CN108764305A - A kind of improved colony intelligence machine learning fault diagnosis system - Google Patents
A kind of improved colony intelligence machine learning fault diagnosis system Download PDFInfo
- Publication number
- CN108764305A CN108764305A CN201810458510.1A CN201810458510A CN108764305A CN 108764305 A CN108764305 A CN 108764305A CN 201810458510 A CN201810458510 A CN 201810458510A CN 108764305 A CN108764305 A CN 108764305A
- Authority
- CN
- China
- Prior art keywords
- indicate
- fault diagnosis
- particle
- value
- module
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Monitoring And Testing Of Nuclear Reactors (AREA)
Abstract
The invention discloses a kind of improved colony intelligence machine learning fault diagnosis systems, for carrying out fault diagnosis, including data preprocessing module, principal component analysis module, Weighted Least Squares Support Vector Machines module and particle cluster algorithm module to Tennessee Yi Siman processes.The present invention carries out fault diagnosis and prediction to the important parameter index of Tennessee Yi Siman chemical processes, overcome the shortcomings of that existing Fault Diagnosis in Chemical Process industrial instrument forecast precision is not high, be easily affected by human factors, it introduces particle cluster algorithm module and Automatic Optimal is carried out to Weighted Least Squares Support Vector Machines parameter, it does not need artificial experience or repeatedly tests to adjust systematic parameter, to obtain the particle cluster algorithm Weighted Least Squares Support Vector Machines Tennessee Yi Siman Fault Diagnosis in Chemical Process systems for having optimal.The robustness of this fault diagnosis system forecast is good, and systematic parameter is few, very easy.
Description
Technical field
The present invention relates to fault diagnosis field, machine learning field and colony intelligence optimization algorithm fields, more particularly to one kind
In conjunction with the Tennessee Yi Siman process chemical industry fault diagnosis systems of machine learning and colony intelligence optimization algorithm.
Background technology
Fault Diagnosis in Chemical Process is always the key points and difficulties in chemical field, and a large amount of technique is contained in chemical process and is become
Amount, causes traditional Fault Diagnosis in Chemical Process technology increasingly to cannot be satisfied the requirement of Fault Diagnosis for Chemical Process.Computer is existing
Appearance in generation industrialization management, greatly increases production efficiency, reduces production cost, all brought for enterprise and country
Huge economic benefit.But chemical engineering industry procedures system is complicated, intercouples between internal system various pieces, certain
A part, which breaks down, just probably causes chain reaction, and whole system is caused to be unable to operate normally.In chemical process
In, if cannot debug in time, it is possible to cause catastrophic event.Such as 1984, one oil plant of Mexico was sent out
Raw oil gas explosion, causes people more than 7000 injured, people more than 500 is dead.The same year, the U.S. combinating carbide company of India's Bhopal
An insecticide factory nearly 450,000 tons of methyl isocyanate leakage has occurred, cause nearly 40,000 people injured, 2374 people are dead.Cause
How this, efficiently monitor chemical process, diagnose in time and be out of order and determine that failure cause is the hot spot faced now and difficulty
Point, the condition monitoring technology and trouble diagnosibility promoted in production process have important practical significance.Weighted least-squares
Support vector machines is a kind of supervised machine learning method, has complete Statistical Learning Theory basis, in number identification, face
Many distinctive advantages are shown in the fields such as identification, are suitable for small sample and non-linear high dimensional data, can carry out data point
Analysis, pattern-recognition and classification return.
Invention content
In order to overcome the shortcomings of that the forecast precision of current existing fault diagnosis technology is not high, easily be affected by human factors,
The purpose of the present invention is to provide a kind of robustness height, the colony intelligence machine learning fault diagnosis systems that parameter is few, very easy.
