CN109241041A - A kind of preprocess method and device of injection molding equipment big data - Google Patents
A kind of preprocess method and device of injection molding equipment big data Download PDFInfo
- Publication number
- CN109241041A CN109241041A CN201810671621.0A CN201810671621A CN109241041A CN 109241041 A CN109241041 A CN 109241041A CN 201810671621 A CN201810671621 A CN 201810671621A CN 109241041 A CN109241041 A CN 109241041A
- Authority
- CN
- China
- Prior art keywords
- data
- injection molding
- molding equipment
- denoising
- degree
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
A kind of preprocess method of injection molding equipment big data, comprising the following steps: obtain the acquisition data of injection molding equipment;Denoising is carried out to the injection molding equipment data got using rolling time horizon method;The similarity of data and default result after judging denoising carries out data replacement processing to the data beyond zone of reasonableness;By treated, injection molding equipment data are sent to database.The method of the present invention, which is first passed through, carries out denoising to injection molding equipment data, then processing is replaced to the injection molding equipment data after denoising again, can satisfactory data be effectively effectively obtained from magnanimity injection molding equipment data, it can be applied to big data to excavate and analyze, to obtain being conducive to analyze injection molding equipment performance and improve the data of production efficiency.
Description
Technical field
The present invention relates to big data analysis technical field, and in particular to a kind of pretreatment of injection molding equipment big data
Method and device.
Background technique
The machine-tool as Plastics Industry is equipped in injection molding, supports the production of the pillars such as household electrical appliances, automobile, consumer electronics
The development of industry.But entire plastic industry still falls within labor-intensive production, information intelligence level falls behind.To protect
Card injection molding machine is able to produce out qualified product, need to influence injection molding machine during production operations plastic products quality because
Element is detected and controlled.In actual production, the project that carry out signal detection includes: the temperature, oil temperature, oil cylinder pressure of barrel
Power, system oil pressure, the stroke of action of injection, speed and time etc..Meanwhile it also being generated when breaking down in injection molding machine production process
Some pictures, video data.The creation data of injection molding machine also constitutes the source of injection molding equipment big data.Existing
These magnanimity injection molding equipment data in, user effectively can not therefrom obtain useful information, cause to be molded into
Type equipment data are discarded.
In conclusion how to provide a kind of technology that can effectively obtain useful data from magnanimity injection molding equipment
Scheme is a problem to be solved by those skilled in the art.
Summary of the invention
Present invention aims to overcome that the shortcomings that prior art and insufficient, a kind of injection molding equipment big data is provided
Preprocess method effectively effectively can obtain useful data from magnanimity injection molding equipment data.
To achieve the above object, The technical solution adopted by the invention is as follows:
A kind of preprocess method of injection molding equipment big data, comprising the following steps:
Obtain the acquisition data of injection molding equipment;
Denoising is carried out to the injection molding equipment data got using rolling time horizon method;
The similarity of data and default result after judging denoising carries out data to the data beyond zone of reasonableness and replaces
Change processing;
By treated, injection molding equipment data are sent to database.
From the foregoing, it will be observed that injection molding equipment generally entails the presence of interference signal when acquiring data, i.e. our usual institutes
When being acquired using signal picker to data, truthful data and noise signal are collected data together and adopt the noise said
It, will be to causing distorted signals if not handling noise jamming in storage.Denoising is being carried out to the data of acquisition
Afterwards, in order to avoid certain data are over treated so as to cause the appearance of unreasonable data, we also need to after denoising
Signal judged, rationally replaced to beyond the data of preset range, form satisfactory data.By denoising
Data are equipped in the injection molding of processing and replacement processing, it is possible to are enough applied to big data and be excavated and analyze, to obtain advantageous
In the data of analysis injection molding equipment performance and raising production efficiency.
In conclusion the method for the present invention, which is first passed through, carries out denoising to injection molding equipment data, then again to denoising
Injection molding that treated equipment data are replaced processing, effectively can effectively obtain from magnanimity injection molding equipment data
Satisfactory data are taken, big data is can be applied to and excavates and analyze, to obtain being conducive to analyze injection molding equipment property
It can and improve the data of production efficiency.
As an improvement of the present invention, in the step, " by treated, injection molding equipment data are sent to data
It before library ", also follows the steps below: according to the precision of data type, in conjunction with the load characteristic of network transmission, using multiple dimensioned quantization
Device carries out quantification treatment to signal.
