CN109782728A - A kind of multivariable monitoring method and system based on Savitzky-Golay filter - Google Patents
A kind of multivariable monitoring method and system based on Savitzky-Golay filter Download PDFInfo
- Publication number
- CN109782728A CN109782728A CN201910250981.8A CN201910250981A CN109782728A CN 109782728 A CN109782728 A CN 109782728A CN 201910250981 A CN201910250981 A CN 201910250981A CN 109782728 A CN109782728 A CN 109782728A
- Authority
- CN
- China
- Prior art keywords
- savitzky
- data
- multivariable
- golay filter
- monitoring method
- 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
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present disclosure proposes a kind of multivariable monitoring methods and system based on Savitzky-Golay filter; it include: the history data for obtaining multi-variable system; include multiple process variables for showing system running state in the history data, and marks the normal or abnormal situation of history data;It is filtered using historical time sequence data of the Savitzky-Golay filter to each process variable, obtains filtered signal amplitude and all-order derivative;By the filtered signal amplitude of each process variable and all-order derivative composition characteristic space;Nonlinear Support Vector Machines model is obtained based on historical data training in feature space, and for judging whether to provide alarm, realizes the monitoring of multivariable.The method proposed has the advantages that high-accuracy, overcomes the defect of traditional single argument alarm threshold design method, can effectively improve monitoring system finiteness, so that missing, the quantity of leakage alarm is all greatly reduced.
Description
Technical field
This disclosure relates to which the fields of automation technology such as Industrial Process Monitoring System, are based on more particularly to one kind
The multivariable monitoring method and system of Savitzky-Golay filter.
Background technique
Industrial Process Monitoring System is the important component of dcs in modern process industry.Due to monitoring
Variable setting is improper, causes monitoring system often to generate a large amount of interference alarm, monitor system performance is seriously reduced, to life
The safety of production and efficiency constitute a serious threat.Cause largely to interfere the existing major reason of alarm to be alarm variable setting
Lack correlation between variable associated therewith, currently, most of process variables all use the alarm of setting signal amplitude upper and lower limit
Design method alerts triggering when the signal amplitude of process variable is more than the upper limit or lower limit of normal range of operation.But it is right
In multi-variable system, since there are correlations for each process variable, only consider that the amplitude variation of each variable itself is inadequate.
Inventor has found under study for action, and a large amount of history number is had accumulated in process industry dcs now
According to setting for multivariable monitoring system can be effectively performed using data-driven modeling method based on these historical datas
Meter.Currently have based on multivariate probability distribution, be based on Bayesian Estimation, based on principal component analysis and based on artificial neural network etc.
Monitoring system design method.But the extracted data characteristics of these methods is usually unaccountable, therefore industrial application mistake
Operations staff is generally difficult to receive these methods in journey.
Summary of the invention
The purpose of this specification embodiment is to provide a kind of multivariable monitoring based on Savitzky-Golay filter
Method reduces interference alarm quantity practical application with higher for improving the validity of the monitoring system in process industry
Value.
This specification embodiment provides a kind of multivariable monitoring method based on Savitzky-Golay filter, uses
Following technical scheme:
Include:
The history data for obtaining multi-variable system shows system running state comprising multiple in the history data
Process variable, and mark the normal or abnormal situation of history data;
It is filtered, is obtained using historical time sequence data of the Savitzky-Golay filter to each process variable
Filtered signal amplitude and all-order derivative;
By the filtered signal amplitude of each process variable and all-order derivative composition characteristic space;
Nonlinear Support Vector Machines model is obtained based on historical data training in feature space, and for judge whether to
It alerts out, realizes the monitoring of multivariable.
The method that this specification embodiment is proposed has the advantages that high-accuracy, overcomes traditional single argument monitoring threshold
It is worth the defect of design method, monitoring system finiteness can be effectively improved, so that misses, the quantity of leakage alarm is all greatly decreased.
