CN115702288B - Engine abnormality diagnosis method, engine abnormality diagnosis program, and engine abnormality diagnosis system - Google Patents
Engine abnormality diagnosis method, engine abnormality diagnosis program, and engine abnormality diagnosis system Download PDFInfo
- Publication number
- CN115702288B CN115702288B CN202180042708.1A CN202180042708A CN115702288B CN 115702288 B CN115702288 B CN 115702288B CN 202180042708 A CN202180042708 A CN 202180042708A CN 115702288 B CN115702288 B CN 115702288B
- Authority
- CN
- China
- Prior art keywords
- engine
- abnormality
- state quantity
- abnormality diagnosis
- model
- 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
- 230000005856 abnormality Effects 0.000 title claims abstract description 213
- 238000003745 diagnosis Methods 0.000 title claims abstract description 96
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000000556 factor analysis Methods 0.000 claims abstract description 38
- 238000010801 machine learning Methods 0.000 claims abstract description 31
- 238000005259 measurement Methods 0.000 claims description 59
- 230000002000 scavenging effect Effects 0.000 claims description 32
- 238000004364 calculation method Methods 0.000 claims description 30
- 238000001514 detection method Methods 0.000 claims description 29
- 239000011159 matrix material Substances 0.000 claims description 25
- 239000000446 fuel Substances 0.000 claims description 21
- 238000001914 filtration Methods 0.000 claims description 12
- 230000003993 interaction Effects 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 7
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 21
- 230000008569 process Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000013178 mathematical model Methods 0.000 description 4
- 238000005312 nonlinear dynamic Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000000498 cooling water Substances 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000006866 deterioration Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000004886 process control Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000001311 chemical methods and process Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D41/00—Electrical control of supply of combustible mixture or its constituents
- F02D41/22—Safety or indicating devices for abnormal conditions
- F02D41/221—Safety or indicating devices for abnormal conditions relating to the failure of actuators or electrically driven elements
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D45/00—Electrical control not provided for in groups F02D41/00 - F02D43/00
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
- G05B23/0254—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D41/00—Electrical control of supply of combustible mixture or its constituents
- F02D41/02—Circuit arrangements for generating control signals
- F02D41/14—Introducing closed-loop corrections
- F02D41/1401—Introducing closed-loop corrections characterised by the control or regulation method
- F02D2041/1413—Controller structures or design
- F02D2041/1415—Controller structures or design using a state feedback or a state space representation
- F02D2041/1417—Kalman filter
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D41/00—Electrical control of supply of combustible mixture or its constituents
- F02D41/02—Circuit arrangements for generating control signals
- F02D41/14—Introducing closed-loop corrections
- F02D41/1401—Introducing closed-loop corrections characterised by the control or regulation method
- F02D2041/1433—Introducing closed-loop corrections characterised by the control or regulation method using a model or simulation of the system
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
本发明的发动机的异常诊断方法、发动机的异常诊断程序及发动机的异常诊断系统执行以下步骤：步骤(S2)：获取发动机模型(10)的初始状态量；步骤(S3)，应用初始状态量并活用发动机模型(10)；步骤(S4)，得到估计状态量；步骤(S5)，获取测量状态量；步骤(S6)，将测量状态量与估计状态量的残差施加到非线性卡尔曼滤波器；步骤(S10)，将卡尔曼增益应用于发动机模型(10)并重复步骤(S4)～(S6)；步骤(S11)，计算测量状态量或残差的相关；步骤(S12)，进行因子分析，求出因子载荷量；步骤(S13)，计算因子得分，检测异常；步骤(S16)，将因子载荷量应用于机器学习；步骤(S17)，基于机器学习诊断异常；步骤(S18)，输出诊断结果。
The engine abnormality diagnosis method, engine abnormality diagnosis program and engine abnormality diagnosis system of the present invention perform the following steps: step (S2): obtaining the initial state quantity of the engine model (10); step (S3): applying the initial state quantity and utilizing the engine model (10); step (S4): obtaining the estimated state quantity; step (S5): obtaining the measured state quantity; step (S6): applying the residual of the measured state quantity and the estimated state quantity to a nonlinear Kalman filter; step (S10): applying the Kalman gain to the engine model (10) and repeating steps (S4) to (S6); step (S11): calculating the correlation of the measured state quantity or the residual; step (S12): performing factor analysis to obtain the factor loading; step (S13): calculating the factor score and detecting the abnormality; step (S16): applying the factor loading to machine learning; step (S17): diagnosing the abnormality based on machine learning; and step (S18): outputting the diagnosis result.
Description
Technical Field
The present invention relates to an engine abnormality diagnosis method, an engine abnormality diagnosis program, and an engine abnormality diagnosis system for diagnosing an abnormality of an engine using a mathematical engine model.
Background
From the viewpoints of safety, efficiency, and the like, it is important to detect an abnormality of an engine mounted on a ship or the like at an early stage.
Patent document 1 discloses a performance estimation system for a gas turbine engine, in which, when a difference occurs between an observed variable and an estimated value of the observed variable, a Constant Gain Extended Kalman Filter (CGEKF) composed of a kalman filter gain designed to be optimal at a certain operating point of the engine and a nonlinear dynamic simulation model faithfully modeling a real engine is used, a model that is always faithful to the real engine is made by adjusting a tuning parameter of a part of state variables constituting the model so as to minimize the difference. Further, in paragraph 0010, an engine performance estimation system and method are described that constantly estimate, detect, and monitor performance degradation due to chronological changes or damage caused by engine usage through a nonlinear dynamic simulation model and an estimation filter of the engine.
Further, patent document 2 discloses a system for determining flight parameters of an aircraft in real time while the aircraft is in flight, which is configured based on a flight mechanics equation that determines an interdependence relationship between at least 2 previously selected flight parameters of the aircraft to be estimated, and includes an extended kalman filter formed to provide a joint estimated value of the selected flight parameters while the aircraft is in flight.
Further, patent document 3 discloses a kalman filter configured to provide a new process variable estimate to a controller in each execution cycle of the controller to enable a control signal for controlling a process to be generated by the controller, and to generate an estimate of a process variable value based on a slow or intermittent process feedback signal.
Prior art literature
Patent literature
Patent document 1 Japanese patent laid-open No. 2009-68359
Patent document 2 Japanese patent application laid-open No. 2013-49408
Patent document 3 Japanese patent application laid-open No. 2018-88289
Disclosure of Invention
Technical problem to be solved by the invention
In patent document 1, as described in paragraph 0005, "a nonlinear dynamic simulation model which can be made to sufficiently reflect the influence of aged deterioration, damage, or the like, which changes the correlation between a measurable item and an unmeasurable item" is described, patent document 1 describes an invention in which a nonlinear dynamic simulation model is tuned in sequence, an unmeasurable item of an engine is estimated from the measurable item, and the estimated result is used to control a gas turbine engine, not to perform early detection of an abnormality of the engine and diagnosis of the cause thereof.