The technical solution adopted by the present invention to solve the technical problems is：A kind of improved colony intelligence machine learning failure is examined
Disconnected system, for carrying out fault diagnosis to Tennessee Yi Siman processes, including data preprocessing module, principal component analysis module,
Weighted Least Squares Support Vector Machines module and particle cluster algorithm module.Wherein：
Data preprocessing module：52 variables of Tennessee Yi Siman processesFor data prediction
The input of module.Since each variable has different units, different dimensions causes the mistake between data magnitude in order to prevent
Difference is first standardized all data, and standardization formula is as follows：
Wherein, mean indicates that the arithmetic mean of instantaneous value of each variable, std indicate the standard deviation of each variable,Indicate input variable
Value, subscript i indicates that ith detection, j indicate that jth ties up variable, x respectivelyijIndicate the value of input variable after standardizing as input
Data.Data after standardization are S={ xi1,xi2,...xi52}。
Principal component analysis module：Ensure to reduce system in the case where not reducing system accuracy by principal component analysis
Complexity.By the data S={ x after standardizationi1,xi2,...xi52Principal component analysis is carried out, retain 85% main component.
Weighted Least Squares Support Vector Machines module：It for establishing diagnostic system, is improved, is introduced by support vector machines
Weight coefficient：
A kind of nonlinear for being input to output is completed by error function minimum, topology is kept in mapping
Invariance.Wherein J indicate object function, w indicate inertia weight, ξ indicate error, C indicate penalty factor, u indicate weight coefficient,Indicate that kernel function mapping, b indicate that bigoted, x is input data, y is output data, subscript i indicates i-th of data, subscript T
The transposition of representing matrix.For the module using the few RBF kernel function K of excellent performance, required parameter, function is as follows：
Wherein, K is kernel function,Indicate that the average value of input data, σ indicate nuclear parameter.
Particle cluster algorithm module：For optimize Weighted Least Squares Support Vector Machines module RBF nuclear parameters σ and punishment because
Sub- C.Particle cluster algorithm is as follows：
(1) initialization of population：The position l of N number of particle is randomly generated in D dimension problem solution spacesi=(li1,li2,...liD),
With speed vi=(vi1,vi2,...viD), i=1,2 ..., N, setting particle learning rate c1=c2=2, setting inertia weight is most
Big value and minimum value wmax=0.8, wmin=0.2, population quantity N=20, setting maximum iteration iter are setmax=100.
Meanwhile iterations k=1 is set；
(2) more new particle：According to the speed of formula (4) more new particle and position；
Wherein, vi(k) it is the speed of particle i at the kth iteration, li(k) it is the positions of particle i at the kth iteration,
PbestIt is the locally optimal solution of particle experience；gbestIt is the globally optimal solution of all particle experiences, w is inertia weight, c1With
c2It is learning rate, r1And r2It is the random number between 0 to 1, wmaxAnd wminIt is the maximum value and minimum value of inertia weight,
itermaxIt is maximum iteration.
(3) P is updatedbest：Compare the fitness value and its individual optimal solution P of some particlebestIf fitness value is better than
Pbest, then the position of using the particle current is as Pbest, wherein fitness value f (x) calculated using following formula；
Wherein,Indicate the actual value of output variable,Indicate that the predicted value of output variable, n indicate number of samples.
(4) g is updatedbest：Compare the fitness value of all particles and the globally optimal solution g of populationbest, select adaptive optimal control
The position of the particle of angle value is as gbest；
(5) end condition judges：Judge whether iterations reach setting value or whether precision is less than 0.001, if reaching
It arrives, iteration ends, if not reaching, turns to (2) and continue iteration；RBF nuclear parameters σ and penalty factor when iteration ends are
The optimized parameter of Weighted Least Squares Support Vector Machines.
Tennessee Yi Siman processes share 21 failures, and the data of different faults are input to colony intelligence Optimized Diagnosis system
In be trained, establish fault diagnosis model.
When the data of unknown failure are input to this fault diagnosis system, diagnostic result display instrument shows diagnostic result.
Beneficial effects of the present invention are mainly manifested in：Important parameter index of the present invention to Tennessee Yi Siman chemical processes
Fault diagnosis and prediction is carried out, overcomes that existing Fault Diagnosis in Chemical Process industrial instrument forecast precision is not high, is easily affected by human factors
Deficiency, introduce particle cluster algorithm module and Automatic Optimal carried out to Weighted Least Squares Support Vector Machines parameter, need not be artificial
Experience is repeatedly tested to adjust systematic parameter, to obtain the particle cluster algorithm weighted least-squares supporting vector for having optimal
Machine Tennessee Yi Siman Fault Diagnosis in Chemical Process systems.The robustness of this fault diagnosis system forecast is good, and systematic parameter is few, very simple
Just.
