side cloud collaborative device state monitoring method based on current ripple, and by using the monitoring method, state analysis such as real-time standby and process can be automatically performed on various devices, and comprehensive efficiency analysis of intelligent devices can be performed.

The term "current ripple" in this application refers to a characteristic waveform or characteristic data that is representative or typical based on a current waveform extracted from a particular node of the plant. Just like the 'human fingerprint' can be used for identity authentication, the 'current ripple' can also snoop the state of the equipment, and the monitoring of the state of the equipment is realized by collecting current waveforms. However, the "fingerprint of a person" enables authentication because two conditions are satisfied: one is that there are no two identical fingerprints and the other is, in principle, life-long invariant. For the equipment, researches show that when the same machining process is carried out, the current waveform change has the same operation trend, but because the equipment of the same type has different states, the waveform details under the same working condition can be slightly different, the targeted filtering treatment or the establishment of a database aiming at the equipment of the same type is needed. In addition, the power point refers to a position point which plays a decisive action or substantial driving role for the working condition of the equipment, for example, for a cold header, the power point refers to a motor in a cold header mechanism. The current data of the power point of the monitored equipment is collected in real time, so that the equipment information is prevented from being acquired from controllers such as a PLC, and the problem of data sources caused by the diversity or unknown limit of protocols of the controllers such as the PLC is avoided.

As shown in fig. 1, the method for monitoring the state of the edge cloud cooperative device based on current ripple includes the steps of:

step S1): selecting a characteristic current ripple library through a terminal according to the type of monitored equipment, wherein the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process, and the characteristic current ripple library is stored in a cloud database and is downloaded to an edge side;

step S2): the edge side collects the current data of the power point of the monitored equipment in real time, and the collection frequency and amplitude unit selects the parameter setting which is the same as the reference current ripple in the characteristic current ripple template; and the number of the first and second groups,

step S3): the edge side carries out pretreatment on the acquired current data to obtain current ripples, and the method adopted by the pretreatment and the method for setting the treatment parameters are the same as the method for constructing the characteristic current ripple library and the parameter setting; and the number of the first and second groups,

step S4): the edge side judges the equipment state or identifies abnormal events to the monitored equipment based on the current ripple according to the equipment preset state threshold and the characteristic current ripple library, and the equipment state comprises: shutdown state, standby state, operating state, abnormal event includes: an abnormal startup event, an unplanned shutdown event, a minor shutdown event, a speed loss event, a processing abnormal event;

step S5): uploading the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data correspondingly collected by the abnormal event to the cloud for storage;

step S6): and downloading and displaying the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data correspondingly acquired by the abnormal event at the terminal.

As shown in fig. 2, before selecting the characteristic current ripple library according to the type of the monitored device, the method further includes: step S0) builds or updates a library of characteristic current ripples from the current data of the sample device.

The characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process, and the characteristic current ripple template is used as a reference standard for identifying the working condition of the equipment. Obtaining the characteristic current ripple template at least comprises: the method comprises the steps of preprocessing the current of a sample device through an edge side to obtain a current ripple, determining a process section of the current ripple through a cloud end, and calculating the mode of sampling points in the current ripple of the process section. The method specifically comprises the following steps:

step S01) the edge side collects multiple sets of current data at the power point of at least one sample device.

The sample devices are the same type of devices to be monitored or devices having the same working principle. In this step, multiple sets of current data are collected at the power point of the sample device. The multiple sets of current data may be current data of devices of the same type, the same specification and the same mechanism, which operate in the same category and are different brands, such as current data of a cold heading machine of the same type or a thread rolling machine of the same type; or may be current data for the same device. The current waveform data of a plurality of operations performed under the same process condition is a 'group', for example, the current data of a plurality of workpieces processed by processing the same workpiece is collected.

Preferably, the current data have the same sampling rate and the same current unit to facilitate later data processing and alignment. Collecting multiple sets of current data may employ a current collection processor.

Step S02), preprocessing the collected current data to obtain current ripples, wherein the preprocessing at least comprises filtering the current data.

In this step, the purpose of the filtering process is to avoid the influence of noise on the position of the current ripple feature point, and the filtering algorithm and parameters need to be the same as those adopted when the device condition identification inference is finally performed. For example, a gaussian low-pass filter can be used for filtering, and the gaussian low-pass filter can quickly filter noise, and meanwhile, the response time is fast, so that real-time processing is facilitated.

The functions of the two steps can be integrated in an intelligent current acquisition processor, and the intelligent current acquisition processor is arranged on the edge side and is connected with sample equipment needing to acquire current.

Step S03) the cloud end determines the equipment process section according to the different characteristics of the current ripple along with the change of different processes.

In the step, the preprocessed current ripple is matched with the process, partition processing is carried out, and different current ripple segments obtained through division represent different process segments. The partition processing can adopt a machine learning mode, manual identification is intervened in the initial stage, and accurate identification and positioning of data are automatically realized along with the increase of the data quantity of collection and identification in the later stage.

Fig. 3 shows the current ripple when the boring machine performs a certain machining process, and it can be seen that the current ripple includes four distinct zoning features: (1) a stage of a gentle and protruding spike, (2) a stage of a sharp and steep peak, (3) a stage of a gentle plateau descending, and (4) a stage of ending a sharp and protruding spike. It was found that the phase of the current ripple feature is related to the process of the workpiece: (1) the method comprises the steps of (1) a process corresponding to a feed process, (2) a process corresponding to a boring process, (3) a process corresponding to a boring section process, and (4) a process corresponding to a tool withdrawal process. Therefore, through the partition processing of the current ripple, the corresponding technological process can be correspondingly obtained.

