CN115270591A - Method and system for monitoring health state of electromechanical actuator based on composite space - Google Patents
Method and system for monitoring health state of electromechanical actuator based on composite space Download PDFInfo
- Publication number
- CN115270591A CN115270591A CN202210271455.1A CN202210271455A CN115270591A CN 115270591 A CN115270591 A CN 115270591A CN 202210271455 A CN202210271455 A CN 202210271455A CN 115270591 A CN115270591 A CN 115270591A
- Authority
- CN
- China
- Prior art keywords
- electromechanical actuator
- target
- fault
- monitoring
- failure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 195
- 230000036541 health Effects 0.000 title claims abstract description 121
- 238000000034 method Methods 0.000 title claims abstract description 60
- 239000002131 composite material Substances 0.000 title claims abstract description 58
- 238000013507 mapping Methods 0.000 claims abstract description 17
- 238000010801 machine learning Methods 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims description 40
- 238000013528 artificial neural network Methods 0.000 claims description 28
- 238000007781 pre-processing Methods 0.000 claims description 17
- 230000007787 long-term memory Effects 0.000 claims description 10
- 230000006403 short-term memory Effects 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000005520 cutting process Methods 0.000 claims description 7
- 230000000306 recurrent effect Effects 0.000 claims description 6
- 230000015654 memory Effects 0.000 claims description 4
- 238000012423 maintenance Methods 0.000 abstract description 3
- 238000003745 diagnosis Methods 0.000 abstract description 2
- 230000005540 biological transmission Effects 0.000 description 35
- 230000007246 mechanism Effects 0.000 description 29
- 238000010586 diagram Methods 0.000 description 25
- 238000006073 displacement reaction Methods 0.000 description 15
- 230000003862 health status Effects 0.000 description 8
- 238000004804 winding Methods 0.000 description 8
- 238000011156 evaluation Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 238000002474 experimental method Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 238000011155 quantitative monitoring Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a method and a system for monitoring the health state of an electromechanical actuator based on a composite space, which relate to the field of dynamic monitoring, diagnosis and maintenance of mechanical systems and mainly comprise the following steps: acquiring operation data of the target electromechanical actuator at the previous moment and acquiring an actual value of the fault monitoring quantity of the target electromechanical actuator at the current moment; predicting a fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the operating data of the target electromechanical actuator at the previous moment and a machine learning algorithm; and monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator in a step-by-step mapping mode based on the fault monitoring amount reference value of the target electromechanical actuator at the current moment and the fault monitoring amount actual value of the target electromechanical actuator at the current moment so as to obtain the health state of the target electromechanical actuator. The invention can monitor the health state of each part of the electromechanical actuator and the health state of the whole electromechanical actuator in real time.
Description
Technical Field
The invention relates to the field of dynamic monitoring, diagnosis and maintenance of mechanical systems, in particular to a method and a system for monitoring the health state of an electromechanical actuator based on a composite space.
Background
Electro-mechanical Actuators (EMA) are a general name of Actuators for realizing position or pressure servo control by controlling the motion of a motor to directly or indirectly control the motion of a load, and are widely applied to the fields of aerospace, military, transportation, industrial and agricultural production and the like. With the advent of the next generation of aerospace systems equipped with fly-by-wire control, electromechanical actuators are rapidly becoming the most critical components for aerospace vehicle safety, driving objects such as airplanes, spacecraft, ground vehicles, etc. for controlling their position, altitude, etc.
The electromechanical actuator is used as a key part in the system, and is often operated under various complex working conditions, so that various faults such as short circuit, deformation, clamping stagnation and the like are easy to occur. Therefore, the health state evaluation of the electromechanical actuator is of great significance to the safe operation of the electromechanical actuator.
The actual working process of the electromechanical actuator shows that except some sudden faults, most faults of the electromechanical actuator are accumulated day by day, the overall health state of the electromechanical actuator is accurately controlled, an alarm is timely provided to an upper-layer control unit when the performance of the electromechanical actuator is reduced, corresponding adjustment or fault-tolerant control is carried out, and serious faults can be well prevented. However, the electromechanical actuator is a complex electromechanical system composed of a plurality of components such as a controller, a driving motor, a reduction gear box, a ball screw and the like, the actuation mechanism of the electromechanical actuator is very complex, the electromechanical actuator comprises a plurality of nonlinear links, the failure modes of the electromechanical actuator can occur in different links, and each component can also occur a plurality of failure modes at the same time, so that the performance of the electromechanical actuator is affected in a complex way. In addition, due to the limited accumulation of fault data of the electromechanical actuator, the fault mode and the influence are based on industrial experience, and the interference of nonlinear factors on the system output in a fault state is not clear, so that the difficulty of evaluating the current health state of the electromechanical actuator is further increased.
The currently available methods for monitoring the health status of electromechanical actuators can be divided into two categories: model-based methods and data-based methods. Although the method can evaluate the health state of the electromechanical actuator to a certain extent, the health state of each component of the electromechanical actuator and the overall health state cannot be evaluated in real time.
Disclosure of Invention
The invention aims to provide a method and a system for monitoring the health state of an electromechanical actuator based on a composite space, which can monitor the health state of each part of the electromechanical actuator and the health state of the whole electromechanical actuator in real time.
In order to achieve the purpose, the invention provides the following scheme:
in a first aspect, the invention provides a method for monitoring the health state of an electromechanical actuator based on a composite space, which is used for monitoring the health state of a target electromechanical actuator; the target electromechanical actuator comprises a plurality of target components, each target component corresponds to one or more fault modes, and each fault mode corresponds to a fault monitoring amount; the composite health state space of the target electromechanical actuator comprises a bottom layer, a middle layer and a top layer; the bottom layer includes all failure modes of the target electromechanical actuator; the intermediate layer includes all target components of the target electromechanical actuator; the top layer is the target electromechanical actuator; the method for monitoring the health state of the electromechanical actuator comprises the following steps:
acquiring operation data of the target electromechanical actuator at the previous moment and acquiring an actual value of a fault monitoring quantity of the target electromechanical actuator at the current moment;
predicting a fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the operation data of the target electromechanical actuator at the previous moment and a machine learning algorithm;
monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator in a step-by-step mapping mode based on the fault monitoring quantity reference value of the target electromechanical actuator at the current moment and the fault monitoring quantity actual value of the target electromechanical actuator at the current moment so as to obtain the health state of the target electromechanical actuator; the state of health includes a degree of failure of each of the failure modes in the target electromechanical actuator, a degree of failure of each of the target components in the target electromechanical actuator, and a degree of overall failure of the target electromechanical actuator.