Description of the drawings
Fig. 1 is a kind of improved colony intelligence machine learning fault diagnosis system structural schematic diagram；
Fig. 2 is colony intelligence Optimized Diagnosis system structure diagram；
Fig. 3 is Tennessee Yi Siman process flow sheets.
Specific implementation mode
The present invention is illustrated below according to attached drawing.
Referring to Fig.1, a kind of improved colony intelligence machine learning fault diagnosis system, including Tennessee Yi Siman processes 1, use
In the measurement easily field intelligent instrument 2 of survey variable, the control station 3 for measuring performance variable, the database 4 for storing data, group
Intelligent optimization diagnostic system 5 and diagnostic result display instrument 6.The field intelligent instrument 2, control station 3 and Tennessee Yi Siman mistakes
Journey 1 connects, and the field intelligent instrument 2, control station 3 are connect with database 4, the database 4 and colony intelligence Optimized Diagnosis system
The input terminal connection of system 5, the output end based on colony intelligence Optimized Diagnosis system 5 are connect with diagnostic result display instrument 6.
Variable with reference to Fig. 3 Tennessee Yi Siman processes is as shown in table 1.
Table 1：Tennessee Yi Siman process variables
| Number | Process variable | Number | Process variable |
| 1 | Feed A (flow tube 1) | 27 | Reactor E chargings (flow tube 6) |
| 2 | Feed D (flow tube 2) | 28 | Reactor F chargings (flow tube 6) |
| 3 | Feed E (flow tube 3) | 29 | Reactor A feeds (flow tube 9) |
| 4 | Combined feed (flow tube 4) | 30 | Reactor B feeds (flow tube 9) |
| 5 | Recirculating mass (flow tube 8) | 31 | Reactor C chargings (flow tube 9) |
| 6 | Reactor feed rate | 32 | Reactor D chargings (flow tube 9) |
| 7 | Reactor pressure | 33 | Reactor E chargings (flow tube 9) |
| 8 | Reactor liquid level | 34 | Reactor F chargings (flow tube 9) |
| 9 | Temperature of reactor | 35 | Reactor G chargings (flow tube 9) |
| 10 | Capacity (flow tube 9) | 36 | Reactor H chargings (flow tube 9) |
| 11 | Gas-liquid separator temperature | 37 | Stripper D flows (flow tube 11) |
| 12 | Gas-liquid separator liquid level | 38 | Stripper E flows (flow tube 11) |
| 13 | Gas-liquid separator temperature | 39 | Stripper F flows (flow tube 11) |
| 14 | Gas-liquid separator bottom of tower flow (stream 10) | 40 | Stripper G flows (flow tube 11) |
| 15 | Stripper liquid level | 41 | Stripper H flows (flow tube 11) |
| 16 | Pressure of stripping tower | 42 | D feed rates |
| 17 | Stripper bottom of tower flow (flow tube 11) | 43 | E feed rates |
| 18 | Stripper temperature | 44 | A feed rates |
| 19 | Stripper steam flow | 45 | Total feed rate |
| 20 | Compressor horsepower | 46 | Compressor recycle valve |
| 21 | Reactor coolant water outlet temperature | 47 | Drain valve |
| 22 | Separator cooling water outlet temperature | 48 | Knockout drum tank flow quantity |
| 23 | Reactor A feeds (flow tube 6) | 49 | Stripper liquid product flow |
| 24 | Reactor B feeds (flow tube 6) | 50 | Stripper water flow |
| 25 | Reactor C chargings (flow tube 6) | 51 | Reactor cooling water flow |
| 26 | Reactor D chargings (flow tube 6) | 52 | Condenser cooling water flow |
Input variable of the Tennessee Yi Siman process datas as colony intelligence Optimized Diagnosis system 5.Pass through manual sampling point
Analysis obtains, and analysis acquisition in every 4 hours is primary.
With reference to Fig. 2, the colony intelligence Optimized Diagnosis system 5 further includes：
Data preprocessing module 7：52 variables of Tennessee Yi Siman processesLocate in advance for data
Manage the input of module.Since each variable has different units, different dimensions causes between data magnitude in order to prevent
Error is first standardized all data, and standardization formula is as follows：
Wherein, mean indicates that the arithmetic mean of instantaneous value of each variable, std indicate the standard deviation of each variable,Indicate input variable
Value, subscript i indicates that ith detection, j indicate that jth ties up variable, x respectivelyijIndicate the value of input variable after standardizing as input
Data.Data after standardization are S={ xi1,xi2,...xi52}。
Principal component analysis module 8：Ensure to reduce system in the case where not reducing system accuracy by principal component analysis
Complexity.By the data S={ x after standardizationi1,xi2,...xi52Principal component analysis is carried out, retain 85% main component.
