CN107066683B - Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics - Google Patents

Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics Download PDF

Info

Publication number
CN107066683B
CN107066683B CN201710080893.9A CN201710080893A CN107066683B CN 107066683 B CN107066683 B CN 107066683B CN 201710080893 A CN201710080893 A CN 201710080893A CN 107066683 B CN107066683 B CN 107066683B
Authority
CN
China
Prior art keywords
power
time
machine tool
working
arithmetic mean
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
Application number
CN201710080893.9A
Other languages
Chinese (zh)
Other versions
CN107066683A (en
Inventor
林文文
许飞
林海
易文凯
陈巍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Mouse Technology Co.,Ltd.
Original Assignee
Nanjing Sichen Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing Sichen Information Technology Co Ltd filed Critical Nanjing Sichen Information Technology Co Ltd
Priority to CN201710080893.9A priority Critical patent/CN107066683B/en
Publication of CN107066683A publication Critical patent/CN107066683A/en
Application granted granted Critical
Publication of CN107066683B publication Critical patent/CN107066683B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/30Circuit design
    • G06F30/36Circuit design at the analogue level
    • G06F30/367Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/06Power analysis or power optimisation

Abstract

The invention discloses a numerical control machine tool reliability model modeling method based on energy consumption characteristics, which comprises the following steps: collecting power data and calculating a power difference sequence; dividing the power difference sequence to obtain power segments; extracting a feature library; the first character string represents a machine state transition path; selecting power data from the power segments, training a CART decision tree based on a feature library, and classifying the power segments by using the decision tree; arranging the classified power segments according to the acquisition time sequence, and representing the power segments by using a second character string; matching the first character string with the second character string to obtain a power sequence, and calculating the duration of the processing period; setting a sliding time window and a threshold value, and counting and calculating a processing period, working time and fault time; and (5) establishing a model. The invention also discloses a numerical control machine reliability model modeling system based on the energy consumption characteristics. The method has higher accuracy, and can avoid the problems of manual error, untimely data, time consumption and labor consumption.