It should be understood here that when extracting the process current ripple, the process current ripple of the adjacent segment is properly included, that is, the process ripple of the upper and lower segments is selected more when the waveform is intercepted, so as to increase the accuracy of identification, and the waveform pattern is selected as shown in the four independent exploded views in fig. 4. However, the definition of the processing time length of the process segment should include and only include the current ripple length of the process segment, and not include the time of the two adjacent process segments selected for identification accuracy.

Of course, the whole current ripple of the whole process of processing a workpiece can be regarded as a process segment, so that although each process step in the process cannot be distinguished, the process formula can be identified according to the overall variation trend characteristic of the total current ripple, and which type of workpiece is currently processed can be identified, so that the method can be used for counting the number of the processed workpieces and analyzing the processing time length and the like.

Step S04), the cloud end selects characteristic points for the data of different process segments so as to align the characteristic points of the different process segments, wherein the characteristic points comprise extreme values.

In this step, feature points of different process segments are aligned to extract a target ripple shape for the process segment from a plurality of sets of current ripples of the same process.

For example, the maximum and minimum values of the current ripple are selected, as shown in fig. 5 by the characteristic point selection corresponding to the current ripple of each process segment of fig. 4. In fig. 4, the diamond points shown in the first, second, and fourth exploded views are the maximum values of the three process segments of the above example, i.e., the characteristic points of the three process segments. In the third decomposition diagram, since there is a gentle plateau, a point at the middle position of the data segment whose derivative result is 0 is selected as the feature point.

Step S05), the cloud end carries out normalization processing on the data of at least two groups of same process segments, and carries out time alignment based on the positions of the feature points.

In this step, after acquiring a plurality of sets of test data, in order to remove noise and other interference factors, a plurality of sets of previously acquired identical process current ripple segments for processing the same kind of workpiece are selected, and normalization processing is performed on the data of the plurality of sets of identical process segments, that is, the data of the identical current ripple process segments have the same data length, and time alignment is performed based on the positions of the feature points. The method comprises the steps of firstly respectively searching characteristic points on each current ripple process section and aligning the corresponding characteristic points, and then enabling the data lengths to be the same. After processing, the multiple sets of current ripple data of the same process segment are made to be the same in length and time aligned, so that further data analysis processing can be performed.

Step S06) the cloud calculates the mode at each sampled data point in the current ripple.

The mode refers to a numerical value with a remarkable concentration trend point on the statistical distribution, and in the step, the interference factors in the current ripple can be filtered out by selecting the mode, so that the target ripple shape of the process section is obtained. In this step, as shown in fig. 6, the sample current ripple is divided into five segments according to a certain percentage of the difference between the maximum value and the minimum value, the segment where the current ripple of all samples has the highest value is the segment where the mode appears, and the RMS arithmetic mean of the data in the segment is calculated to obtain the mode, that is, the value that the current ripple of the real-time current is expected to be close to at the position finally.

Step S07), the cloud determines a characteristic current ripple template corresponding to the process section according to the mode of the sampling points in the current ripple, and the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process.

Through step S06), the mode of each point is calculated for the current ripple data of the same process segment, and then the mode in the order of sampling points is used to obtain the characteristic current ripple template of the process segment, where the mode is the data with the largest occurrence frequency in a group of data, and is the original data in the group of data, but not the corresponding frequency, that is, the current ripple data set formed by all the modes is the characteristic current ripple template. In this step, each process segment is processed in the above manner to obtain a set of characteristic current ripple templates corresponding to typical processes, so as to form a characteristic current ripple library, and the characteristic library is used as a reference for identifying the working condition of subsequent equipment.

When the characteristic current ripple library is constructed, the functions of the five steps can be integrated in a cloud server to be completed, limited calculation resources of the edge side are not occupied, and the functions are downloaded to the edge side when scheduling requirements exist.

The preset state threshold value of the equipment can be used for judging the states of a shutdown state, a standby state and the like, and can be pre-stored in advance according to the state of the equipment, and the characteristic current ripple library and the preset state threshold value of the equipment of the same equipment are set in a set.

After the characteristic current ripple library of a certain device is established, the state of the device can be automatically identified and pre-warned, and the realization of the device state monitoring based on the current ripple will be described in detail according to the flow chart of fig. 1.

Step S1), selecting a characteristic current ripple library through a terminal according to the type of the monitored equipment, wherein the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process, and the characteristic current ripple library is stored in a cloud database and downloaded to the edge side.

In this step, a library of matching characteristic current ripples may be manually selected by the terminal. This matching process may be omitted when the current collection processor and the monitored device are not initially in use, and will not be described in detail herein.

Step S2), the edge side collects the current data of the power point of the monitored equipment in real time, and the collection frequency and the amplitude unit are selected to have the same parameter setting as the reference current ripple in the characteristic current ripple template.

In this step and the subsequent steps, configuration information such as the acquisition frequency and amplitude unit of the reference current ripple in the characteristic current ripple template, the method adopted by the preprocessing, the processing parameter setting and the like can be issued to the edge side through the cloud in advance.

In this step, after obtaining the cloud-based characteristic current ripple library, the intelligent current collection processor may be configured to automatically match the collection frequency and amplitude unit to select a parameter setting that is the same as the reference current ripple in the characteristic current ripple template.

Step S3), the edge side preprocesses the acquired current data to obtain current ripples, and the preprocessing method and the processing parameter setting are the same as the method and the parameter setting for constructing the characteristic current ripple library.

In this step, after the characteristic current ripple library of the cloud is obtained, the method and parameter setting for automatic matching of the intelligent current acquisition processor can be set, so that the method for selecting and constructing the characteristic current ripple library is the same as the parameter setting.

