CN109799802B - Sensor fault diagnosis and fault tolerance control method in molecular weight distribution control - Google Patents
Sensor fault diagnosis and fault tolerance control method in molecular weight distribution control Download PDFInfo
- Publication number
- CN109799802B CN109799802B CN201811489086.3A CN201811489086A CN109799802B CN 109799802 B CN109799802 B CN 109799802B CN 201811489086 A CN201811489086 A CN 201811489086A CN 109799802 B CN109799802 B CN 109799802B
- Authority
- CN
- China
- Prior art keywords
- fault
- molecular weight
- output
- representing
- probability density
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000006870 function Effects 0.000 claims abstract description 66
- 238000006243 chemical reaction Methods 0.000 claims abstract description 10
- 238000013528 artificial neural network Methods 0.000 claims abstract description 9
- 238000005516 engineering process Methods 0.000 claims abstract description 4
- 238000012360 testing method Methods 0.000 claims abstract description 4
- 230000009466 transformation Effects 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 20
- 238000013461 design Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000004519 manufacturing process Methods 0.000 description 5
- PPBRXRYQALVLMV-UHFFFAOYSA-N Styrene Chemical compound C=CC1=CC=CC=C1 PPBRXRYQALVLMV-UHFFFAOYSA-N 0.000 description 2
- 238000001311 chemical methods and process Methods 0.000 description 2
- 238000002485 combustion reaction Methods 0.000 description 2
- 238000000227 grinding Methods 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 239000003999 initiator Substances 0.000 description 2
- 239000000178 monomer Substances 0.000 description 2
- 238000006116 polymerization reaction Methods 0.000 description 2
- 239000000376 reactant Substances 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012662 bulk polymerization Methods 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000003756 stirring Methods 0.000 description 1
- 230000001988 toxicity Effects 0.000 description 1
- 231100000419 toxicity Toxicity 0.000 description 1
Images
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides a sensor fault diagnosis and fault tolerance control method in molecular weight distribution control, which comprises the following steps: obtaining the molecular weight distribution in the chemical reaction process on line by a mass spectrometer or a laser test technology to obtain an output probability density function, and approximating the output probability density function by a B-spline neural network; establishing a state space model of the system, obtaining parameters of the state space model after transformation by a scanning identification method, and designing a learning fault diagnosis observer; monitoring whether a sensor has a fault or not in real time by using a learning fault diagnosis observer, and estimating fault information in time when the sensor has the fault; compensating the output weight of the system by using the estimated fault information; and inputting the designed control signal into the system to enable the system to output the molecular weight distribution to track the distribution of the expected molecular weight. The invention can provide guarantee for reliability and safety in the system operation process and protect the life safety and property safety of related enterprise personnel.
Description
Technical Field
The invention relates to the technical field of sensor fault diagnosis and fault-tolerant control, in particular to a sensor fault diagnosis and fault-tolerant control method in molecular weight distribution control.
Background
The molecular weight distribution control of the system in chemical process control is the application of a typical random distribution control system and is paid much attention in the related fields. The concept of the random distribution system is provided by the production process of the paper industry, and is extended to the fields of ore grinding, boiler combustion, chemical reaction and the like, and because the random distribution system is widely applied, the random distribution system is paid attention by relevant experts and scholars in relevant fields. Because the related industries limited by production environments mostly adopt mechanized and intelligent production lines, a large number of components such as sensors and actuators are needed, and the long-time uninterrupted work in the production environments is difficult to avoid faults, and serious consequences are caused if the faults cannot be diagnosed in time and fault-tolerant control is carried out. The control object of the random distribution system is an output probability density function of the whole system, the input, noise and fault types of the random distribution system do not necessarily obey Gaussian distribution, and the random variable of the random distribution system is non-Gaussian and is a non-Gaussian random distribution system. The disturbance of the random distribution system is mostly non-gaussian, the distribution of the disturbance is difficult to describe, and the performance of fault diagnosis is influenced.