Patent document 2 discloses an invention for simultaneously estimating at least two flight parameters related to each other by a flight mechanics equation by extending a kalman filter and for controlling the flight of an aircraft, and does not disclose early detection of an abnormality of an engine and diagnosis of the cause thereof.
Patent document 3 relates to an improvement of a kalman filter in a process control system for a chemical process, a petroleum process, or the like, and is not an invention for performing early detection of an abnormality of an engine and diagnosis of the cause thereof.
Accordingly, an object of the present invention is to provide an engine abnormality diagnosis method, an engine abnormality diagnosis program, and an engine abnormality diagnosis system that enable early detection of an abnormality of an engine and diagnosis of a cause thereof.
Solution to the above technical problems
An abnormality diagnosis method for an engine corresponding to the description of the embodiment 1 is a method for performing abnormality diagnosis on an abnormality of the engine by using a mathematical engine model, and is characterized by comprising an initial state quantity acquisition step of acquiring an initial state quantity of the engine model, an engine model utilization step of applying the initial state quantity to the engine model and utilizing the engine model, an engine state estimation step of calculating a state of the engine based on the initial state quantity and obtaining an estimated state quantity by using the engine model, a measured state quantity acquisition step of acquiring a measured state quantity of the engine, a Kalman filtering step of applying a residual error of the acquired measured state quantity and the calculated estimated state quantity to a nonlinear Kalman filter, a repetition step of applying a Kalman gain obtained by applying the obtained Kalman gain to the nonlinear Kalman filter to the engine model, and repeating the engine state estimation step, the measured state quantity acquisition step and the Kalman filtering step, a correlation calculation step of calculating a correlation of the measured state quantity or the residual error when applied to the nonlinear Kalman filter, a factor analysis step of performing factor analysis on the correlation of the measured state quantity or the residual error, a factor obtaining a factor load quantity, a detection step of the factor, a detection of the residual error, and a machine learning step of diagnosing the abnormality, and a machine learning step of the abnormality load, and a machine learning step of the abnormality, and an abnormality diagnosis based on the abnormality factor.
According to the present invention described in claim 1, the abnormality of the engine can be detected early based on the calculated factor score by performing factor analysis using the measured state quantity or the residual error between the measured state quantity and the calculated estimated state quantity. Further, the factor load amount can be applied to machine learning to diagnose the cause of the abnormality of the engine.
The present invention described in claim 2 is characterized in that the model updating step is further executed, and the engine model is updated by applying the result obtained in the kalman filtering step or the result obtained by processing the acquired measurement state quantity to the engine model utilization step.
According to the present invention described in claim 2, the engine model can be updated and the calculation accuracy of the estimated state quantity can be kept high at all times.
The present invention described in claim 3 is characterized in that the correlation of the measurement state quantity or the residual error in the correlation calculation step is calculated based on the correlation matrix.
According to the invention described in claim 3, the calculation accuracy of the correlation between the measurement state quantity and the residual error can be improved by simple calculation.
In the present invention described in claim 4, in the factor analysis step, a Singular Value Decomposition (SVD) is performed based on a covariance matrix as a correlation matrix, thereby deriving the factor load.
According to the invention described in claim 4, it is possible to extract essential matters without being limited by the shape of the covariance matrix, and detect abnormality of the engine early.
The invention described in claim 5 is characterized in that the machine learning uses self-organizing map (SOM).
According to the invention described in claim 5, the cause of the abnormality of the engine can be classified by using self-organizing map (SOM) which is an unsupervised machine learning.
The invention described in claim 6 is characterized in that the initial state quantity acquired in the initial state quantity acquisition step is a load (Q p) of the engine and a fuel supply quantity including a fuel pump rack position (h p).
According to the invention described in claim 6, the estimated state quantity can be obtained based on the load (Q p) of the engine, which is important in estimating the state of the engine, and the fuel supply quantity including the fuel pump rack position (h p).
The invention described in claim 7 is characterized in that the measurement state quantity acquired in the measurement state quantity acquisition step is the rotation speed (n e) of the engine.
According to the invention described in claim 7, the engine speed (n e) which is a high-probability measure for the engine can be used for abnormality diagnosis of the engine, and the measurement accuracy can be improved.
The present invention described in claim 8 is characterized in that the supercharger rotation speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s) and the exhaust gas temperature (T e) are obtained as estimated state amounts in the engine state estimating step.
According to the invention described in claim 8, it is possible to obtain the diagnosis result of the supercharger rotation speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), or the exhaust gas temperature (T e) which are useful for estimating the state of the engine.
The abnormality diagnosis program of the engine according to the description of claim 9 is a program for performing abnormality diagnosis of an abnormality of the engine using a mathematical engine model, and is characterized in that an initial state quantity acquisition step, an engine model utilization step, an engine state estimation step, a measurement state quantity acquisition step, a kalman filter step, a repetition step, a correlation calculation step, a factor analysis step, an abnormality detection step, a machine learning application step, an abnormality diagnosis step, and an output step in the abnormality diagnosis method of the engine are executed by a computer.
According to the invention described in claim 9, the abnormality of the engine can be detected early based on the calculated factor score by performing factor analysis using the measured state quantity or the residual error between the measured state quantity and the calculated estimated state quantity. Further, the factor load amount can be applied to machine learning to diagnose the cause of the abnormality of the engine.
An abnormality diagnosis system for an engine according to claim 10 is characterized by comprising an engine, a condition input means for inputting an initial state quantity of an engine model, a state quantity measuring means for measuring a state of the engine to obtain a measured state quantity, a computer for executing an abnormality diagnosis method for the engine or an abnormality diagnosis program for the engine, and an information providing means for providing abnormality information including a diagnosis result of an abnormality of the engine, which is output from the computer.
According to the invention described in claim 10, it is possible to provide abnormality information including results of early detection of an abnormality of an engine and diagnosis of a cause thereof by a computer.
The invention described in claim 11 is characterized in that the updating of the engine model is performed by a computer.
According to the invention described in claim 11, the engine model is updated by the computer, whereby the accuracy of calculating the estimated state quantity can be easily improved.
The invention described in claim 12 is characterized in that the rotation speed (n e) of the engine is obtained as the measurement state quantity by the state quantity measuring means.
According to the invention described in claim 12, the engine speed (n e) which is a high-probability measure for the engine can be used for abnormality diagnosis of the engine, and the measurement accuracy can be improved.
The invention described in claim 13 is characterized in that the information providing means provides the results of at least 1 of the supercharger speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s) and the exhaust temperature (T e) of the engine as the diagnosis result of the abnormality.
According to the invention described in claim 13, it is possible to obtain the diagnosis result of the supercharger rotation speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), or the exhaust gas temperature (T e) which are useful for estimating the state of the engine.
The invention described in claim 14 is characterized by comprising an abnormality time control means for controlling the engine at the time of abnormality based on the output of the abnormality information.
According to the invention described in claim 14, by controlling the engine at the time of abnormality, it is possible to prevent the abnormality of the engine from becoming serious or causing a failure.
