CN117214591A - Fault diagnosis system and method for deep-diving propeller - Google Patents
Fault diagnosis system and method for deep-diving propeller Download PDFInfo
- Publication number
- CN117214591A CN117214591A CN202311483722.2A CN202311483722A CN117214591A CN 117214591 A CN117214591 A CN 117214591A CN 202311483722 A CN202311483722 A CN 202311483722A CN 117214591 A CN117214591 A CN 117214591A
- Authority
- CN
- China
- Prior art keywords
- voltage
- module
- fault
- propeller
- data
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000004891 communication Methods 0.000 claims abstract description 27
- 230000009467 reduction Effects 0.000 claims abstract description 13
- 239000011159 matrix material Substances 0.000 claims description 27
- 238000003066 decision tree Methods 0.000 claims description 15
- 238000004088 simulation Methods 0.000 claims description 11
- 238000005259 measurement Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000004804 winding Methods 0.000 claims description 8
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 241000657949 Elderberry carlavirus D Species 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000001934 delay Effects 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 5
- 239000002131 composite material Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 230000001105 regulatory effect Effects 0.000 description 4
- 238000002955 isolation Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000017525 heat dissipation Effects 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000007789 sealing Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 229910001220 stainless steel Inorganic materials 0.000 description 2
- 239000010935 stainless steel Substances 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000001994 activation Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 230000001808 coupling effect Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a fault diagnosis system and a fault diagnosis method for a deep-diving apparatus propeller, which relate to the technical field of deep-diving apparatus propulsion system fault diagnosis and comprise the following steps: outer box, display screen, keyboard and electrical equipment, outer box have the upper cover and with the box body of the one end connection of upper cover, the display screen is embedded in the upper cover, the keyboard is located box body upper surface, the inside electrical equipment that has of box body, electrical equipment includes: the system comprises a power supply module, a stabilized voltage power supply, a voltage reduction module, an upper computer, a communication module, a voltage output module and a current and voltage acquisition module; the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are connected with the upper computer through the communication module; the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are respectively connected with the propeller. The technical scheme of the invention solves the problem that the fault diagnosis system of the deep submersible propeller in the prior art cannot adapt to complex climate conditions, so that high-precision and high-efficiency fault diagnosis cannot be performed.
Description
Technical Field
The invention relates to the technical field of fault diagnosis of a deep-diving propeller system, in particular to a fault diagnosis system and method for a deep-diving propeller.
Background
With the rapid development of the fields of artificial intelligence, materials, energy sources and the like, the mission capability of the submersible is continuously improved. The working scene is mainly a complex and changeable marine environment, and is easy to be subjected to the coupling effect of loads such as wind, wave and current. How to ensure that the safety and reliability of the submersible vehicle can finish various tasks is an important precondition for continuous development of the submersible vehicle, and how to accurately and effectively monitor the state and diagnose faults of the submersible vehicle is a key technology for ensuring the safety of the submersible vehicle. The submersible propeller is an important part of the submersible, and is closely related to any voyage of the aircraft, so that fault diagnosis of the propeller is also extremely important.
The propeller has heavy load, more parts and complex and changeable fault reasons, and the research on fault diagnosis of the propeller has been widely focused on the ocean science community in the fields of motion modeling, information processing, data mining, deep learning and the like, and has obtained a plurality of research results. However, compared with the fault diagnosis and health management systems of the propulsion systems such as industrial automation and aircrafts, the basic theory of fault diagnosis and the research and development of intelligent health monitoring systems are different, and research and development of fault diagnosis equipment of the propulsion systems are lacking.
With the continuous complexity of the running environment of the submersible, the task types are continuously rich, and whether the propeller is capable of working normally or not has an important influence on the submersible by taking the propeller as a main component. The fault diagnosis system is used as an important tool for fault diagnosis, and the current fault diagnosis mainly depends on a personal portable handheld or desktop computer, but as the working environment of a submersible vehicle tends to be complicated, the personal portable computer can not adapt to severe weather conditions such as high temperature, high salt, extremely cold, storm and the like, so that the working capacity is lost, and the accurate and efficient fault diagnosis can not be realized.
Accordingly, there is a need for a fault diagnosis system and method that accommodates complex climate conditions and detects, identifies and locates the operational status of a deep-submersible vehicle propulsion.
Disclosure of Invention
The invention mainly aims to provide a fault diagnosis system and method for a deep-diving propeller, which are used for solving the problem that the fault diagnosis system of the deep-diving propeller in the prior art cannot adapt to complex climate conditions so as to be incapable of performing high-precision and high-efficiency fault diagnosis.
To achieve the above object, the present invention provides a fault diagnosis system for a deep-submersible propeller, comprising: outer box, display screen, keyboard and electrical equipment, outer box have the upper cover and with the box body of the one end connection of upper cover, the display screen is embedded in the upper cover, the keyboard is located box body upper surface, the inside electrical equipment that has of box body, electrical equipment includes: the system comprises a power supply module, a stabilized voltage supply, a voltage reduction module, an upper computer, a communication module, a voltage output module, a current and voltage acquisition module and a rotating speed acquisition module; the stabilized voltage power supply is also connected with the propeller; the voltage reducing module is connected with the upper computer, and the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are connected with the upper computer through the communication module; the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are respectively connected with the propeller.
Further, the power supply module converts the input 220V alternating voltage into 24V direct voltage, wherein one path of 24V direct voltage supplies power to the upper computer and the display screen through the voltage reduction module, and the 24V direct voltage also supplies power to the communication module, the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module respectively.
Further, the power supply module is connected with the stabilized power supply and then is connected with the upper computer through the 2-core plug connector respectively, the stabilized power supply is connected with the 2-core plug connector through a lead and then is connected with two cores of the 10-core plug connector, and finally is connected with the propeller, and the 10-core plug connector penetrates out of the inner part of the outer box body; the regulated power supply converts 220V alternating current into 110V direct current to supply power to the propeller, and the upper computer receives a current voltage signal output by the regulated power supply to the propeller through the 485 module; the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are connected with the propeller through a part of wires of the 10-core plug-in connector and transmit signals.
Further, the voltage output module outputs the input digital signal into a continuous voltage signal through digital-to-analog conversion.
Further, the rotating speed acquisition module is a PWM acquisition module.
Further, the communication module converts the RS485 serial port signal into an Ethernet signal, the upper computer receives data through the Ethernet, and the communication module performs data transmission with the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module through the 485 module.
The invention also provides a fault diagnosis method for the deep submersible vehicle propeller, which comprises the following steps:
s1, acquiring operation data of the deep-diving propeller, and reducing the dimension of the operation data by using a PCA method.
S2, classifying the operation data by adopting a DBSCAN algorithm to extract fault data.
And S3, carrying out data enhancement on fault data by adopting an RFR algorithm to obtain fault sample data, preprocessing the sample data, and endowing corresponding fault class labels.
And S4, performing fault diagnosis based on the convolutional neural network.
Further, the step S1 specifically includes the following steps:
s1.1, matrixIs composed of propeller operation data, matrixThe definition is as follows:
（1）。
s1.2, centralizing the data of each column to obtain a new data matrix, wherein the centralization formula is as follows:
（2）。
in the formula (2)Representing the data being centred on the data being processed,representative traversal NoEach of the rows of data.
S1.3, calculating covariance matrix of the sample：
（3）；
In (3)Represented as a centered matrix.
S1.4, decomposing the covariance matrix based on the eigenvalues or SVDs, and solving eigenvalues and corresponding eigenvectors of the covariance matrix.
S1.5, arranging the eigenvectors into a matrix according to the corresponding eigenvalues from top to bottom and obtaining k rows to form the matrix。
S1.6, obtaining new data after dimension reduction：
（4）；
Namely, obtaining the data after dimension reduction:
（5）。
further, the step S3 specifically includes the following steps:
s3.1, obtaining sample data of three fault categories including normal state of the propeller, damage of the driver, blade loss and foreign matter winding by constructing an experimental platform.
S3.2, preprocessing the data through manual or program screening, identifying and removing abnormal points and outliers in the measured data, and generating a fault sample.
S3.3, training RFRs by using the preprocessed fault samples, respectively obtaining the number of decision trees by using a grid search method for the fault samples in different propeller states, and identifying various parameters of the RFR models to obtain 3 RFR models capable of representing the propeller performances in different fault states.
S3.4, adding Gaussian noise to simulate the conditions of ocean noise and environmental interference, and generating various simulation fault samples with the same scale as the actual measurement samples according to the missing degree of each fault sample by utilizing the trained RFR in proportion, so that the ratio of the actual measurement samples to each simulation fault sample is 1:1.
S3.5, integrating the actual measurement sample and the simulation sample according to the requirements of fault detection and classification so as to ensure the relative diversity of the fault sample, preprocessing the generated sample data and marking corresponding fault class labels.
Further, the step S4 specifically includes the following steps:
s4.1, calculating the voltage-current correlation coefficientAnd voltage rotation speed related coefficientTo eliminate delays of the control signal and the action signal;
（6）。
in (7)Representing the first of the voltagesThe number of observations made is a function of the number of observations,represents the average value of the voltage and,representing the i-th observation of the current,representing the average value of the current.
（7）；
In (8)Represents the first rotation speedThe number of observations made is a function of the number of observations,representing the average value of the rotational speed.
And S4.2, inputting the control signal and the action signal into a convolutional neural network to perform fault diagnosis.
The invention has the following beneficial effects:
the fault diagnosis system provided by the invention has the advantages that the electrical equipment is packaged in the outer box body, the outer box body is made of the composite material, and the composite material has high strength, good heat dissipation performance, excellent heat resistance, weather resistance and impact resistance. The fault diagnosis system is provided with a human engineering lifting handle, is convenient to carry when going out, adopts a high-density protection pad design, and has strong body anti-falling capability. The inner sealing performance is good, dust can be effectively prevented from entering, and the waterproof performance is excellent. Therefore, the device can be more suitable for complex climate conditions, the display screen of the fault diagnosis system is arranged in the upper box body of the fault diagnosis system, the TFT LCD display screen is adopted, the resolution is 1920x1080, and the rotation within the range of less than or equal to 90 degrees can be realized in the YOZ plane. The fault diagnosis system is provided with three USB interfaces, and can use a USB flash disk to import or export data, and can also be externally connected with other experimental equipment, so that the data transmission and experimental efficiency are improved. The keyboard is made of stainless steel material, has excellent waterproof, dustproof and anti-corrosion properties, and has good sensitivity and stability.
The fault diagnosis method provided by the invention can be used for stably running the system and obtaining a better data set through a pool experiment in a laboratory environment.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art. In the drawings:
fig. 1 shows a schematic view of a part of the fault diagnosis system for a deep-submersible propeller according to the invention.
Fig. 2 shows a result diagram of optimizing the number of rotational speed signals of a blade loss fault RFR decision tree by using the fault diagnosis method for the deep submersible propeller provided by the invention.
FIG. 3 shows a result diagram of optimizing the number of current signals of a blade loss fault RFR decision tree by using the fault diagnosis method for the deep submersible propeller.
FIG. 4 shows a result diagram of optimizing the number of rotation speed signals of the RFR decision tree for foreign object winding faults by using the fault diagnosis method for the deep submersible vehicle propeller.
FIG. 5 shows a result diagram of optimizing the number current signals of the RFR decision tree of the foreign object winding fault by using the fault diagnosis method for the deep submersible vehicle propeller.
FIG. 6 shows a graph of the result of optimizing the number of rotational speed signals of the RFR decision tree for the damage to the driver by using the fault diagnosis method for the deep submersible vehicle propeller.
FIG. 7 shows a graph of the result of optimizing the number of current signals of the RFR decision tree for the damage to the driver by using the fault diagnosis method for the deep submersible vehicle propeller.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A fault diagnosis system for a deep-submersible vehicle as shown in fig. 1, comprising: outer box, display screen, keyboard and electrical equipment, outer box have the upper cover and with the box body of the one end connection of upper cover, the display screen is embedded in the upper cover, the keyboard is located box body upper surface, the inside electrical equipment that has of box body, electrical equipment includes: the system comprises a power supply module, a stabilized voltage supply, a voltage reduction module, an upper computer, a communication module, a voltage output module, a current and voltage acquisition module and a rotating speed acquisition module; the stabilized voltage power supply is also connected with the propeller; the voltage reducing module is connected with the upper computer, and the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are connected with the upper computer through the communication module; the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are respectively connected with the propeller.
The power module adopts an LRS-200-24 power supply, and can realize RS485 communication isolation and input optical coupling isolation; the communication interface supports RS485; support a standard modbus protocol; and automatically storing the transmitted data. And an LRS-200-24 power supply is utilized to convert the input alternating voltage 220V into the output direct voltage 24V, and power is supplied to equipment in the whole system such as a voltage output module, a current and voltage acquisition module, a rotating speed acquisition module, a display screen and the like.
The following describes the selection of the current and voltage acquisition module: by detecting the supply current, the actual power consumption of the propeller can be known, which is helpful for judging whether the propeller is operating within the rated power range. Therefore, the invention adopts the 0-5A analog input module to realize current collection, and supplies power to the analog input module through the 24V direct current power supply, the AD collection digit is 12 digits, the current resolution is 1mA, the precision is +/-1 per mill, the sampling frequency is 200KHz, and the data update rate is 30Hz. The input mode is differential input, the corresponding input impedance is 10mΩ, and the Modbus RTU communication protocol is used, and the communication mode is isolation type RS485, so that static electricity, lightning surge and interference can be effectively prevented.
In addition, the detection of the power supply voltage is also of great significance for fault diagnosis of the propeller, and performance degradation or damage caused by too high or too low voltage can be avoided. The voltage acquisition in the invention also adopts an analog input module which needs to be connected in parallel to a measured circuit. The voltage range is 0-500V, the voltage resolution is 10mV, so that the voltage signal of an external 110V power supply can be received, the input impedance is more than 10mΩ, and the rest parameters are the same as those of the current acquisition module.
The fault diagnosis system provided by the invention has the advantages that the electrical equipment is packaged in the outer box body, the outer box body is made of the composite material, and the composite material has high strength, good heat dissipation performance, excellent heat resistance, weather resistance and impact resistance. The fault diagnosis system is provided with a human engineering lifting handle, is convenient to carry when going out, adopts a high-density protection pad design, and has strong body anti-falling capability. The inner sealing performance is good, dust can be effectively prevented from entering, and the waterproof performance is excellent. Therefore, the device can be more suitable for complex climate conditions, the display screen of the fault diagnosis system is arranged in the upper box body of the fault diagnosis system, the TFT LCD display screen is adopted, the resolution is 1920x1080, and the rotation within the range of less than or equal to 90 degrees can be realized in the YOZ plane. The fault diagnosis system is provided with three USB interfaces, and can use a USB flash disk to import or export data, and can also be externally connected with other experimental equipment, so that the data transmission and experimental efficiency are improved. The keyboard is made of stainless steel material, has excellent waterproof, dustproof and anti-corrosion properties, and has good sensitivity and stability.
Specifically, the power supply module converts the input 220V alternating voltage into 24V direct voltage, wherein one path of 24V direct voltage supplies power to the upper computer and the display screen through the voltage reduction module, and the 24V direct voltage also supplies power to the communication module, the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module respectively. The voltage reducing module in the invention adopts the LM2596S high-power voltage reducing module to realize the function of the voltage reducing module, reduces the input voltage from 24V to 12V output voltage, connects two paths of 12V voltages in parallel through the wire connector, and respectively serves as the input voltage of a display screen and an upper computer, thereby improving the stability and the working efficiency of a circuit, protecting and reducing the loss of electronic equipment, preventing the occurrence of the conditions of overheat overload and the like, and further ensuring the reliability of a system.
Specifically, the power supply module is connected with the stabilized power supply and then is connected with the upper computer through the 2-core plug connector respectively, the stabilized power supply is connected with the 2-core plug connector through a lead and then is connected with two cores of the 10-core plug connector, and finally is connected with the propeller, and the 10-core plug connector penetrates out of the inner part of the outer box body; the regulated power supply converts 220V alternating current into 110V direct current to supply power to the propeller, and the upper computer receives a current voltage signal output by the regulated power supply to the propeller through the 485 module; the voltage output module (sending out control signals), the current and voltage acquisition module (acquiring current and voltage signals fed back by the propeller) and the rotating speed acquisition module (acquiring rotating speed signals) are connected with the propeller through a part of wires of the 10-core plug-in connector and perform signal transmission.
Specifically, the voltage output module outputs the input digital signal into a continuous voltage signal through digital-to-analog conversion. The voltage output module adopts the MB8 AO+8 paths of analog quantity output modules, can output 8 paths of 0-20mA, 4-20mA, 0-5V, 0-10V, +/-5V and +/-10V analog quantity signals, can be independently configured for each channel output signal, and is very suitable for being applied to industrial sites with complex signal specifications; 8 paths of analog quantity output can be controlled through an isolated RS485 interface; the module adopts Modbus-RTU communication, and can be directly adapted to PLC, DCS, domestic various configuration software and the like. Based on the performance, the fault diagnosis system adopts the MB8 AO+8 paths of analog quantity output modules, the signal output and the power supply of the fault diagnosis system are isolated from the RS485 communication electrical signals, various serial mode and common mode interferences are effectively inhibited, the accuracy of data is ensured, and meanwhile, the reliable operation of the modules is also ensured.
Specifically, the rotation speed acquisition module is a PWM acquisition module. The power supply voltage of the selected PWM acquisition module is 7-24V, the current is not more than 35mA, the input amplitude is 3-24V, the frequency range is 1-100 kHz, the number, period, frequency and duty ratio of the input PWM can be read through RS232 or RS485 communication, and the update time of the duty ratio is 0.035s.
Specifically, the communication module converts the RS485 serial port signal into an Ethernet signal, the upper computer receives data through the Ethernet, the communication module is connected with the upper computer through an RJ45 connector, and the communication module carries out data transmission with the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module through the 485 module. The communication module of the invention selects UT-6804MT serial server.
The invention also provides a fault diagnosis method for the deep submersible vehicle propeller, which comprises the following steps:
s1, acquiring operation data of the deep-diving propeller, and reducing the dimension of the operation data by using a PCA method.
S2, classifying the operation data by adopting a DBSCAN density-based clustering algorithm to extract fault data.
The data are classified into a training set and a testing set after DBSCAN clustering, then normalization processing is carried out on the data, and the original data are mapped on the basis of a common linear normalization methodWithin the interval, the normalization formula is as follows:
。
and S3, carrying out data enhancement on fault data by adopting an RFR random forest algorithm to obtain fault sample data, preprocessing the sample data, and endowing corresponding fault class labels.
And S4, performing fault diagnosis based on the convolutional neural network.
The convolutional neural network model building process sequentially comprises the following steps: convolution calculation, batch normalization, activation, pooling and rejection.
Specifically, the step S1 specifically includes the steps of:
s1.1, matrixIs composed of propeller operation data, matrixThe definition is as follows:
（1）。
s1.2, centralizing the data of each column to obtain a new data matrix, wherein the centralization formula is as follows:
（2）。
in the formula (2)Representing the data being centred on the data being processed,representative traversal NoEach of the rows of data.
S1.3, calculating covariance matrix of the sample：
（3）。
In (3)Represented as a centered matrix.
S1.4, decomposing the covariance matrix based on the eigenvalues or SVDs, and solving eigenvalues and corresponding eigenvectors of the covariance matrix.
S1.5, arranging the eigenvectors into a matrix according to the corresponding eigenvalues from top to bottom and obtaining k rows to form the matrix。
S1.6, obtaining new data after dimension reduction：
（4）。
Namely, obtaining the data after dimension reduction:
（5）。
specifically, the step S3 specifically includes the following steps:
s3.1, obtaining sample data of three fault categories including normal state of the propeller, damage of the driver, blade loss and foreign matter winding by constructing an experimental platform.
S3.2, preprocessing the data through manual or program screening, identifying and removing abnormal points and outliers in the measured data, and generating a fault sample.
S3.3, training RFRs by using the preprocessed fault samples, respectively obtaining the number of decision trees by using a grid search method for the fault samples in different propeller states, and identifying various parameters of the RFR models to obtain 3 RFR models capable of representing the propeller performances in different fault states.
S3.4, adding Gaussian noise to simulate the conditions of ocean noise and environmental interference, and generating various simulation fault samples with the same scale as the actual measurement samples according to the missing degree of each fault sample by utilizing the trained RFR in proportion, so that the ratio of the actual measurement samples to each simulation fault sample is 1:1.
S3.5, integrating the actual measurement sample and the simulation sample according to the requirements of fault detection and classification so as to ensure the relative diversity of the fault sample, preprocessing the generated sample data and marking corresponding fault class labels.
Specifically, step S4 specifically includes the following steps:
s4.1, calculating the voltage-current correlation coefficientAnd voltage rotation speed related coefficientTo eliminate delays of the control signal and the action signal;
（6）。
in (7)Representing the first of the voltagesThe number of observations made is a function of the number of observations,represents the average value of the voltage and,representing the i-th observation of the current,representing the average value of the current.
（7）。
In (8)Represents the first rotation speedThe number of observations made is a function of the number of observations,representing the average value of the rotational speed.
If it isClose to 1, there is a strong positive correlation between the twoThe method comprises the steps of carrying out a first treatment on the surface of the If it isClose to-1, then there is a strong negative correlation between them; if it isClose to 0, there is no linear relationship between them.
And S4.2, inputting the control signal and the action signal into a convolutional neural network to perform fault diagnosis.
The following is a verification of the method for generating data provided by the present invention:
and carrying out a series of pool tests, and obtaining the number of samples of the normal state of the propeller, blade loss faults, foreign matter winding faults and drive damage faults in the test, wherein the number is shown in table 1.
Table 1 number of experimental samples in each state
。
First, 3 RFR models were trained with the propeller control signals of the three fault samples as RFR inputs and the current and rotational speed as RFR outputs. Aiming at different fault samples, the number of decision trees is debugged by a grid search method, and the average value of root mean square errors of 5-fold cross validation is taken as an evaluation index. The training results of the number of the optimal decision trees are shown in fig. 2-7, when the number of the blade loss fault optimal decision trees is 38, the number of the foreign object winding fault optimal decision trees is 30, and the number of the driver damage fault optimal decision trees is 38, the root mean square error is minimum, and at the moment, the sample data of the RFR simulation is most consistent with the distribution of the original experimental data.
To verify that the RFR generated data is reliable, tables 2, 3 show the results of the comparison between the two fault sample real data and the RFR generated data. Taking the blade loss fault RFR as an example, when the number of decision trees is 37, the root mean square error of the rotation speed signals of the generated samples and the original samples is 84.4363, and the root mean square error of the current signals is 0.0258, as can be seen from table 2, in 400 sampling point data, the analog current rotation speed signals and the original current rotation speed signals are very close, and the change trend and the distribution characteristics of the analog current rotation speed signals and the original current rotation speed signals are relatively similar. Therefore, the data enhancement method based on RFR is effective, and new data constructed based on a small number of propeller failure samples by RFR can represent the failure characteristics thereof, thereby completing the establishment of the balanced sample library.
Table 2 comparative table of real and simulated data for blade loss failure samples
。
Table 3 foreign object winding failure sample real data VS simulation data
。
It should be understood that the above description is not intended to limit the invention to the particular embodiments disclosed, but to limit the invention to the particular embodiments disclosed, and that the invention is not limited to the particular embodiments disclosed, but is intended to cover modifications, adaptations, additions and alternatives falling within the spirit and scope of the invention.
Claims (10)
1. A fault diagnosis system for a deep-submersible vehicle propeller, comprising: outer box, display screen, keyboard and electrical equipment, the outer box have the upper cover and with the box body that the one end of upper cover is connected, the display screen embedded in the upper cover, the keyboard is located the box body upper surface, the box body is inside to have electrical equipment, electrical equipment includes: the system comprises a power supply module, a stabilized voltage supply, a voltage reduction module, an upper computer, a communication module, a voltage output module, a current and voltage acquisition module and a rotating speed acquisition module;
the display screen and the keyboard are connected with the upper computer;
the power module is respectively connected with the stabilized voltage power supply, the voltage reduction module, the communication module, the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module;
the stabilized voltage power supply is also connected with the propeller; the voltage reducing module is connected with the upper computer, and the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are connected with the upper computer through communication modules; the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are respectively connected with the propeller.
2. The fault diagnosis system for the deep submersible vehicle propeller according to claim 1, wherein the power module converts an input 220V ac voltage into a 24V dc voltage, one 24V dc voltage is used for supplying power to the upper computer and the display screen through the step-down module, and the 24V dc voltage is used for supplying power to the communication module, the voltage output module, the current voltage acquisition module and the rotation speed acquisition module respectively.
3. The fault diagnosis system for a deep submersible vehicle propeller according to claim 1, wherein the power supply module is connected with the stabilized power supply and then is connected with the upper computer through a 2-core plug connector respectively, the stabilized power supply is connected with the 2-core plug connector through a wire and then is connected with two cores of a 10-core plug connector, and finally is connected with the propeller, and the 10-core plug connector penetrates out of the inner part of the outer box body; the stabilized voltage power supply converts 220V alternating current into 110V direct current to supply power to the propeller, and the upper computer receives a current voltage signal output by the stabilized voltage power supply to the propeller through a 485 module; the voltage output module, the current and voltage acquisition module and the rotating speed acquisition module are connected with the propeller through a part of wires of the 10-core plug-in connector and transmit signals.
4. The fault diagnosis system for a deep-submersible vehicle according to claim 1, wherein the voltage output module outputs the input digital signal as a continuous voltage signal via digital-to-analog conversion.
5. The fault diagnosis system for a deep-submersible vehicle of claim 1, wherein the rotational speed acquisition module is a PWM acquisition module.
6. The fault diagnosis system for a deep submersible vehicle according to claim 1, wherein the communication module converts an RS485 serial signal into an ethernet signal, the upper computer receives data through the ethernet, and the communication module performs data transmission with the voltage output module, the current voltage acquisition module and the rotation speed acquisition module through 485 modules.
7. A fault diagnosis method for a deep-submersible propeller, using a system according to any of claims 1-6, characterized in that it comprises in particular the following steps:
s1, acquiring operation data of a deep submersible propeller, and reducing the dimension of the operation data by using a PCA method;
s2, classifying the operation data by adopting a DBSCAN algorithm to extract fault data;
s3, carrying out data enhancement on fault data by adopting an RFR algorithm to obtain fault sample data, preprocessing the sample data, and endowing corresponding fault class labels;
and S4, performing fault diagnosis based on the convolutional neural network.
8. The fault diagnosis method for a deep-submersible vehicle according to claim 6, wherein step S1 specifically comprises the steps of:
s1.1, matrixConsists of propeller operation data, matrix +.>The definition is as follows:
（1）；
s1.2, centralizing the data of each column to obtain a new data matrix, wherein the centralization formula is as follows:
（2）；
in the formula (2)Representing the data being centred, +.>Representative traversal->Each data of a row;
s1.3, calculating covariance matrix of the sample：
（3）；
In (3)Represented as a centered matrix;
s1.4, decomposing the covariance matrix based on the eigenvalues or SVDs, and solving eigenvalues and corresponding eigenvectors of the covariance matrix;
s1.5, arranging the eigenvectors into a matrix according to the corresponding eigenvalues from top to bottom and obtaining k rows to form the matrix；
S1.6, obtaining new data after dimension reduction：
（4）；
Namely, obtaining the data after dimension reduction:
（5）。
9. the fault diagnosis method for a deep-submersible vehicle according to claim 6, wherein step S3 specifically comprises the steps of:
s3.1, acquiring sample data of three fault categories, namely normal state of the propeller, damage of the driver, blade loss and foreign matter winding, by constructing an experimental platform;
s3.2, preprocessing the data through manual or program screening, identifying and removing abnormal points and outliers in the measured data, and generating a fault sample;
s3.3, training RFRs by using the preprocessed fault samples, respectively obtaining the number of decision trees by using a grid search method for the fault samples in different propeller states, and identifying various parameters of the RFR models to obtain 3 RFR models capable of representing the propeller performances in different fault states;
s3.4, adding Gaussian noise to simulate the conditions of ocean noise and environmental interference, and generating various simulation fault samples with the same scale as the actual measurement samples according to the missing degree of each fault sample by utilizing the trained RFR in proportion, so that the ratio of the actual measurement samples to each simulation fault sample is 1:1;
s3.5, integrating the actual measurement sample and the simulation sample according to the requirements of fault detection and classification so as to ensure the relative diversity of the fault sample, preprocessing the generated sample data and marking corresponding fault class labels.
10. The fault diagnosis method for a deep-submersible vehicle according to claim 6, wherein step S4 specifically comprises the steps of:
s4.1, calculating the voltage-current correlation coefficientAnd a voltage rotational speed dependence coefficient->To eliminate delays of the control signal and the action signal;
（6）；
in (7)Represents the first part of the voltage>Individual observations->Representing the average value of the voltage,/">Representing the i-th observation of the current,represents the average value of the current;
（7）；
in (8)Represents the first part of the rotational speed>Individual observations->Represents an average value of the rotational speeds;
and S4.2, inputting the control signal and the action signal into a convolutional neural network to perform fault diagnosis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311483722.2A CN117214591A (en) | 2023-11-09 | 2023-11-09 | Fault diagnosis system and method for deep-diving propeller |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311483722.2A CN117214591A (en) | 2023-11-09 | 2023-11-09 | Fault diagnosis system and method for deep-diving propeller |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117214591A true CN117214591A (en) | 2023-12-12 |
Family
ID=89049702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311483722.2A Pending CN117214591A (en) | 2023-11-09 | 2023-11-09 | Fault diagnosis system and method for deep-diving propeller |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117214591A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117647758A (en) * | 2024-01-30 | 2024-03-05 | 青岛哈尔滨工程大学创新发展中心 | AUV propeller state monitoring system and method based on pre-detection |
| CN117710759A (en) * | 2024-02-06 | 2024-03-15 | 青岛哈尔滨工程大学创新发展中心 | System and method for predicting faults and service lives of manned submersible vehicles |
| CN117697764A (en) * | 2024-02-05 | 2024-03-15 | 青岛哈尔滨工程大学创新发展中心 | Fault diagnosis system and method for flexible mechanical arm for submersible operation |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110125950A (en) * | 2019-05-28 | 2019-08-16 | 哈尔滨工程大学 | A kind of underwater caterpillar type robot for becoming appearance |
| CN209979777U (en) * | 2019-05-05 | 2020-01-21 | 上海工程技术大学 | Electronic and electrical fault detection device with excellent air tightness |
| CN111275164A (en) * | 2020-01-13 | 2020-06-12 | 浙江大学 | Underwater robot propulsion system fault diagnosis method |
| CN111751673A (en) * | 2020-07-14 | 2020-10-09 | 湖南科技大学 | Fault fixed-point protection type detection device based on power distribution network |
| CN213292672U (en) * | 2020-07-01 | 2021-05-28 | 上海海洋大学 | Underwater operation robot water is precious |
| CN216146117U (en) * | 2021-07-26 | 2022-03-29 | 深圳市朗非创新科技有限公司 | Underwater propeller |
| CN114357663A (en) * | 2022-03-15 | 2022-04-15 | 中国长江三峡集团有限公司 | Method for training gearbox fault diagnosis model and gearbox fault diagnosis method |
| CN114544155A (en) * | 2022-01-28 | 2022-05-27 | 江苏科技大学 | AUV propeller multi-information-source fusion fault diagnosis method and system based on deep learning |
| CN116643517A (en) * | 2023-05-17 | 2023-08-25 | 青岛哈尔滨工程大学创新发展中心 | Accident prevention equipment for underwater manned submersible vehicle |
-
2023
- 2023-11-09 CN CN202311483722.2A patent/CN117214591A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN209979777U (en) * | 2019-05-05 | 2020-01-21 | 上海工程技术大学 | Electronic and electrical fault detection device with excellent air tightness |
| CN110125950A (en) * | 2019-05-28 | 2019-08-16 | 哈尔滨工程大学 | A kind of underwater caterpillar type robot for becoming appearance |
| CN111275164A (en) * | 2020-01-13 | 2020-06-12 | 浙江大学 | Underwater robot propulsion system fault diagnosis method |
| CN213292672U (en) * | 2020-07-01 | 2021-05-28 | 上海海洋大学 | Underwater operation robot water is precious |
| CN111751673A (en) * | 2020-07-14 | 2020-10-09 | 湖南科技大学 | Fault fixed-point protection type detection device based on power distribution network |
| CN216146117U (en) * | 2021-07-26 | 2022-03-29 | 深圳市朗非创新科技有限公司 | Underwater propeller |
| CN114544155A (en) * | 2022-01-28 | 2022-05-27 | 江苏科技大学 | AUV propeller multi-information-source fusion fault diagnosis method and system based on deep learning |
| CN114357663A (en) * | 2022-03-15 | 2022-04-15 | 中国长江三峡集团有限公司 | Method for training gearbox fault diagnosis model and gearbox fault diagnosis method |
| WO2023035869A1 (en) * | 2022-03-15 | 2023-03-16 | 中国长江三峡集团有限公司 | Gearbox fault diagnosis model training method and gearbox fault diagnosis method |
| CN116643517A (en) * | 2023-05-17 | 2023-08-25 | 青岛哈尔滨工程大学创新发展中心 | Accident prevention equipment for underwater manned submersible vehicle |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117647758A (en) * | 2024-01-30 | 2024-03-05 | 青岛哈尔滨工程大学创新发展中心 | AUV propeller state monitoring system and method based on pre-detection |
| CN117697764A (en) * | 2024-02-05 | 2024-03-15 | 青岛哈尔滨工程大学创新发展中心 | Fault diagnosis system and method for flexible mechanical arm for submersible operation |
| CN117697764B (en) * | 2024-02-05 | 2024-04-30 | 青岛哈尔滨工程大学创新发展中心 | Fault diagnosis system and method for flexible mechanical arm for submersible operation |
| CN117710759A (en) * | 2024-02-06 | 2024-03-15 | 青岛哈尔滨工程大学创新发展中心 | System and method for predicting faults and service lives of manned submersible vehicles |
| CN117710759B (en) * | 2024-02-06 | 2024-04-26 | 青岛哈尔滨工程大学创新发展中心 | System and method for predicting faults and service lives of manned submersible vehicles |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117214591A (en) | Fault diagnosis system and method for deep-diving propeller | |
| Wang et al. | Modified multiscale weighted permutation entropy and optimized support vector machine method for rolling bearing fault diagnosis with complex signals | |
| CN116383636A (en) | Coal mill fault early warning method based on PCA and LSTM fusion algorithm | |
| CN111273125A (en) | RST-CNN-based power cable channel fault diagnosis method | |
| CN116956215A (en) | Fault diagnosis method and system for transmission system | |
| CN114154779A (en) | Health assessment and fault diagnosis method and system based on EMD-SOM | |
| CN117077077A (en) | Voltage transformer error state prediction method, device, equipment and storage medium | |
| Xiong et al. | Fault diagnosis of UAV based on adaptive Siamese network with limited data | |
| Gong et al. | Photovoltaic arrays fault diagnosis based on an improved dilated convolutional neural network with feature-enhancement | |
| Attouri et al. | Fault detection in photovoltaic systems using machine learning technique | |
| CN114117923A (en) | High-voltage parallel reactor state judgment system and method based on chaotic feature space | |
| CN117708707A (en) | Intelligent early warning method and system for abnormal line loss rate of big data lower station area | |
| Hichri et al. | Fault detection and diagnosis in grid-connected photovoltaic systems | |
| Wang et al. | Fault Diagnosis of Highway Machinery Hydraulic System Based on LS-TF | |
| CN115392339A (en) | Method, system, device and medium for assisting fault data determination and fault diagnosis | |
| CN113673162A (en) | Transformer body state evaluation method based on fuzzy evaluation and DSmT | |
| CN111428780A (en) | Power grid abnormal operation state identification method based on data driving | |
| CN112231982A (en) | Photovoltaic panel fault detection method based on distributed soft measurement model | |
| CN105021914B (en) | Detect analysis expert instrument in portable Ship Power Station | |
| CN117368639B (en) | Power transmission line fault identification system and method based on artificial intelligence | |
| CN117950381B (en) | Digital production and manufacturing management system | |
| He et al. | Circuit Fault Detection System of Simulation and Neural Network | |
| CN117647758A (en) | AUV propeller state monitoring system and method based on pre-detection | |
| CN211554259U (en) | Battery simulator | |
| Zhang et al. | Power grid equipment operation status monitoring system based on Internet of Things technology |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20231212 |
|
| RJ01 | Rejection of invention patent application after publication |