The traditional fault diagnosis method is mainly based on a model design observer or a filter to carry out fault diagnosis. The existing fault diagnosis and fault tolerance control method only aims at the actuator of a non-Gaussian random distribution system to carry out fault diagnosis and fault tolerance control on the sensor, and does not consider the fault diagnosis and fault tolerance control on the sensor when the sensor fails.
Disclosure of Invention
The invention provides a sensor fault diagnosis and fault-tolerant control method in molecular weight distribution control, aiming at the technical problems that the existing non-Gaussian random distribution system only carries out fault diagnosis and fault-tolerant control on an actuator and does not consider the current situation that a sensor has faults, the problem of sensor fault diagnosis of the non-Gaussian random distribution control system is solved, and the problem of fault-tolerant control when the sensor of the non-Gaussian random distribution control system has faults is solved.
In order to achieve the purpose, the technical scheme of the invention is realized as follows: a sensor fault diagnosis and fault tolerance control method in molecular weight distribution control comprises the following steps:
step 100: obtaining the molecular weight distribution in the chemical reaction process on line by a mass spectrometer or a laser test technology to obtain an output probability density function, and approximating the output probability density function by a B-spline neural network;
step 200: establishing a state space model of the system according to the approximated output probability density function, obtaining parameters of the state space model after transformation by a scanning identification method, and designing a learning fault diagnosis observer;
step 300: the designed learning fault diagnosis observer is implemented on a computer, whether a fault occurs in a sensor is monitored in real time, and when the fault occurs, the learning fault diagnosis observer estimates fault information in time;
step 400: when the system has a fault, the fault diagnosis observer accurately estimates fault information, and compensates the output weight of the system by using the estimated fault information;
step 500: and inputting a designed control signal into the system to enable the system to output a molecular weight distribution tracking the distribution of the expected molecular weight.
The output probability density function in step 100 is given by:
wherein phi isi(y) represents the ith B-spline basis function, n represents the number of B-spline basis functions, y represents the output of the system, t represents time, u (t) is the control input, ω isi(t) is the weight of the ith B-spline basis function;
since the integral of the output probability density function γ (y, u (t)) in its defined interval [ a, B ] is equal to 1, the output probability density function is approximated by a linear B-spline basis function neural network using a B-spline neural network as follows:
γ(y,u(t))=C(y)V(t)+T(y);
wherein the linear radial basis functionAndparameter(s)φ1(y),φ2(y),φ3(y) are linear B-spline basis functions of the system output y; v (t) ═ ω1(t),ω1(t),…ω3(t)]TTo output a weight vector, omegai(t) is the ith B-spline basis functionThe weight, i, is 1,2, 3.
The establishment of the state space model of the system according to the approximated probability density function in step 200 is as follows:
introducing a new state variable by using the output weight of the systemThe state space model of the system becomes:
wherein x (t) is the state variable of the original system, u (t) is the control input, f (t) is the fault vector, V (t) is the weight vector,representing the first derivative, A, of the state variable x (t)sIdentity matrix, x, representing the same dimension as the state variables(t) represents a state variable related to the output weight,represents a state variable xs(t) first derivative, A, B, D and G are parameter matrices, respectively;
wherein the state variable of the new system isThe parameter matrices are respectivelyIpRepresenting a P-dimensional identity matrix;
the design learning fault diagnosis observer is as follows:
wherein,representing the first derivative of the new system state variable observations,an observed value representing a new system state variable,representing the observer gain matrix, epsilon (t) representing the residual between the observed value and the actual value,the weight value of the observation is shown,representing an observed probability density function, Z (t) representing a vector with the same dimension as the fault, Z (t-tau) representing the value of the vector Z (t) before time tau, epsilon (t-tau) representing the residual error before time tau, K1、K2And W represents an observation system parameter,representing the derivative of the fault observation.
In step 400, the estimated fault information is used to compensate the output weight of the system as follows:wherein, vc (t) is the compensated output weight.