In this step, the current data is preprocessed to obtain a current ripple. The filtering method and the parameters of the filter adopted for the current data preprocessing need the same method and the same parameters for constructing the characteristic current ripple library sampling, and the subsequent steps are all based on the current ripple after the current data preprocessing. In the application process, once the characteristic current ripple library and the characteristic current ripple template thereof are selected, an automatic setting processing mode and processing parameters thereof can be selected in the current data acquisition process, or the processing mode and the processing parameters thereof can be manually set according to prompts, which is not described in detail herein.

Step S4), the edge side judges the device state or identifies the abnormal event to the monitored device based on the current ripple according to the device preset state threshold and the characteristic current ripple library, the device state includes: shutdown state, standby state, operating state, abnormal event includes: an abnormal start-up event, an unplanned shutdown event, a minor shutdown event, a speed loss event, a process exception event.

Typical device states include: shutdown state, standby state, working state, abnormal state. The method for identifying the equipment state comprises the step of comparing the current ripple value with an equipment preset state threshold value, wherein the equipment preset state threshold value can be issued to an edge side through a cloud end, and the equipment preset state threshold value comprises a shutdown current ripple threshold value Ioff, a standby current ripple threshold value Istandby and a correlation threshold value.

As shown in fig. 7, in this step, if the current ripple is smaller than the shutdown current ripple threshold, the monitored device status is determined as the shutdown status. And the current in the shutdown state is almost 0, the current ripple value of the point after the preprocessing is compared with a shutdown current ripple threshold Ioff, and if the current ripple value I is smaller than the shutdown current ripple threshold Ioff (i.e. I < Ioff), the equipment state is the shutdown state, and the identification is finished.

If the current ripple is smaller than the standby current ripple threshold and larger than the shutdown current ripple threshold, the current ripple is judged to be in the standby state by the state of the monitoring equipment. And comparing the current ripple value of the point after the pretreatment with a standby current ripple threshold Istandby, if the current ripple value I is smaller than the standby current ripple threshold Istandby but larger than a shutdown current ripple threshold Ioff (i.e. Ioff < I < Istandby), the equipment state is in a standby state, and the identification is finished.

And if the correlation coefficient of the current ripple and the reference current ripple in the characteristic current ripple template is greater than the correlation threshold, judging the state of the monitored equipment to be a normal state. And respectively selecting reference characteristic current ripples from the characteristic current ripple library, calculating a correlation coefficient value of the real-time current ripples, and if the correlation degree is high and is larger than a correlation degree threshold value a (for example, a is 0.9), determining that the current equipment is performing a corresponding process, and the equipment state is a normal state, and further identifying the working condition of the equipment, even performing operations such as workpiece counting and automatic work reporting.

If the current ripple does not belong to any one of the situations, considering whether the length of the current ripple is smaller than that of the reference current ripple, if so, indicating that the length of the acquired current data is not enough, and judging not yet, and waiting for data acquisition and preprocessing; and if the length of the current ripple is greater than or equal to the length of the reference current ripple and the current ripple of the matched process does not exist, the equipment state of the monitored equipment is an abnormal state.

Preferably, on the basis of identifying that the monitored equipment is in the working state, the method can further comprise identifying the current equipment working condition of the monitored equipment based on the characteristic current ripple template. The working condition of the equipment refers to the working state of the equipment under the condition directly related to the action of the equipment, namely the state of the equipment process. Judging the working condition of the equipment according to the current ripple template, and specifically comprising the following steps:

step S41) calculating Pearson correlation coefficient values of the current ripple of the current monitored device and the reference current ripples of the characteristic current ripple templates of various processes in the characteristic current ripple library.

Calculating Pearson correlation coefficient values of the current ripple X of the current monitored equipment and reference current ripples of current ripple templates Yk of various processes in the characteristic current ripple library, wherein the Pearson correlation coefficient value formula is as follows:

wherein the content of the first and second substances,

xi is the current ripple of the current monitored equipment, i is the serial number of the data point position, i is 1,2, … Nk, and the Xi represents the position of the current ripple and the past total Nk data points;

m is the number of a certain process characteristic current ripple template in the characteristic current ripple library.

Further, step S42) determines the current process of the monitored device according to the Pearson correlation coefficient values of the current ripple of the monitored device and the reference current ripple of the characteristic current ripple template.

In this step, if the Pearson correlation coefficient value of the current ripple of the monitored equipment and the reference current ripple of the characteristic current ripple template of a certain process section is between the preset state threshold value of the equipment and 1.0, the monitored equipment is currently carrying out the process; otherwise, the monitored equipment is not currently performing such a process.

The value of the Pearson correlation coefficient value calculated according to the formula (1) is usually between 0 and 1.0, the Pearson correlation coefficient value is close to 1, the similarity of the current ripple form and the current ripple form of the process participating in calculation in the characteristic current ripple library is high, and the process can be considered to be currently carried out; conversely, a Pearson correlation coefficient value near 0 indicates that the current ripple pattern is very different from the current ripple pattern of the process currently participating in the calculation, and that the ongoing process is not selected to participate in the calculation. For example, when the Pearson correlation coefficient value between the current ripple X of the currently monitored device and the reference current ripple in the current ripple template Yk of a certain process in the characteristic current ripple library is greater than a set threshold value R, for example, R is 0.9, it indicates that the process operation is currently performed.

It should be understood herein that the characteristic current ripple library includes current ripple characteristic current ripple templates of reference current ripples of various processes, where a process may be an entirety including a set of process steps, or may be one of the process steps, and multiple operations such as feeding, cutting, rotating, retracting, etc. performed to process a workpiece are a set of processing operations, which can be regarded as a process, and only one characteristic current ripple template including the entire process in the characteristic current ripple library corresponds to the characteristic current ripple template; each process step can be regarded as a process, feed-cutting-rotation-withdrawal are regarded as four independent processes, and four characteristic current ripple templates respectively corresponding to the characteristic current ripple templates are arranged in the characteristic current ripple library.