Optionally, the method further includes:
and performing fault characteristic analysis on each fault mode in the target electromechanical actuator, and determining a fault monitoring amount used for monitoring the fault mode.
Optionally, the method further includes: and constructing a regression model based on the long-term and short-term memory neural network.
Optionally, the constructing a regression model based on the long-term and short-term memory neural network specifically includes:
constructing a recurrent neural network; the recurrent neural network comprises a long-short term memory neural network layer and a full-connection neural network layer;
constructing a training sample set; the training sample set comprises a plurality of sample input data and label data corresponding to each sample input data; the sample input data is the running data of the target electromechanical actuator in a normal state; the tag data is an actual value of the fault monitoring amount of the target electromechanical actuator in a normal state;
and training the regression neural network based on the training sample set to obtain a regression model.
Optionally, the predicting the reference value of the fault monitoring amount of the target electromechanical actuator at the current moment based on the operation data of the target electromechanical actuator at the previous moment and a machine learning algorithm specifically includes:
preprocessing the operation data of the target electromechanical actuator at the last moment; the preprocessing comprises data splicing operation, normalization operation and sliding window cutting operation;
and predicting the fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the preprocessed operating data of the target electromechanical actuator at the last moment and the regression model.
Optionally, based on the current time, the reference value of the fault monitoring amount of the target electromechanical actuator and the current time, the actual value of the fault monitoring amount of the target electromechanical actuator, and monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator by adopting a step-by-step mapping manner, so as to obtain the health state of the target electromechanical actuator, specifically including:
calculating the fault degree of each fault mode based on the fault monitoring quantity reference value of the target electromechanical actuator at the current moment and the fault monitoring quantity actual value of the target electromechanical actuator at the current moment;
determining a relationship of the failure mode to the target component based on a composite state-of-health space of the target electromechanical actuator, and determining a degree of failure for each of the target components based on the relationship of the failure mode to the target component and the degree of failure for each of the failure modes;
determining an overall degree of failure of the target electromechanical actuator based on the degree of failure of each of the target components.
In a second aspect, the invention provides a system for monitoring the health state of an electromechanical actuator based on a composite space, which is used for monitoring the health state of a target electromechanical actuator; the target electromechanical actuator comprises a plurality of target components, each target component corresponds to one or more fault modes, and each fault mode corresponds to a fault monitoring amount; the composite health state space of the target electromechanical actuator comprises a bottom layer, a middle layer and a top layer; the bottom layer includes all failure modes of the target electromechanical actuator; the intermediate layer includes all of the target components of the target electromechanical actuator; the top layer is the target electromechanical actuator; the electromechanical actuator state of health monitoring system includes:
the data acquisition module is used for acquiring the operation data of the target electromechanical actuator at the previous moment and acquiring the actual value of the fault monitoring amount of the target electromechanical actuator at the current moment;
the fault monitoring quantity reference value prediction module is used for predicting the fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the operation data of the target electromechanical actuator at the previous moment and a machine learning algorithm;
the electromechanical actuator health state determining module is used for monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator by adopting a step-by-step mapping mode based on the fault monitoring amount reference value of the target electromechanical actuator at the current moment and the fault monitoring amount actual value of the target electromechanical actuator at the current moment so as to obtain the health state of the target electromechanical actuator; the state of health includes a degree of failure of each of the failure modes in the target electromechanical actuators, a degree of failure of each of the target components in the target electromechanical actuators, and a degree of failure of the target electromechanical actuators as a whole.
Optionally, the method further includes: and the regression model building module is used for building a regression model based on the long-term and short-term memory neural network.
Optionally, the module for predicting the reference value of the monitored fault quantity specifically includes:
the preprocessing unit is used for preprocessing the operation data of the target electromechanical actuator at the last moment; the preprocessing comprises data splicing operation, normalization operation and sliding window cutting operation;
and the fault monitoring quantity reference value prediction unit is used for predicting the fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the preprocessed operation data of the target electromechanical actuator at the last moment and the regression model.
Optionally, the health state determination module of the electromechanical actuator specifically includes:
a failure degree calculation unit of a failure mode, configured to calculate a failure degree of each failure mode based on a reference value of a failure monitoring amount of the target electromechanical actuator at a current time and an actual value of the failure monitoring amount of the target electromechanical actuator at the current time;
a failure degree calculation unit of a target component for determining a relationship of the failure mode with the target component based on a composite state-of-health space of the target electromechanical actuator, and determining a failure degree of each of the target components based on the relationship of the failure mode with the target component and the failure degree of each of the failure modes;
and the overall fault degree determining unit of the target electromechanical actuator is used for determining the overall fault degree of the target electromechanical actuator based on the fault degree of each target component.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
compared with the traditional electromechanical actuator health state monitoring method, the electromechanical actuator health state monitoring method and system based on the composite space can acquire the fault monitoring quantity reference value of the electromechanical actuator in real time through a machine learning algorithm; through the compound health state space of the electromechanical actuator, the reference value of the fault monitoring amount and the actual value of the fault monitoring amount, calculation is carried out step by step, the fault degree of a typical fault mode of the electromechanical actuator can be analyzed, the fault degree of each key part in the electromechanical actuator can be analyzed, the integral health state index of the electromechanical actuator is obtained on the basis, namely the fault degree, and the actual running condition of the electromechanical actuator can be better reflected.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a method for monitoring the health status of an electromechanical actuator based on a composite space according to the present invention;
FIG. 2 is a flowchart illustrating a method for monitoring the health status of an electromechanical actuator based on a composite space according to an embodiment of the present invention;
FIG. 3 is a schematic representation of a composite health state space constructed for an electromechanical actuator in accordance with the present invention;
FIG. 4 is a schematic diagram of the training results of the LSTM-based method for training the output displacement reference model of the electromechanical actuator of a certain type; FIG. 4 (a) is a schematic diagram of the training results of the test set 1 for training the output displacement reference model of a certain type of electromechanical actuator based on LSTM according to the present invention; FIG. 4 (b) is a schematic diagram of the training result of test set 2 for training the output displacement reference model of a certain type of electromechanical actuator based on LSTM according to the present invention;
FIG. 5 is a schematic diagram of the training results of the present invention for training a control board current reference model of a model electromechanical actuator motor based on LSTM; FIG. 5 (a) is a schematic diagram of a training result of a test set 1 for training a control board current reference model of a motor of an electromechanical actuator of a certain type based on LSTM according to the present invention; FIG. 5 (b) is a schematic diagram of the training results of test set 2 for training the control board current reference model of a certain type of electromechanical actuator motor based on LSTM according to the present invention;
FIG. 6 is a schematic diagram of a training result of a standard deviation reference model for training a ball screw vibration signal of an electromechanical actuator of a certain type based on LSTM according to the present invention; FIG. 6 (a) is a schematic diagram of a training result of a test set 1 for training a standard deviation reference model of a ball screw vibration signal of an electromechanical actuator of a certain type based on LSTM; FIG. 6 (b) is a schematic diagram of the training result of test set 2 of the standard deviation reference model for training the vibration signal of the ball screw of the electromechanical actuator of a certain type based on LSTM according to the present invention;
FIG. 7 is a schematic illustration of the monitoring results of the present invention for a normal state of an electromechanical actuator of a certain type; FIG. 7 (a) is a schematic diagram showing the current monitoring result of a control board of the present invention in a normal state for a certain type of electromechanical actuator; FIG. 7 (b) is a schematic diagram illustrating the output displacement monitoring results of a certain type of electromechanical actuator in a normal state according to the present invention; FIG. 7 (c) is a diagram illustrating the standard deviation monitoring result of the vibration signal when the electromechanical actuator of a certain type is in a normal state according to the present invention; FIG. 7 (d) is a schematic diagram illustrating the results of monitoring the stuck failure of the transmission mechanism when the electromechanical actuator of a certain type is in a normal state; FIG. 7 (e) is a schematic diagram illustrating the monitoring result of the excessive clearance fault of the transmission mechanism when the electromechanical actuator of a certain type is in a normal state; FIG. 7 (e) is a schematic diagram illustrating a monitoring result of a damage fault degree of a surface of a ball screw when an electromechanical actuator of a certain type is in a normal state according to the present invention; FIG. 7 (g) is a schematic view of a health indicator monitoring result of an electromechanical actuator of a certain type when the electromechanical actuator is in a normal state according to the present invention;
FIG. 8 is a schematic view of the monitoring results of the present invention for a stuck failure condition of the injection drive mechanism for a particular type of electromechanical actuator; FIG. 8 (a) is a schematic diagram illustrating the monitoring results of the control plate current when a stuck fault condition of a transmission mechanism is injected into an electromechanical actuator of a certain type according to the present invention; FIG. 8 (b) is a schematic view of the output displacement monitoring result of the present invention when injecting a transmission jamming fault state into a certain type of electromechanical actuator; FIG. 8 (c) is a schematic diagram illustrating the vibration signal standard deviation monitoring results when a stuck fault condition of a transmission mechanism is injected into an electromechanical actuator of a certain type in accordance with the present invention; FIG. 8 (d) is a schematic view of the transmission sticking fault level monitoring of the present invention when a transmission sticking fault condition is injected into an electromechanical actuator of a certain type; FIG. 8 (e) is a schematic diagram illustrating the result of monitoring the excessive clearance fault condition of the transmission mechanism when the electro-mechanical actuator of a certain type is injected into the stuck fault condition of the transmission mechanism according to the present invention; FIG. 8 (e) is a schematic diagram illustrating a monitoring result of a damage fault level of a surface of a ball screw when a clamping-stagnation fault state of a transmission mechanism is injected into an electromechanical actuator of a certain type according to the present invention; FIG. 8 (g) is a schematic view of the health index monitoring results of an electromechanical actuator according to the present invention when a stuck failure state of a transmission mechanism is injected into an electromechanical actuator of a certain type;
FIG. 9 is a schematic view of the monitoring results of the present invention for a ball screw surface damage fault condition injected into a model electromechanical actuator; FIG. 9 (a) is a schematic diagram illustrating the current monitoring result of a control board when a damage fault condition is injected into the surface of a ball screw for a certain type of electromechanical actuator according to the present invention; FIG. 9 (b) is a schematic diagram illustrating the output displacement monitoring results when the fault state of damage to the surface of a ball screw is injected into an electromechanical actuator of a certain type according to the present invention; FIG. 9 (c) is a schematic diagram illustrating the standard deviation monitoring result of vibration signals when a damage fault condition is injected into the surface of a ball screw for a certain type of electromechanical actuator according to the present invention; FIG. 9 (d) is a schematic view of the transmission mechanism sticking fault degree monitoring result when a damage fault state is injected into the surface of a ball screw by an electromechanical actuator of a certain type; FIG. 9 (e) is a schematic diagram illustrating the monitoring result of the excessive clearance fault state of the transmission mechanism when the damage fault state is injected to the surface of the ball screw by the electromechanical actuator of a certain type; FIG. 9 (e) is a schematic diagram illustrating a result of monitoring a damage fault level of a surface of a ball screw when an electromechanical actuator of a certain type is injected into a damage fault state of the surface of the ball screw according to the present invention; FIG. 9 (g) is a schematic view of a health indicator monitoring result of an electromechanical actuator when a damage fault state is injected into a surface of a ball screw of the electromechanical actuator of a certain type;
FIG. 10 is a schematic structural diagram of a health status monitoring system for an electromechanical actuator based on a complex space according to 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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a method and a system for monitoring the health state of an electromechanical actuator based on a composite space, aiming at solving the requirement that the electromechanical actuator monitors the health degree of all parts of a system to the whole body through a quantitative index on the premise of keeping high real-time performance, and provides the method and the system for monitoring the health state of the electromechanical actuator, which have high real-time performance and strong universality and comprise the complete quantitative monitoring of all fault modes and all parts of the system to the whole body.
The invention provides a method and a system for monitoring the health state of an electromechanical actuator based on a composite space, which mainly comprise two parts: and constructing a composite health state space of the electromechanical actuator, and monitoring the health state of the electromechanical actuator based on the composite space.
In order to make the aforementioned objects, features and advantages of the present invention more comprehensible, the present invention is described in detail with reference to the accompanying drawings and the detailed description thereof.
Example one
The method for monitoring the health state of the electromechanical actuator based on the composite space is used for monitoring the health state of a target electromechanical actuator; the target electromechanical actuator comprises a plurality of target components, each target component corresponds to one or more fault modes, and each fault mode corresponds to a fault monitoring amount; the composite health state space of the target electromechanical actuator comprises a bottom layer, a middle layer and a top layer; the bottom layer includes all failure modes of the target electromechanical actuator; the intermediate layer includes all target components of the target electromechanical actuator; the top layer is the target electromechanical actuator.
As shown in fig. 1, a method for monitoring a state of health of an electromechanical actuator according to an embodiment of the present invention includes:
step 101: and acquiring the operation data of the target electromechanical actuator at the previous moment and acquiring the actual value of the fault monitoring quantity of the target electromechanical actuator at the current moment.
Step 102: and predicting the fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the operation data of the target electromechanical actuator at the last moment and a machine learning algorithm.
Step 103: monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator in a step-by-step mapping mode based on the fault monitoring quantity reference value of the target electromechanical actuator at the current moment and the fault monitoring quantity actual value of the target electromechanical actuator at the current moment so as to obtain the health state of the target electromechanical actuator; the state of health includes a degree of failure of each of the failure modes in the target electromechanical actuators, a degree of failure of each of the target components in the target electromechanical actuators, and a degree of failure of the target electromechanical actuators as a whole.
On the basis of the embodiment shown in fig. 1, the method for monitoring the health states of the electromechanical actuators provided by the embodiment of the invention further includes performing fault feature analysis on each fault mode in the target electromechanical actuator, and determining a fault monitoring amount used for monitoring the fault mode.
On the basis of the embodiment shown in fig. 1, the method for monitoring the health state of the electromechanical actuator according to the embodiment of the present invention further includes: and constructing a regression model based on the long-term and short-term memory neural network.
The regression model is constructed by the following specific steps:
step A: constructing a recurrent neural network; the recurrent neural network comprises a long-short term memory neural network layer and a fully connected neural network layer.
And B: constructing a training sample set; the training sample set comprises a plurality of sample input data and label data corresponding to each sample input data; the sample input data is the running data of the target electromechanical actuator in a normal state; and the tag data is an actual value of the fault monitoring amount of the target electromechanical actuator in a normal state.
And C: and preprocessing the data in the training sample set. The preprocessing comprises a data splicing operation, a normalization operation and a sliding window cutting operation.
Step D: and training the regression neural network based on the preprocessed training sample set to obtain a regression model.
Further, on the basis of the regression model, predicting a reference value of a fault monitoring amount of the target electromechanical actuator at the current moment based on the operation data of the target electromechanical actuator at the previous moment and a machine learning algorithm specifically includes:
firstly, preprocessing the operation data of the target electromechanical actuator at the last moment; the preprocessing comprises a data splicing operation, a normalization operation and a sliding window cutting operation. And secondly, predicting a fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the preprocessed operating data of the target electromechanical actuator at the last moment and the regression model.
On the basis of the embodiment shown in fig. 1, based on the reference value of the fault monitoring amount of the target electromechanical actuator at the current time and the actual value of the fault monitoring amount of the target electromechanical actuator at the current time, monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator in a step-by-step mapping manner to obtain the health state of the target electromechanical actuator specifically includes:
a, step a: and calculating the fault degree of each fault mode based on the fault monitoring quantity reference value of the target electromechanical actuator at the current moment and the fault monitoring quantity actual value of the target electromechanical actuator at the current moment.
Step b: determining a relationship of the failure mode to the target component based on a composite state of health space of the target electromechanical actuator, and determining a degree of failure for each of the target components based on the relationship of the failure mode to the target component and the degree of failure for each of the failure modes.
Step c: determining an overall degree of failure of the target electromechanical actuator based on the degree of failure of each of the target components.
Example two
The embodiment of the invention provides a method for monitoring the health state of an electromechanical actuator based on a composite space; the electromechanical actuator is a complex electromechanical system consisting of a plurality of components such as a controller, a driving motor, a transmission mechanism, a ball screw and the like.
As shown in FIG. 2, the method for monitoring the state of health of the electromechanical actuator based on the composite space provided by the embodiment of the invention comprises the following steps.
Step 201: firstly, selecting several typical faults which are common or have larger influence when the electromechanical actuator works, and then carrying out characteristic analysis on the typical faults of the electromechanical actuator to obtain a system state quantity capable of representing the fault degree of the electromechanical actuator as a fault monitoring quantity. The method comprises the following steps:
based on the generation mechanism of several common fault modes in the operation process of the electromechanical actuator, certain fault characteristic analysis is carried out, and fault monitoring quantity used for evaluating the fault modes, namely several state parameters which can represent the fault degree of each current fault mode in the operation process and a fault threshold value when a fault occurs, are determined. The greater the deviation of the measured fault monitoring quantity from its reference value when the electromechanical actuator is normal, the more severe the degree of fault representing the corresponding fault mode.
Typical failure modes of the electromechanical actuator selected by the embodiment of the invention include a motor winding turn-to-turn short circuit fault, a motor bearing surface damage fault, a ball screw surface damage fault, a transmission mechanism clamping stagnation fault and a transmission mechanism gap oversize fault. Wherein,
the fault monitoring quantity corresponding to the turn-to-turn short circuit fault of the motor winding is the difference of three-phase currents of the motor;
the fault monitoring quantity corresponding to the damage fault of the surface of the motor bearing is a vibration amplitude on the characteristic frequency of the motor bearing fault;
the fault monitoring amount corresponding to the damage fault of the surface of the ball screw is the standard deviation (representing the vibration intensity) of the vibration signal of the ball screw;
the fault monitoring quantity corresponding to the jamming fault of the transmission mechanism is the control panel current of the motor;
the fault monitoring amount of the fault of the overlarge clearance of the transmission mechanism is the output displacement/control surface deflection angle of the electromechanical actuator.
Step 202: the electromechanical actuator which is completely normal and does not have faults is tested, and the running data and the system state quantity of the electromechanical actuator under various different input signals and different working conditions are collected; the operation data comprises command signals of the electromechanical actuator with different waveforms, different frequencies and different amplitudes, such as sine waves, square waves and trapezoidal waves, various input signals, motor control input voltage under different working conditions, load data under different load conditions and the like. The system state quantity comprises motor control panel current, motor three-phase current, motor vibration signals, lead screw vibration signals and output displacement.
Step 203: and (3) constructing a regression model based on the long-term and short-term memory neural network, and inputting the preprocessed input signals into the regression model to predict the reference state quantity corresponding to each fault monitoring quantity at the next moment.
The regression network is trained according to collected experimental data of different input signals and different working conditions to obtain a regression model, and the regression model is used as a reference model of a normal electromechanical actuator. The regression model includes a long-short term memory neural network layer and a fully connected neural network layer. The input data are an electromechanical actuator instruction signal, motor control input voltage and load data corresponding to the current moment, and the output data are control panel current of the motor, difference of three-phase current of the motor, vibration amplitude on motor bearing fault characteristic frequency, standard deviation of ball screw vibration signals and output displacement of the electromechanical actuator at the next moment.
The training process of the regression model is as follows:
and performing data splicing on the input data according to the acquisition time for the data of different channels acquired in the step 202, namely the data of different components acquired by different sensors. Assuming that the length of the acquired data is t, the signal after splicing can be regarded as an n-dimensional time domain data, which includes information of n sensor channels. For the regression network input signals, n =3, including the electromechanical actuator command signal, the motor control input voltage, and the load data. The data of different sensor channels have different dimensions, so that the n-dimensional time domain data needs to be normalized by the maximum and minimum normalization. In order to increase the number of samples, n-dimensional time domain data after normalization is cut through sliding windows, each sliding window contains time domain data in a short period of time, the data are cut into 0-1s,0.01-1.01s and 0.02-1.02s, and the like, a training set of a regression network is obtained, and the training set comprises processed electromechanical actuator command signals, motor control input voltage, load data, namely input samples, control plate currents of a motor, difference of three-phase currents of the motor, vibration amplitude values on motor bearing fault characteristic frequencies, standard difference of ball screw vibration signals and output displacement of an electromechanical actuator, namely label data.
Training the model through a training set; because the data in the training set are experimental data of the normal electromechanical actuator in working under different experimental conditions, the data represent a working model of the normal electromechanical actuator. The trained model, namely the regression model, is used as a reference model of the electromechanical actuator, and the reference value of the failure monitoring quantity of each channel when the electromechanical actuator does not fail can be calculated through input data, namely, the value which is supposed to be output by the failure monitoring quantity of the electromechanical actuator in the normal state at the next moment can be obtained through input in a period of time.
Step 204: based on the constructed composite health state space of the electromechanical actuator, the fault degree of each layer is evaluated layer by layer according to a step-by-step mapping mode through the collected multichannel state data of the electromechanical actuator and a reference model obtained through training, and finally the overall health index of the electromechanical actuator is comprehensively considered. The method comprises the following steps:
first, data of each channel of the electromechanical actuator to be evaluated during working is measured, and the type of the measured data of each channel is the same as that of the data collected in step 202.
Then, based on the electromechanical actuator reference model trained in step 203, the reference value of the control board current of the motor, the reference value of the difference between the three-phase currents of the motor, the reference value of the vibration amplitude on the motor bearing fault characteristic frequency, the reference value of the standard deviation of the ball screw vibration signal and the reference value of the output displacement of the electromechanical actuator at the next moment are calculated by using the processed electromechanical actuator instruction signal, the motor control input voltage and the load data as input data.
And finally, comparing the reference value of the next moment with the actual measurement value of the next moment, and evaluating the fault degree of each layer by layer according to a step-by-step mapping mode to finally calculate the overall health index of the electromechanical actuator.
The health assessment method for layer-by-layer mapping is totally divided into three layers, the fault mode layer of the bottom layer is based on the analysis in the step 201, and the typical fault modes of motor winding turn-to-turn short circuit, motor bearing surface damage, ball screw surface damage, overlarge transmission gap, clamping stagnation of a transmission mechanism and five electromechanical actuators are selected as five subspaces of the bottom layer; the fault monitoring amount corresponding to the motor winding turn-to-turn short circuit fault is the difference of three phase currents of the motor, the fault monitoring amount corresponding to the motor bearing surface damage fault is the vibration amplitude on the motor bearing fault characteristic frequency, the fault monitoring amount corresponding to the ball screw surface damage fault is the standard deviation (representing the vibration intensity) of a ball screw vibration signal, the fault monitoring amount corresponding to the transmission mechanism jamming fault is the control panel current of the motor, and the fault monitoring amount corresponding to the transmission mechanism clearance oversize fault is the output displacement/control surface deflection angle of the electromechanical actuator. After the fault monitoring amount of each fault mode is determined, the fault degree of the corresponding fault mode is obtained according to the deviation between the fault monitoring amount and the reference value, mapping is carried out on the upper key component layer, and the calculation formula of the fault degree of the fault mode is as follows:
wherein u is an actual value of a fault monitoring quantity obtained by measuring through a sensor on an electromechanical actuator to be evaluated, and u isnomFor fault monitoring quantity reference values, u, output by regression modelsmaxThe fault threshold value at fault, i.e. the value at which the deviation from the normal value is maximal, is monitored for each fault determined in step 201.
Different possible failure modes are comprehensively considered for each important part of the electromechanical actuator, namely the driving motor, the ball screw and the transmission mechanism in the important part layer, and the driving motor is considered to be possible to have faults of motor winding turn-to-turn short circuit and motor bearing surface damage; for the ball screw, a failure in which damage to the surface of the ball screw may occur is considered; the transmission mechanism is wide in concept, transmission-related faults possibly occurring in the whole transmission chain are considered, the fault that transmission gaps are too large or the transmission mechanism is stuck is considered, the fault degree of a component for each typical fault mode is used as the base of a key component layer subspace, the fault degree of the component is obtained through Euclidean distance evaluation function mapping, and the fault degree is output to the integral layer of the electromechanical actuator.
Wherein, tjDegree of failure, x, of the jth critical componentiDegree of failure, x, of the ith failure mode in the jth critical sectionmax-iHere, the maximum value of the failure degree corresponding to the ith failure mode in the jth important component is set to 1,n, which represents the sum of the failure modes corresponding to the jth important component.
The electromechanical actuator monolithic layer divides the electromechanical actuator into three key parts: the driving motor, the ball screw and the transmission mechanism are used for obtaining the integral health index of the electromechanical actuator by mapping Euclidean distance evaluation functions by taking the fault degrees of three key parts as the base of the space of the layer.
Wherein y is the health index of the electromechanical actuator, tjDegree of failure, t, of the jth important componentmax-jThe maximum value of the failure degree of the jth key component is set as 1; m represents the sum of the number of important parts in the electromechanical actuator, and is set to 3 here.
The embodiment of the invention has the advantages and positive effects that:
firstly, the reference model construction method based on data does not need to establish a complex electromechanical actuator model, and avoids the interference caused by various nonlinear links and error terms.
Secondly, the health state assessment method has high real-time performance and can reflect the health indexes of the electromechanical actuator in real time in the operation process of the electromechanical actuator.
Thirdly, compared with the traditional feature extraction, the provided health state assessment method directly uses the classifier to classify the current data, and can reflect the current dangerous state of the system through the larger difference between the reference value and the actual value of each channel when a fault which does not contain prior information is faced.
Fourthly, the health assessment of the electromechanical actuator not only comprises the assessment of the overall health of the system, but also comprises the analysis of the failure degree of each key component in the system and the typical failure mode, so that the actual operation condition of the electromechanical actuator can be better reflected.
EXAMPLE III
Because the experiment platform of the temporary inorganic electric actuator, the method provided by the embodiment of the invention is realized by the open source electromechanical actuator experiment data set based on NASA, namely the design of the electromechanical actuator health state monitoring method based on the composite space, which is specifically as follows:
(1) Analysis of typical failure characteristics of electromechanical actuators
Firstly, typical faults in the working process of the electromechanical actuator are analyzed, and for turn-to-turn short circuit faults of a motor winding, three phases of the motor are not symmetrical any more, so that three-phase currents are not symmetrical any more; for the surface damage fault of a motor bearing and the surface damage fault of a ball screw, a sudden impact pulse force is generated each time when a damage point is contacted with the surface of an element, so that the amplitude of the vibration signal on the characteristic frequency is increased; for the fault of overlarge transmission clearance, the output displacement of the fault can be influenced by the clearance to generate deviation; for a transmission mechanism jamming fault, the jamming torque can cause the increase of the control panel current of the motor. Therefore, the fault monitoring quantity of each typical fault is determined, the fault monitoring quantity corresponding to the turn-to-turn short circuit fault of the motor winding is the difference of three-phase currents of the motor, the fault monitoring quantity corresponding to the surface damage fault of the motor bearing is the vibration amplitude on the characteristic frequency of the motor bearing, the fault monitoring quantity corresponding to the surface damage fault of the ball screw is the standard deviation (representing the vibration intensity) of a vibration signal of the ball screw, the fault monitoring quantity of the clamping fault of the transmission mechanism is the control panel current of the motor, and the fault monitoring quantity of the transmission clearance fault is the output displacement/control plane deflection angle of the electromechanical actuator.
(2) Constructing a composite health State space
Based on the above analysis, a composite health state space of the electromechanical actuator as shown in fig. 3 is constructed, and the state of the electromechanical actuator as a whole, i.e., HI, is evaluated through a three-layer space of electromechanical actuator, important parts, failure modesEMA。
(3) Reference model for training electromechanical actuator
The method is adopted to evaluate the NASA open source electromechanical actuator experimental data set, and the reference model of the electromechanical actuator is trained through the electromechanical actuator experimental data in the normal state so as to obtain the fault monitoring quantity reference value at the next moment. And due to the defect that the data set collects data, vibration signals of the bearing and three-phase currents of the motor are not included. The training results of the regression model of the three-channel signal of the output displacement, the control panel current and the vibration signal standard deviation are shown in fig. 4-6.
(4) Assessing electromechanical actuator health status
Because the defect that the data set collects data does not contain vibration signals of the bearing and three-phase currents of the motor, the damage fault of the surface of the bearing of the motor and the short circuit fault of the turn-to-turn winding of the motor are not considered. And obtaining a reference value of the fault monitoring quantity through the trained regression model, and evaluating the health state through a composite space.
The evaluation objects comprise an electromechanical actuator in a normal state, an electromechanical actuator injected with a jamming failure state of a transmission mechanism and an electromechanical actuator injected with a surface damage failure state of a ball screw, and the evaluation results are shown in fig. 7-9. It can be observed that, for the electromechanical actuator in the normal state, the actual value and the reference value of each fault monitoring amount are very close to each other; for the electromechanical actuator injected with the clamping stagnation fault state of the transmission mechanism, the standard deviation of the control plate current of the motor generating obvious deviation and the ball screw vibration signal with smaller deviation can be observed; for an electromechanical actuator injected with a ball screw surface damage fault condition, the standard deviation of the ball screw vibration signal, which produces a significant deviation, can be observed. By adopting the method, the health state evaluation of the electromechanical actuator under different states is realized.
Example four
As shown in fig. 10, a health status monitoring system for an electromechanical actuator based on a complex space according to an embodiment of the present invention is used for monitoring the health status of a target electromechanical actuator; the target electromechanical actuator comprises a plurality of target components, each target component corresponds to one or more fault modes, and each fault mode corresponds to a fault monitoring amount; the composite health state space of the target electromechanical actuator comprises a bottom layer, a middle layer and a top layer; the bottom layer includes all failure modes of the target electromechanical actuator; the intermediate layer includes all of the target components of the target electromechanical actuator; the top layer is the target electromechanical actuator; the electromechanical actuator state of health monitoring system includes:
the data obtaining module 1001 is configured to obtain operation data of the target electromechanical actuator at a previous time, and obtain an actual value of a fault monitoring amount of the target electromechanical actuator at a current time.
And a fault monitoring amount reference value predicting module 1002, configured to predict a fault monitoring amount reference value of the target electromechanical actuator at the current time based on the operation data of the target electromechanical actuator at the previous time and a machine learning algorithm.
The electromechanical actuator health state determining module 1003 is configured to monitor the fault degree of each layer in the composite health state space of the target electromechanical actuator in a step-by-step mapping manner based on the fault monitoring amount reference value of the target electromechanical actuator at the current moment and the fault monitoring amount actual value of the target electromechanical actuator at the current moment, so as to obtain the health state of the target electromechanical actuator; the state of health includes a degree of failure of each of the failure modes in the target electromechanical actuator, a degree of failure of each of the target components in the target electromechanical actuator, and a degree of overall failure of the target electromechanical actuator.
On the basis of the embodiment described in fig. 10, the system provided by this embodiment further includes: and the regression model building module is used for building a regression model based on the long-term and short-term memory neural network.
On the basis of the embodiment shown in fig. 10, the fault monitoring amount reference value predicting module 1002 specifically includes:
the preprocessing unit is used for preprocessing the operation data of the target electromechanical actuator at the last moment; the preprocessing comprises a data splicing operation, a normalization operation and a sliding window cutting operation.
And the fault monitoring quantity reference value prediction unit is used for predicting the fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the preprocessed running data of the target electromechanical actuator at the last moment and the regression model.
On the basis of the embodiment shown in fig. 10, the health status determining module 1003 for the electromechanical actuator specifically includes:
and the fault degree calculation unit of the fault modes is used for calculating the fault degree of each fault mode based on the reference value of the fault monitoring amount of the target electromechanical actuator at the current moment and the actual value of the fault monitoring amount of the target electromechanical actuator at the current moment.
A target component failure degree calculation unit configured to determine a relationship of the failure mode to the target component based on a composite state of health space of the target electromechanical actuator, and determine a failure degree of each of the target components based on the relationship of the failure mode to the target component and the failure degree of each of the failure modes.
And the overall fault degree determining unit of the target electromechanical actuator is used for determining the overall fault degree of the target electromechanical actuator based on the fault degree of each target component.
The invention provides a method and a system for evaluating the health state of an electromechanical actuator based on a composite space, and provides the method and the system for evaluating the health state of the electromechanical actuator, which have high real-time performance and strong universality and comprise complete quantitative evaluation on each fault mode and all parts of a system to the whole body. According to the method and the system, the corresponding fault characteristics are extracted by analyzing the typical fault mode of the electromechanical actuator, and the composite health state space of the electromechanical actuator is constructed. The method comprises the steps of carrying out experiments on a completely normal electromechanical actuator, collecting experimental data of the electromechanical actuator under each channel, constructing a regression model based on a long-term and short-term memory neural network, using the regression model as a reference model of the electromechanical actuator, and calculating a reference value of the fault monitoring quantity of each channel when the electromechanical actuator does not fail through input data. On the basis, the reference value is compared with the actual measurement value, the fault degree of each layer is evaluated layer by layer according to a step-by-step mapping mode, and finally the overall health index of the electromechanical actuator system is calculated. The method realizes the real-time health state evaluation of the electromechanical actuator, can evaluate the health state of the electromechanical actuator from the fault to the part to each level of the whole electromechanical actuator based on the constructed composite health state space, and provides information for maintenance decision. In practical application, a reference value of the fault monitoring amount needs to be obtained, so that a reference model of the electromechanical actuator is trained through actual experimental data based on an LSTM neural network, and the reference value of each fault monitoring amount at the next moment is predicted.
In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the foregoing, the description is not to be taken in a limiting sense.
Claims (10)
1. A method for monitoring the health state of an electromechanical actuator based on a composite space is characterized in that the method for monitoring the health state of the electromechanical actuator is used for monitoring the health state of a target electromechanical actuator; the target electromechanical actuator comprises a plurality of target components, each target component corresponds to one or more fault modes, and each fault mode corresponds to a fault monitoring amount; the composite health state space of the target electromechanical actuator comprises a bottom layer, a middle layer and a top layer; the bottom layer includes all failure modes of the target electromechanical actuator; the intermediate layer includes all of the target components of the target electromechanical actuator; the top layer is the target electromechanical actuator; the method for monitoring the health state of the electromechanical actuator comprises the following steps:
acquiring operation data of the target electromechanical actuator at the previous moment and acquiring an actual value of a fault monitoring quantity of the target electromechanical actuator at the current moment;
predicting a fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the operation data of the target electromechanical actuator at the previous moment and a machine learning algorithm;
monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator in a step-by-step mapping mode based on the fault monitoring amount reference value of the target electromechanical actuator at the current moment and the fault monitoring amount actual value of the target electromechanical actuator at the current moment so as to obtain the health state of the target electromechanical actuator; the state of health includes a degree of failure of each of the failure modes in the target electromechanical actuators, a degree of failure of each of the target components in the target electromechanical actuators, and a degree of failure of the target electromechanical actuators as a whole.
2. The method for monitoring the state of health of an electromechanical actuator based on a composite space according to claim 1, further comprising:
and carrying out fault characteristic analysis on each fault mode in the target electromechanical actuator, and determining a fault monitoring amount used for monitoring the fault mode.
3. The method for monitoring the state of health of an electromechanical actuator based on a composite space according to claim 1, further comprising: and constructing a regression model based on the long-term and short-term memory neural network.
4. The method for monitoring the health state of the electromechanical actuator based on the composite space as claimed in claim 3, wherein the constructing of the regression model based on the long-term and short-term memory neural network specifically comprises:
constructing a recurrent neural network; the recurrent neural network comprises a long-short term memory neural network layer and a full-connection neural network layer;
constructing a training sample set; the training sample set comprises a plurality of sample input data and label data corresponding to each sample input data; the sample input data is the running data of the target electromechanical actuator in a normal state; the tag data is an actual value of the fault monitoring amount of the target electromechanical actuator in a normal state;
and training the regression neural network based on the training sample set to obtain a regression model.
5. The method for monitoring the health state of the electromechanical actuator based on the composite space as claimed in claim 3, wherein the predicting the reference value of the fault monitoring amount of the target electromechanical actuator at the current time based on the operation data of the target electromechanical actuator at the previous time and a machine learning algorithm specifically comprises:
preprocessing the operation data of the target electromechanical actuator at the last moment; the preprocessing comprises data splicing operation, normalization operation and sliding window cutting operation;
and predicting a fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the preprocessed operation data of the target electromechanical actuator at the last moment and the regression model.
6. The method for monitoring the health state of the electromechanical actuator based on the composite space as claimed in claim 1, wherein the step-by-step mapping method is used to monitor the fault degree of each layer in the composite health state space of the target electromechanical actuator based on the reference value of the fault monitoring amount of the target electromechanical actuator at the current moment and the actual value of the fault monitoring amount of the target electromechanical actuator at the current moment, so as to obtain the health state of the target electromechanical actuator, and specifically comprises:
calculating the fault degree of each fault mode based on the fault monitoring quantity reference value of the target electromechanical actuator at the current moment and the fault monitoring quantity actual value of the target electromechanical actuator at the current moment;
determining a relationship of the failure mode to the target component based on a composite state-of-health space of the target electromechanical actuator, and determining a degree of failure for each of the target components based on the relationship of the failure mode to the target component and the degree of failure for each of the failure modes;
determining an overall degree of failure of the target electromechanical actuator based on the degree of failure of each of the target components.