Description

Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics
Technical Field
The invention relates to a reliability model modeling method and a system, in particular to a numerical control machine reliability model modeling method and a system based on energy consumption characteristics.
Background
The reliability model of the numerically controlled machine tool is a probability density function for describing a machining cycle and working and failure times of the numerically controlled machine tool. The reliability model of the numerical control machine is the basis for the management, production line performance evaluation and performance optimization of the numerical control machine. Generally, before the machine tool leaves the factory, the factory will give the reference parameters of the machine tool in the production specification. For the user, the staff gives the nominal values of the parameters by recording and the like in the earlier stage. However, since machine maintenance personnel may keep track of adjustments to the numerically controlled machine tool, these adjustments are likely to affect the parameters. Over time, these variations accumulate over time and eventually lead to a situation where the nominal value does not match the actual value. Therefore, they need to be carefully measured to ensure the accuracy of the reliability model. At present, the measurement of the parameters is mostly completed by manual regular statistics, and the manual regular statistics method causes the problems of untimely data feedback, easy error, human resource consumption and the like.
Disclosure of Invention
In view of the defects of the prior art, the invention aims to provide a numerical control machine reliability model modeling method based on energy consumption characteristics. Another object of the present invention is to provide a numerical control machine reliability model modeling system based on energy consumption characteristics.
The technical scheme of the invention is as follows: a numerical control machine reliability model modeling method based on energy consumption characteristics comprises the following steps,
s01, collecting power data of the machine tool, and calculating a power difference sequence;
s02, dividing the power difference sequence to obtain a power segment based on the machine tool state;
s03, extracting a group of feature libraries capable of representing the power characteristics of each power segment, wherein the feature libraries comprise a plurality of features which are related to different running states of the machine tool and change along with the change of the states;
s04, representing a machine tool state transition path corresponding to one process by using a first character string;
s05, selecting power data from all power segments according to a proportion to carry out state identification, training a CART decision tree based on the feature library, and classifying the power segments by the trained decision tree;
s06, arranging the power fragments classified in the step S05 according to the collection time sequence and representing the power fragments by a second character string;
s07, matching the first character string and the second character string by adopting a Knuth-Morris-Pratt algorithm to obtain a power sequence during processing in the power data, and subtracting the initial time from the end time of the power sequence during processing to obtain the processing period duration;
s08, setting a sliding time window, counting the number of processing periods in the time window and the time lengths of all the processing periods, and calculating the arithmetic mean of the time lengths of all the processing periods in the time window as the processing period of the current working day; moving the sliding time window backwards by one working day to obtain the processing period of the next working day;
s09, setting a threshold, considering the working state that the power data of the machine tool at the working time is greater than the threshold, considering the fault state that the power data of the machine tool at the working time is less than the threshold, counting the continuous working time and the number thereof in the sliding time window and calculating the arithmetic mean, and counting the fault time and the number thereof and calculating the arithmetic mean; moving the sliding time window backwards for one working day to obtain the working time arithmetic mean and the fault time arithmetic mean of the next working day;
and S10, taking the working time arithmetic mean and the fault time arithmetic mean of the working cycles of different working days obtained in the step S08 and the working time arithmetic mean and the fault time arithmetic mean of different working days obtained in the step S09 as the reliability model parameters of the machine tool.
Further, the machine tool states are divided into four states of standby, starting, no-load and cutting.
Further, the calculating power difference sequence is represented by P ═ PiWhere i ═ 1,2, …, n } represents power data, n represents the number of power collection points, PiRepresenting the ith power acquisition point, and D ═ D is used for the power difference sequenceiI is 1,2, …, n, Di=Pi+1-Pi
Further, when the power difference sequence is divided in step S02, a power data critical point when the machine tool state transitions is determined according to an empirical bayesian threshold method, where the power data critical point is a dividing point of the power difference sequence.
Further, the features in step S03 include maximum, minimum, mean absolute deviation, geometric mean, and third-order autoregressive coefficients.
Further, the maximum value calculation formula is pmax=max(piI is 1,2, …, m), the minimum value calculation formula is pmin=min(piI is 1,2, …, m), and the average absolute deviation is calculated as pmad=mediani(|pi-medianj(pj) |) the geometric mean calculation formula isThe third-order autoregressive coefficient is calculated by the formulam is the number of power data acquisitions in a power segment, piDenotes the ith power acquisition point, y is the geometric mean of the power segment, ω is the intercept, βtIs the coefficient of the regression, and is,is a noise parameter.
A numerical control machine tool reliability model modeling system based on energy consumption characteristics comprises a field data acquisition module: collecting machine tool power data, packaging and uploading to a server; the knowledge base module is used for establishing a characteristic base of the power characteristics of the machine tool in the operation process and a machine tool state transfer path of a process; the analysis module is used for carrying out power difference sequence segmentation, classification of power segments and calculation of machine tool reliability model parameters; and the setting module is used for setting a sliding time window and a threshold value.
Furthermore, the field data acquisition module comprises a plurality of intelligent electric meters to form an RS-485 network.
The technical scheme provided by the invention has the advantages that the automatic calculation of the reliability model parameters of the machine tool by utilizing the intelligent ammeter to acquire the power data of the machine tool is higher in accuracy compared with manual regular statistical processing, and the problems of manual error, untimely data, time consumption and labor consumption can be avoided.
Drawings
FIG. 1 is a diagram of a numerical control machine reliability model modeling system based on energy consumption characteristics.
FIG. 2 is a flow chart of a numerical control machine reliability model modeling method based on energy consumption characteristics.
FIG. 3 is a graph showing the variation of power with time in a certain process of the NC machine tool.
FIG. 4 shows the status recognition result of the turning process of the NC machine tool according to the embodiment.
Detailed Description
The present invention is further illustrated by the following examples, which are not to be construed as limiting the invention thereto.
Referring to fig. 1, the system for modeling the reliability model of the numerical control machine based on the energy consumption characteristics includes: the field data acquisition module: each numerical control machine tool is provided with an intelligent electric meter, power data are collected at a certain frequency, the power data are collected in a data collector and then packed and uploaded to a server, and the intelligent electric meter and the data collector are connected through an RS-485 bus; the knowledge base module is used for establishing a characteristic base of the power characteristics of the machine tool in the operation process and a machine tool state transfer path of a process; the analysis module is used for carrying out power difference sequence segmentation, classification of power segments and calculation of machine tool reliability model parameters; and the setting module is used for setting a sliding time window and a threshold value. The data collector and the server can adopt three modes of wired networking, wireless networking and GPRS networking. And the server is in wired connection with the analysis module, the knowledge base module and the setting module.
Referring to fig. 2, taking a numerical control lathe as an example, the modeling method of the numerical control machine reliability model based on the energy consumption characteristics is as follows:
s01, the field data acquisition module acquires power data of the machine tool and uploads the power data to the server, and the analysis module extracts the power data from the server to calculate a power difference sequence: with P ═ PiWhere i is 1,2, …, n represents power data, and n represents workNumber of rate acquisition points, PiRepresenting the ith power acquisition point, and D ═ D is used for the power difference sequenceiI is 1,2, …, n, Di=Pi+1-Pi
S02, determining a power data critical point when the machine tool state is transferred by an Empirical Bayes threshold method proposed by a document of ' Empirical Bayes Selection of wavelet thresholds ' annals of Statistics ' (2005) as an observation value, and dividing to obtain a power segment based on the machine tool state.
And S03, establishing a power characteristic library of the machine tool operation process by the knowledge base module. According to the document "Multi-objective machining-based Optimization for reducing carbon emission and Optimization in engineering Optimization" (2015), the numerical control machine tool can be divided into four states of standby, starting, idle load and cutting. The machine tool can be divided into states of standby, starting, idle load, cutting and the like. According to the characteristics of the power of the machine tool in the operation process, a plurality of characteristics which are related to different operation states of the machine tool and change along with the change of the states are extracted to form a power characteristic library. For each power segment, extracting a group of feature libraries capable of characterizing the power characteristics of the power segment, including a maximum value (p)max) Minimum value (p)min) Mean absolute deviation (p)mad) Geometric mean (p)gm) Third order autoregressive coefficient (β)123) And seven features. M acquisition points in the power segment, and p is used for each acquisition pointi(i ═ 1,2, …, m), the calculation formula of the characteristics is as follows:
pmax=max(pi,i=1,2,…,m)
pmin=min(pi,i=1,2,…,m)
pmad=mediani(|pi-medianj(pj)|)
m is the number of power data acquisitions in a power segment, piDenotes the ith power acquisition point, y is the geometric mean of the power segment, ω is the intercept, βtIs the coefficient of the regression, and is,is a noise parameter.
S04, the state transition path is obtained by the processing procedure of the steps. Fig. 3 shows a curve of the power over time for one step of the machine tool, the state transition path of which is standby- > start- > idle- > cut- > idle. The four processing states are indicated by one character, respectively, as shown in table 1. The state transition path in fig. 3 can be denoted as a first string R: "ITACA"
TABLE 1 processing status and character correspondence table
Status of state Character(s)
Standby I
Starting up T
No load A
Cutting of C
S05, selecting a certain proportion of all power segments to carry out state identification, training a CART decision tree, and using the trained decision tree for classification of the power segments. Fig. 4 shows the result of the state recognition of a certain turning process.
S06, arranging the classified power segments according to the acquisition time sequence, and using a second character string CpAnd marking.
S07, matching character strings R and C by using Knuth-Morris-Pratt algorithmp,CpThe matched substring in the sequence table is the power sequence in processing. The end time of the power sequence during machining is subtracted by the start time to obtain the machining period.
And S08 and S09, a setting module, a sliding time window length of relevant parameters of the reliability model of the numerical control machine tool and a threshold value a are set. And extracting a sliding time window from the setting module, starting from the starting time to the ending time of the time window, if the sliding time window is matched with the ending time, adding 1 to the counter value, and simultaneously recording the duration of the substring, namely a processing period. Thus, the number of processing cycles (in n) in the time window can be obtained1Expressed), and all machining cycle durations τi(i=1,2,…,n1). And calculating the arithmetic mean of all the processing periods in the time window, and taking the arithmetic mean as the processing period of the current working day.
Similarly, the sliding time window is moved backward by one working day, and the processing cycle of the next working day can be obtained.
Calculating the working time and the repair time: a threshold value a is given, and when the power data at a certain normal working moment is greater than a, the machine tool is considered to be in a working state; and when the power data is less than a, the machine tool is considered to be in a fault state. A sliding time window is given, and the continuous working time t in the time window is countedup,iAnd a number nupAlso the time of failure tdown,iAnd number ndown. An arithmetic mean of the operating time and the fault time within the time window is calculated. Similarly, the sliding time window is moved backward by one working day, and t of the next working day can be obtainedup,tdown
S10, the reliability model related parameters of the numerical control machine tool obtained through the steps are shown in the table 2. As can be seen from the table, the machining cycle remains substantially constant in a small time dimension, but there is some fluctuation in the mean on-time and mean off-time. Therefore, the modeling method for the reliability model of the numerical control machine tool of the industrial workshop, which is described by the invention, has significance for the performance evaluation of the production line of the industrial workshop by continuously monitoring and analyzing relevant parameters of the model.
TABLE 2 relevant parameters of reliability model of numerical control machine
Related parameter Day 1 Day 2 Day 3 Day 4 Day 5
τ(s) 72 73 72 72 72
tup(h) 38.4 39.1 40.5 39.7 41.3
tdown(h) 4.8 4.3 4.9 5.2 4.8

Claims (5)

1. A numerical control machine tool reliability model modeling method based on energy consumption characteristics is characterized in that: comprises the following steps of (a) carrying out,
s01, collecting power data of the machine tool, and calculating a power difference sequence;
s02, dividing the power difference sequence to obtain a power segment based on the machine tool state;
s03, extracting a group of feature libraries capable of representing the power characteristics of each power segment, wherein the feature libraries comprise a plurality of features which are related to different running states of the machine tool and change along with the change of the states;
s04, representing a machine tool state transition path corresponding to one process by using a first character string;
s05, selecting power data from all power segments according to a proportion to carry out state identification, training a CART decision tree based on the feature library, and classifying the power segments by the trained decision tree;
s06, arranging the power fragments classified in the step S05 according to the collection time sequence and representing the power fragments by a second character string;
s07, matching the first character string and the second character string by adopting a Knuth-Morris-Pratt algorithm to obtain a power sequence during processing in the power data, and subtracting the initial time from the end time of the power sequence during processing to obtain the processing period duration;
s08, setting a sliding time window, counting the number of processing periods in the time window and the time lengths of all the processing periods, and calculating the arithmetic mean of the time lengths of all the processing periods in the time window as the processing period of the current working day; moving the sliding time window backwards by one working day to obtain the processing period of the next working day;
s09, setting a threshold, considering the working state that the power data of the machine tool at the working time is greater than the threshold, considering the fault state that the power data of the machine tool at the working time is less than the threshold, counting the continuous working time and the number thereof in the sliding time window and calculating the arithmetic mean, and counting the fault time and the number thereof and calculating the arithmetic mean; moving the sliding time window backwards for one working day to obtain the working time arithmetic mean and the fault time arithmetic mean of the next working day;
and S10, taking the working time arithmetic mean and the fault time arithmetic mean of the working cycles of different working days obtained in the step S08 and the working time arithmetic mean and the fault time arithmetic mean of different working days obtained in the step S09 as the reliability model parameters of the machine tool.
2. The method for modeling the reliability model of a numerical control machine tool based on energy consumption characteristics according to claim 1, wherein the machine tool states are divided into four states of standby, start-up, idle and cutting.
3. The method according to claim 1, wherein the sequence of differential power calculations is represented by P ═ PiWhere i ═ 1,2, …, n } represents power data, n represents the number of power collection points, PiRepresenting the ith power acquisition point, and D ═ D is used for the power difference sequenceiI is 1,2, …, n, Di=Pi+1-Pi
4. The method as claimed in claim 1, wherein when the power difference sequence is divided in step S02, a critical point of power data is determined according to an empirical bayesian threshold method when the machine tool is in a state transition, and the critical point of power data is the divided point of the power difference sequence.
5. The numerical control machine reliability model modeling method based on energy consumption characteristics according to claim 1, characterized in that said characteristics in said step S03 include maximum value, minimum value, mean absolute deviation, geometric mean and third-order autoregressive coefficients.
CN201710080893.9A 2017-02-15 2017-02-15 Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics Active CN107066683B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710080893.9A CN107066683B (en) 2017-02-15 2017-02-15 Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710080893.9A CN107066683B (en) 2017-02-15 2017-02-15 Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics

Publications (2)

Publication Number Publication Date
CN107066683A CN107066683A (en) 2017-08-18
CN107066683B true CN107066683B (en) 2020-06-16

Family

ID=59599277

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710080893.9A Active CN107066683B (en) 2017-02-15 2017-02-15 Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics

Country Status (1)

Country Link
CN (1) CN107066683B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103235555A (en) * 2013-04-18 2013-08-07 重庆大学 On-line machine tool equipment utilization state monitoring method and device based on power information
KR101329996B1 (en) * 2012-08-08 2013-11-15 한밭대학교 산학협력단 Manufacturing information extraction method and equipment, machine tool monitoring method and system using the same
CN104794492A (en) * 2015-04-27 2015-07-22 重庆大学 Online machine tool equipment machining and running state recognizing method based on power feature models
CN104808587A (en) * 2015-04-27 2015-07-29 重庆大学 Utilization statistical approach based on operation states of machining equipment

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101329996B1 (en) * 2012-08-08 2013-11-15 한밭대학교 산학협력단 Manufacturing information extraction method and equipment, machine tool monitoring method and system using the same
CN103235555A (en) * 2013-04-18 2013-08-07 重庆大学 On-line machine tool equipment utilization state monitoring method and device based on power information
CN104794492A (en) * 2015-04-27 2015-07-22 重庆大学 Online machine tool equipment machining and running state recognizing method based on power feature models
CN104808587A (en) * 2015-04-27 2015-07-29 重庆大学 Utilization statistical approach based on operation states of machining equipment

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
"ELECTRE I Decision Model of Reliability Design Scheme for Computer Numerical Control Machine";Jihong Pang et.al;《JOURNAL OF SOFTWARE》;20110531;第6卷(第5期);第894-900页 *
"一种新的信号控制干道行程时间实时估计模型";张勇 等;《自动化学报》;20090930;第35卷(第9期);第1151-1158页 *
"数控机床可靠性概述";张根保 等;《制造技术与机床》;20141231(第7期);第8-14页 *
"数控机床智能化状态监测与故障诊断系统";赵敏;《中国优秀硕士学位论文全文数据库工程科技辑》;20120415(第4期);全文 *
"面向重型数控机床的服役可靠性评估方法及增长技术研究";熊尧;《中国博士学位论文全文数据库工程科技辑》;20131015(第10期);全文 *

Also Published As

Publication number Publication date
CN107066683A (en) 2017-08-18

Similar Documents

Publication Publication Date Title
CN105631596B (en) Equipment fault diagnosis method based on multi-dimensional piecewise fitting
JP6129028B2 (en) Energy consumption prediction method for building power equipment
CN109396953B (en) Machine tool working state intelligent identification system based on signal fusion
CN105574615B (en) wavelet-BP neural network wind power prediction method based on spatial correlation and GA
CN108107838A (en) A kind of numerical control equipment tool wear monitoring method based on cloud knowledge base and machine learning
CN104808587A (en) Utilization statistical approach based on operation states of machining equipment
CN109396576B (en) Deep learning-based electric spark machining stability and energy consumption state optimization decision system and decision method
CN107292502B (en) Power distribution network reliability assessment method
CN110594107B (en) Wind turbine generator fault detection method and device based on rapid gradient elevator
CN109685314B (en) Non-intrusive load decomposition method and system based on long-term and short-term memory network
CN107516135A (en) A kind of automation monitoring learning method for supporting multi-source data
CN111275288A (en) XGboost-based multi-dimensional data anomaly detection method and device
CN107066683B (en) Numerical control machine tool reliability model modeling method and system based on energy consumption characteristics
CN109647899B (en) Method for forecasting power consumption of multi-specification rolled pieces in hot rolling and finish rolling process of strip steel
CN111475921A (en) Tool residual life prediction method based on edge calculation and L STM network
CN111047082A (en) Early warning method and device for equipment, storage medium and electronic device
CN104408525B (en) The quantitative evaluation and control method of solving job shop scheduling problem risk
CN108305002A (en) A kind of decreasing failure rate height recognition methods based on monitoring Weibull form parameter
CN110109431B (en) Intelligent acquiring system for OEE information of die casting machine
CN111275367B (en) Regional comprehensive energy system energy efficiency state evaluation method
CN112613646A (en) Equipment state prediction method and system based on multi-dimensional data fusion
CN111858231A (en) Single index abnormality detection method based on operation and maintenance monitoring
CN109469919B (en) Power station air preheater ash blocking monitoring method based on weight clustering
CN111414957A (en) D-PMU disturbance classification method based on PCA algorithm and XGboost algorithm
CN112032003B (en) Method for monitoring operation performance of large wind turbine generator

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20190107

Address after: 210000 2nd Floor, Block C, Gupinggang 4, Gulou District, Nanjing City, Jiangsu Province

Applicant after: Nanjing Sichen Information Technology Co., Ltd.

Address before: 215500 Tongji Science and Technology Plaza, 333 Hushan Road, Changshu High-tech Industrial Development Zone, Suzhou City, Jiangsu Province, 308

Applicant before: Suzhou Wenhai IOT Technology Co. Ltd.

GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20200619

Address after: Room 1428, unit 1, unit 1, building 10 (original 3), Guanggu Chuangye street, Dongxin Road, Donghu New Technology Development Zone, Wuhan, Hubei Province 430070

Patentee after: Wuhan Mouse Technology Co.,Ltd.

Address before: 210000 2nd Floor, Block C, Gupinggang 4, Gulou District, Nanjing City, Jiangsu Province

Patentee before: Nanjing Sichen Information Technology Co.,Ltd.