The invention described in claim 15 is characterized in that the abnormality information is provided using the man-machine interaction means as the information providing means.
According to the invention described in claim 15, by providing the abnormality information of the engine from the man-machine interaction means, a crew member or the like can cope with the abnormality promptly and appropriately.
The invention described in claim 16 is characterized by comprising a transmission means for transmitting the abnormality information provided by the information providing means to another place.
According to the invention described in claim 16, the diagnosis result including the abnormality of the engine can be obtained in real time even at a location remote from the engine.
The invention described in claim 17 is characterized by comprising a connection means for connecting the state measuring means, the computer, and the information providing means on line.
According to the invention described in claim 17, it is possible to transmit and provide a diagnosis result including an abnormality of the engine on line in real time.
Effects of the invention
According to the abnormality diagnosis program for an engine of the present invention, the abnormality of the engine can be detected early on based on the calculated factor score by performing factor analysis using the measured state quantity or the residual of the measured state quantity and the calculated estimated state quantity. Further, the factor load amount can be applied to machine learning to diagnose the cause of the abnormality of the engine.
Further, in the case where the model updating step of applying the result obtained in the kalman filtering step or the result obtained by processing the acquired measurement state quantity to the engine model utilizing step to update the engine model is further performed, the engine model can be updated so that the calculation accuracy of the estimated state quantity is always kept in a high state.
In addition, in the case of performing calculation of correlation of residuals in the correlation calculation step based on the correlation matrix, the calculation accuracy of correlation of the measurement state quantity or residuals can be improved by simple calculation.
In the factor analysis step, when the factor load is derived by performing Singular Value Decomposition (SVD) based on the covariance matrix as the correlation matrix, it is possible to extract essential content without being limited by the shape of the covariance matrix, and detect the abnormality of the engine early.
In addition, in the case of using self-organizing map (SOM) as machine learning, the cause of the abnormality of the engine can be classified using self-organizing map (SOM) as unsupervised machine learning.
Further, in the case where the initial state quantity acquired in the initial state quantity acquisition step is the load (Q p) of the engine and the fuel supply quantity including the fuel pump rack position (h p), the estimated state quantity can be obtained based on the load (Q p) of the engine and the fuel supply quantity including the fuel pump rack position (h p) that are vital in estimating the state of the engine.
In addition, when the measurement state quantity acquired in the measurement state quantity acquisition step is the rotational speed (n e) of the engine, the rotational speed (n e) of the engine, which has a high chance of being measured for the engine, can be used for abnormality diagnosis of the engine, and the measurement accuracy can be improved.
Further, when the supercharger rotation speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), and the exhaust gas temperature (T e) are obtained as the estimated state amounts of the engine state estimating step, the diagnosis result of the supercharger rotation speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), or the exhaust gas temperature (T e) that are useful for estimating the state of the engine can be obtained.
Further, according to the abnormality diagnosis program of the engine of the present invention, the abnormality of the engine can be detected early based on the calculated factor score by performing factor analysis using the measured state quantity or the residual of the measured state quantity and the calculated estimated state quantity. Further, the factor load amount can be applied to machine learning to diagnose the cause of the abnormality of the engine.
Further, according to the abnormality diagnosis system for an engine of the present invention, it is possible to provide abnormality information including results of performing early detection of an abnormality of the engine and diagnosis of a cause thereof, using a computer.
In addition, in the case where the engine model is updated by the computer, whereby the calculation accuracy of the estimated state quantity can be easily improved.
In addition, when the rotational speed (n e) of the engine is obtained as the measurement state quantity by the state quantity measuring means, the rotational speed (n e) of the engine, which has a high chance of being measured for the engine, can be used for abnormality diagnosis of the engine, and the measurement accuracy can be improved.
Further, in the case where the result of at least 1 of the supercharger speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), and the exhaust gas temperature (T e) of the engine is provided as the diagnosis result of the abnormality by the information providing unit, the diagnosis result of the supercharger speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), or the exhaust gas temperature (T e) useful in estimating the state of the engine can be obtained.
Further, in the case of providing an abnormality time control unit that controls the engine at the time of abnormality based on the output of the abnormality information, by controlling the engine at the time of abnormality, it is possible to prevent the abnormality of the engine from becoming serious or causing a failure.
In addition, in the case where abnormality information is provided using the man-machine interaction unit as the information providing unit, by providing abnormality information of the engine from the man-machine interaction unit, a crew member or the like can cope with the abnormality promptly and appropriately.
Further, in the case of providing the transmission means for transmitting the abnormality information provided by the information providing means to another place, the diagnosis result including the abnormality of the engine can be obtained in real time even in a place remote from the engine.
In addition, in the case of the connection unit that connects the state quantity measuring unit, the computer, and the information providing unit on-line, the diagnosis result including the abnormality of the engine can be transmitted and provided on-line in real time.
Drawings
Fig. 1 is a block diagram of an abnormality diagnosis system of an engine according to an embodiment of the present invention.
Fig. 2 is a flowchart of an abnormality diagnosis routine of the engine thereof.
Fig. 3 is a schematic diagram of abnormality diagnosis of the engine.
Fig. 4 is a conceptual diagram of factor analysis thereof.
Fig. 5 is a diagram showing an example of factor analysis using measurement data.
Fig. 6 is a diagram showing an example of the factor score thereof.
Fig. 7 is a graph showing the change in the factor load amount of each measured value.
Fig. 8 is a diagram showing an example of a mathematical model of its engine.
Fig. 9 is a diagram showing parameters of an engine model thereof.
Fig. 10 is a diagram showing the concept and calculation formula of the prediction (estimation) step and the update (correction) step of the kalman filter.
Fig. 11 is a diagram showing a relationship between a kalman filter and factor analysis.
Fig. 12 is a diagram showing an example of abnormality detection by factor score.
Detailed Description
An abnormality diagnosis method for an engine, an abnormality diagnosis program for an engine, and an abnormality diagnosis system for an engine according to an embodiment of the present invention are described below.
Fig. 1 is a block diagram of an abnormality diagnosis system of an engine.
The engine abnormality diagnosis system diagnoses an abnormality of an engine 1 mounted on a ship or the like using a mathematical engine model 10.
The engine abnormality diagnosis system includes a condition input unit 2 for inputting an initial state quantity of an engine model 10, a state quantity measuring unit 3 for measuring a state of an engine 1 and obtaining a measured state quantity, a computer 4 for executing an engine abnormality diagnosis method or program, an information providing unit (human-machine interaction unit: HMI) 5 for providing abnormality information including a diagnosis result of an abnormality of the engine 1 outputted by the computer 4, an abnormality time control unit 6 for controlling the engine 1 at the time of abnormality based on an output of the abnormality information, a transmitting unit 7 for transmitting the abnormality information provided by the information providing unit 5 to other places, and a connecting unit 8 for connecting the respective devices on line. The engine model 10 is preferably installed in the computer 4 corresponding to the engine 1 from the beginning, but may be changed in the middle via the condition input means 2 or the like when, for example, a part or all of the engine 1 is changed to a different state in association with maintenance of a ship or the like.