7. A health state monitoring system of an electromechanical actuator based on a composite space is characterized in that the health state monitoring system of the electromechanical actuator is used for monitoring the health state of a target electromechanical actuator; the target electromechanical actuator comprises a plurality of target components, each target component corresponds to one or more fault modes, and each fault mode corresponds to a fault monitoring amount; the composite health state space of the target electromechanical actuator comprises a bottom layer, a middle layer and a top layer; the bottom layer includes all failure modes of the target electromechanical actuator; the intermediate layer includes all of the target components of the target electromechanical actuator; the top layer is the target electromechanical actuator; the electromechanical actuator state of health monitoring system includes:
the data acquisition module is used for acquiring the operation data of the target electromechanical actuator at the previous moment and acquiring the actual value of the fault monitoring amount of the target electromechanical actuator at the current moment;
the fault monitoring quantity reference value prediction module is used for predicting the fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the operation data of the target electromechanical actuator at the previous moment and a machine learning algorithm;
the electromechanical actuator health state determining module is used for monitoring the fault degree of each layer in the composite health state space of the target electromechanical actuator by adopting a step-by-step mapping mode based on the fault monitoring amount reference value of the target electromechanical actuator at the current moment and the fault monitoring amount actual value of the target electromechanical actuator at the current moment so as to obtain the health state of the target electromechanical actuator; the state of health includes a degree of failure of each of the failure modes in the target electromechanical actuator, a degree of failure of each of the target components in the target electromechanical actuator, and a degree of overall failure of the target electromechanical actuator.
8. The composite space-based electromechanical actuator state of health monitoring system of claim 7, further comprising: and the regression model building module is used for building a regression model based on the long-term and short-term memory neural network.
9. The system for monitoring the health state of the electromechanical actuator based on the composite space as claimed in claim 8, wherein the module for predicting the reference value of the fault monitoring amount specifically comprises:
the preprocessing unit is used for preprocessing the operation data of the target electromechanical actuator at the last moment; the preprocessing comprises data splicing operation, normalization operation and sliding window cutting operation;
and the fault monitoring quantity reference value prediction unit is used for predicting the fault monitoring quantity reference value of the target electromechanical actuator at the current moment based on the preprocessed running data of the target electromechanical actuator at the last moment and the regression model.
10. The system for monitoring the state of health of an electromechanical actuator based on a composite space of claim 7, wherein the module for determining the state of health of the electromechanical actuator comprises:
a failure degree calculation unit of a failure mode, configured to calculate a failure degree of each failure mode based on a reference value of a failure monitoring amount of the target electromechanical actuator at a current time and an actual value of the failure monitoring amount of the target electromechanical actuator at the current time;
a target component failure degree calculation unit for determining a relationship of the failure mode to the target component based on a composite state-of-health space of the target electromechanical actuator, and determining a failure degree of each of the target components based on the relationship of the failure mode to the target component and the failure degree of each of the failure modes;
and the overall fault degree determining unit of the target electromechanical actuator is used for determining the overall fault degree of the target electromechanical actuator based on the fault degree of each target component.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210271455.1A CN115270591A (en) | 2022-03-18 | 2022-03-18 | Method and system for monitoring health state of electromechanical actuator based on composite space |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210271455.1A CN115270591A (en) | 2022-03-18 | 2022-03-18 | Method and system for monitoring health state of electromechanical actuator based on composite space |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115270591A true CN115270591A (en) | 2022-11-01 |
Family
ID=83758439
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210271455.1A Pending CN115270591A (en) | 2022-03-18 | 2022-03-18 | Method and system for monitoring health state of electromechanical actuator based on composite space |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115270591A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115755835A (en) * | 2022-11-04 | 2023-03-07 | 四川大学 | Online health factor optimization method for electromechanical actuator |
| CN116449135A (en) * | 2023-04-19 | 2023-07-18 | 北京航空航天大学 | Method and system for determining health state of electromechanical system component and electronic equipment |
-
2022
- 2022-03-18 CN CN202210271455.1A patent/CN115270591A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115755835A (en) * | 2022-11-04 | 2023-03-07 | 四川大学 | Online health factor optimization method for electromechanical actuator |
| CN116449135A (en) * | 2023-04-19 | 2023-07-18 | 北京航空航天大学 | Method and system for determining health state of electromechanical system component and electronic equipment |
| CN116449135B (en) * | 2023-04-19 | 2024-01-30 | 北京航空航天大学 | Method and system for determining health state of electromechanical system component and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Byington et al. | Data-driven neural network methodology to remaining life predictions for aircraft actuator components | |
| Long et al. | Attitude data-based deep hybrid learning architecture for intelligent fault diagnosis of multi-joint industrial robots | |
| CN115270591A (en) | Method and system for monitoring health state of electromechanical actuator based on composite space | |
| EP2204778B1 (en) | Prognostics and health monitoring for electro-mechanical systems and components | |
| Jardine et al. | A review on machinery diagnostics and prognostics implementing condition-based maintenance | |
| Tibaduiza et al. | Damage classification in structural health monitoring using principal component analysis and self‐organizing maps | |
| Camci et al. | Feature evaluation for effective bearing prognostics | |
| CN112884051B (en) | Data-driven multi-component online complex fault diagnosis method for lightweight unmanned aerial vehicle | |
| CN105718876B (en) | A kind of appraisal procedure of ball-screw health status | |
| RU2626231C1 (en) | Method of diagnostics of technical condition and electromechanical device remaining lifetime estimation with asynchronous motor | |
| CN112785091B (en) | Method for carrying out fault prediction and health management on oil field electric submersible pump | |
| Kordestani et al. | A new fault prognosis of MFS system using integrated extended Kalman filter and Bayesian method | |
| EP2409398B1 (en) | Device for monitoring step switches | |
| CN115526375A (en) | Space flight equipment predictive maintenance system | |
| Lahmadi et al. | A data-driven method for estimating the remaining useful life of a composite drill pipe | |
| CN114356897A (en) | Electromechanical actuator fault diagnosis method based on data fusion | |
| Veresnikov et al. | The Electromechanical Actuator Technical Condition Monitoring System Based on Data Mining Methods | |
| US20050278146A1 (en) | Method for simplified real-time diagnoses using adaptive modeling | |
| Valceschini et al. | Model-based fault diagnosis of sliding gates electro-mechanical actuators transmission components with motor-side measurements | |
| Veresnikov et al. | The development of data mining methods criteria to identify failures of aircraft control surface actuator | |
| DE102008021248A1 (en) | Unidirectionally reinforced fiber composite material's structural condition determining method for use in e.g. automobile manufacturing field, involves testing material and determining viscoelastic material characteristics of material | |
| He et al. | Probabilistic model based algorithms for prognostics | |
| Zhang | Condition monitoring of a fan using neural networks | |
| Baïkeche et al. | On parametric and nonparametric fault detection in linear closed-loop systems | |
| Craig | Data driven approach to non-stationary EMA fault detection and investigation into remaining useful life |
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 |