If the current ripple is not matched with all reference current ripples in the characteristic current ripple library, the current equipment can be judged to be in an abnormal state and the working condition of the equipment is abnormal because the equipment which is excluded from being in a standby state and a shutdown state before is.

It should be understood herein that the above-described process relating to the current ripple calculation and comparison may be referred to in the specific process of constructing the characteristic current ripple template process (including mode) and will not be described in detail herein.

Further, on the basis of the general equipment status, the step S43) may be further implemented to identify an abnormal event based on the equipment status according to the production schedule, the abnormal event including: an abnormal power-on start event, an unplanned shutdown event, a minor shutdown event, a speed loss event. The management personnel can set production arrangement in advance through the terminal and upload the production arrangement to the cloud end, before the abnormal event is identified based on the equipment state, the production arrangement can be issued to the edge side through the cloud end in advance, and the production arrangement generally comprises a planned shutdown period plan and the like.

Abnormal boot event: and if the current state of the monitored equipment is a non-shutdown state, such as a standby state or a working state, sending an alarm so as to confirm and coordinate problems through an upper management layer. One possible scenario at this time is: due to the smoothness of information communication, professional maintenance personnel overhaul, and other people do not know to illegally open equipment, so that safety risks exist in the equipment and personnel, and therefore, immediate measures are very necessary.

Unplanned shutdown event: if the time length exceeds the set shutdown time threshold value in the production process, a long-time shutdown event is indicated to occur, and the time is an unscheduled shutdown time period. The duration of the unplanned downtime period is counted, and the length of the planned downtime and the unplanned downtime period are summed into an Availability Loss (Availability Loss) time.

A small stop event: if the current time is in a shutdown state and the time period of the current time is a scheduled production time period, a shutdown event occurs, and if the shutdown time is less than a set shutdown threshold value in the production process, a small shutdown occurs. The small stops are usually less than 5mins, as defined by the plant specifications, and usually do not require the involvement of maintenance personnel. With regard to small stops, an undetected problem is that these stops are repetitive, sometimes with the same problem occurring repeatedly at different times, resulting in the operator seeing it small and unseen, rendering most companies unable to accurately track the small stops. Because the time is relatively short, it is not easy to count accurately in time, but the time is much less and not a little. In the embodiment, the basic data of the small stop can be counted in real time, so that performance analysis can be performed on operators, equipment, types of machined workpieces and the like.

Speed loss event: in the scheduled production time, if the time interval between two times of workpiece processing exceeds the normal workpiece circulation time, namely exceeds the normal process beat, the production efficiency degradation is caused, which may be caused by insufficient personnel operation proficiency, equipment performance degradation and the like, and the data can be used for the management layer to obtain the real-time yield evaluation, so that the production task delivery can be dynamically adjusted. The small landing event time and the speed Loss event time add up to a Performance Loss (Performance Loss) time.

Processing abnormal events: in the normal production process, if the identification process error occurs, the reason possibly caused is at least one of the error of a new processing process file, the abnormality of a workpiece and the abnormality of equipment, urgent intervention is needed, or else, waste products can be continuously generated. The process exception event time is a Quality Loss (Quality Loss) time.

Step S44) calculating the comprehensive efficiency of the equipment according to the production arrangement, the equipment state and the abnormal events.

Further, the comprehensive efficiency of the equipment can be calculated according to the production arrangement and the equipment state and abnormal events obtained by the analysis. The Equipment comprehensive efficiency (OEE) is essentially the ratio of the actual qualified output to the theoretical output in the load time, and the ratio of the actual production capacity of the Equipment to the theoretical output can be expressed by determining the percentage of the actual effective production time to the planned production time.

As mentioned above, the production schedule, i.e. the planned production time of the device within a certain time, can be set in advance by the terminal and stored in the terminal or the cloud. The actual effective production time can be obtained by judging the equipment state or identifying abnormal events of the monitored equipment based on the current ripple.

Specifically, if the complete process for each workpiece includes P process steps, 1,2, … P, respectively. And each step obtains the current ripple of the real-time process in the workpiece machining process, if the obtained current ripple has the same variation trend with the reference current ripple in the characteristic current ripple template and the Pearson correlation coefficient value shows high correlation, the complete machining of the whole workpiece is confirmed to be completed, the workpiece is counted into a qualified workpiece set X, and the X in the set of the qualified workpieces obtained in the analysis period contains L qualified workpieces. The time length of the characteristic current ripple corresponding to the process segment of each current ripple is Tlp, the Tlp can be extracted from the current ripple, and the ratio of the total processing time of qualified workpieces in the OEE analysis period to the total analysis period can be regarded as OEE. OEE is calculated as follows:

wherein the content of the first and second substances,

l is the total number of workpieces produced by the monitored equipment in the analysis period;

p is the number of process steps to produce the first workpiece;

tlp is the time length of the pth process for producing the pth workpiece;

and T is the planned production time.

Step S5), the device state and the related time thereof, the related time of the abnormal event and the current data correspondingly collected are uploaded to the cloud for storage.

Step S6), the device state and the relevant time thereof, the relevant time of the abnormal event and the current data correspondingly collected are downloaded and displayed on the terminal.

The terminal may be a display screen with an operable interface, a mobile phone or other intelligent devices, which is not limited herein.

According to the rule of the collected equipment current data, under the condition that the density of repeated information value information in the current data is low, the equipment state, abnormal events and the like are judged on the collected data at the edge side, and the equipment state and the relevant time are saved; when the abnormal event is judged to occur, the related time of the abnormal event and the current data correspondingly collected by the abnormal event are saved and uploaded to the cloud, so that valuable information is reserved, and the whole data volume is reduced; correspondingly, a current data storage and reference basis is arranged at the cloud end, so that a reference current ripple is provided as a judgment basis for monitoring the equipment state; furthermore, the abnormal positioning and checking of the equipment can be quickly realized, the health state and the working condition of the whole equipment can be evaluated, reference can be provided for the operation working condition of the similar equipment, and the comprehensive efficiency of the equipment can be calculated.

By way of example, the monitoring method will be described in detail below by taking an example of equipment state monitoring of the cold header, wherein functions of automatic counting of workpieces, real-time equipment working condition analysis, comprehensive equipment efficiency analysis and the like are described.

The cold header is special equipment which takes a pier as a main part and is specially used for producing fasteners such as nuts, bolts and the like in batches, and is widely applied to the processing of producing various fasteners such as automobiles, ships, machinery and the like. In the machining mechanism, the cold header is a forging forming method for thickening the top of a bar or a wire, a crank connecting rod and a slide block mechanism are driven by a motor to perform linear motion, and a male die and a female die are utilized to enable a blank of a machined part to be subjected to plastic deformation or separation so as to realize forging forming.

According to the regulations of a production line, the cold header works for A hours every day, wherein Bmins are required for working such as meeting early, preparation and inspection every day; during operation, Cmins is serviced during the morning and afternoon hours. However, in actual operation, the time lengths such as the inspection time and the maintenance time may have a slight deviation in the execution process, so that the actual load time may deviate from the theoretical load time. In the embodiment, a current signal is collected on an electric wire on the power supply side of a motor of the cold header through a current collection processor, filtering processing is carried out, and comparison is carried out according to a data collection point relative to a preset state threshold value and current ripple of the equipment. Therefore, the edge side can calculate the finished number of workpieces and the real-time working condition of equipment in real time and finally obtain the OEE_{Not tested}，OEE_{Not tested}And sending the data to a cloud end, and combining the reported real-time qualified rate (possibly related to other product quality detection results) by the cloud end to finally obtain an accurate real-time OEE value.

Equation (2) is modified as follows:

OEE time starting rate × performance starting rate × yield ═ OEE undetected × yield formula (3)

Wherein:

OEE_{not tested}: for OEE without quality inspection, the final OEE is obtained by correcting the qualified rate coefficient_{Not tested}Time starting rate is multiplied by performance starting rate;

because:

the time starting rate is equal to the equipment starting time (sixth)/load time (fifth);

performance starting rate (c) is equal to theoretical processing period (c) multiplied by processing quantity (c)/equipment starting time (c);

thus:

OEE_{not tested}(theoretical machining period of three x plus)Work quantity (r)/load time (v);

wherein:

loading time (calendar working time) ("calendar working time") ("planned downtime");

calendar working time (c): the time for starting production in the factory plan is taken as the working time A hours per day in the embodiment; planned downtime ((c)): in this embodiment, work such as morning meeting, preparation inspection and the like of work hours Bmins every day, and overhaul time of Cmins during the morning and afternoon in the working process are counted to (B +2C) mins for fixed downtime every day;

processing quantity (iv): through current ripple analysis, the current state is a normal working state, and the Pearson correlation coefficient value of the current ripple and the process segment of the reference current ripple in the characteristic current ripple template of the cold header is between a preset state threshold value of equipment and 1.0, which indicates that a matched machining process is obtained through identification, namely, one workpiece machining action is completed, the machining quantity of the workpieces is accumulated to be 1, and the total machining accumulated quantity of the workpieces in the analysis period is the final machining quantity of the workpieces;

theoretical machining cycle (c): for a cold heading machine, the time length of the reference current ripple in the cold heading machine characteristic current ripple template represents the duration of one complete cold heading action, namely the theoretical machining period.

Therefore, according to the machining quantity (r), the theoretical machining period (c), the calendar working time (c) and the planned downtime (c), the OEE can be obtained in real time_{Not tested}. In this embodiment, the machining number (r) can be obtained by accumulating the matching number of the current ripple and the characteristic current ripple template, and the theoretical machining period (c) is the time of the process section extracted from the reference current ripple in the characteristic current ripple template, the calendar working time (c) and the planned downtime (r) can be obtained by the production arrangement which is set by the terminal and stored in the cloud.

The qualified rate is as follows: the current qualified product quantity and the current total product quantity are related, and can be obtained through a quality inspection system of a factory, for example, a real-time qualified product rate can be directly obtained when an intelligent quality inspection system is adopted, and the real-time qualified product rate is uploaded to an OEE of a cloud end_{Not tested}X is qualifiedThe yield can obtain the real-time OEE.

In addition, various times can be accurately counted, such as: the actual planned downtime difference is the actual planned downtime-the planned downtime. When the cold heading machine is operated manually, the time deviation between the planned downtime and the actual execution downtime can be revealed through the actual planned downtime difference specified every day, and the time deviation can reflect the time loss of attendance checking of operators to a certain extent. According to the current ripple analysis, the downtime of Dmins is about 8 hours in the morning, and the analysis is the morning meeting and the preparation time; downtime of Emins and E' mins is respectively found at 11:00 am and 3:00 pm, and analyzed as planned rest time in the morning and planned rest time in the afternoon; at noon 12: and F, the downtime of Fmins is analyzed to be the noon break time from 00 to 13:00 in the afternoon, and the loss of the attendance time can be obtained by analyzing and comparing the time difference.