The connection means 8 is, for example, a router, LAN, or the like, and connects the state quantity measuring means 3, the computer 4, and the information providing means 5 on line. Thereby, the diagnosis result including the abnormality of the engine 1 can be transmitted and provided on line in real time. The online connection can be made by wireless or wired connection via the connection unit 8.
The transmission unit 7 is provided in the computer 4. The computer 4 further includes a control unit 11, an initial state quantity acquisition unit 12, an engine state estimation unit 13, a measurement state quantity acquisition unit 14, a kalman filter unit 15, a repetition unit 16, a correlation calculation unit 17, a factor analysis unit 18, an abnormality detection unit 19, a machine learning application unit 20, an abnormality diagnosis unit 21, an engine model update unit 22, a main memory 23, an auxiliary memory 24, an engine model utilization unit 27, an output unit 28, and the like.
The auxiliary memory 24 is, for example, a hard disk or the like. The auxiliary memory 24 stores the engine model 10, the factor load space 25, and the mechanical device failure database 26. The engine model 10 is a model constructed in advance based on the specifications and characteristics of the engine 1. In the factor load amount space 25, factor load amount data is accumulated from the detection of abnormality of the engine 1. In the mechanical failure database 26, data collected by simulating the failure of the engine 1 by a simulation program is collected. In addition, if there is an engine state value at the time of actual failure, the engine state value may be collected in the mechanical failure database 26.
The condition input unit 2 is a mouse, a keyboard, a touch panel, or the like. The operator or user 101 of the computer 4 inputs the initial state quantity using the condition input unit 2. As the initial state quantity, the load (Q p) of the engine 1 and the fuel supply quantity including the fuel pump rack position (h p) are preferably input. Thus, the estimated state quantity can be obtained based on the load (Q p) of the engine 1, which is important in estimating the state of the engine 1, and the fuel supply quantity including the fuel pump rack position (h p). The input initial state quantity is transmitted to the initial state quantity acquisition unit 12 of the computer 4 and acquired. Further, when the operating condition is changed sufficiently to change the engine model 10, a condition signal can also be input through the condition input unit 2.
The state quantity measuring unit 3 is various sensors or the like. In the state quantity measuring unit 3, the rotation speed (n e) of the engine 1 is preferably obtained as a measured state quantity. This makes it possible to use the rotation speed (n e) of the engine 1, which is a high-probability measurement for the engine 1, for abnormality diagnosis of the engine 1, and to improve the measurement accuracy. In addition to the measurement state quantity measured by the state quantity measuring means 3, there are, for example, the scavenging pressure (P s) of the engine 1, the exhaust gas temperature (T e), the load (Q p) of the engine 1, and the like. The measurement state quantity measured by the state quantity measuring unit 3 is sent to the measurement state quantity acquiring section 14 of the computer 4. The measurement state quantity may be, in addition to the measurement value itself, an average value of a certain period of the measurement value, a difference from the average value, or the like, which is processed based on the measurement value.
The output unit 28 outputs the diagnosis result of the engine 1 obtained by the computer 4 to the transmission unit 7, the information providing unit (man-machine interaction unit) 5, and the abnormality-time control unit 6.
The information providing unit 5, the abnormality-time control unit 6, and the transmitting unit 7 may be provided outside the computer 4.
The information providing unit 5 preferably provides the result of at least 1 of the supercharger speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), and the exhaust gas temperature (T e) of the engine 1 as a diagnosis result of abnormality of the engine 1. As a result, the diagnosis result for the supercharger rotation speed (n tc), the scavenging pressure (P s), the scavenging temperature (T s), or the exhaust gas temperature (T e) which are useful in estimating the state of the engine 1 can be obtained. Further, the information providing unit 5 can provide not only the diagnosis result of the abnormality of the engine 1 but also all the information associated with the abnormality including the abnormality detection information.
The transmission unit 7 transmits the outputted diagnosis result to a device provided at a location different from the computer 4 through a wired or wireless LAN via the connection unit 8. Thus, even in a place remote from the engine 1, abnormality information including the diagnosis result of the abnormality of the engine 1 can be obtained in real time. In fig. 1, a bridge 100 and a user 101 on land such as a ship company are shown as places for receiving abnormality information. For example, in the case where the computer 4 is in the engine room, the information providing unit 5 transmits the abnormality information to the bridge 100 through the in-ship LAN.
The information providing unit (man-machine interaction unit) 5 is, for example, a monitor, a speaker, or the like. By providing abnormality information of the engine 1 from the information providing unit (man-machine interaction unit) 5, a crew member or the like can cope with the abnormality promptly and appropriately. The information providing unit 5 includes, in addition to the man-machine interaction unit, a storage unit that temporarily stores abnormality information, a transmission unit that transmits abnormality information to, for example, a smart phone, and other units related to the provision of information.
The abnormality-time control unit 6 controls the engine 1 automatically at the time of abnormality or according to an operation from a crew member based on the abnormality detection result or the diagnosis result of the abnormality. By controlling the engine 1 at the time of abnormality, the abnormality of the engine 1 can be prevented from becoming serious or causing a malfunction. The information or signal supplied to the abnormality time control unit 6 may be unprocessed information or signal of the information supplied to the information supply unit 5.
Fig. 2 is a flowchart of an engine abnormality diagnosis method, and fig. 3 is a schematic diagram of an engine abnormality diagnosis. Since the abnormality diagnosis method of the engine can be provided as a program, the abnormality diagnosis program of the engine will be described below as a program for causing a computer to execute each step in the abnormality diagnosis method of the engine.
When the personal computer starts executing the abnormality diagnosis program of the engine, the control unit 11 reads the engine model 10 stored in the auxiliary memory 24 (engine model reading step S1).
After the engine model reading step S1, the initial state quantity acquisition unit 12 acquires the initial state quantity of the engine model 10 input by the condition input unit 2 (initial state quantity acquisition step S2).
After the initial state quantity acquisition step S2, the control unit 11 applies the initial state quantity to the read engine model 10 in the engine model utilization unit 27, and utilizes the engine model 10 (engine model utilization step S3). In the engine model utilization step S3, the engine model 10 is initially set or updated. In order to use the engine model 10, it is necessary to apply an arbitrary initial state quantity, and if an initial state quantity (for example, a load or a fuel quantity) as close as possible to the current state is applied in the engine model using step S3 of using the engine model 10, it is possible to realize the use of the engine model 10, such as a reduction in the calculation time in the engine model 10, an improvement in the accuracy of the estimation calculation based on the engine model 10, or the like. In addition, in the engine model using step S3, the engine model 10 can be updated by using the residual error input in the kalman filtering step S6 or the result obtained by processing the acquired measurement state quantity, and the engine model 10 can faithfully reproduce the engine 1 of the real object and use the engine.