Weighted Least Squares Support Vector Machines module 9：It for establishing diagnostic system, is improved, is drawn by support vector machines
Enter weight coefficient：
A kind of nonlinear for being input to output is completed by error function minimum, topology is kept in mapping
Invariance.Wherein J indicate object function, w indicate inertia weight, ξ indicate error, C indicate penalty factor, u indicate weight coefficient,Indicate that kernel function mapping, b indicate that bigoted, x is input data, y is output data, subscript i indicates i-th of data, subscript T
The transposition of representing matrix.For the module using the few RBF kernel function K of excellent performance, required parameter, function is as follows：
Wherein, K is kernel function,Indicate that the average value of input data, σ indicate nuclear parameter.
Particle cluster algorithm module 10：RBF nuclear parameters σ for optimizing Weighted Least Squares Support Vector Machines module and punishment
Factor C.Particle cluster algorithm is as follows：
(1) initialization of population：The position l of N number of particle is randomly generated in D dimension problem solution spacesi=(li1,li2,...liD),
With speed vi=(vi1,vi2,...viD), i=1,2 ..., N, setting particle learning rate c1=c2=2, setting inertia weight is most
Big value and minimum value wmax=0.8, wmin=0.2, population quantity N=20, setting maximum iteration iter are setmax=100.
Meanwhile iterations k=1 is set；
(2) more new particle：According to the speed of formula (4) more new particle and position；
Wherein, vi(k) it is the speed of particle i at the kth iteration, li(k) it is the positions of particle i at the kth iteration,
PbestIt is the locally optimal solution of particle experience；gbestIt is the globally optimal solution of all particle experiences, w is inertia weight, c1With
c2It is learning rate, r1And r2It is the random number between 0 to 1, wmaxAnd wminIt is the maximum value and minimum value of inertia weight,
itermaxIt is maximum iteration.
(3) P is updatedbest：Compare the fitness value and its individual optimal solution P of some particlebestIf fitness value is better than
Pbest, then the position of using the particle current is as Pbest, wherein fitness value f (x) calculated using following formula；
Wherein,Indicate the actual value of output variable,Indicate that the predicted value of output variable, n indicate number of samples.
(4) g is updatedbest：Compare the fitness value of all particles and the globally optimal solution g of populationbest, select adaptive optimal control
The position of the particle of angle value is as gbest；
(5) end condition judges：Judge whether iterations reach setting value or whether precision is less than 0.001, if reaching
It arrives, iteration ends, if not reaching, turns to (2) and continue iteration；RBF nuclear parameters σ and penalty factor when iteration ends are
The optimized parameter of Weighted Least Squares Support Vector Machines.
Tennessee Yi Siman processes share 21 failures, and the data of different faults are input to colony intelligence Optimized Diagnosis system
It is trained in 5, establishes fault diagnosis model.
When the data of unknown failure are input to this fault diagnosis system, diagnostic result display instrument 6 shows diagnostic result.
The embodiment of the present invention is used for illustrating the present invention, rather than limits the invention, in the spirit of the present invention
In scope of the claims, to any modifications and changes that the present invention makes, protection scope of the present invention is both fallen within.
Claims (5)
1. a kind of improved colony intelligence machine learning fault diagnosis system is examined for carrying out failure to Tennessee Yi Siman processes
It is disconnected, it is characterised in that：Including data preprocessing module, principal component analysis module, Weighted Least Squares Support Vector Machines module and
Particle cluster algorithm module.
2. improved colony intelligence machine learning fault diagnosis system according to claim 1, which is characterized in that the data are pre-
The input of processing module is 52 variables of Tennessee Yi Siman processesSince each variable has not
With unit, in order to prevent different dimensions cause the error between data magnitude, first all data are standardized,
It is as follows to standardize formula：
Wherein, mean indicates that the arithmetic mean of instantaneous value of each variable, std indicate the standard deviation of each variable,Indicate the value of input variable,
Subscript i indicates that ith detection, j indicate that jth ties up variable, x respectivelyijIndicate the value of input variable after standardizing as input data.
Data after standardization are S={ xi1,xi2,...xi52}。
3. improved colony intelligence machine learning fault diagnosis system according to claim 1, which is characterized in that the principal component
Analysis module ensures the complexity of the reduction system in the case where not reducing system accuracy by principal component analysis.It will standardization
Data S={ x afterwardsi1,xi2,...xi52Principal component analysis is carried out, retain 85% main component.
4. improved colony intelligence machine learning fault diagnosis system according to claim 1, which is characterized in that the weighting is most
Small two multiply support vector machines module for establishing diagnostic system, are improved by support vector machines, introduce weight coefficient：
A kind of nonlinear for being input to output is completed by error function minimum, topological novariable is kept in mapping
Property.Wherein J indicate object function, w indicate inertia weight, ξ indicate error, C indicate penalty factor, u indicate weight coefficient,Table
Show that kernel function mapping, b indicate that bigoted, x is input data, y is output data, subscript i indicates that i-th of data, subscript T indicate square
The transposition of battle array.For the module using the few RBF kernel function K of excellent performance, required parameter, function is as follows：
Wherein, K is kernel function,Indicate that the average value of input data, σ indicate nuclear parameter.
5. improved colony intelligence machine learning fault diagnosis system according to claim 1, which is characterized in that the population
Algoritic module is used to optimize the RBF nuclear parameters σ and penalty factor of Weighted Least Squares Support Vector Machines module.Particle cluster algorithm
It is as follows：
(1) initialization of population：The position l of N number of particle is randomly generated in D dimension problem solution spacesi=(li1,li2,...liD), and speed
Spend vi=(vi1,vi2,...viD), i=1,2 ..., N, setting particle learning rate c1=c2=2, inertia weight maximum value is set
With minimum value wmax=0.8, wmin=0.2, population quantity N=20, setting maximum iteration iter are setmax=100.Meanwhile
Iterations k=1 is set；
(2) more new particle：According to the speed of formula (4) more new particle and position；
Wherein, vi(k) it is the speed of particle i at the kth iteration, li(k) it is the positions of particle i at the kth iteration, Pbest
It is the locally optimal solution of particle experience；gbestIt is the globally optimal solution of all particle experiences, w is inertia weight, c1And c2It is
Learning rate, r1And r2It is the random number between 0 to 1, wmaxAnd wminIt is the maximum value and minimum value of inertia weight, itermax
It is maximum iteration.
(3) P is updatedbest：Compare the fitness value and its individual optimal solution P of some particlebestIf fitness value is better than Pbest,
The position for then using the particle current is as Pbest, wherein fitness value f (x) calculated using following formula；
Wherein,Indicate the actual value of output variable,Indicate that the predicted value of output variable, n indicate number of samples.
(4) g is updatedbest：Compare the fitness value of all particles and the globally optimal solution g of populationbest, select adaptive optimal control angle value
Particle position as gbest；
(5) end condition judges：Judge whether iterations reach setting value or whether precision is less than 0.001, if reaching, repeatedly
In generation, terminates, if not reaching, turns to (2) and continues iteration；RBF nuclear parameters σ and penalty factor when iteration ends are to weight most
Small two multiply the optimized parameter of support vector machines.
Tennessee Yi Siman processes share 21 failures, by the data of different faults be input in colony intelligence Optimized Diagnosis system into
Row training, establishes fault diagnosis model.
When the data of unknown failure are input to this fault diagnosis system, diagnostic result display instrument shows diagnostic result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810458510.1A CN108764305A (en) | 2018-05-14 | 2018-05-14 | A kind of improved colony intelligence machine learning fault diagnosis system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810458510.1A CN108764305A (en) | 2018-05-14 | 2018-05-14 | A kind of improved colony intelligence machine learning fault diagnosis system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108764305A true CN108764305A (en) | 2018-11-06 |
Family
ID=64006756
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810458510.1A Pending CN108764305A (en) | 2018-05-14 | 2018-05-14 | A kind of improved colony intelligence machine learning fault diagnosis system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108764305A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109635880A (en) * | 2019-01-08 | 2019-04-16 | 浙江大学 | Malfunction Diagnosis for Coal-Mining Machine system based on strong robust adaptive beamforming algorithm |
| CN109635879A (en) * | 2019-01-08 | 2019-04-16 | 浙江大学 | A kind of Malfunction Diagnosis for Coal-Mining Machine system that parameter is optimal |
| CN110008533A (en) * | 2019-03-18 | 2019-07-12 | 浙江大学 | A kind of air separation plant energy-consumption monitoring system based on hybrid modeling |
| CN110750756A (en) * | 2019-10-01 | 2020-02-04 | 深圳市行健自动化股份有限公司 | Method for checking and diagnosing real-time online instrument by optimal support vector machine algorithm |
| CN112069567A (en) * | 2020-08-07 | 2020-12-11 | 湖北交投十巫高速公路有限公司 | Method for predicting compressive strength of concrete based on random forest and intelligent algorithm |
| CN116203907A (en) * | 2023-03-27 | 2023-06-02 | 淮阴工学院 | Chemical process fault diagnosis alarm method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101021723A (en) * | 2006-12-22 | 2007-08-22 | 浙江大学 | Melt index detection fault diagnozing system and method in propylene polymerization production |
| CN102364501A (en) * | 2011-09-14 | 2012-02-29 | 哈尔滨工程大学 | Method for reproducing two-dimensional defect of petroleum pipeline PSO-BP (Particle Swarm Optimization-Back-Propagation) neural network |
| CN102778538A (en) * | 2012-07-06 | 2012-11-14 | 广东电网公司电力科学研究院 | Soft measuring method based on improved SVM (Support Vector Machine) for measuring boiler unburned carbon content in fly ash |
| CN104463381A (en) * | 2015-01-04 | 2015-03-25 | 福州大学 | Building energy consumption predication method based on KPCA and WLSSVM |
| CN107133642A (en) * | 2017-04-25 | 2017-09-05 | 东北大学 | A kind of priori method for diagnosing faults based on Tennessee Yi Siman processes |
| CN107632592A (en) * | 2017-09-01 | 2018-01-26 | 南通大学 | Nonlinear time-varying procedure fault monitoring method based on efficient recursion core pivot element analysis |
-
2018
- 2018-05-14 CN CN201810458510.1A patent/CN108764305A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101021723A (en) * | 2006-12-22 | 2007-08-22 | 浙江大学 | Melt index detection fault diagnozing system and method in propylene polymerization production |
| CN102364501A (en) * | 2011-09-14 | 2012-02-29 | 哈尔滨工程大学 | Method for reproducing two-dimensional defect of petroleum pipeline PSO-BP (Particle Swarm Optimization-Back-Propagation) neural network |
| CN102778538A (en) * | 2012-07-06 | 2012-11-14 | 广东电网公司电力科学研究院 | Soft measuring method based on improved SVM (Support Vector Machine) for measuring boiler unburned carbon content in fly ash |
| CN104463381A (en) * | 2015-01-04 | 2015-03-25 | 福州大学 | Building energy consumption predication method based on KPCA and WLSSVM |
| CN107133642A (en) * | 2017-04-25 | 2017-09-05 | 东北大学 | A kind of priori method for diagnosing faults based on Tennessee Yi Siman processes |
| CN107632592A (en) * | 2017-09-01 | 2018-01-26 | 南通大学 | Nonlinear time-varying procedure fault monitoring method based on efficient recursion core pivot element analysis |
Non-Patent Citations (3)
| Title |
|---|
| SHIMING HE 等: "A novel fault diagnosis method based on optimal relevance vector machine", 《NEUROCOMPUTING》 * |
| SHIMING HE 等: "An Effective Fault Diagnosis Approach Based on Optimal Weighted Least Squares Support Vector Machine", 《THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING》 * |
| 王斌 等: "基于改进 KFDA 和 RW ν -SVM 的化工生产***故障快速诊断", 《中国安全科学学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109635880A (en) * | 2019-01-08 | 2019-04-16 | 浙江大学 | Malfunction Diagnosis for Coal-Mining Machine system based on strong robust adaptive beamforming algorithm |
| CN109635879A (en) * | 2019-01-08 | 2019-04-16 | 浙江大学 | A kind of Malfunction Diagnosis for Coal-Mining Machine system that parameter is optimal |
| CN109635879B (en) * | 2019-01-08 | 2023-04-07 | 浙江大学 | Coal mining machine fault diagnosis system with optimal parameters |
| CN109635880B (en) * | 2019-01-08 | 2023-06-27 | 浙江大学 | Coal mining machine fault diagnosis system based on robust self-adaptive algorithm |
| CN110008533A (en) * | 2019-03-18 | 2019-07-12 | 浙江大学 | A kind of air separation plant energy-consumption monitoring system based on hybrid modeling |
| CN110750756A (en) * | 2019-10-01 | 2020-02-04 | 深圳市行健自动化股份有限公司 | Method for checking and diagnosing real-time online instrument by optimal support vector machine algorithm |
| CN110750756B (en) * | 2019-10-01 | 2023-06-20 | 深圳市行健自动化股份有限公司 | Real-time on-line instrument checksum diagnosis method through optimal support vector machine algorithm |
| CN112069567A (en) * | 2020-08-07 | 2020-12-11 | 湖北交投十巫高速公路有限公司 | Method for predicting compressive strength of concrete based on random forest and intelligent algorithm |
| CN112069567B (en) * | 2020-08-07 | 2024-01-12 | 湖北交投十巫高速公路有限公司 | Method for predicting compressive strength of concrete based on random forest and intelligent algorithm |
| CN116203907A (en) * | 2023-03-27 | 2023-06-02 | 淮阴工学院 | Chemical process fault diagnosis alarm method and system |
| CN116203907B (en) * | 2023-03-27 | 2023-10-20 | 淮阴工学院 | Chemical process fault diagnosis alarm method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764305A (en) | A kind of improved colony intelligence machine learning fault diagnosis system | |
| CN110320892B (en) | Sewage treatment equipment fault diagnosis system and method based on L asso regression | |
| CN106355030B (en) | A kind of fault detection method based on analytic hierarchy process (AHP) and Nearest Neighbor with Weighted Voting Decision fusion | |
| CN103914064B (en) | Based on the commercial run method for diagnosing faults that multi-categorizer and D-S evidence merge | |
| CN106647718B (en) | Nonlinear industrial processes fault detection method based on the slow signature analysis of Bayes's core | |
| CN105046402B (en) | A kind of state evaluating method applied to secondary equipment of intelligent converting station | |
| CN111553590A (en) | Radar embedded health management system | |
| CN108536128A (en) | A kind of machine learning fault diagnosis system of parameter optimization | |
| CN108921230A (en) | Method for diagnosing faults based on class mean value core pivot element analysis and BP neural network | |
| CN112036089A (en) | Coal mill fault early warning method based on DPC-MND and multivariate state estimation | |
| CN108898311A (en) | A kind of data quality checking method towards intelligent distribution network repairing dispatching platform | |
| CN105334823B (en) | Industrial process fault detection method based on the linear dynamic system model for having supervision | |
| CN108536130A (en) | A kind of Fault Diagnosis in Chemical Process system of colony intelligence optimizing | |
| CN109583520A (en) | A kind of state evaluating method of cloud model and genetic algorithm optimization support vector machines | |
| CN101021723A (en) | Melt index detection fault diagnozing system and method in propylene polymerization production | |
| CN109710661A (en) | Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis | |
| CN110443481B (en) | Power distribution automation terminal state evaluation system and method based on hybrid K-nearest neighbor algorithm | |
| CN109298633A (en) | Chemical production process fault monitoring method based on adaptive piecemeal Non-negative Matrix Factorization | |
| CN116383636A (en) | Coal mill fault early warning method based on PCA and LSTM fusion algorithm | |
| CN108345284A (en) | A kind of quality dependent failure detection method becoming gauge block based on two | |
| CN109507972A (en) | Industrial processes fault monitoring method based on layer-stepping non-gaussian monitoring algorithm | |
| WO2021114320A1 (en) | Wastewater treatment process fault monitoring method using oica-rnn fusion model | |
| CN111767657B (en) | Nuclear power system fault diagnosis method and system | |
| CN108830006B (en) | Linear-nonlinear industrial process fault detection method based on linear evaluation factor | |
| CN105741184B (en) | Transformer state evaluation method and device |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181106 |
|
| WD01 | Invention patent application deemed withdrawn after publication |