In the step 500, the probability density function of the desired molecular weight is represented as γ (y, u (t)) ═ c (y) Vg + t (y), and Vg is the desired weight; the following switching function and control law are designed to realize the molecular weight distribution tracking expected distribution after the fault:
the switching function is designed as follows:
where H and K both represent parameters of the switching function, S (t) represents the switching function,representing a weight error vector;
when the system fault is detected to occur, the control law of the system is adjusted to be as follows:
wherein, K3M is a parameter satisfying a system stability condition,alpha represents a parameter that satisfies a stable condition as a derivative of the fault observation;
solving, by a computer, the following linear matrix inequalities:
0<(6+3σ)K1 TK1≤I;
P2(A1+MK3)+(A1+MK3)TP2+Q2≤0;
solving parameters of the fault diagnosis observer meeting the conditions and parameters meeting control input conditions for tracking given distribution, and achieving the purposes of fault-tolerant control and tracking expected molecular weight distribution; wherein R is1,Q1,Q2,P1,P2For proper dimension positive determination of the symmetric matrix, sigma, gamma1Is a small normal number, A1=DAD-1I is a one-dimensional identity matrix; Σ denotes a 1 × 2 matrix.
The invention has the beneficial effects that: the designed fault diagnosis observer carries out fault diagnosis on the non-Gaussian random distribution system sensor, and the designed fault-tolerant control scheme can effectively compensate after the system has sensor faults, so that the probability density function output by the system tracks expected output. The invention can be used for the fault diagnosis and fault-tolerant control of the sensor in the random process described by the output available probability density function of chemical reaction, paper making, ore grinding, boiler combustion and the like, and can provide guarantee for the reliability and safety in the system operation process and protect the life safety and property safety of related enterprise personnel because the invention relates to the production environments of high temperature, high pressure, toxicity and the like mostly.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic view of the closed-loop control of molecular weight distribution in a polymerization process.
FIG. 3 is a fault and fault estimation diagram of the present invention;
FIG. 4 is a graph of a desired output Probability Density Function (PDF) for the system of the present invention;
FIG. 5 is a graph of the Probability Density Function (PDF) of the output of the system without fault tolerance control of the present invention;
FIG. 6 is a graph of the output Probability Density Function (PDF) with fault tolerance control for the system of the present invention;
FIG. 7 is a two-dimensional effect graph of the expected probability density function and the actual output probability density function for fault-tolerant control of the system of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.
As shown in fig. 1, a method for diagnosing and fault-tolerant molecular weight distribution control sensor faults in a chemical process includes the following steps:
step 100: the probability density function of the molecular weight distribution in the chemical reaction process is obtained on line by a mass spectrometer or a laser test technology, and the probability density function is approximated by a B-spline neural network.
The approximation principle is as follows, and the probability density function is given by:
wherein phi isi(y) represents the ith B-spline basis function, n represents the number of B-spline basis functions, y represents the output of the system, t represents time, u (t) is the control input, ω isiAnd (t) is the weight of the ith B spline basis function. Because the output probability density function gamma (y, u (t)) is in the defined interval [ a, b ]]Is equal to 1, i.e.:
ω1(t)φ1(y)+ω2(t)φ2(y)…+ωn(t)φn(y)=1 (2)
specifically, the following 3 linear B-spline basis function neural networks are used to approximate the probability density function, that is, n in formula (2) is 3, and the linear B-spline basis function is expressed as follows:
wherein,due to the fact thatLinear radial basis functionAndwherein the parameters The probability density function can also be expressed as:
γ(y,u(t))=C(y)V(t)+T(y) (4)
wherein v (t) ═ ω1(t),ω1(t),ω3(t)]TFor outputting the weight vector, a and b represent the defined interval of the probability density function.
Step 200: and establishing a state space model of the system according to the probability density function, obtaining parameters of the state space model after transformation by a scanning identification method, and designing and learning a fault diagnosis observer.