Unscheduled downtime: hmins for 9:30 AM and H 'mins for 17:30 PM, which were not within the scheduled downtime period, were identified as unscheduled downtime events, corresponding to unscheduled downtime (H + H') mins. Unplanned downtime is a major cause of low time-to-time drive rates.

Small landing event and speed loss event: there are numerous standby states and short down times during the process, and the cumulative time associated with these events is the primary cause of low performance start-up rates.

The current work reporting mode adopted for performance evaluation of operators mainly depends on a manual mode, and the efficiency is low. In addition, the deviation between the plan and the actual condition and the failure to obtain the fine processing time length in the traditional OEE manual calculation mode can cause the analysis deviation of the OEE, and meanwhile, the data is collected and analyzed in a manual mode, the result cannot be analyzed in real time, and the corrected strategy is often greatly behind the actual operation. When digital factory modification is carried out, a photoelectric sensor or a magnetic sensor is often required to be additionally arranged to acquire the current processing quantity. However, these methods are limited in installation location and require equipment downtime, which interferes with normal production. The traditional equipment management method cannot accurately analyze the actual load time, so that the OEE calculation generates deviation. In addition, the collection of the number of workpieces is often a troublesome problem, for example, the machine is often started in advance during the machining process, the workpiece material is fed later, and the number of actions directly acquired from the PLC of the machine is different from the actual number of machined workpieces. The above reasons may cause that the traditional method cannot obtain an accurate and real-time analysis result of the equipment efficiency.

Researches show that the motor current can correspondingly change along with the extrusion stress change of a workpiece in the cold heading process, and the current shows certain regular change. However, since the current change shows a fluctuation within a certain range due to the external electromagnetic noise interference and the workpiece material difference, there is a large error in the recognition of the machining process by a simple threshold judgment method. By adopting the side cloud cooperative type equipment state monitoring method based on the current ripple, the current can be collected in real time in a non-stop mode, and the equipment working condition and process can be analyzed in real time, so that a real-time and accurate equipment efficiency analysis result can be obtained, and the method is suitable for cold headers of different manufacturers.

The edge cloud cooperative equipment state monitoring method based on the current ripple is realized based on the framework: the edge side continuously processes and uploads current data including equipment states and relevant time thereof, relevant time of abnormal events and correspondingly acquired current data to a cloud end by acquiring current data of power points of monitored equipment in real time; the current ripple of the equipment can be identified through an artificial intelligence algorithm to obtain the current state or different working conditions of the equipment due to different current ripples in different processes. The cloud end performs continuous big data autonomous learning to obtain and correct a characteristic current ripple library judged by the equipment state or an equipment preset state threshold value, and sends the characteristic current ripple library or the equipment preset state threshold value to the edge side at any time, so that the equipment state identification strategy can be corrected at any time, multiple equipment states can be monitored for the same equipment, and comprehensive equipment state monitoring can be performed on the basis of a set of hardware platform; and the edge side performs new equipment state judgment and original current data capture by optimizing the judgment threshold, so that the final characteristic current ripple library or the equipment preset state threshold of the edge side is more and more accurate, and the equipment state monitoring accuracy is higher and higher.

Example 2:

as another aspect of the present invention, this embodiment provides a current ripple-based edge cloud collaborative device state monitoring system, which implements the current ripple-based edge cloud collaborative device state monitoring method of embodiment 1.

As shown in fig. 8, when monitoring the device status of the monitored device 10 of the device end 1, the edge cloud collaborative device status monitoring system based on current ripple includes an edge side 2, a cloud side 3 and a terminal 4, the cloud side 3 includes a library modeling unit 31, the edge side 2 includes a data acquisition unit 21, a data processing unit 22 and a judgment unit 23, the terminal 4 includes a library selection unit 41 and a display unit 42, the edge side 2, the cloud side 3 and the terminal 4 are connected to each other in a communication manner, wherein:

the library modeling unit 31 is configured to construct or update a characteristic current ripple library according to the current data of the sample device, where the characteristic current ripple library includes a device preset state threshold and at least one characteristic current ripple template corresponding to the process, and obtaining the characteristic current ripple template at least includes: preprocessing the current of the sample equipment through an edge side to obtain current ripple, determining a process section of the current ripple through a cloud end, and calculating the mode of sampling points in the current ripple of the process section;

a library selection unit 41 configured to select a characteristic current ripple library through the terminal 4 according to the type of the monitored device 10, wherein the characteristic current ripple library is stored in the database of the cloud 3 and downloaded to the edge side 2;

the data acquisition unit 21 is configured to acquire current data of a power point of the monitored equipment 10 in real time, and the acquisition frequency and amplitude unit selects parameter settings which are the same as the reference current ripple in the characteristic current ripple template;

the data processing unit 22 is configured to preprocess the acquired current data to obtain current ripples, and the preprocessing adopts the same method and processing parameter setting as those of the method and parameter setting for constructing the characteristic current ripple library;

the determining unit 23 is configured to determine a device state or identify an abnormal event for the monitored device 10 based on the current ripple according to a device preset state threshold and a characteristic current ripple library, where the device state includes: shutdown state, standby state, operating state, abnormal event includes: an abnormal startup event, an unplanned shutdown event, a minor shutdown event, a speed loss event, a processing abnormal event;

and the display unit 42 is configured to display the equipment state and the relevant time thereof, the relevant time of the abnormal event and the corresponding collected current data thereof.

The terminal 4 may be a smart device 40 such as a mobile phone, a computer, etc. with a display screen and an interactive interface. The cloud 3 may be a centralized or distributed cloud server 30, and has functions of computing, storing, library modeling, and the like.

In constructing a library of characteristic current ripples from current data of a sample device, the following operations are performed:

collecting a plurality of groups of current data at the power point of at least one sample device through a data collecting unit 21;

preprocessing the acquired current data to obtain current ripples through a data processing unit 22, wherein the preprocessing at least comprises filtering the current data;

the library modeling unit 31 is further configured to:

determining equipment process sections according to different characteristics of current ripple along with different process changes; and the number of the first and second groups,

selecting characteristic points for the data of different process segments to align the characteristic points of the different process segments, wherein the characteristic points comprise extreme values; and the number of the first and second groups,

normalizing the data of at least two groups of same process segments, and aligning time based on the positions of the characteristic points; and, calculating a mode at each sampled data point in the current ripple; and the number of the first and second groups,

and determining a characteristic current ripple template corresponding to the process section according to the mode of sampling points in the current ripple, wherein the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process, and the characteristic current ripple template is used as a reference standard for identifying the working condition of the equipment.

It is easily understood that the cloud 3 includes a storage unit 32, which stores various device information including, but not limited to, device states and their associated times, time associated with an abnormal event and its corresponding collected current data, and other necessary drivers. The format or manner of storing information may be designed or customized as required, and is not limited herein.

When judging the device state or recognizing an abnormal event for the monitored device 10 based on the current ripple according to the device preset state threshold and the characteristic current ripple library, the device preset state threshold and the characteristic current ripple library are issued to the judging unit 23 of the edge side 2 through the cloud 3, the device preset state threshold includes a shutdown current ripple threshold, a standby current ripple threshold and a correlation threshold, and in the judging unit 23:

if the current ripple is less than the shutdown current ripple threshold, determining that the monitored device 10 is in a shutdown state;

if the current ripple is less than the standby current ripple threshold and greater than the shutdown current ripple threshold, the state of the monitored device 10 is determined as the standby state;

if the correlation coefficient of the current ripple and the reference current ripple in the characteristic current ripple template is greater than the correlation threshold, the state of the monitored equipment 10 is judged to be a normal state;

if the present current ripple does not fall in any of the above-described situations, the monitored device 10 status is determined to be an abnormal status.

The edge side 2 further comprises a calculation unit 24 connected to the determination unit 23, the calculation unit 24 being configured to, when determining whether the monitored device 10 is in an operative state:

calculating Pearson correlation coefficient values of the current ripple of the current monitored equipment 10 and the reference current ripple of the characteristic current ripple template of various processes in the characteristic current ripple library; calculating the comprehensive efficiency of the equipment according to the production arrangement, the equipment state and the abnormal events;

the determining unit 23 further determines the current process of the monitored device 10 according to the Pearson correlation coefficient value of the current ripple of the monitored device 10 and the reference current ripple of the characteristic current ripple template:

if the Pearson correlation coefficient value of the current ripple of the monitored equipment 10 and the reference current ripple of the characteristic current ripple template of a certain process section is between the preset state threshold value of the equipment and 1.0, the monitored equipment 10 is currently carrying out the process;

otherwise, the monitored equipment 10 is not currently performing such a process.

The current collection processor 20 integrates current data collection, processing, judgment and calculation, can realize equipment state monitoring, equipment working condition identification and OEE calculation, carries out early warning based on the information, and reports the information to the cloud end 3 through network communication, so as to realize equipment information filing. When the equipment state changes, reporting parameters such as the current working condition, the current process, the current OEE and the like; when the abnormity is found, the abnormal event and the original current data or current ripple are reported at the same time.

In this embodiment, a current ripple-based edge cloud collaborative device state monitoring system is implemented based on the above architecture, wherein specific functions, setting manners or calculation manners of the functional modules refer to steps or process descriptions corresponding to embodiment 1, and detailed descriptions thereof are omitted here.

According to the edge cloud cooperative type equipment state monitoring system based on the current ripple, the data acquisition processor of the edge side 2 acquires current data of a power point of monitored equipment in real time, continuously processes the current data including the equipment state and the related time thereof, the related time of an abnormal event and the corresponding acquired current data and uploads the current data to the cloud side 3, the cloud side 3 performs continuous big data autonomous learning, obtains a characteristic current ripple library judged by correcting the equipment state or an equipment preset state threshold value and sends the characteristic current ripple library or the equipment preset state threshold value to the edge side 2 at any time, so that an equipment state identification strategy can be corrected at any time, the same equipment can be monitored in various equipment states, and therefore comprehensive equipment state monitoring is performed on the basis of a set of hardware platform; the edge side 2 performs new equipment state judgment and original current data capture by optimizing the judgment threshold, and along with the improvement of the accuracy of the characteristic current ripple library of the edge side 2 or the threshold of the preset state of the equipment, the accuracy of monitoring the equipment state can also be improved.

Example 3:

as another aspect of the present invention, the present embodiment provides a distributed storage medium, in which a plurality of instructions are stored, which can be set in different location spaces or areas according to different processing functions, including:

set up in the terminal, be applicable to by the loading of processor and carry out:

selecting a characteristic current ripple library according to the type of the monitored equipment, and displaying the relevant time, the relevant time of the abnormal event and the current data acquired correspondingly; the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to the process;

preprocessing the current of the sample equipment to obtain current ripple when a characteristic current ripple template is obtained; and the number of the first and second groups,

acquiring current data of a power point of monitored equipment in real time, wherein acquisition frequency and amplitude units are selected to have the same parameter setting as reference current ripples in a characteristic current ripple template; and the number of the first and second groups,

preprocessing the acquired current data to obtain current ripples, wherein the preprocessing method and the processing parameter setting are the same as the method and the parameter setting for constructing a characteristic current ripple library; and the number of the first and second groups,

according to the preset state threshold value and the characteristic current ripple library of the equipment, judging the equipment state or identifying abnormal events for the monitored equipment based on the current ripple, wherein the equipment state comprises the following steps: shutdown state, standby state, operating state, abnormal event includes: an abnormal startup event, an unplanned shutdown event, a minor shutdown event, a speed loss event, a processing abnormal event;

preprocessing the current of the sample equipment to obtain current ripple when a characteristic current ripple template is obtained, determining a process section of the current ripple and calculating the mode of sampling points in the current ripple of the process section; and storing the characteristic current ripple library, the equipment state and the relevant time thereof, the relevant time of the abnormal event and the current data correspondingly acquired in the database of the cloud.

The storage medium shown in this embodiment may be a hard disk or a storage unit of a control system, and a computer program (i.e., a program product) is stored on the storage medium, and when the computer program is executed by a processor, according to different settings and locations, the steps described in the foregoing method embodiments are implemented, for example, the computer program is set in a cloud and is adapted to be loaded and executed by the processor: the cloud organizes a plurality of levels of data layout including devices, event maps, and signature waveforms according to the index data set. The specific implementation of each step is not repeated here.

It should be noted that examples of the storage medium may also include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory, or other optical and magnetic storage media, which are not described in detail herein.

The distributed storage medium provided in this embodiment stores the implementation program of the edge cloud collaborative device state monitoring method based on current ripple provided in embodiment 1, and the edge cloud collaborative device state monitoring method based on current ripple may be distributed on the edge side, the cloud side and the terminal, and the operation and update of the control software logic on the edge side and the cloud side are implemented through a communication network and the like based on the current ripple of the device in cooperation with the current acquisition processor, so that the device state monitoring is performed comprehensively on the basis of a set of hardware platform.

It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The specific embodiments are specific examples of implementing the technical solutions of the present invention. Also, the term "comprises/comprising" when used herein refers to the presence of a feature, integer or component, but does not preclude the presence or addition of one or more other features, integers or components.

It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.

Claims

1. A side cloud cooperative type equipment state monitoring method based on current ripple is characterized by comprising the following steps:

selecting a characteristic current ripple library through a terminal according to the type of the monitored equipment, wherein the characteristic current ripple library is stored in a cloud database and is downloaded to the edge side;

preprocessing the acquired current data to obtain current ripples, wherein the preprocessing method and the processing parameter setting are the same as the method and the parameter setting for constructing the characteristic current ripple library; and the number of the first and second groups,

according to the preset state threshold value of the equipment and the characteristic current ripple library, judging the equipment state or identifying an abnormal event for the monitored equipment based on the current ripple, wherein the equipment state comprises the following steps: shutdown state, standby state, working state, abnormal state, the abnormal event includes: an abnormal startup event, an unplanned shutdown event, a minor shutdown event, a speed loss event, a processing abnormal event;

2. The current ripple-based edge cloud collaborative device state monitoring method according to claim 1, wherein constructing a library of characteristic current ripples from current data of sample devices comprises the steps of:

the cloud end determines an equipment process section according to different characteristics of the current ripple along with different process changes; and the number of the first and second groups,

3. The current ripple-based edge cloud collaborative device state monitoring method of claim 1, wherein the device preset state threshold and the characteristic current ripple library are issued to the edge side through the cloud, and the device preset state threshold comprises a shutdown current ripple threshold, a standby current ripple threshold and a correlation threshold; on the side of the edge, it is,

4. The current ripple-based edge cloud cooperative equipment state monitoring method according to claim 3, wherein when the monitored equipment is in an operating state, judging the equipment working condition of the monitored equipment comprises the following steps:

wherein the content of the first and second substances,

otherwise, the monitored equipment is not currently performing such a process.

5. The current ripple-based edge cloud collaborative device state monitoring method of claim 3, further comprising: calculating the comprehensive efficiency of the equipment according to the production arrangement, the equipment state and the abnormal events, wherein the formula for calculating the comprehensive efficiency of the equipment is as follows:

wherein the content of the first and second substances,

OEE is the equipment comprehensive efficiency of the monitored equipment;

p is the number of process steps to produce the first workpiece;

tlp is the time length of the pth process for producing the pth workpiece;

t is the planned production time;

6. The utility model provides a limit cloud collaborative device state monitoring system based on current ripple, its characterized in that includes edge side, high in the clouds and terminal, the high in the clouds includes storehouse modeling unit, the edge side includes data acquisition unit, data processing unit and judgement unit, the terminal includes storehouse selection unit and display element, the edge side the high in the clouds the terminal communication connection of each other, wherein:

7. The current ripple based edge cloud collaborative device state monitoring system of claim 6,

acquiring a plurality of groups of current data at a power point of at least one sample device through the data acquisition unit;

the library modeling unit is configured to:

8. The current ripple-based edge cloud collaborative device state monitoring system of claim 6, wherein a device preset state threshold and the characteristic current ripple library are issued to the determination unit at an edge side through a cloud, the device preset state threshold includes a shutdown current ripple threshold, a standby current ripple threshold and a correlation threshold, and in the determination unit:

9. The current ripple-based edge cloud collaborative device state monitoring system of claim 8, wherein the edge side further comprises a computing unit connected to the determining unit, the computing unit configured to, when determining whether the monitored device is in an operational state:

otherwise, the monitored equipment is not currently performing such a process.

10. A distributed storage medium having a plurality of instructions stored therein,

selecting a characteristic current ripple library according to the type of the monitored equipment, and displaying the equipment state and the related time thereof, the related time of an abnormal event and the current data correspondingly acquired; the characteristic current ripple library comprises an equipment preset state threshold and at least one characteristic current ripple template corresponding to a process;