The step " denoising is carried out to the injection molding equipment data got using rolling time horizon method " include with
Lower sub-step:
A. System State Model is established according to acquisition data dynamic characteristic:
Wherein x (t), u (t) and ω (t) are respectively state, input and the noise jamming of t moment data, y (t) and v (t)
The respectively collection value and noise jamming of t moment data;
B. rolling window length L is established, objective function is established to System State Model;
C. minimization objective function seeks the data value after current time denoising based on acquisition data in window;
D. rolling window is moved forward into a step, return step c, so circulation is until traversing all data points.
Further, the objective function are as follows:
Wherein,For in denoising of the t moment to t-L time data,For the t-L moment
Prediction to data mode, α are weight coefficient.
Further, the step " similarity of data and default result after judging denoising, to beyond reasonable model
Similarity in the data progress data replacement processing enclosed " are as follows:Wherein μ is data precision,For the value of data t moment after denoising.
Further, the step " similarity of data and default result after judging denoising, to beyond reasonable model
Data replacement processing concrete mode in the data progress data replacement processing enclosed " are as follows:
IfThen replacement data
IfThen replacement dataWherein μ is data precision,For by replacement treated data value,For the value of data t moment after denoising.
As an improvement of the present invention, the multiple dimensioned quantizer is obtained by following steps:
The bit number τ of each data packet is analyzed according to transmission network1, while defining the data bit of t moment actual transmissions
Number is τ2(t), the degree of load F (t) of transmission channel t moment is calculated;
Data variation window value K is counted according to respective historical data, calculates the average load degree based on window value
After finding out the average load degree of each window value, 9 grades will be divided into from big to small by average load degree, i.e., it is all
Average load degree can be referred in 9 ranks, adjust corresponding quantization scale ρ according to rank, i.e., when average load degree is
At the 9th grade, quantization scale ρ is adjusted to 0.9, and when average load degree is the 8th grade, quantization resolution ρ is adjusted to 0.8, and so on,
When average load degree is the 1st grade, quantization resolution is adjusted to 0.1;
Based on different loads grade, the multiple dimensioned quantizer in corresponding data source is designedWherein,ρ is quantization scale,J ∈ 0, ± 1,
± 2 ... },C is preset positive constant.
The present invention also provides a kind of pretreatment units of injection molding equipment big data.
To achieve the above object, The technical solution adopted by the invention is as follows:
A kind of pretreatment unit of injection molding equipment big data, including data acquisition module, denoising module, data
Replacement module and data outputting module;
Data acquisition module, for obtaining the acquisition data of injection molding equipment;
Denoising module, for being carried out at denoising using rolling time horizon method to the injection molding equipment data got
Reason;
Data replacement module, for judging the similarity of the data after denoising and default result, to exceeding reasonable model
The data enclosed carry out data replacement processing;
Data outputting module, for injection molding equipment data to be sent to database by treated.
As an improvement of the present invention, further include quantification treatment module, data are being equipped into treated injection molding
Before being sent to database, quantification treatment is carried out to signal first with quantification treatment module, specifically, the quantification treatment module is used
In the precision according to data type, in conjunction with the load characteristic of network transmission, signal is carried out at quantization using multiple dimensioned quantizer
Reason.
Further, the multiple dimensioned quantizer is obtained by following steps:
The bit number τ of each data packet is analyzed according to transmission network1, while defining the data bit of t moment actual transmissions
Number is τ2(t), the degree of load F (t) of transmission channel t moment is calculated;
Data variation window value K is counted according to respective historical data, calculates the average load degree based on window value
After finding out the average load degree of each window value, 9 grades will be divided into from big to small by average load degree, i.e., it is all
Average load degree can be referred in 9 ranks, adjust corresponding quantization scale ρ according to rank, i.e., when average load degree is
At the 9th grade, quantization scale ρ is adjusted to 0.9, and when average load degree is the 8th grade, quantization resolution ρ is adjusted to 0.8, and so on,
When average load degree is the 1st grade, quantization resolution is adjusted to 0.1;
Based on different loads grade, the multiple dimensioned quantizer in corresponding data source is designedWherein,ρ is quantization scale,J ∈ 0, ± 1,
± 2 ... },C is preset positive constant.
Compared with prior art, the innovative point of technical solution of the present invention and beneficial effect are:
The method of the present invention, which is first passed through, carries out denoising to injection molding equipment data, then again to the note after denoising
Modeling molding equipment data are replaced processing, effectively can effectively obtain and meet the requirements from magnanimity injection molding equipment data
Data, can be applied to big data and excavate and analysis, to obtain being conducive to analyzing injection molding equipment performance and improve life
Produce the data of efficiency.
Detailed description of the invention
Fig. 1 is the flow chart for the preprocess method that big data is equipped in injection molding of the present invention.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples.It is understood that tool described herein
Body embodiment is used only for explaining the present invention rather than limiting the invention.It also should be noted that for the ease of retouching
It states, only some but not all contents related to the present invention are shown in the drawings.
Embodiment
Referring to FIG. 1, a kind of preprocess method of injection molding equipment big data, comprising the following steps:
S1. the acquisition data of injection molding equipment are obtained.
S2. denoising is carried out to the injection molding equipment data got using rolling time horizon method;
Wherein, step " denoising is carried out to the injection molding equipment data got using the rolling time horizon method " packet
Include following sub-step:
A. System State Model is established according to acquisition data dynamic characteristic:
Wherein x (t), u (t) and ω (t) are respectively state, input and the noise jamming of t moment data, y (t) and v (t)
The respectively collection value and noise jamming of t moment data;
B. rolling window length L is established, objective function is established to System State Model;
Wherein, the objective function are as follows:
Wherein,For in denoising of the t moment to t-L time data,For the t-L moment
Prediction to data mode, α are weight coefficient;
C. minimization objective function seeks the data value after current time denoising based on acquisition data in window;
D. rolling window is moved forward into a step, return step c, so circulation is until traversing all data points.
Rolling time horizon method is newly entered output data information and last moment to this when based on one section in rolling window
The predictive information at quarter is implemented to optimize to current time data, it does not need to comprehend the distribution situation of noise, in each time optimization
Performance indicator relates only to from the moment the following limited time, and arrives subsequent time, this optimization time is simultaneously to being pushed forward
It moves, constantly carries out on-line optimization, there is very strong robustness.
S3. the similarity of data and default result after judging denoising counts the data beyond zone of reasonableness
It is handled according to replacement;
Wherein, the step " similarity of data and default result after judging denoising, to beyond zone of reasonableness
Similarity in data progress data replacement processing " are as follows:Wherein μ is data precision,For
The value of data t moment after denoising;
Step " the similarity of data and default result after judging denoising, to the data for exceeding zone of reasonableness
Data replacement processing concrete mode in progress data replacement processing " are as follows:
IfThen replacement data
IfThen replacement dataWherein μ is data precision,For by replacement treated data value,For the value of data t moment after denoising.
After carrying out denoising to the signal of acquisition, in order to avoid certain data are over treated so as to cause unreasonable
The appearance of data, we also need to judge the signal after denoising, and it is reasonable to carry out to the data beyond preset range
Replacement, forms satisfactory data.
S4. by treated, injection molding equipment data are sent to database.
From the foregoing, it will be observed that injection molding equipment generally entails the presence of interference signal when acquiring data, i.e. our usual institutes
When being acquired using signal picker to data, truthful data and noise signal are collected data together and adopt the noise said
It, will be to causing distorted signals if not handling noise jamming in storage.Denoising is being carried out to the data of acquisition
Afterwards, in order to avoid certain data are over treated so as to cause the appearance of unreasonable data, we also need to after denoising
Signal judged, rationally replaced to beyond the data of preset range, form satisfactory data.By denoising
Data are equipped in the injection molding of processing and replacement processing, it is possible to are enough applied to big data and be excavated and analyze, to obtain advantageous
In the data of analysis injection molding equipment performance and raising production efficiency.
In conclusion the method for the present invention, which is first passed through, carries out denoising to injection molding equipment data, then again to denoising
Injection molding that treated equipment data are replaced processing, effectively can effectively obtain from magnanimity injection molding equipment data
Satisfactory data are taken, big data is can be applied to and excavates and analyze, to obtain being conducive to analyze injection molding equipment property
It can and improve the data of production efficiency.
In the present embodiment, before the step S4 " by treated, injection molding equipment data are sent to database ",
It also follows the steps below: according to the precision of data type, in conjunction with the load characteristic of network transmission, using multiple dimensioned quantizer to letter
Number carry out quantification treatment;
Wherein, the multiple dimensioned quantizer is obtained by following steps:
The bit number τ of each data packet is analyzed according to transmission network1, while defining the data bit of t moment actual transmissions
Number is τ2(t), the degree of load F (t) of transmission channel t moment is calculated;
Data variation window value K is counted according to respective historical data, calculates the average load degree based on window value
After finding out the average load degree of each window value, 9 grades will be divided into from big to small by average load degree, i.e., it is all
Average load degree can be referred in 9 ranks, adjust corresponding quantization scale ρ according to rank, i.e., when average load degree is
At the 9th grade, quantization scale ρ is adjusted to 0.9, and when average load degree is the 8th grade, quantization resolution ρ is adjusted to 0.8, and so on,
When average load degree is the 1st grade, quantization resolution is adjusted to 0.1;
Based on different loads grade, the multiple dimensioned quantizer in corresponding data source is designedWherein,ρ is quantization scale,J ∈ 0, ± 1,
± 2 ... },C is preset positive constant.
After obtaining satisfactory data, the load characteristic based on transmission network need to design multi-density quantizer, amount
Changing device is quantified to data, to facilitate data to be transmitted by wireless network.Existing method mainly uses some fixations
The method of step-length quantifies data, is then compressed to quantized data in order to transmit.But under big data environment,
The fixed quantizer of step-length is difficult to make full use of the bandwidth of transmission channel.In order to make full use of network transmission bandwidth, transmission is as far as possible
More useful information, it is therefore desirable in conjunction with the load capacity of network transmission, multiple dimensioned multidate information quantizer is designed, to improve number
According to efficiency of transmission.
A kind of pretreatment unit of injection molding equipment big data, including data acquisition module, denoising module, data
Replacement module and data outputting module;
Data acquisition module, for obtaining the acquisition data of injection molding equipment;
Denoising module, for being carried out at denoising using rolling time horizon method to the injection molding equipment data got
Reason;
Data replacement module, for judging the similarity of the data after denoising and default result, to exceeding reasonable model
The data enclosed carry out data replacement processing;
Data outputting module, for injection molding equipment data to be sent to database by treated.
Injection molding is equipped when acquiring data, generally entails the presence of interference signal, i.e. our usually said noises
When being acquired using signal picker to data, truthful data and noise signal are collected data collector together and work as
In, it, will be to causing distorted signals if not handling noise jamming.After carrying out denoising to the data of acquisition, it is
Certain data are avoided to be over treated so as to cause the appearance of unreasonable data, we also need to the signal after denoising
Judged, the data beyond preset range are rationally replaced, satisfactory data are formed.By denoising and
Data are equipped in the injection molding of replacement processing, it is possible to are enough applied to big data and be excavated and analyze, to obtain being conducive to analyze
Injection molding equipment performance and the data for improving production efficiency.
In the present embodiment, the pretreatment unit further includes quantification treatment module, will treated injection molding dress
Before standby data are sent to database, quantification treatment, specifically, the quantification treatment are carried out to signal first with quantification treatment module
Module is used for the precision according to data type, in conjunction with the load characteristic of network transmission, is carried out using multiple dimensioned quantizer to signal
Quantification treatment;
Wherein, the multiple dimensioned quantizer is obtained by following steps:
The bit number τ of each data packet is analyzed according to transmission network1, while defining the data bit of t moment actual transmissions
Number is τ2(t), the degree of load F (t) of transmission channel t moment is calculated;
Data variation window value K is counted according to respective historical data, calculates the average load degree based on window value
After finding out the average load degree of each window value, 9 grades will be divided into from big to small by average load degree, i.e., it is all
Average load degree can be referred in 9 ranks, adjust corresponding quantization scale ρ according to rank, i.e., when average load degree is
At the 9th grade, quantization scale ρ is adjusted to 0.9, and when average load degree is the 8th grade, quantization resolution ρ is adjusted to 0.8, and so on,
When average load degree is the 1st grade, quantization resolution is adjusted to 0.1;
Based on different loads grade, the multiple dimensioned quantizer in corresponding data source is designedWherein,ρ is quantization scale,J ∈ 0, ± 1,
± 2 ... },C is preset positive constant.
The above embodiment is a preferred embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment
Limitation, other any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the present invention,
It should be equivalent substitute mode, be included within the scope of the present invention.
Claims (10)
1. a kind of preprocess method of injection molding equipment big data, it is characterised in that the following steps are included:
Obtain the acquisition data of injection molding equipment;
Denoising is carried out to the injection molding equipment data got using rolling time horizon method;
The similarity of data and default result after judging denoising carries out at data replacement the data beyond zone of reasonableness
Reason;
By treated, injection molding equipment data are sent to database.
2. the preprocess method of injection molding equipment big data according to claim 1, it is characterised in that: in the step
Before " treated injection molding equipment data are sent to database ", also follow the steps below: according to the precision of data type,
In conjunction with the load characteristic of network transmission, quantification treatment is carried out to signal using multiple dimensioned quantizer.
3. the preprocess method of injection molding equipment big data according to claim 1, it is characterised in that: the step " benefit
Denoising is carried out to the injection molding equipment data got with rolling time horizon method " include following sub-step:
A. System State Model is established according to acquisition data dynamic characteristic:
Wherein x (t), u (t) and ω (t) are respectively state, input and the noise jamming of t moment data, and y (t) and v (t) are respectively
For the collection value and noise jamming of t moment data;
B. rolling window length L is established, objective function is established to System State Model;
C. minimization objective function seeks the data value after current time denoising based on acquisition data in window;
D. rolling window is moved forward into a step, return step c, so circulation is until traversing all data points.
4. the preprocess method of injection molding equipment big data according to claim 3, it is characterised in that: the target letter
Number are as follows:
Wherein,For in denoising of the t moment to t-L time data,For t-L moment logarithm
According to the prediction of state, α is weight coefficient.
5. the preprocess method of injection molding equipment big data according to claim 3, it is characterised in that: the step " is sentenced
Data beyond zone of reasonableness are carried out data replacement processing by the similarity of data and default result after disconnected denoising " in
Similarity are as follows:Wherein μ is data precision,For data after denoising t moment
Value.
6. the preprocess method of injection molding equipment big data according to claim 3, it is characterised in that: the step " is sentenced
Data beyond zone of reasonableness are carried out data replacement processing by the similarity of data and default result after disconnected denoising " in
Data replacement processing concrete mode are as follows:
IfThen replacement data
IfThen replacement dataWherein μ is data precision,
For by replacement treated data value,For the value of data t moment after denoising.
7. the preprocess method of injection molding equipment big data according to claim 2, it is characterised in that: the multiple dimensioned amount
Change device to obtain by following steps:
The bit number τ of each data packet is analyzed according to transmission network1, while the data bit number for defining t moment actual transmissions is τ2
(t), the degree of load F (t) of transmission channel t moment is calculated;
Data variation window value K is counted according to respective historical data, calculates the average load degree based on window value
After finding out the average load degree of each window value, 9 grades will be divided into from big to small by average load degree, i.e., it is all average
Degree of load can be referred in 9 ranks, adjust corresponding quantization scale ρ according to rank, i.e., when average load degree is the 9th grade
When, quantization scale ρ is adjusted to 0.9, and when average load degree is the 8th grade, quantization resolution ρ is adjusted to 0.8, and so on, when flat
When equal degree of load is the 1st grade, quantization resolution is adjusted to 0.1;
Based on different loads grade, the multiple dimensioned quantizer in corresponding data source is designedWherein,ρ is quantization scale, C is preset positive constant.
8. a kind of pretreatment unit of injection molding equipment big data, it is characterised in that: including data acquisition module, denoising
Module, data replacement module and data outputting module;
Data acquisition module, for obtaining the acquisition data of injection molding equipment;
Denoising module, for carrying out denoising to the injection molding equipment data got using rolling time horizon method;
Data replacement module, for judging the similarity of the data after denoising and default result, to beyond zone of reasonableness
Data carry out data replacement processing;
Data outputting module, for injection molding equipment data to be sent to database by treated.
9. the pretreatment unit of injection molding equipment big data according to claim 8, it is characterised in that: further include at quantization
Manage module, will before treated injection molding equipment data are sent to database, first with quantification treatment module to signal into
Row quantification treatment, specifically, the quantification treatment module are used for the precision according to data type, and the load in conjunction with network transmission is special
Property, quantification treatment is carried out to signal using multiple dimensioned quantizer.
10. the pretreatment unit of injection molding equipment big data according to claim 9, it is characterised in that: described multiple dimensioned
Quantizer is obtained by following steps:
The bit number τ of each data packet is analyzed according to transmission network1, while the data bit number for defining t moment actual transmissions is τ2
(t), the degree of load F (t) of transmission channel t moment is calculated;
Data variation window value K is counted according to respective historical data, calculates the average load degree based on window value
After finding out the average load degree of each window value, 9 grades will be divided into from big to small by average load degree, i.e., it is all average
Degree of load can be referred in 9 ranks, adjust corresponding quantization scale ρ according to rank, i.e., when average load degree is the 9th grade
When, quantization scale ρ is adjusted to 0.9, and when average load degree is the 8th grade, quantization resolution ρ is adjusted to 0.8, and so on, when flat
When equal degree of load is the 1st grade, quantization resolution is adjusted to 0.1;
Based on different loads grade, the multiple dimensioned quantizer in corresponding data source is designedWherein,ρ is quantization scale, C is preset positive constant.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810671621.0A CN109241041B (en) | 2018-06-26 | 2018-06-26 | Preprocessing method and device for big data of injection molding equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810671621.0A CN109241041B (en) | 2018-06-26 | 2018-06-26 | Preprocessing method and device for big data of injection molding equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109241041A true CN109241041A (en) | 2019-01-18 |
CN109241041B CN109241041B (en) | 2021-05-11 |
Family
ID=65071968
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810671621.0A Active CN109241041B (en) | 2018-06-26 | 2018-06-26 | Preprocessing method and device for big data of injection molding equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109241041B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150197164A1 (en) * | 2014-01-14 | 2015-07-16 | Ford Global Technologies, Llc | Receding horizon regression analysis for battery impedance parameter estimation |
CN107578822A (en) * | 2017-07-25 | 2018-01-12 | 广东工业大学 | A kind of pretreatment and feature extracting method for the multi-modal big data of medical treatment |
CN108133734A (en) * | 2017-12-21 | 2018-06-08 | 广东工业大学 | A kind of analysis method, device and the equipment of medical expense big data |
-
2018
- 2018-06-26 CN CN201810671621.0A patent/CN109241041B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150197164A1 (en) * | 2014-01-14 | 2015-07-16 | Ford Global Technologies, Llc | Receding horizon regression analysis for battery impedance parameter estimation |
CN107578822A (en) * | 2017-07-25 | 2018-01-12 | 广东工业大学 | A kind of pretreatment and feature extracting method for the multi-modal big data of medical treatment |
CN108133734A (en) * | 2017-12-21 | 2018-06-08 | 广东工业大学 | A kind of analysis method, device and the equipment of medical expense big data |
Non-Patent Citations (1)
Title |
---|
鲁仁全 等: "《连续系统的状态重置动态量化反馈控制器》", 《控制理论与应用》 * |
Also Published As
Publication number | Publication date |
---|---|
CN109241041B (en) | 2021-05-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110263474A (en) | A kind of cutter life real-time predicting method of numerically-controlled machine tool | |
CN102056018B (en) | Method and system for providing TV guide and method for providing program-requesting information | |
CN103246922A (en) | Method for video abstract generation | |
CN105224690B (en) | Generate and select the method and system of the executive plan of the corresponding sentence containing ginseng | |
CN110967974B (en) | Coal flow balance self-adaptive control method based on rough set | |
CN108548822A (en) | A kind of wide cut continuous surface defective vision detecting system | |
CN114842371B (en) | Unsupervised video anomaly detection method | |
CN110868221B (en) | Multi-mode data automatic compression method | |
CN106682225A (en) | Big data collecting and storing method and system | |
CN117376430B (en) | Industrial data rapid transmission method and system based on DCS | |
CN111958321A (en) | Numerical control machine tool cutter wear degree identification method based on deep neural network | |
CN109241041A (en) | A kind of preprocess method and device of injection molding equipment big data | |
CN113221674B (en) | Video stream key frame extraction system and method based on rough set reduction and SIFT | |
Liu et al. | Multimedia image and video retrieval based on an improved HMM | |
CN114049368A (en) | Intelligent industrial design software of IID + ginseng tree network | |
CN116561230B (en) | Distributed storage and retrieval system based on cloud computing | |
CN109359724A (en) | A kind of method and device of compression storage convolution neural network model | |
CN116563293B (en) | Photovoltaic carrier production quality detection method and system based on machine vision | |
CN115269659B (en) | Remote monitoring system for energy consumption detection of combined air conditioning box | |
CN108550053B (en) | User consumption data acquisition and analysis system and method based on platform technology | |
CN115327942A (en) | Intelligent environment monitoring system | |
CN114800049B (en) | Grating ruler processing operation signal error compensation system | |
CN115186574A (en) | Tool residual life prediction method based on gated cyclic residual error network | |
CN103942759A (en) | Three-dimensional noise reduction method and device based on Gaussian background model in fixed scene | |
CN112270183B (en) | News propagation effect monitoring system based on text |
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 |