The engine state estimating unit 13 calculates the state of the engine 1 based on the initial state quantity by using the engine model 10 and obtains an estimated state quantity (engine state estimating step S4).
Further, the measurement state quantity acquisition section 14 acquires the measurement state quantity of the engine 1 obtained by the measurement by the state quantity measurement unit 3 (measurement state quantity acquisition step S5).
The kalman filter unit 15 inputs the residual error between the acquired measurement state quantity and the calculated estimated state quantity to a nonlinear kalman filter (kalman filter step S6). In the kalman filtering step S6, the state estimation and the estimation of the model parameters are performed.
After the kalman filtering step S6, the engine model updating section 22 determines whether or not the engine model 10 needs to be updated (model update determination step S7).
When the model update determination step S7 determines that "yes" is required, that is, when the engine model 10 needs to be updated, the engine model updating unit 22 applies the result obtained in the kalman filtering step S6 or the result obtained by processing the acquired measurement state quantity to the engine model using step S3, and updates the model parameters to update the engine model 10 (model updating step S8). By updating the engine model 10, the calculation accuracy of the estimated state quantity can be kept always in a high state in accordance with the aged deterioration or the like of the engine 1. Further, the engine model 10 is updated by the computer 4, whereby the calculation accuracy of the estimated state quantity can be easily improved. The determination in the model update determination step S7 is performed by setting a threshold value for the model parameter in advance.
On the other hand, when the model update determination step S7 is determined as no, that is, when the engine model update unit 22 determines that the engine model 10 does not need to be updated, the repeating unit 16 determines whether or not a predetermined time has elapsed (time elapsed determination step S9). The repetition unit 16 is divided into, for example, 0.1 seconds, and repeatedly performed at k, k+1, and k+2.
When the time lapse determination step S9 is "no", that is, when the predetermined time has not elapsed, the repeating unit 16 applies the kalman gain obtained by the nonlinear kalman filter to the engine model 10, and repeats the engine state estimation step S4, the measurement state quantity acquisition step S5, and the kalman filter step S6 (repeats step S10).
In this way, the update of the engine model 10 and the state estimation are repeated using two functions of the kalman filter, such as the state estimation and the model parameter estimation. In addition, when the measured state quantity is reliable, the model parameters are identified using a tracking filter, and the kalman filter can perform only state estimation.
The nonlinear kalman filter is preferably an unscented kalman filter or an extended kalman filter. This makes the kalman gain more appropriate for the engine 1 as a nonlinear system, and improves the accuracy of calculating the estimated state quantity.
On the other hand, when the time lapse determining step S9 determines that "yes", that is, the repeating unit 16 determines that the predetermined time has elapsed, the correlation calculating unit 17 calculates the correlation of the residual obtained when the input to the nonlinear kalman filter is performed (correlation calculating step S11). The calculation of the correlation of the residuals in the correlation calculation step S11 is preferably performed based on a correlation matrix. This can improve the calculation accuracy of the correlation of the residual errors by simple calculation.
In addition, the correlation may be calculated using the acquired measurement state quantity instead of the residual error obtained when the input to the nonlinear kalman filter is performed.
After the correlation calculation step S11, the factor analysis unit 18 performs factor analysis on the correlation of the residuals to determine the factor load amount (factor analysis step S12).
Instead of the correlation of the residual error, the correlation of the acquired measurement state quantity may be used to perform factor analysis to determine the factor load quantity.
After the factor analysis step S12, the abnormality detection unit 19 calculates a factor score from the factor load amount, and detects an abnormality (abnormality detection step S13). The abnormality detected in the abnormality detection step S13 may be outputted as abnormality information.
In the factor analysis step S12, singular Value Decomposition (SVD) is performed based on the covariance matrix as the correlation matrix, thereby deriving the factor load amount. Thus, it is possible to extract essential content without being limited by the shape of the covariance matrix, and to detect abnormality of the engine 1 early.
The factor load amount obtained in the factor analysis step S12 is accumulated in the factor load amount space 25 (factor load amount accumulating step S14). The machine learning application section 20 reads out the data stored in the mechanical device failure database 26 (mechanical device failure data reading out step S15), and applies the factor load amount to machine learning (machine learning application step S16). As machine learning, self-organizing map (SOM) is used. This makes it possible to clearly classify the cause of the abnormality of the engine 1 by the SOM that is the unsupervised machine learning. Further, as the machine learning algorithm, SVM (Supportvectormachine: support vector machine), fuzzy C-means (fuzzy C-means) and the like can be used in addition to SOM.
The abnormality diagnosis section 21 diagnoses an abnormality based on the machine learning (abnormality diagnosis step S17).
After the abnormality detection step S13 and the abnormality diagnosis step S17, the output unit 28 outputs abnormality information including the diagnosis result of the abnormality of the engine 1 (output step S18). The output of the abnormality information in the output step S18 may include abnormality detection information and incidental information in addition to the diagnosis result of the abnormality. The output destination is the information providing unit (man-machine interaction unit) 5, the abnormality-time control unit 6, and the transmitting unit 7 as described above.
The abnormality diagnosis of the engine 1 using the factor analysis will be described in detail below.
Fig. 4 is a conceptual diagram of factor analysis. The factor analysis is an analysis for searching for a hidden (unmeasurable) variable factor F, which is common to all the measurement values (measurement state amounts) y m and whose relationship is represented by a im. Which uses a linear relationship parameter a as shown in the following formula (1).
[ Number 1]
In the formula (1), f is a common factor, a im is a linear coefficient, and the remaining u i is measurement error or noise due to an unexplained factor.
Each measurement y m is linearly connected to a certain number of common factors f. The linear coefficient a im is called the factor load. Factor analysis refers to the exploration of hidden variables (factors), but the factors can also be considered as some kind of abnormality (cause of accident) when a measured value changes.
Fig. 5 is a diagram showing an example of factor analysis using measurement data. As shown in fig. 5, when an example of calculation of the factor load a and the main factor of a specific follow-up change based on the measured value is given as an example of factor analysis using the measured data Y, the measured value Y 1,y2,……ym represents, for example, the scavenging pressure, the exhaust gas temperature, and the engine load of the engine 1, and the intervals (for example, 0~t intervals) at a certain time are combined to create a matrix (fig. 5 a). The covariance matrix of this matrix was calculated, and R was obtained by dividing the standard deviation of each variable (fig. 5 (b)). Singular Value Decomposition (SVD) is performed on R to obtain singular value S. Next, the factor 1 load a is obtained (fig. 5 (c)). The value obtained by adding the squares of the respective factor 1 load amounts a and dividing by the sum of variances is a factor score D (index) (fig. 5 (D)).
The Singular Value Decomposition (SVD) is also performed on the correlation matrix of the covariance matrix to obtain the factor load in the 1 st stage, and the EM (ExpectationMaximization: expectation maximization) method is used to obtain the significant factor load having characteristics.