Another embodiment of this specification provides a kind of multivariable monitoring system based on Savitzky-Golay filter,
Using following technical scheme:
Data selecting module is configured as: being obtained the history data of multi-variable system, is wrapped in the history data
Containing multiple process variables for showing system running state, and mark the normal or abnormal situation of history data;
Filter module is configured as: using Savitzky-Golay filter to the historical time sequence of each process variable
Column data is filtered, and obtains filtered signal amplitude and all-order derivative;
Feature space module is configured as: the filtered signal amplitude and all-order derivative of each process variable being formed special
Levy space;
Alarm module is configured as: obtaining Nonlinear Support Vector Machines mould based on historical data training in feature space
Type, and for judging whether to provide alarm, realize multivariable monitoring.
The method that this specification technical solution is proposed has the advantages that high-accuracy, improves tradition and single argument is arranged
The method for monitoring threshold value, so that missing, the quantity of leakage alarm is all greatly decreased.
Another embodiment of this specification provides a kind of computer equipment, including memory, processor and is stored in storage
On device and the computer program that can run on a processor, which is characterized in that the processor realizes one when executing described program
Multivariable monitoring method of the kind based on Savitzky-Golay filter.
Another embodiment of this specification provides a kind of computer readable storage medium, is stored thereon with computer program,
It is characterized in that, the program realizes a kind of multivariable monitoring side based on Savitzky-Golay filter when being executed by processor
Method.
Compared with prior art, the beneficial effect of the disclosure is:
Disclosed technique scheme combined data drives advantage and the multivariable of qualitiative trends analysis advantage to monitor, and is proposed
Method has the advantages that high-accuracy, overcomes the defect of traditional single argument monitoring threshold design method, can effectively improve monitoring
System finiteness, so that missing, the quantity of leakage alarm is all greatly decreased.
Detailed description of the invention
The Figure of description for constituting a part of this disclosure is used to provide further understanding of the disclosure, and the disclosure is shown
Meaning property embodiment and its explanation do not constitute the improper restriction to the disclosure for explaining the disclosure.
A kind of Fig. 1: the stream of multivariable monitoring method based on Savitzky-Golay filter of embodiment of the present disclosure
Cheng Tu;
A kind of Fig. 2: multivariable monitoring system block diagram based on Savitzky-Golay filter of embodiment of the present disclosure;
Fig. 3: the historical data time series chart of the feed-water pump of embodiment of the present disclosure;
Fig. 4: the time sequence of amplitude, first derivative and second dervative after the filtering of the fore pump electric current of embodiment of the present disclosure
Column figure.
Specific embodiment
It is noted that following detailed description is all illustrative, it is intended to provide further instruction to the disclosure.Unless another
It indicates, all technical and scientific terms used herein has usual with disclosure person of an ordinary skill in the technical field
The identical meanings of understanding.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the disclosure.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
Existing qualitiative trends analytic approach, the feature changed over time by extracting single process variable, such as increases, no
The qualitiative trends information such as change, reduction, since qualitiative trends information has good visual interpretation, qualitiative trends information pair
Also there is intuitively interpretation in real process monitoring, while the experience and intuitive reason of field operator can be met well
Solution.
Examples of implementation one
The examples of implementation provide a kind of multivariable monitoring method based on Savitzky-Golay filter, pass through
Savitzky-Golay filter extracts filtered signal amplitude and all-order derivative as qualitative features, and in feature space
Whether training Nonlinear Support Vector Machines model should provide alarm by Nonlinear Support Vector Machines Model checking, realize more
The alarm monitoring of variable.
In an examples of implementation, a kind of multivariable monitoring method based on Savitzky-Golay filter, referring to attached drawing 1
It is shown, it specifically includes:
Step S1 obtains enough historical datas, and has marked the normal or abnormal situation of historical data;
Step S2 is filtered using historical time sequence data of the Savitzky-Golay filter to each process variable
Wave obtains filtered signal amplitude and all-order derivative;
Step S3, by the filtered signal amplitude of each variable and all-order derivative composition characteristic space;
Step S4 obtains Nonlinear Support Vector Machines model based on historical data training in feature space, and for sentencing
It is disconnected whether to provide alarm.