The state space model of the system is as follows:
Wherein x (t) is the state variable of the original system, u (t) is the control input, f (t) is the fault vector, V (t) is the weight vector,representing the first derivative, A, of the state variable x (t)sIdentity matrix, x, representing the same dimension as the state variables(t) represents a state variable related to the output weight,represents a state variable xsThe first derivatives of (t), A, B, D and G are parameter matrices, respectively.
The design learning fault diagnosis observer is as follows:
wherein,representing the first derivative of the new system state variable observations,an observed value representing a new system state variable,representing the observer gain matrix, epsilon (t) representing the residual between the observed value and the actual value,the weight value of the observation is shown,representing an observed probability density function, Z (t) representing a vector with the same dimension as the fault, Z (t-tau) representing the value of the vector Z (t) before time tau, epsilon (t-tau) representing the residual error before time tau, K1、K2And W represents an observation system parameter,representing the derivative of the fault observation.
Step 300: the method is implemented by a designed learning fault diagnosis observer on a computer, and the specific principle is that an observation system is constructed through input and output information of an original system, the state of the observation system approaches the state of the original system, when a fault occurs, a residual error is constructed based on the output difference of the original system and the observation system, and fault information is estimated by utilizing the residual error. And monitoring whether the sensor has faults or not in real time through a computer, and learning a fault diagnosis observer to estimate fault information in time when the sensor has the faults.
Step 400: when the system has a fault, the fault diagnosis observer accurately estimates fault information, and as shown in fig. 3, compensates the output weight of the system by using the estimated fault information:
step 500: and inputting a designed control signal into the system to enable the system to output a molecular weight distribution tracking expected distribution. As shown in fig. 4-7.
Vg is a desired weight, the probability density function of the desired molecular weight can be expressed as γ (y, u (t) ═ c (y) Vg + t (y), and the following switching function and control law are designed to realize the tracking of the molecular weight distribution after the failure to the desired distribution:
the switching function is designed as follows:
h and K both represent parameters of the switching function, S (t) represents the switching function,representing the weight error vector.
When a system fault is detected, the control input of the system is adjusted to the formula (10), so that the purposes of fault-tolerant control and tracking expected molecular weight distribution are realized:
wherein, K3In order to satisfy the parameters of the system stability condition,for the fault observation, α represents a parameter satisfying the stable condition.
Solving linear matrix inequalities (11), (12), (13) and (14) by a computer to obtain the parameters of the fault diagnosis observer satisfying the conditions and the parameters satisfying the control input conditions for tracking the given distribution as follows:
0<(6+3σ)K1 TK1≤I (12)
P2(A1+MK3)+(A1+MK3)TP2+Q2≤0 (14)
wherein R is1,Q1,Q2,P1,P2For proper dimension positive determination of the symmetric matrix, sigma, gamma1Is a small normal number, A1=DAD-1And M is DB, and I is a one-dimensional identity matrix. Solving the parameters of the fault diagnosis observer meeting the conditions and the parameters of the control input conditions meeting the tracking given distribution, realizing the purposes of fault-tolerant control and tracking the expected molecular weight distribution, and obtaining the following results:
K1=0.07,K2=-0.5,H=[0.4 -0.1],K3=[12.1635 0.0389]
the modeling and control method is applied to controlling the molecular weight distribution shape in the dynamic process of styrene bulk polymerization. The molecular weight distribution closed-loop control schematic of the polymerization process is shown in FIG. 2.
The system model may be represented as:
wherein,is the average residence time of the reactants; i is0Is the initial concentration (mol. ml) of the initiator-1) (ii) a I is the initiator concentration (mol. ml)-1);M0Is the initial concentration (mol. ml) of the monomer-1) (ii) a M is the concentration of the monomer (mol. ml)-1);Ki,Kp,KtrmIs the reaction rate constant; kd,KI,KMIs in phase with the control inputA constant of off; ri(i ═ 1, 2.., q) is a radical.