Fig. 6 is a diagram showing an example of the factor score, fig. 6 (a) shows the original factor score F1 (D1), and fig. 6 (b) shows the filtered factor score F1 (D1). Fig. 7 is a graph showing a change in the factor load amount of each measured value, where (a) of fig. 7 is the scavenging pressure, (b) of fig. 7 is the supercharger speed, (c) of fig. 7 is the exhaust gas temperature, and (d) of fig. 7 is the scavenging temperature. If the factor scores D1 (main indexes) are arranged in time series, for example, as shown in fig. 6, when an abnormality (some kind of change) occurs in the engine 1, such as when the flow rate of the cooling water in the air cooler of the engine 1 starts to decrease.
In other words, as shown in fig. 7, the information contained in the factor load amount matrix a indicates an abnormality caused by a certain change in the propulsion system. Further, the factor load amount (row of matrix a) represents the strength of the relationship between the propulsion system parameters as the element of the error, and represents the characteristic of the cause of the abnormality. Therefore, by machine learning the factor load amount using a machine learning algorithm, such as self-organizing map (SOM), the factor load amount can be used for the reason classification of the accident of the propulsion system. In addition, the factor score D1 may be used for early detection of abnormalities.
As shown in fig. 3, in the present invention, a kalman filter observer is used as the kalman filter unit 15, and factor analysis is performed using the residual error between the measured state quantity and the estimated state quantity. That is, in the present invention, as another method, the same factor analysis as described above is performed using the residual error E inherently calculated by the kalman filter unit 15 between the measured state quantity and the estimated state quantity obtained by the engine state estimating unit 13 instead of the measured value (measured state quantity) Y. The residual E is a deviation from the estimated state quantity, and reflects the deviation from the normal state if the estimated state quantity is considered to be the normal state. Therefore, it is possible to detect that some abnormality has occurred in the engine 1 based on the residual E.
The Kalman filter observer is based on a digitally twinned engine model 10. In the kalman filter observer, a residual error E of a measured state quantity obtained by a dynamic process of the engine 1 and an estimated state quantity calculated by a process of the engine model 10 based on the initial state quantity is input to a nonlinear kalman filter. Thereby obtaining the Kalman gain. Kalman gain is applied to the engine model 10 for mathematical process control of the engine model 10.
In this way, by monitoring the engine state using the engine model 10 as a digital twin model of the engine 1, it is possible to detect a failure of the engine 1 at an early stage and diagnose the cause.
Fig. 8 is a diagram showing an example of a mathematical model of an engine.
The left side of the arrow in fig. 8 is a diagram of the relation between the actual fuel supply system (fuel pump rack position (h p)) in the engine 1 and the measurement system of the rotational speed (n e) of the engine 1, etc., and shows the measurement points and measurement values in the engine 1. The right side in fig. 8 shows an engine model 10 as a mathematical model of the engine 1.
In fig. 8, the state quantity measuring unit 3 is shown by the english letters enclosed. The thermometer was defined as "T" surrounded by O, the manometer was defined as "P" surrounded by O, the tachometer was defined as "n" surrounded by O, and the axiometer was defined as "Q" surrounded by O. The load variation is measured by an axiom.
For each step (k, k+1, k+2), a measurement is measured, and each step calculation (estimation) is performed using engine model 10. In order to improve the estimation accuracy of the engine model 10, the engine model 10 is gradually corrected by acquiring the rotational speed (n e) of the engine 1 that can be measured most reliably and with high accuracy to calculate the kalman gain. This is the kalman filter.
In the mathematical model of the engine 1 shown in fig. 8, the behavior in the normal state can be represented by a nonlinear state space model. The nonlinear state space model is composed of an equation expressed by a state equation X, parameters expressed by a state quantity X, and inputs expressed by an input quantity u. The right side of the state equation X is expressed by functions shown by a box at the right center, and the relation between the state quantity n e、ntc、Ps、Te、Pe、Gf and the input quantity h p、Qp is expressed by 5 equations.
For example, the output function H of each system function F and observation equation y (T) is shown taking the fuel pump rack position (H p) and the load of the engine (Q p) as inputs, and taking the rotation speed (n e) of the engine 1, the supercharger rotation speed (n tc), the scavenging pressure (P s), the exhaust pressure (P e), the exhaust temperature (T e), and the fuel flow rate (G f) as states (x) and outputs (y).
The observation value y output in the example of fig. 8 is equal to the state quantity x.
Fig. 9 is a diagram showing parameters of the engine model.
Examples of the model parameters include a load (Q p) of an engine such as a propeller torque, an engine torque (Q e), a moment of inertia (I e、Itc), a fuel pump rack position (h p), Engine speed (n e), supercharger speed (n tc), atmospheric pressure (P a), atmospheric temperature (T a), Scavenging pressure (P s), scavenging temperature (T s), maximum compression pressure in cylinder (P c), maximum combustion pressure in cylinder (P z), Average effective pressure in cylinder (P i), scavenging tank volume (V a.r), exhaust tank volume (V e.r), thermodynamic constant (R a、Re、ke、Cpe、Cpa), Cooling water temperature (T w), compressor outlet temperature (T c), discharge pressure (P e), discharge temperature (T e), Turbine outlet temperature (T out), turbine outlet pressure (P out), fuel flow (G f), scavenge flow (G a), Compressor air flow (G c), exhaust flow (G e).
Fig. 10 is a diagram showing concepts and calculation formulas of a prediction (estimation) step (kalman filter step S6) and an update (correction) step (model update step S8) of the kalman filter.
If the difference between the engine model 10 and the entity and the uncertainty of the measurement performed by the state quantity measuring unit 3 are taken into consideration, an error (residual) may occur between the estimated state quantity obtained based on the engine model 10 and the measured state quantity. This error is calculated, corrected by measuring the state quantity, and the estimated state quantity is made as close as possible to the correct value, which is an unscented kalman filter as a nonlinear kalman filter. The basis of the unscented kalman filter is shown in fig. 10.
By applying a kalman filter, the uncertainty of measurement and modeling is taken into account in discrete sampling times k, and the behavior of the propulsion system is repeatedly represented with the calculated error (formula (2) below).
[ Number 2]
In the normal operation, the error distribution is regarded as a normal distribution of 0 average. To detect an abnormal state, an error generated by the kalman filter is provided to a factor analysis.
Fig. 11 is a diagram showing a relationship between a kalman filter and factor analysis.
Parameters a of the factor analysis model (y=af+u) are estimated from the correlation matrix (covariance matrix). As described above, the kalman filter observer performs covariance estimation each time based on the digitally twinned engine model 10, and uses the kalman gain as an indicator (indicator) of which state quantity should be corrected.
Fig. 12 is a diagram showing an example of abnormality detection by factor score.
Fig. 12 shows an example of a rapid detection of an abnormality in the abnormality diagnosis system of the engine by performing an experiment of simulating the clogging of the supercharger suction filter of the actual engine 1. The horizontal axis of the graph indicates the number of samples, and the 1 st time is 0.1 second. Although the supercharger suction filter gradually closes and the pressure loss increases, the factor score F1 rapidly increases in the initial stage of the pressure loss starting to increase, so that an abnormality can be detected.