In specific embodiment, in step S1, if multi-variable system is
X=[X1,X2,…,XR]。 (1)
Process variable XrIn t0The amplitude at moment is xr(t0), 1≤r≤R, R are the quantity of process variable.
The normal or abnormal situation of historical data is indicated by sequence label y (t).If t0There is exception in the moment, then y (t0)
Take+1.On the contrary, if t0Exception is not present in moment, then y (t0) take -1.
In specific embodiment, in step S2, Savitzky-Golay filter is a kind of based on multinomial minimum two
The filtering method of multiplication fitting, is widely used in data flow smoothing denoising.Embodiment of the present disclosure uses Savitzky-Golay
The filtered signal amplitude of filter extraction process time series variation data and the principle of all-order derivative are as follows:
A column time series data x (t) is considered, for t0Data point x (t before and after moment0+ n), n is time difference ,-M≤n
≤M.2M+1 data point is had altogether to these and carries out fitting of a polynomial.Polynomial fit function form is as follows,
Wherein, K is the maximum order of fitting of a polynomial, akFor undetermined coefficient.
In least square method, the accumulated error of fitting of a polynomial is
Wherein, by minimizing LKAvailable each coefficient ai, 0≤i≤K, even LKTo aiPartial derivative be zero,
Above formula can be transformed to
It can be seen that this is about akMulti head linear equation group, share K+1 equation.Meanwhile akFor x (t0+ n) line
Property combination, which can be solved by following process.Order matrix A={ αn,i, wherein
αn,i=ni, (6)
Wherein-M≤n≤M, i=0,1 ..., K.Remember B=ATA, then the element β of matrix Bi,kFor
Wherein i=0,1 ..., K, k=0,1 ..., K.Vector is remembered again
With coefficient vector to be solved
Then formula (5) can be expressed as
Ba=ATAa=ATx。 (10)
It can must solve and be
A=(ATA)-1ATX=Hx, (11)
Wherein matrix H={ hi,n, i=0,1 ..., K, n=-M ,-M+1 ..., M-1, M,
H=(ATA)-1AT。 (12)
Then for t0Shi Keyou n=0, filtered sequence s (t) is in t0The signal amplitude at moment is
It can be in the hope of t by (2)0The all-order derivative at moment is as follows,
In specific embodiment, in step S3, for multi-variable system X=[X1,X2,…,XR], feature space is
S=[S1,S2,…,SR], (15)
Wherein SrFor process variable XRFiltered signal amplitude and all-order derivative, i.e.,
In specific embodiment, in step S4, non-linear support is obtained based on historical data training in feature space
Vector machine model, wherein support vector machines is a kind of common disaggregated model in machine Learning Theory.Support vector machines passes through structure
A hyperplane is made to separate the data point of two classifications.Savitzky-Golay filter is mentioned using supporting vector machine model
The feature taken is classified, and judges whether that alarm should be provided.
It include the training data set of N number of data in given n dimensional feature space S
T={ (x1,y1),(x2,y2),…,(xN,yN), (17)
Wherein, n dimensional feature vector xi∈ S (t), label yi∈ y (t), i=1,2 ..., N.Linear SVM will be
An Optimal Separating Hyperplane is constructed in feature space S, if the hyperplane form is as follows:
Wx+b=0, (18)
And corresponding categorised decision function is
F (x)=sign (wx+b). (19)
Wherein vector w and variable b is parameter needed for defining hyperplane.
The hyperplane is obtained by solving following optimization problem
Wherein ξiFor slack variable, C is punishment parameter.The restricted problem is obtained by solving its dual problem,
If the optimal solution of the problem isThen the slope of hyperplane (16) is
And select positive componentIt calculates
The case where being linearly inseparable for training data, can be reflected original sample space by construction kernel function K (x, z)
It is mapped to new feature space, realizes linear separability in new space.(21) formula is rewritten into following form,
Corresponding decision function is
Gaussian kernel function is used in this method:
Nonlinear Support Vector Machines model is obtained based on historical data training.Judge whether to need to provide announcement using formula (25)
It is alert, i.e.,
Wherein, whenAlarm is generated when being+1, whenNo alarm is indicated when being -1.
Examples of implementation two
Referring to shown in attached drawing 2, the disclosure also discloses a kind of multivariable monitoring based on Savitzky-Golay filter
System, comprising:
Data selecting module is configured as: being obtained the history data of multi-variable system, is wrapped in the history data
Containing multiple process variables for showing system running state, and mark the normal or abnormal situation of history data;
Filter module is configured as: using Savitzky-Golay filter to the historical time sequence of each process variable
Column data is filtered, and obtains filtered signal amplitude and all-order derivative;
Feature space module is configured as: the filtered signal amplitude and all-order derivative of each process variable being formed special
Levy space;
Alarm module is configured as: obtaining Nonlinear Support Vector Machines mould based on historical data training in feature space
Type, and for judging whether to provide alarm, realize the monitoring of multivariable.
In data selecting module, the normal or abnormal situation of history data is indicated by sequence label, if certain moment
There are exceptions, then sequence label takes+1, on the contrary, sequence label takes -1 if exception is not present in certain moment.
It should be noted that although being referred to several modules or submodule of equipment in the detailed description above, it is this
Division is only exemplary rather than enforceable.In fact, in accordance with an embodiment of the present disclosure, two or more above-described moulds
The feature and function of block can embody in a module.Conversely, the feature and function of an above-described module can be with
Further division is to be embodied by multiple modules.
Examples of implementation three
Another embodiment of this specification provides a kind of computer readable storage medium, is stored thereon with computer program,
It is characterized in that, the program realizes a kind of multivariable monitoring side based on Savitzky-Golay filter when being executed by processor
Method.It can be found in examples of implementation one about a kind of multivariable monitoring method based on Savitzky-Golay filter, herein no longer
Detailed description.
In the present embodiment, computer program product may include computer readable storage medium, containing for holding
The computer-readable program instructions of row various aspects of the disclosure.Computer readable storage medium, which can be, can keep and store
By the tangible device for the instruction that instruction execution equipment uses.Computer readable storage medium for example can be-- but it is unlimited
In-- storage device electric, magnetic storage apparatus, light storage device, electric magnetic storage apparatus, semiconductor memory apparatus or above-mentioned
Any appropriate combination.
Examples of implementation four
Another embodiment of this specification is to utilize the multivariable monitoring method based on Savitzky-Golay filter
Specific example, this example data are the operation datas of feed pump common in process industry.The related feed pump operating status
Five process variables are as follows: fore pump electric current X1, inlet flow rate X2, inlet pressure X3, rotary speed actual value X4, outlet pressure X5, composition
Multi-variable system
X=[X1,X2,X3,X4,X5]。
In this embodiment, referring to attached drawing 3, the time series chart of one section of typical history data is illustrated.At this section
It is interior that there are the abnormal datas that multistage has been marked via operations staff.Abnormal data segment mark is denoted as dark back in figure
Scape.
It is filtered using historical data of the Savitzky-Golay filter to each process variable.Filter parameter is
M=30, K=2.Filtered amplitude s is obtained based on formula (13)~(14)rAnd all-order derivativeWithDeng.Referring to attached drawing 4,
Each subgraph therein illustrates fore pump electric current X1Filtering after amplitude s1、WithTime series chart.
Feature space S=[S in this example1,S2,…,SR] tieed up comprising 5 × 3=15.It is obtained based on historical data training non-thread
Property supporting vector machine model.
Traditional single argument alarm method will benchmark as a comparison.Variable is X in this example3It is configured with a high alarm threshold
Variable, i.e.,
Single argument monitoring method (28) are compared in test data set with proposed based on the changeable of signal derivative
Measure the performance of monitoring method (27).Table 1 gives the mistake announcement rate β of two methods1With leakage announcement rate β2The comparing result of two indices.
The circular of the two indexs is as follows,
Count () is counting function in formula.After comparison it can be found that compared to single argument monitoring method, the disclosure is real
It applies the multivariable monitoring method based on signal derivative that the technical solution of example is proposed and makes the mistake of monitoring system, leakage alarm rate
All it is substantially reduced.The multivariable monitoring method performance that thus embodiment of the present disclosure is proposed is more excellent.
1 rate of false alarm of table and rate of failing to report Comparative result
β1 | β2 | |
Single argument monitoring | 28.6% | 64.7% |
Multivariable monitoring | 4.6% | 5.1% |
The disclosure is based on the multivariable monitoring method of Savitzky-Golay filter for improving the prison in process industry
The validity of control system reduces interference alarm quantity practical application value with higher.
Examples of implementation five
The examples of implementation disclose a kind of multivariable monitoring method, and the alarm device that the monitoring method uses is configured as holding
Multivariable monitoring method of one of row above-described embodiment one based on Savitzky-Golay filter.
The alarm device that multivariable monitoring method uses can be the light alarm device comprising control unit in the form of expression, or
The alarm equipment of person's other forms.
It is understood that in the description of this specification, reference term " embodiment ", " another embodiment ", " other
The description of embodiment " or " first embodiment~N embodiment " etc. means specific spy described in conjunction with this embodiment or example
Sign, structure, material or feature are contained at least one embodiment or example of the disclosure.In the present specification, to above-mentioned
The schematic representation of term may not refer to the same embodiment or example.Moreover, the specific features of description, structure, material
Person's feature can be combined in any suitable manner in any one or more of the embodiments or examples.
The foregoing is merely preferred embodiment of the present disclosure, are not limited to the disclosure, for the skill of this field
For art personnel, the disclosure can have various modifications and variations.It is all within the spirit and principle of the disclosure, it is made any to repair
Change, equivalent replacement, improvement etc., should be included within the protection scope of the disclosure.
Claims (10)
1. a kind of multivariable monitoring method based on Savitzky-Golay filter, characterized in that include:
The history data of multi-variable system is obtained, includes multiple mistakes for showing system running state in the history data
Cheng Bianliang, and mark the normal or abnormal situation of history data;
It is filtered, is filtered using historical time sequence data of the Savitzky-Golay filter to each process variable
Signal amplitude and all-order derivative afterwards;
By the filtered signal amplitude of each process variable and all-order derivative composition characteristic space;
Nonlinear Support Vector Machines model is obtained based on historical data training in feature space, and for judging whether to provide announcement
It is alert, realize the monitoring of multivariable.
2. a kind of multivariable monitoring method based on Savitzky-Golay filter as described in claim 1, characterized in that
The normal or abnormal situation of history data is indicated by sequence label, if certain moment has exception, sequence label takes+1,
On the contrary, sequence label takes -1 if exception is not present in certain moment.
3. a kind of multivariable monitoring method based on Savitzky-Golay filter as described in claim 1, characterized in that
Obtain the process of filtered signal amplitude and all-order derivative are as follows:
Fitting of a polynomial is carried out for multiple data points of a column time series data;
Using least square method, the accumulated error of fitting of a polynomial is obtained;
Accumulated error by minimizing fitting of a polynomial obtains each coefficient in accumulated error expression;
Enabling the accumulated error of fitting of a polynomial is zero to the partial derivative of each coefficient, obtains the multi head linear equation group about coefficient;
Solve multi head linear equation group, the signal amplitude under obtaining filtered sequence at a time;
To signal amplitude derivation, the all-order derivative at the moment is obtained.
4. a kind of multivariable monitoring method based on Savitzky-Golay filter as described in claim 1, characterized in that
By constructing, a hyperplane will be present abnormal the supporting vector machine model and there is no the data points of two abnormal classifications
It separates;
Classified using the feature that supporting vector machine model extracts Savitzky-Golay filter, judging whether should be to
It alerts out.
5. a kind of multivariable monitoring method based on Savitzky-Golay filter as described in claim 1, characterized in that
About the supporting vector machine model:
Given feature space includes the training data set of multiple data;
Linear SVM will construct an Optimal Separating Hyperplane and corresponding categorised decision function;
It is the case where being linearly inseparable for training data, by construction kernel function that original sample space reflection is empty to new feature
Between, linear separability is realized in new space;
Nonlinear Support Vector Machines model is obtained for based on historical data training, judges whether to need to alert.
6. a kind of multivariable monitoring system based on Savitzky-Golay filter, characterized in that include:
Data selecting module is configured as: the history data of multi-variable system is obtained, comprising more in the history data
A process variable for showing system running state, and mark the normal or abnormal situation of history data;
Filter module is configured as: using Savitzky-Golay filter to the historical time sequence number of each process variable
According to being filtered, filtered signal amplitude and all-order derivative are obtained;
Feature space module is configured as: the filtered signal amplitude of each process variable and all-order derivative composition characteristic is empty
Between;
Alarm module is configured as: Nonlinear Support Vector Machines model is obtained based on historical data training in feature space, and
For judging whether to provide alarm, the monitoring of multivariable is realized.
7. a kind of multivariable monitoring system based on Savitzky-Golay filter as claimed in claim 6, characterized in that
In data selecting module, the normal or abnormal situation of history data is indicated by sequence label, if there is exception in certain moment,
Then sequence label takes+1, on the contrary, sequence label takes -1 if exception is not present in certain moment.
8. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that realize that any described one kind of claim 1-6 is based on when the processor executes described program
The multivariable monitoring method of Savitzky-Golay filter.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
A kind of claim 1-6 any multivariable monitoring method based on Savitzky-Golay filter is realized when row.
10. a kind of multivariable monitoring method, which is characterized in that used alarm device is configured as perform claim and 1-6 is required to appoint
A kind of multivariable monitoring method based on Savitzky-Golay filter described in one.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910250981.8A CN109782728B (en) | 2019-03-29 | 2019-03-29 | Multivariable monitoring method and system based on Savitzky-Golay filter |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910250981.8A CN109782728B (en) | 2019-03-29 | 2019-03-29 | Multivariable monitoring method and system based on Savitzky-Golay filter |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109782728A true CN109782728A (en) | 2019-05-21 |
CN109782728B CN109782728B (en) | 2021-06-04 |
Family
ID=66490746
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910250981.8A Expired - Fee Related CN109782728B (en) | 2019-03-29 | 2019-03-29 | Multivariable monitoring method and system based on Savitzky-Golay filter |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109782728B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114187970A (en) * | 2021-11-30 | 2022-03-15 | 清华大学 | Lithium ion battery internal and external characteristic simulation method based on electrochemical mechanism |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
BRPI1104468A2 (en) * | 2011-09-21 | 2013-08-13 | Univ Estadual Da Paraiba | biodiesel grading process and biodiesel grading instrument |
CN104502103A (en) * | 2014-12-07 | 2015-04-08 | 北京工业大学 | Bearing fault diagnosis method based on fuzzy support vector machine |
CN104596767A (en) * | 2015-01-13 | 2015-05-06 | 北京工业大学 | Method for diagnosing and predicating rolling bearing based on grey support vector machine |
CN108007881A (en) * | 2017-11-30 | 2018-05-08 | 中国农业大学 | A kind of aquaculture water quality total nitrogen content detection method based on spectral technique |
-
2019
- 2019-03-29 CN CN201910250981.8A patent/CN109782728B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
BRPI1104468A2 (en) * | 2011-09-21 | 2013-08-13 | Univ Estadual Da Paraiba | biodiesel grading process and biodiesel grading instrument |
CN104502103A (en) * | 2014-12-07 | 2015-04-08 | 北京工业大学 | Bearing fault diagnosis method based on fuzzy support vector machine |
CN104596767A (en) * | 2015-01-13 | 2015-05-06 | 北京工业大学 | Method for diagnosing and predicating rolling bearing based on grey support vector machine |
CN108007881A (en) * | 2017-11-30 | 2018-05-08 | 中国农业大学 | A kind of aquaculture water quality total nitrogen content detection method based on spectral technique |
Non-Patent Citations (2)
Title |
---|
宋知用: "《MATLAB数字信号处理85个实用案例精讲 入门到进阶》", 30 November 2016, 北京航空航天大学出版社 * |
张立等: "基于核函数的非线性支持向量机", 《科技展望》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114187970A (en) * | 2021-11-30 | 2022-03-15 | 清华大学 | Lithium ion battery internal and external characteristic simulation method based on electrochemical mechanism |
WO2023098715A1 (en) * | 2021-11-30 | 2023-06-08 | 清华大学 | Electrochemical-mechanism-based simulation method for internal and external characteristics of lithium ion battery |
Also Published As
Publication number | Publication date |
---|---|
CN109782728B (en) | 2021-06-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10824139B2 (en) | Equipment maintenance method, equipment maintenance device, and storage medium for the same | |
CN106447210A (en) | Distribution network equipment health degree dynamic diagnosis method involving credibility evaluation | |
CN109501834A (en) | A kind of point machine failure prediction method and device | |
EP3026518A1 (en) | Method for Root analysis of an alarm flood sequence | |
CN113673600B (en) | Industrial signal abnormality early warning method, system, storage medium and computing device | |
CN107679089A (en) | A kind of cleaning method for electric power sensing data, device and system | |
CN113344133B (en) | Method and system for detecting abnormal fluctuation of time sequence behaviors | |
JP6368031B2 (en) | Abnormality prediction / recovery support system, abnormality prediction / recovery support method, and water treatment system | |
CN107272667A (en) | A kind of industrial process fault detection method based on parallel PLS | |
CN103926490A (en) | Power transformer comprehensive diagnosis method with self-learning function | |
US11120350B2 (en) | Multilevel pattern monitoring method for industry processes | |
CN106096789A (en) | A kind of based on machine learning techniques can be from the abnormal industry control security protection of perception and warning system | |
CN111090939A (en) | Early warning method and system for abnormal working condition of petrochemical device | |
CN112632845B (en) | Data-based mini-reactor online fault diagnosis method, medium and equipment | |
CN109737045A (en) | Air compressor fault early warning method and system applied to chip production and related device | |
CN111768022A (en) | Equipment detection method and device for coal machine production equipment | |
CN107368054A (en) | Performance analysis management system and factory management system | |
CN109782728A (en) | A kind of multivariable monitoring method and system based on Savitzky-Golay filter | |
CN108762242A (en) | A kind of distributed fault detection method based on polylith canonical correlation analysis model | |
CN108241894A (en) | Fault Locating Method, equipment and storage medium | |
WO2018154558A1 (en) | Methods and systems for problem-alert aggregation | |
US12038742B2 (en) | Method for alarm handling in a processing system | |
Rao et al. | Alarm correlation analysis with applications to industrial alarm management | |
CN116778688A (en) | Machine room alarm event processing method, device, equipment and storage medium | |
CN111934903A (en) | Docker container fault intelligent prediction method based on time sequence evolution genes |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210604 |
|
CF01 | Termination of patent right due to non-payment of annual fee |