Selecting state variablesThe system parameter matrix is respectively expressed asAndK1,K2is a fault-related parameter. The parameters of the state space model obtained by the scan recognition method are as follows:
if the temperature sensor in the reaction kettle breaks down, the measured temperature is not consistent with the actual temperature in the kettle, the injection amount of each reaction component or the dosage of heating oil can be influenced, and the molecular weight distribution of products in the kettle is further influenced. When a fault occurs, the observer can quickly acquire the fault information and feed back the fault information to the control system. The control system will switch to the pre-designed control input (10) for fault tolerant control. The molecular weight distribution of the reaction product is adjusted to a desired distribution by adjusting the temperature and the ratio of the reactants in the chemical reaction vessel by changing the amount of monomer or initiator injected, the amount of heating oil used, the stirring rate, and the like.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (4)
1. A sensor fault diagnosis and fault tolerance control method in molecular weight distribution control is characterized by comprising the following steps:
step 100: obtaining the molecular weight distribution in the chemical reaction process on line by a mass spectrometer or a laser test technology to obtain an output probability density function, and approximating the output probability density function by a B-spline neural network;
step 200: establishing a state space model of the system according to the approximated output probability density function, obtaining parameters of the state space model after transformation by a scanning identification method, and designing a learning fault diagnosis observer;
step 300: the designed learning fault diagnosis observer is implemented on a computer, whether a fault occurs in a sensor is monitored in real time, and when the fault occurs, the learning fault diagnosis observer estimates fault information in time;
step 400: when the system has a fault, the fault diagnosis observer accurately estimates fault information, and compensates the output weight of the system by using the estimated fault information;
step 500: inputting the designed control signal into the system to enable the system to output molecular weight distribution to track the distribution of the expected molecular weight;
the establishment of the state space model of the system according to the approximated probability density function in step 200 is as follows:
introducing a new state variable by using the output weight of the systemThe state space model of the system becomes:
wherein x (t) is the state variable of the original system, u (t) is the control input, f (t) is the fault vector, V (t) is the weight vector,representing the first derivative of the state variable x (t),AsIdentity matrix, x, representing the same dimension as the state variables(t) represents a state variable related to the output weight,represents a state variable xs(t) first derivative, A, B, D and G are parameter matrices, respectively;
wherein the state variable of the new system isThe parameter matrices are respectivelyIpRepresenting a P-dimensional identity matrix;
the design learning fault diagnosis observer is as follows:
wherein,representing the first derivative of the new system state variable observations,an observed value representing a new system state variable,representing the observer gain matrix, epsilon (t) representing the residual between the observed value and the actual value,the weight value of the observation is shown,representing an observed probability density function, Z (t) representing a vector with the same dimension as the fault, Z (t-tau) representing the value of the vector Z (t) before time tau, epsilon (t-tau) representing the residual error before time tau, K1、K2And W represents an observation system parameter,representing the derivative of the fault observation.
2. The method for diagnosing and controlling the failure of a sensor in the molecular weight distribution control according to claim 1, wherein the step 100 outputs a probability density function given by:
wherein phi isi(y) represents the ith B-spline basis function, n represents the number of B-spline basis functions, y represents the output of the system, t represents time, u (t) is the control input, ω isi(t) is the weight of the ith B-spline basis function;
since the integral of the output probability density function γ (y, u (t)) in its defined interval [ a, B ] is equal to 1, the output probability density function is approximated by a linear B-spline basis function neural network using a B-spline neural network as follows:
γ(y,u(t))=C(y)V(t)+T(y);
3. The method for diagnosing and controlling the failure of a sensor in the molecular weight distribution control according to claim 2, wherein the step 400 of compensating the output weight of the system by using the estimated failure information comprises:wherein, vc (t) is the compensated output weight.
4. The method for diagnosing and controlling the failure of a sensor in the molecular weight distribution control according to claim 1, wherein the probability density function of the desired molecular weight in step 500 is represented by γ (y, u (t)) ═ c (y) Vg + t (y), Vg being a desired weight; the following switching function and control law are designed to realize the molecular weight distribution tracking expected distribution after the fault:
the switching function is designed as follows:
where H and K both represent parameters of the switching function, S (t) represents the switching function,representing a weight error vector;
when the system fault is detected to occur, the control law of the system is adjusted to be as follows:
wherein, K3M is a parameter satisfying a system stability condition,alpha represents a parameter that satisfies a stable condition as a derivative of the fault observation;
solving, by a computer, the following linear matrix inequalities:
0<(6+3σ)K1 TK1≤I;
P2(A1+MK3)+(A1+MK3)TP2+Q2≤0;
solving parameters of the fault diagnosis observer meeting the conditions and parameters meeting control input conditions for tracking given distribution, and achieving the purposes of fault-tolerant control and tracking expected molecular weight distribution; wherein R is1,Q1,Q2,P1,P2For proper dimension positive determination of the symmetric matrix, sigma, gamma1Is a small normal number, A1=DAD-1I is a one-dimensional identity matrix; Σ denotes a 1 × 2 matrix.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811489086.3A CN109799802B (en) | 2018-12-06 | 2018-12-06 | Sensor fault diagnosis and fault tolerance control method in molecular weight distribution control |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811489086.3A CN109799802B (en) | 2018-12-06 | 2018-12-06 | Sensor fault diagnosis and fault tolerance control method in molecular weight distribution control |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109799802A CN109799802A (en) | 2019-05-24 |
CN109799802B true CN109799802B (en) | 2021-06-25 |
Family
ID=66556516
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811489086.3A Active CN109799802B (en) | 2018-12-06 | 2018-12-06 | Sensor fault diagnosis and fault tolerance control method in molecular weight distribution control |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109799802B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110579963A (en) * | 2019-09-03 | 2019-12-17 | 浙江工业大学 | Networked motion control system state estimation method based on adaptive state observer |
CN111190429B (en) * | 2020-01-13 | 2022-03-18 | 南京航空航天大学 | Unmanned aerial vehicle active fault-tolerant control method based on reinforcement learning |
CN112684707B (en) * | 2020-12-25 | 2022-04-26 | 扬州大学 | Styrene bulk polymerization anti-interference distribution shape control method based on interference observer |
CN115437359B (en) * | 2022-11-07 | 2023-02-28 | 西北工业大学 | Sensor fault-oriented airplane fault-tolerant control method, system and device |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101481019A (en) * | 2009-02-20 | 2009-07-15 | 华中科技大学 | Fault tolerant observing method of sensor for satellite attitude control system |
CN103399493A (en) * | 2013-08-07 | 2013-11-20 | 长春工业大学 | Real-time diagnosis and tolerant system for sensor faults of reconfigurable mechanical arm and method thereof |
CN107023367A (en) * | 2017-03-29 | 2017-08-08 | 北京航空航天大学 | A kind of SCR system of diesel engine ammonia input pickup fault diagnosis and fault tolerant control method |
CN107035490A (en) * | 2017-03-29 | 2017-08-11 | 北京航空航天大学 | A kind of SCR system of diesel engine nitrogen oxides input pickup method for diagnosing faults |
JP2017167599A (en) * | 2016-03-14 | 2017-09-21 | 日本電気株式会社 | Fault diagnosis apparatus, fault diagnosis system, fault diagnosis method, and program |
CN107329083A (en) * | 2017-07-28 | 2017-11-07 | 南京航空航天大学 | For the method for diagnosing faults of bullet train traction electric machine nonlinear system sensor |
CN108536017A (en) * | 2018-05-03 | 2018-09-14 | 山东师范大学 | Random distribution interacted system cooperation fault tolerant control method based on dynamic feedback control |
-
2018
- 2018-12-06 CN CN201811489086.3A patent/CN109799802B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101481019A (en) * | 2009-02-20 | 2009-07-15 | 华中科技大学 | Fault tolerant observing method of sensor for satellite attitude control system |
CN103399493A (en) * | 2013-08-07 | 2013-11-20 | 长春工业大学 | Real-time diagnosis and tolerant system for sensor faults of reconfigurable mechanical arm and method thereof |
JP2017167599A (en) * | 2016-03-14 | 2017-09-21 | 日本電気株式会社 | Fault diagnosis apparatus, fault diagnosis system, fault diagnosis method, and program |
CN107023367A (en) * | 2017-03-29 | 2017-08-08 | 北京航空航天大学 | A kind of SCR system of diesel engine ammonia input pickup fault diagnosis and fault tolerant control method |
CN107035490A (en) * | 2017-03-29 | 2017-08-11 | 北京航空航天大学 | A kind of SCR system of diesel engine nitrogen oxides input pickup method for diagnosing faults |
CN107329083A (en) * | 2017-07-28 | 2017-11-07 | 南京航空航天大学 | For the method for diagnosing faults of bullet train traction electric machine nonlinear system sensor |
CN108536017A (en) * | 2018-05-03 | 2018-09-14 | 山东师范大学 | Random distribution interacted system cooperation fault tolerant control method based on dynamic feedback control |
Also Published As
Publication number | Publication date |
---|---|
CN109799802A (en) | 2019-05-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109799802B (en) | Sensor fault diagnosis and fault tolerance control method in molecular weight distribution control | |
Naidu et al. | Use of neural networks for sensor failure detection in a control system | |
Wang et al. | Fault detection for systems with model uncertainty and disturbance via coprime factorization and gap metric | |
Shashoa et al. | Sensor fault detection and isolation in a thermal power plant steam separator | |
Peng et al. | New kernel independent and principal components analysis‐based process monitoring approach with application to hot strip mill process | |
Lu et al. | Fault detection, diagnosis, and performance assessment scheme for multiple redundancy aileron actuator | |
Sobchuk et al. | Ensuring the Properties of Functional Stability of Manufacturing Processes Based on the Application of Neural Networks. | |
Hongmei et al. | Fault diagnosis based on improved Elman neural network for a hydraulic servo system | |
Chang et al. | A C-means clustering based fuzzy modeling method | |
CN110889646B (en) | Method for checking correctness of operating parameters of generator set adapting to working condition change | |
Ren et al. | Collaborative fault tolerant control for non‐gaussian stochastic distribution systems based on adaptive control strategy | |
Patel et al. | Passive fault-tolerant tracking for nonlinear system with intermittent fault and time delay | |
Abougarair et al. | Supervisory Control and Simulation of Boiler Drum Water Level | |
Bernardi et al. | Reduced Order Observer Applied to a Linear Parameter Varying System with Unknown Input | |
Li et al. | Parameter intervals used for fault isolation in non-linear dynamic systems | |
Guo et al. | Fault diagnosis for actuators in a class of nonlinear systems based on an adaptive fault detection observer | |
Skitt et al. | Process monitoring using auto-associative, feed-forward artificial neural networks | |
Guan et al. | Fault diagnosis and statistical information tracking fault tolerant control for non-gaussian non-linear stochastic systems | |
Kousar et al. | Sensor fault detection in coupled liquid tanks system | |
Skaf et al. | Fault detection and diagnosis for general discrete-time stochastic systems using output probability density estimation | |
Curry et al. | Robust fault detection using robust l/sub 1/estimation and fuzzy logic | |
Ren et al. | Fault tolerant control for sequentially connected stochastic distribution continuous systems | |
Subramanian et al. | Fault diagnosis of batch reactor using machine learning methods | |
Liu et al. | Fault detection method for nonlinear systems based on probabilistic neural network filtering | |
Geetha et al. | Comparative performance analysis of extended Kalman filter and neural observer for state estimation of continuous stirred tank reactor |
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 |