In this way, by performing factor analysis using the residuals of the measured state quantity and the calculated estimated state quantity, it is possible to detect an abnormality of the engine 1 early on the basis of the calculated factor score. Further, the factor load amount can be applied to machine learning to diagnose the cause of the abnormality of the engine 1. Further, the computer 4 can provide abnormality information including the results of early detection of abnormality of the engine 1 and diagnosis of the cause thereof.
The respective components and peripheral units of the computer 4 may be provided externally or internally as appropriate, or may share the functions of the computer 4 by a plurality of computers, or may be partially provided as discrete circuits.
Industrial applicability
The present invention can perform, for example, early detection and diagnosis of an abnormality of an engine of a launch vessel, and can obtain a diagnosis result including an abnormality of the engine in real time even in a remote place, thereby contributing to safe and efficient sailing. The present invention can also be used for early detection of an abnormality of an engine other than a ship and diagnosis of the abnormality.
Description of the reference numerals
1 Engine
2 Condition input unit
3 State quantity measuring unit
4 Computer
5 Information providing unit (man-machine interaction unit)
6 Abnormal time control unit
7 Transmitting unit
8 Connection unit
10 Engine model
S2 initial State quantity acquisition step
S3 engine model utilization step
S4 engine state estimation step
S5 measurement state quantity acquisition step
S6 Kalman filtering step
S8 model updating step
S10 repeating the steps
S11 correlation calculation step
S12 factor analysis step
S13 abnormality detection step
S16 machine learning application step
S17 abnormality diagnosis step
S18, outputting.
Claims (16)
1. An abnormality diagnosis method for an engine, which is a method for diagnosing an abnormality of the engine using a mathematical engine model, comprising:
an initial state quantity acquisition step of acquiring an initial state quantity of the engine model constructed in advance based on the specification and characteristics of the engine;
an engine model utilization step of applying the initial state quantity to the engine model and utilizing the engine model;
An engine state estimation step of calculating a state of the engine based on the initial state quantity by using the engine model and obtaining an estimated state quantity;
a measurement state quantity acquisition step of acquiring a measurement state quantity of the engine which becomes a basis for construction of the engine model;
A Kalman filtering step of applying the obtained residual error of the measurement state quantity and the calculated estimated state quantity to a nonlinear Kalman filter;
Repeating the step of applying a kalman gain obtained by applying the nonlinear kalman filter to the engine model, and repeating the engine state estimating step, the measurement state quantity acquiring step, and the kalman filtering step;
a correlation calculation step of calculating a correlation matrix of the residual error of the estimated state quantity and the measured state quantity when applied to the nonlinear kalman filter;
A factor analysis step, namely performing factor analysis on the calculation result of the correlation matrix to obtain a factor load;
An abnormality detection step of calculating a factor score according to the factor load amount and detecting the abnormality;
a machine learning application step of applying the factor load amount to machine learning;
an abnormality diagnosis step of diagnosing the abnormality based on the machine learning;
And outputting a detection result of the abnormality of the engine and a diagnosis result of the abnormality.
2. The abnormality diagnosis method of an engine according to claim 1, characterized in that a model updating step of applying a result obtained in the kalman filtering step or a result obtained by processing the obtained measurement state quantity as an average value or a difference from the average value for a certain period of measurement values to the engine model using step to update the engine model is further performed.
3. The abnormality diagnosis method of an engine according to claim 1, characterized in that in the factor analysis step, singular value decomposition, SVD, is performed based on a covariance matrix as the correlation matrix, thereby deriving the factor load amount.
4. The abnormality diagnosis method of an engine according to claim 1, characterized in that the machine learning uses a self-organizing map, SOM.
5. The abnormality diagnosis method of an engine according to claim 1, characterized in that the initial state quantity acquired in the initial state quantity acquisition step is a load (Q p) of the engine and a fuel supply quantity including a fuel pump rack position (h p).
6. The abnormality diagnosis method of an engine according to claim 1, characterized in that the measured state quantity acquired in the measured state quantity acquisition step is a rotation speed (n e) of the engine.
7. The abnormality diagnosis method of an engine according to claim 1, characterized in that a supercharger speed (n tc), a scavenging pressure (P s), a scavenging temperature (T s) and an exhaust temperature (T e) are obtained as the estimated state quantities of the engine state estimation step.
8. A computer-readable storage medium storing an abnormality diagnosis program for an engine, the computer-readable storage medium storing a program for diagnosing an abnormality of the engine using a mathematical engine model,
The initial state quantity acquisition step, the engine model utilization step, the engine state estimation step, the measurement state quantity acquisition step, the kalman filter step, the repetition step, the correlation calculation step, the factor analysis step, the abnormality detection step, the machine learning application step, the abnormality diagnosis step, and the output step in the abnormality diagnosis method of an engine according to any one of claims 1 to 7 are executed by a computer.
9. An abnormality diagnosis system for an engine, comprising:
An engine;
A condition input unit that inputs an initial state quantity of the engine model;
a state quantity measuring unit that measures a state of the engine and obtains a measured state quantity;
a computer configured to execute the abnormality diagnosis method of an engine according to any one of claims 1 to 7 or the abnormality diagnosis program of an engine according to claim 8;
an information providing unit that provides abnormality information including a diagnosis result of abnormality of the engine, which is output by the computer.
10. The abnormality diagnosis system of an engine according to claim 9, characterized in that updating of the engine model is performed by the computer.
11. The abnormality diagnosis system of an engine according to claim 9, characterized in that a rotation speed (n e) of the engine is obtained as the measured state quantity by the state quantity measurement unit.
12. The abnormality diagnostic system of an engine according to claim 9, characterized in that a result of at least 1 of a supercharger speed (n tc), a scavenging pressure (P s), a scavenging temperature (T s) and an exhaust gas temperature (T e) of the engine is provided as a diagnostic result of the abnormality by the information providing unit.
13. The abnormality diagnosis system of an engine according to claim 9, characterized by comprising an abnormality time control unit that performs control to prevent an abnormality of the engine from becoming serious or causing a failure at the time of an abnormality based on the output of the abnormality information.
14. The abnormality diagnostic system of an engine according to claim 9, characterized in that a man-machine interaction unit is used as the information providing unit to provide the abnormality information.
15. The abnormality diagnosis system of an engine according to claim 9, characterized by comprising a transmission unit that transmits the abnormality information provided by the information providing unit to another place.
