CN113568383B - Non-invasive identification method suitable for machining center - Google Patents
Non-invasive identification method suitable for machining center Download PDFInfo
- Publication number
- CN113568383B CN113568383B CN202110732418.1A CN202110732418A CN113568383B CN 113568383 B CN113568383 B CN 113568383B CN 202110732418 A CN202110732418 A CN 202110732418A CN 113568383 B CN113568383 B CN 113568383B
- Authority
- CN
- China
- Prior art keywords
- data
- processing
- machining center
- clustering
- center
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
- G05B19/4184—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by fault tolerance, reliability of production system
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/31—From computer integrated manufacturing till monitoring
- G05B2219/31088—Network communication between supervisor and cell, machine group
-
- 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]
Abstract
The invention relates to a non-invasive identification method suitable for a machining center, which comprises the steps of S1, collecting power utilization data of the machining center in a non-invasive mode in real time, and clustering main data in the power utilization data to obtain power utilization rule cluster maps of m devices of the machining center in each time period; s2, respectively carrying out corresponding manual training on m devices according to corresponding clustering data in the clustering chart in the S1 to obtain a product type-processing flow-time sequence-electricity utilization data identification model of the ith device, i belongs to (0, m), S3, collecting electricity utilization data of the current processing center in real time, combining a production task workshop device scheduling scheme, and obtaining the product type, the product quantity and the current processing flow state of each processing device of the processing center by using the identification model obtained in the S2.
Description
Technical Field
The invention relates to the field of intelligent manufacturing, in particular to a non-invasive identification method suitable for a machining center.
Background
The main solution at present is to increase data acquisition points on processing equipment by deploying a wireless local area network, and to transmit data back to a monitoring center through a communication technology, or to interactively communicate with an upper computer through an existing remote communication control expansion interface of the existing processing equipment, so that the running data of the equipment per se is synchronized to the monitoring point of the processing center in real time through an intranet, and is displayed on a screen in real time.
No matter the data acquisition points are added on the processing equipment or the existing remote communication control expansion interface is utilized, the modes all need to carry out communication network layout in the processing center, so that the one-time investment cost is higher, and the application range is small.
Aiming at the mode of adding the collecting points on the equipment, the collecting points are easy to damage during processing operation and need to be supported by professional technicians, so that the maintenance cost is high; for a communication interface expansion link mode, the existing equipment group has a small quantity of possession, so that the types of data acquisition points introduced by the machining center equipment are more, the technical support dimension is higher, and the maintenance is not suitable.
Moreover, the aforementioned methods are actually perceived by the operator, which results in poor user operation experience, and such methods have the problem of data opacity.
Disclosure of Invention
In view of the foregoing problems, an object of the present invention is to provide a non-invasive identification method suitable for a machining center, which can monitor plant equipment in real time without adding data acquisition points and a communication network, effectively reduce investment cost, have a wide application range, provide transparent data, and provide a good user experience.
In order to realize the purpose, the technical scheme of the invention is as follows: a non-invasive identification method suitable for a machining center is characterized by comprising the following steps:
s1, collecting power consumption data of a machining center in real time in a non-invasive mode, and clustering main data in the power consumption data to obtain a power consumption law cluster diagram of m devices of the machining center in each time period;
s2, respectively carrying out corresponding manual training on m devices according to corresponding clustering data in the clustering diagram in the S1 to obtain a product type-processing flow-time sequence-electricity utilization data identification model of the ith device, wherein i belongs to (0, m);
and S3, acquiring the current power utilization data of the machining center in real time, combining a production task workshop equipment scheduling scheme, and obtaining the type and the number of the currently processed products of each machining equipment of the machining center and the current machining process state of the machining equipment by using the identification model obtained in the S2.
Further, the non-intrusive mode in S1 specifically refers to accessing a branch metering device of a factory from an access power line of each processing workshop of the processing center.
Further, the step S1 of collecting power consumption data of the machining center in a non-invasive manner in real time specifically means collecting power consumption data of equipment from branch metering equipment of a factory in real time.
Further, the main data in S1 includes current data and power data.
Further, the cluster map in S2 is a processing flow-power-time sequence cluster map of m pieces of equipment in the processing center obtained in S1.
Further, the step S3 specifically includes collecting branch power consumption data of each processing workshop of the current processing center in real time, clustering the power consumption data to obtain a power consumption data clustering map of different time periods, superposing the power consumption data of each processing equipment in corresponding time periods in combination with a scheduling scheme, and matching according to the identification model in the step S2 to obtain the type and quantity of the currently processed products of the processing equipment and the current processing flow state of the processing equipment.
Further, the clustering process in the method adopts a K-Mcans algorithm for processing.
Further, the artificial training is a deep learning training model DSSM.
Further, the input data of the deep learning training model DSSM includes product type, time sequence, and power consumption data.
Furthermore, the electricity consumption data further comprise ms-level voltage, harmonic data, s-level electric energy total attack, active power, power factors, a voltage phase sequence and a current phase sequence.
Compared with the prior art, the invention has the advantages that:
data are collected in a non-invasive mode, and a clustering algorithm and a neural network algorithm are combined, so that the real-time dynamic monitoring on the types and the quantities of products processed by each processing device in a processing workshop and a processing flow is realized efficiently and quickly, and the problems that in the past, data collection points or communication networks need to be additionally added, the input cost is too high, and the user experience feeling is not good are avoided.
Drawings
Fig. 1 is an overall structural diagram of a start time sequence data cluster and a processing time sequence data cluster obtained by the method of the present application by clustering.
Fig. 2 is a processing timing data cluster of the same processing tool at different processing stages.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
The non-invasive identification method applicable to the machining center comprises the following steps,
s1, collecting power consumption data of a machining center in a non-invasive mode in real time, and clustering main data in the power consumption data to obtain power consumption rule cluster maps of m devices of the machining center in each time period;
s2, respectively carrying out corresponding manual training on m devices according to corresponding clustering data in the clustering diagram in the S1 to obtain a product type-processing flow-time sequence-electricity utilization data identification model of the ith device, wherein i belongs to (0, m);
and S3, acquiring the current power utilization data of the processing center in real time, combining a production task workshop equipment scheduling scheme, and obtaining the type and the number of the currently processed products of each processing equipment of the processing center and the current processing flow state of the processing equipment by using the identification model obtained in the S2.
The non-intrusive mode in S1 refers to accessing branch metering equipment of a factory from access power lines of each processing workshop of the processing center, and the non-intrusive mode for acquiring the power consumption data of the processing center in real time in S1 specifically refers to acquiring the power consumption data of the equipment from the branch metering equipment of the factory in real time.
Because the branch metering equipment of the factory is necessary to be equipped for the existing live line of the current factory, the cost investment of additionally adding a data acquisition point or a communication network does not exist, and because the branch metering equipment of the factory is positioned at the inlet end of the line and is far away from the processing equipment, the branch metering equipment of the factory can be operated noninductively by an operator, the user experience is improved, and the problems of the existing method are overcome.
The method mainly adopts current data and power data to perform clustering processing to obtain a starting time sequence data cluster and a processing time sequence data cluster, as shown in fig. 1, wherein a curve 1 represents the starting time sequence data cluster of the equipment 1, a curve 2 represents the processing time sequence data cluster of the equipment 1, a curve 3 represents the starting time sequence data cluster of the equipment 2, a curve 4 represents the processing time sequence data cluster of the equipment 2, a curve 5 represents the starting time sequence data cluster of the equipment 3, and a curve 6 represents the processing time sequence data cluster of the equipment 3.
The cluster map in S2 is a processing flow-power consumption-time sequence cluster map, i.e. a processing time sequence data cluster, of m devices in the processing center obtained by the cluster method in S1, fig. 2 shows a time sequence cluster map corresponding to the same device in different processing time periods, where the processing time periods include such processing procedures as a drilling time period and a cutting time period, in other words, the processing time periods shown in fig. 2 are such timing charts (shown in a curve 8) as a drilling-time sequence data cluster (shown in a curve 7) and a cutting-time sequence data cluster (shown in a curve 8) of the same device in the drilling time period, for example, the specific processing procedures represented by the curves 7 and 8 are determined according to an actual workshop, and the description is only used here for exemplification, and the protection range is not limited.
It should be mentioned that the collected electricity data includes auxiliary data besides the main data, such as ms-level voltage, harmonic data, s-level total power, active power, power factor, voltage phase sequence, and current phase sequence, and the collected electricity data is synchronized to the monitoring center in real time after being collected, and the clustering process is obtained by collecting and processing the collected data of the monitoring center.
The S3 specifically comprises the steps of collecting branch power utilization data of each processing workshop of the current processing center in real time, clustering the power utilization data to obtain a power utilization data clustering chart in different time periods, superposing the power utilization data of each processing equipment in corresponding time periods by combining a scheduling scheme, and carrying out matching identification according to the identification model in the S2 to obtain the type, the number and the current processing flow state of the currently processed products of the processing equipment.
Specifically, for the processing time sequence data cluster of each processing device, which is sorted out through the clustering algorithm in the S1, the processing flow and the collected data are subjected to associated division in a manual training mode, and the product type is added into the input data through manual intervention, so that the product type-processing flow-time sequence-electricity utilization data model of the processing device is obtained through multiple times of training. It is apparent that the collected data herein includes timing and electricity usage data. In the application, the clustering algorithm is a K-Mmans algorithm, and the artificial training network adopts a deep learning training model DSSM.
In order to make the obtained model more accurate, obviously, the training operation needs to be performed for multiple times, so that the multidimensional matching is improved, and the recognition rate is increased.
After the identification model is obtained, collecting branch power utilization data of each workshop of the whole machining center in real time, processing the power utilization data by the same clustering method to obtain a power utilization data clustering graph of each time interval, combining a corresponding workshop equipment scheduling scheme combined with a production task, superposing the power utilization data of each machining equipment at the corresponding time interval according to time interval curve data in the identification model, and inputting the superposed data into the identification model, so that the final machining product type, product quantity and machining flow state of the current machining equipment are obtained through analysis and processing of the identification model.
According to the method and the device, data are collected in a non-invasive mode, a clustering algorithm and a neural network algorithm are combined, the real-time dynamic monitoring of the product types, the number and the processing flows of each processing device in a processing workshop is realized efficiently and quickly, and the problem that in the past, data collection points or communication networks need to be additionally added is avoided, so that the input cost is too high, and the user experience feeling is not good.
While embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims (6)
1. A non-invasive identification method suitable for a machining center is characterized in that:
s1, collecting power consumption data of a machining center in a non-invasive mode in real time, and clustering main data in the power consumption data to obtain power consumption rule cluster maps of m devices of the machining center in each time period; the non-invasive mode for acquiring the electricity consumption data of the processing center in real time refers to the step of acquiring the electricity consumption data of branch metering equipment of a factory, which is accessed into an electricity consumption access line of each processing workshop, wherein each processing workshop belongs to the processing center; the electricity utilization rule cluster diagram is a processing flow-electricity utilization-time sequence cluster diagram of m devices;
s2, respectively carrying out corresponding manual training on m devices according to corresponding clustering data in the clustering chart in the S1 to obtain a product type-processing flow-time sequence-electricity utilization data identification model of the ith device, wherein i belongs to (0, m);
s3, acquiring current power utilization data of the machining center in real time, combining a production task workshop equipment scheduling scheme, and obtaining the type and the number of products currently machined by each machining equipment of the machining center and the current machining process state of the machining equipment by using the identification model obtained in the S2; the process of obtaining the type and the number of the currently processed products of each processing device of the processing center and the current processing flow state of the processing device is as follows: the branch power utilization data of each processing workshop of the current processing center are collected in real time, the power utilization data are clustered to obtain power utilization data clustering maps in different time periods, power utilization data in corresponding time periods of each processing device are superposed in combination with a scheduling scheme, and matching is carried out according to the recognition model in S2, so that the type and the number of products currently processed by the processing device and the current processing flow state of the processing device are obtained.
2. The non-intrusive identification method for a machining center as defined in claim 1, wherein:
the main data in S1 includes current data and power data.
3. The non-invasive identification method applicable to a machining center according to claim 1, characterized in that:
the clustering process in the method adopts a K-Mcans algorithm for processing.
4. The non-invasive identification method applicable to a machining center according to claim 1, characterized in that:
the manual training is a deep learning training model DSSM.
5. The non-intrusive identification method for a machining center as defined in claim 4, wherein:
the input data of the deep learning training model DSSM comprises product type, time sequence and power consumption data.
6. The non-invasive identification method applicable to a machining center according to claim 1, characterized in that:
the electricity utilization data further comprises ms-level voltage, harmonic data, s-level electric energy total attack, active power, power factors, a voltage phase sequence and a current phase sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110732418.1A CN113568383B (en) | 2021-06-30 | 2021-06-30 | Non-invasive identification method suitable for machining center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110732418.1A CN113568383B (en) | 2021-06-30 | 2021-06-30 | Non-invasive identification method suitable for machining center |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113568383A CN113568383A (en) | 2021-10-29 |
| CN113568383B true CN113568383B (en) | 2022-11-04 |
Family
ID=78163122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110732418.1A Active CN113568383B (en) | 2021-06-30 | 2021-06-30 | Non-invasive identification method suitable for machining center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113568383B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109840691A (en) * | 2018-12-31 | 2019-06-04 | 天津求实智源科技有限公司 | Non-intrusion type subitem electricity estimation method based on deep neural network |
| CN111275069A (en) * | 2019-12-10 | 2020-06-12 | 深圳供电局有限公司 | Non-invasive load monitoring method |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8209062B2 (en) * | 2009-12-16 | 2012-06-26 | Robert Bosch Gmbh | Method for non-intrusive load monitoring using a hybrid systems state estimation approach |
| US10018997B2 (en) * | 2013-06-28 | 2018-07-10 | Fisher-Rosemount Systems, Inc. | Non-intrusive data analytics in a process control system |
| US11526827B2 (en) * | 2018-09-28 | 2022-12-13 | The Regents Of The University Of California | Non-intrusive workflow assessment (NIWA) for manufacturing optimization |
| CN110188826A (en) * | 2019-05-31 | 2019-08-30 | 湖南科技大学 | Household electrical appliance operating status non-invasive inspection methods based on intelligent electric meter data |
| US11782401B2 (en) * | 2019-08-02 | 2023-10-10 | Aspentech Corporation | Apparatus and methods to build deep learning controller using non-invasive closed loop exploration |
| CN112613542B (en) * | 2020-12-14 | 2024-01-12 | 国网甘肃省电力公司营销服务中心 | Bidirectional LSTM-based load identification method for enterprise decontamination equipment |
| CN112924779A (en) * | 2021-01-25 | 2021-06-08 | 河北科技大学 | Non-invasive load identification method and device and terminal equipment |
-
2021
- 2021-06-30 CN CN202110732418.1A patent/CN113568383B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109840691A (en) * | 2018-12-31 | 2019-06-04 | 天津求实智源科技有限公司 | Non-intrusion type subitem electricity estimation method based on deep neural network |
| CN111275069A (en) * | 2019-12-10 | 2020-06-12 | 深圳供电局有限公司 | Non-invasive load monitoring method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113568383A (en) | 2021-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104112242B (en) | Power system 10kV busbar voltage yield analysis system and methods | |
| CN103177404B (en) | Can data analysis system based on using of data mining | |
| CN105809260A (en) | Method for intelligently generating monitoring information point list of dispatching automation system | |
| CN109185917B (en) | Boiler combustion state online diagnosis method and system based on flame intensity signal | |
| CN115908046A (en) | Visual power distribution system based on airport terminal building BIM | |
| CN101598927B (en) | Control system of soda carbonization technique based on neural network and control method thereof | |
| CN113568383B (en) | Non-invasive identification method suitable for machining center | |
| CN213754105U (en) | Photovoltaic power plant intelligent management system based on big data | |
| CN112215410B (en) | Power load prediction method based on improved deep learning | |
| CN111445141B (en) | Load distribution method, system and device of heat supply unit | |
| CN205844432U (en) | Intelligent transformer based on VXworks monitoring and micro-water integrated processing system | |
| CN205317843U (en) | Electric information synchronization collection system of non -intruding multichannel | |
| CN115983548A (en) | Power distribution network intelligent planning method based on artificial intelligence technology | |
| CN115102164A (en) | Multi-unit collaborative optimization load distribution method, system, equipment and storage medium | |
| CN115037037A (en) | Wisdom garden electric energy monitoring management system | |
| CN103400215A (en) | Method associating PMU measurement equipment with electrical-network operational data equipment | |
| CN103953502A (en) | Data collection method and data collection system of wind generating set | |
| CN114741447A (en) | Distributed energy station data processing method and device | |
| CN210742736U (en) | Integrated information platform for power plant | |
| CN111882284A (en) | Operation monitoring method and system for distributed power supply grid connection | |
| CN108255142B (en) | Quality of production control method and device | |
| CN116310258B (en) | Three-dimensional construction management system and method based on transformer substation live-action | |
| CN109193949A (en) | A kind of photovoltaic plant monitors cloud platform and system automatically | |
| CN115051714A (en) | Waveform data compression method based on FFT (fast Fourier transform) | |
| CN203250205U (en) | Manufacturing equipment acquisition interacted system |
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 |