16. The abnormality diagnosis system of an engine according to claim 9, characterized by comprising a connection means for connecting the state quantity measurement means, the computer, and the information providing means on-line.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2020-116438 | 2020-07-06 | ||
| JP2020116438 | 2020-07-06 | ||
| PCT/JP2021/025526 WO2022009904A1 (en) | 2020-07-06 | 2021-07-06 | Engine abnormality diagnosis method, engine abnormality diagnosis program, and engine abnormality diagnosis system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115702288A CN115702288A (en) | 2023-02-14 |
| CN115702288B true CN115702288B (en) | 2025-01-07 |
Family
ID=79553234
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202180042708.1A Active CN115702288B (en) | 2020-07-06 | 2021-07-06 | Engine abnormality diagnosis method, engine abnormality diagnosis program, and engine abnormality diagnosis system |
Country Status (4)
| Country | Link |
|---|---|
| JP (1) | JP7450238B2 (en) |
| KR (1) | KR102692731B1 (en) |
| CN (1) | CN115702288B (en) |
| WO (1) | WO2022009904A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102645702B1 (en) * | 2023-04-10 | 2024-03-11 | 사단법인 한국선급 | Failure diagnosis system for marine engine combustion chamber amd method for diagnosing failure of marine engine combustion chamber using the same |
| CN118966790A (en) * | 2024-10-12 | 2024-11-15 | 临沂大学 | An intelligent detection and early warning method and system for workshop equipment operation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10339664A (en) * | 1997-06-10 | 1998-12-22 | Babcock Hitachi Kk | Monitor and monitoring method |
| JP2009068359A (en) * | 2007-09-11 | 2009-04-02 | Japan Aerospace Exploration Agency | Gas turbine engine performance estimation system |
| WO2019175707A1 (en) * | 2018-03-16 | 2019-09-19 | 株式会社半導体エネルギー研究所 | Charge state estimation apparatus for secondary battery, abnormality detection apparatus for secondary battery, abnormality detection method for secondary battery, and management system for secondary battery |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5341800B2 (en) * | 2010-03-11 | 2013-11-13 | アズビル株式会社 | Control model update device, control model update method, air conditioning control system, data validity judgment device |
| FR2978858B1 (en) | 2011-08-01 | 2013-08-30 | Airbus Operations Sas | METHOD AND SYSTEM FOR DETERMINING FLIGHT PARAMETERS OF AN AIRCRAFT |
| CN102855349B (en) * | 2012-08-06 | 2015-07-01 | 南京航空航天大学 | Quick prototype design method and platform for gas path fault diagnosis for aeroengine |
| US9405286B2 (en) * | 2013-03-01 | 2016-08-02 | Fisher-Rosemount Systems, Inc. | Use of predictors in process control systems with wireless or intermittent process measurements |
| US9436174B2 (en) | 2013-03-01 | 2016-09-06 | Fisher-Rosemount Systems, Inc. | Kalman filters in process control systems |
| KR101505975B1 (en) | 2014-05-30 | 2015-03-26 | 충남대학교산학협력단 | Method and system for fault dignosis of engine |
| KR102393418B1 (en) * | 2017-03-30 | 2022-05-03 | 삼성전자주식회사 | Data learning server and method for generating and using thereof |
| KR20170080556A (en) * | 2017-06-27 | 2017-07-10 | 현대중공업 주식회사 | An engine monitoring system and a method for the same |
-
2021
- 2021-07-06 JP JP2022535361A patent/JP7450238B2/en active Active
- 2021-07-06 WO PCT/JP2021/025526 patent/WO2022009904A1/en active Application Filing
- 2021-07-06 CN CN202180042708.1A patent/CN115702288B/en active Active
- 2021-07-06 KR KR1020227043363A patent/KR102692731B1/en active IP Right Grant
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10339664A (en) * | 1997-06-10 | 1998-12-22 | Babcock Hitachi Kk | Monitor and monitoring method |
| JP2009068359A (en) * | 2007-09-11 | 2009-04-02 | Japan Aerospace Exploration Agency | Gas turbine engine performance estimation system |
| WO2019175707A1 (en) * | 2018-03-16 | 2019-09-19 | 株式会社半導体エネルギー研究所 | Charge state estimation apparatus for secondary battery, abnormality detection apparatus for secondary battery, abnormality detection method for secondary battery, and management system for secondary battery |
Also Published As
| Publication number | Publication date |
|---|---|
| JPWO2022009904A1 (en) | 2022-01-13 |
| KR20230009484A (en) | 2023-01-17 |
| KR102692731B1 (en) | 2024-08-08 |
| CN115702288A (en) | 2023-02-14 |
| WO2022009904A1 (en) | 2022-01-13 |
| JP7450238B2 (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5575792B2 (en) | Aircraft engine abnormality detection | |
| CN109857094B (en) | Aero-engine fault diagnosis method based on two-stage Kalman filter algorithm | |
| CN102246110B (en) | Standardization of data used for monitoring an aircraft engine | |
| US8417432B2 (en) | Method for calculating confidence on prediction in fault diagnosis systems | |
| CN104756029B (en) | A kind of system of the parts group of monitoring device | |
| CN115702288B (en) | Engine abnormality diagnosis method, engine abnormality diagnosis program, and engine abnormality diagnosis system | |
| WO2017209179A1 (en) | Monitoring device, method for monitoring target device, and program | |
| WO2016133121A1 (en) | Abnormality diagnosis method and abnormality diagnosis system | |
| EP1656597B1 (en) | method and flight management system for mathematically modeling the performance characteristics of an aircraft | |
| CN103324155A (en) | System monitoring | |
| JP5939439B2 (en) | Abnormality diagnosis device | |
| CN107076018A (en) | Surge decision maker, surge decision method and program | |
| JP6803788B2 (en) | Information processing equipment, information processing methods and programs | |
| Mosallam et al. | Integrated bayesian framework for remaining useful life prediction | |
| EP2583146A1 (en) | Asset health monitoring using stable distributions for heavy-tailed data | |
| CN109308484A (en) | Minimal risk diagnosis method and device for aero-engine multi-type faults | |
| JP6404700B2 (en) | Sensor combination determination device and sensor combination determination method | |
| WO2021140942A1 (en) | Diagnosing device, diagnosing method, and program | |
| CN118946710A (en) | Method for monitoring the health status of an aircraft turbine engine | |
| JP2018142200A (en) | State estimation method and state estimation device | |
| US20230415921A1 (en) | Method for determining an efficiency fault of an aircraft turboshaft engine module | |
| Schadler et al. | Distance-Based Data Selection for Regression Models Used in an Adaptive Fault Detection Approach | |
| WO2005050337A2 (en) | Intelligent database for performance predictions | |
| JP2025011455A (en) | Abnormality sign determination device and abnormality sign determination method | |
| Seyedhosseini et al. | Direct remaining useful life prediction based on multi-sensor information integrations by using evidence theory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20231117 Address after: Tokyo, Japan Applicant after: National Research and Development Corporation Marine. Harbor. Aviation Technology Research Institute Applicant after: Mitsui Yiaisi Co.,Ltd. Address before: Tokyo, Japan Applicant before: National Research and Development Corporation Marine. Harbor. Aviation Technology Research Institute Applicant before: MITSUI ENGINEERING & SHIPBUILDING Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |