CN116383615A - Multisource sensing information fusion and expansion method based on system state characterization - Google Patents
Multisource sensing information fusion and expansion method based on system state characterization Download PDFInfo
- Publication number
- CN116383615A CN116383615A CN202310344544.9A CN202310344544A CN116383615A CN 116383615 A CN116383615 A CN 116383615A CN 202310344544 A CN202310344544 A CN 202310344544A CN 116383615 A CN116383615 A CN 116383615A
- Authority
- CN
- China
- Prior art keywords
- state evaluation
- state
- parameter
- parameters
- index
- 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
- 238000000034 method Methods 0.000 title claims abstract description 76
- 230000004927 fusion Effects 0.000 title claims abstract description 32
- 238000012512 characterization method Methods 0.000 title claims description 25
- 238000011156 evaluation Methods 0.000 claims abstract description 232
- 238000005096 rolling process Methods 0.000 claims abstract description 43
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000000605 extraction Methods 0.000 claims abstract description 23
- 238000009826 distribution Methods 0.000 claims abstract description 16
- 238000004140 cleaning Methods 0.000 claims abstract description 14
- 230000001133 acceleration Effects 0.000 claims description 47
- 238000005070 sampling Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 14
- 238000005259 measurement Methods 0.000 claims description 13
- 238000000513 principal component analysis Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 239000010687 lubricating oil Substances 0.000 claims description 12
- 230000002441 reversible effect Effects 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 8
- 239000000498 cooling water Substances 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000001953 sensory effect Effects 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 230000003416 augmentation Effects 0.000 claims description 2
- 230000008859 change Effects 0.000 abstract description 9
- 230000008901 benefit Effects 0.000 abstract description 4
- 238000006731 degradation reaction Methods 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000007500 overflow downdraw method Methods 0.000 abstract description 2
- 239000003921 oil Substances 0.000 description 13
- 238000005461 lubrication Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000003786 synthesis reaction Methods 0.000 description 3
- 230000000670 limiting effect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000010729 system oil Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Probability & Statistics with Applications (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a multisource sensing information fusion and expansion method based on system state representation, which takes only irreparable key fault factors in a sliding bearing system, a gear box system and a rolling bearing system as a ship power system state evaluation index system, can better reflect the running state of a ship power system, and is beneficial to realizing the effective evaluation of the power system state; the method has the advantages that the fusion method of the multi-state parameter characteristics is adopted, the data missing value filling and denoising processing is carried out through the data cleaning operation, the multi-source data redundant information is removed through the characteristic selection and characteristic extraction operation, the problem that an existing ship power system depends on the single parameter threshold alarming method can be solved, the degradation process before the failure of the ship power system occurs can be recognized in advance, the state change of the ship power system during operation can be accurately estimated, the weight distribution accuracy is guaranteed through the comprehensive weighting method, and the accuracy of the subsequently constructed state evaluation reference model is effectively improved.
Description
Technical Field
The invention relates to the technical field of ship state evaluation, in particular to a multisource sensing information fusion and expansion method based on system state characterization.
Background
With the continuous development of science and technology, the structure of ships becomes more and more complex, and simultaneously, the connection between each subsystem in the ships becomes more and more compact, and once a subsystem fails, the whole ship is caused to fail, which brings serious threat to life and property safety of people and even brings disastrous results. The ship power system is a core subsystem in the ship, and the source of sensing data is wide, complex and various, so that the multi-source sensing information is fused and processed, and the accurate control of the state evaluation of the ship power system is particularly necessary, so that the ship power system is not only an important basis for state maintenance, but also a premise of carrying out fault prediction and fault diagnosis on the ship power system.
At present, a small amount of monitoring information is often used for state evaluation aiming at a ship power system at home and abroad, a fixed threshold value is usually set depending on subjective experience of a person, or a large amount of monitoring information is used, but the state evaluation method mainly shows data instability, high coupling and strong correlation because of high complexity, the state information of the system is difficult to obtain from the data, and the change range of the threshold value of the conventional state evaluation characteristic parameter is larger because of the complex and changeable working condition of the ship power system, so that the accuracy of state evaluation is influenced.
Disclosure of Invention
In view of the above, the invention provides a multisource sensing information fusion and expansion method, a multisource sensing information fusion and expansion device, electronic equipment and a storage medium based on system state characterization, so as to solve the problem that the accuracy of state evaluation is affected due to the fact that the conventional state evaluation characteristic parameter threshold value is large in change range due to the fact that the working condition of a ship power system is complex and changeable in the prior art.
In order to solve the above technical problems, in a first aspect, an embodiment of the present invention provides a method for fusing and expanding multisource sensing information based on system state characterization, including:
acquiring running state parameters of all subsystems in a ship power system, and performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system;
dividing the operation working conditions of the ship power system into various types, extracting state evaluation characteristic parameters under each type of operation working conditions, and constructing a state evaluation reference model corresponding to the operation working conditions based on the state evaluation characteristic parameters;
acquiring an operation condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the operation condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
Further, the operational state parameters of the plain bearing system include temperature parameters and pressure parameters; the temperature parameters comprise a first bearing temperature parameter of the drive end sliding bearing and a second bearing temperature parameter of the non-drive end sliding bearing, and the running state parameters of the sliding bearing system further comprise lubricating oil temperature and cooling water temperature; the pressure parameters comprise a first lubricating oil pressure parameter of the non-driving sliding bearing and a second lubricating oil pressure parameter of the non-driving sliding bearing;
the operational state parameters of the gearbox system include a first acceleration parameter including acceleration collected at a plurality of points on the gearbox;
the operating state parameters of the rolling bearing system include a second acceleration parameter including an acceleration collected at each rolling bearing;
after the operation state parameters of all subsystems in the ship power system are obtained, the method further comprises the following steps:
establishing a parameter group based on a time sequence for each subsystem, wherein the parameter group comprises the time sequence and the numerical value of each running state parameter corresponding to the time sequence; the parameter sets include a plain bearing system parameter set, a gearbox system parameter set, and a rolling bearing system parameter set.
Further, performing data cleaning, feature extraction and feature analysis on the running state parameters to obtain state evaluation feature parameters, which specifically include:
interpolation processing is carried out on the missing values in the parameter group based on a Lagrange interpolation method, noise data are screened out and removed based on a 3 sigma criterion, and interpolation filling is carried out on the missing values after the noise data are removed based on the Lagrange interpolation method;
performing feature selection on each running state parameter in the parameter group based on a Pearson correlation coefficient method, and extracting state evaluation feature parameters;
and carrying out standardization processing and secondary feature extraction processing on the state evaluation feature parameters, and determining the weight of each state evaluation feature parameter in the parameter group.
Further, feature selection is performed on each operation state parameter in the parameter group based on the Pearson correlation coefficient method, and state evaluation feature parameters are extracted, specifically including:
in a sliding bearing system parameter group, taking the bearing temperature of a driving end sliding bearing as a reference temperature, determining a first Pearson correlation coefficient of a residual temperature parameter and the reference temperature, and eliminating a temperature parameter of which the first Pearson correlation coefficient is lower than a preset first correlation coefficient threshold value to obtain a state evaluation characteristic parameter corresponding to the sliding bearing system parameter group;
Randomly selecting one acceleration from the gear box system parameter group and the rolling bearing system parameter group as a reference acceleration, determining second Pearson correlation coefficients of residual accelerations in the gear box system parameter group and the rolling bearing system parameter group and the reference acceleration, and eliminating accelerations of which the second Pearson correlation coefficients are higher than a second correlation coefficient threshold value; and extracting time domain characteristic parameters and frequency domain characteristic parameters of each acceleration in the gear box system and the rolling bearing system to obtain state evaluation characteristic parameters corresponding to the gear box system parameter group and the rolling bearing system parameter group.
Further, the state evaluation feature parameters are subjected to standardization processing and secondary feature extraction processing, and the weight of each running state parameter in the parameter group is determined, which specifically comprises:
based on a multi-index comprehensive evaluation method and a preset index interval, classifying the state evaluation characteristic parameters in each parameter group into a forward index, a moderate index and a reverse index; carrying out standardization processing based on the corresponding strategy;
if the state evaluation characteristic parameter is judged to be a forward index, and the actual measurement value of the running state parameter is not more than an average value; or if the state evaluation characteristic parameter is judged to be a reverse index, and the actual measurement value of the running state parameter is larger than the average value; determining that the running state parameter after standard deviation processing is 0;
If the state evaluation characteristic parameter is judged to be a forward index, and the actual measurement value of the running state parameter is larger than an average value; or if the state evaluation characteristic parameter is judged to be a moderate index; or if the state evaluation characteristic parameter is judged to be a reverse index, and the actual measurement value of the running state parameter is not more than the average value; the running state parameters after standard deviation processing are determined as follows: the difference between the measured value and the average value is divided by the standard deviation of the running state parameter;
and carrying out secondary feature extraction on the parameter group based on a principal component analysis method, calculating initial weights of all state evaluation feature parameters in the parameter group based on an entropy method, and correcting the initial weights based on a weighted average method.
Further, constructing a state evaluation reference model corresponding to the operation condition based on the state evaluation characteristic parameters, specifically including:
acquiring a set of state evaluation characteristic parameters corresponding to the current class of operation conditions, and forming a state evaluation characteristic vector based on the state evaluation characteristic parameters;
determining the weight of each state evaluation characteristic parameter in the state evaluation characteristic vector, and modeling the distribution of the state evaluation characteristic parameters as a Gaussian probability density function;
And constructing a state evaluation reference model of the current class of operation conditions based on the weight of each state evaluation characteristic parameter and the corresponding Gaussian probability density function.
Further, constructing a state evaluation index based on the operation condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index, wherein the method specifically comprises the following steps:
based on the state evaluation feature vector at the current moment, the corresponding weight, the mean value vector of the Gaussian distribution of the state evaluation reference model under each operation condition, and the covariance matrix, determining the first Markov distance between the state evaluation feature vector at the current moment and each Gaussian distribution;
determining a second mahalanobis distance between the state evaluation feature vector at the current moment and the state evaluation reference model under each operation condition based on the weight coefficient of the Gaussian probability density function and the first mahalanobis distance;
determining a third mahalanobis distance between a state evaluation feature vector at the current moment and a state evaluation reference model of T operating conditions based on the second mahalanobis distance and the probability that the operating conditions of the ship power system at the current moment belong to each category, wherein T is the number of the operating condition categories;
Constructing a state evaluation index based on the product of the third mahalanobis distance and a normal running state constant value of the ship power system;
carrying out state evaluation on the ship power system based on the average state evaluation index in the sliding window, wherein if the average state evaluation index of the continuous multiple sampling points is judged to be larger than a first index threshold value or the state evaluation index of at least one sampling point is judged to be larger than a second index threshold value, the abnormal state is judged; if the average state evaluation index of the continuous multiple sampling points is larger than the second index threshold or the state evaluation index of at least one sampling point is larger than the third index threshold, judging that the immediate shutdown processing is needed; wherein the first indicator threshold is less than the second indicator threshold, and the second indicator threshold is less than the third indicator threshold.
In a second aspect, an embodiment of the present invention provides a multi-source sensing information fusion and expansion device based on system state characterization, including:
the system comprises a multisource data fusion module, a state evaluation module and a state evaluation module, wherein the multisource data fusion module is used for acquiring operation state parameters of all subsystems in a ship power system, and performing data cleaning, feature extraction and principal component analysis on the operation state parameters to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system;
The operating condition dividing module divides the operating condition of the ship power system into various types, extracts state evaluation characteristic parameters under each type of operating condition, and constructs a state evaluation reference model corresponding to the operating condition based on the state evaluation characteristic parameters;
the state evaluation module is used for acquiring the running condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the running condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
In a third aspect, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the steps of the method for fusing and expanding multisource sensing information based on system state characterization according to the embodiment of the first aspect of the present invention are implemented when the processor executes the program.
In a fourth aspect, embodiments of the present invention provide a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a system state characterization based multisource sensory information fusion and expansion method according to the embodiments of the first aspect of the present invention.
According to the multisource sensing information fusion and expansion method, device, electronic equipment and storage medium based on system state characterization, which are provided by the embodiment of the invention, only key irreparable fault factors such as a sliding bearing system parameter group, a gear box system parameter group and a rolling bearing system parameter group are used as a ship power system state evaluation index system, so that the running state of a ship power system can be better reflected, and the effective evaluation of the power system state can be realized; the method has the advantages that the defect that the existing ship power system depends on a single parameter threshold alarming method can be overcome, the degradation process before the failure of the ship power system can be recognized in advance, and the state change of the ship power system during operation can be accurately estimated; and meanwhile, the method adopts a comprehensive weighting method based on principal component analysis and entropy weighting summation to ensure the accuracy of weight distribution, thereby effectively improving the accuracy of a state reference model constructed subsequently, further improving the accuracy of state evaluation and further effectively improving the rationality of an evaluation result.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a multi-source sensing information fusion and expansion method based on system state characterization according to an embodiment of the invention;
FIG. 2 is a diagram of a multi-source data group according to an embodiment of the present invention;
FIG. 3 is a flowchart of a method for multi-source sensing information fusion and expansion based on system state characterization according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a physical structure according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. 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.
Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor systems and/or microcontroller systems.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
In order to effectively analyze the state of the ship power system, the possibility of failure of the ship power system is predicted, and operations such as data cleaning, feature selection, feature extraction, feature fusion and expansion are required to be performed on multi-source sensing information of the ship power system to perform state evaluation, so that the ship power system can be ensured to run continuously, safely and reliably. The method reduces the experience requirements of testers, improves the state evaluation efficiency and accuracy, and has certain engineering application value.
Therefore, the embodiment of the invention provides a multisource sensing information fusion and expansion method, a multisource sensing information fusion and expansion device, electronic equipment and a storage medium based on system state characterization, which take only irreparable key fault factors such as a sliding bearing system parameter group, a gear box system parameter group and a rolling bearing system parameter group as a ship power system state evaluation index system, can better reflect the running state of a ship power system, and is beneficial to realizing effective evaluation of the power system state. The following description and description will be made with reference to various embodiments.
Fig. 1 is a system state representation-based multi-source sensing information fusion and expansion method, that is, a system state representation-based multi-source sensing information fusion and expansion method, provided by an embodiment of the invention, including:
step S1, acquiring running state parameters of all subsystems in a ship power system, and performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system.
S11, acquiring operation state parameters of all subsystems in a ship power system;
In order to drive a ship to navigate at a certain speed, it must be given thrust. This thrust is generated by the operation of the propeller, which may be a propeller, a flat wheel, a paddle wheel, or the like. The propeller is driven by a prime motor, and the prime motor comprises a diesel engine, a steam engine, a gas turbine, a combined power device formed by the diesel engine, the steam engine, the gas turbine and the gas turbine, and the like. The prime movers are called main machines, and the main machines, together with auxiliary machines, pipelines and equipment required for ensuring the operation of the main machines, shaft systems for transmitting the power of the main machines to the propeller, and the like are collectively called ship power systems. The ship power system mainly comprises a main engine, transmission equipment, a shafting, cabin mechanical equipment and a power pipeline; because the working condition of the ship power system is complex and changeable, the change range of the conventional state characteristic parameter threshold value is larger, and the accuracy of state evaluation is affected.
The sliding bearing is an important component of a ship power system, and whether the bearing system works normally can directly influence the economic benefit, the power performance, the service life and the like of the whole ship. Compared with the general working condition environment of the actual use of the radial sliding bearing, the marine sliding bearing is more complex, mainly because the ship is continuously influenced by factors such as wind waves and the like to cause the ship body to swing and incline, the working plane of the sliding bearing is also influenced by the swing of the ship body to different degrees, and the safety and the reliability of the whole sliding bearing system are more required. In the sliding bearing, an oil ring is usually adopted to bring oil between a rotating shaft and a bearing bush, and if the oil ring is deformed or worn seriously, the oil ring does not rotate or rotates slowly, so that the oil quantity between the shaft and the bearing bush is reduced, and overheat occurs; as long as the power system runs, the lubrication system needs to ensure certain system oil pressure, and once the pressure of the lubricating oil is insufficient, the lubrication system cannot ensure that the lubricating oil plays the roles of lubrication, cooling, cleaning, sealing, corrosion resistance, shock absorption and the like, so that all friction pair parts such as a crank connecting rod mechanism, a valve mechanism and the like are excessively worn, and even a cylinder pulling and a tile holding are caused, so that the ship power system is scrapped in advance; the excessive pressure of the lubricating oil can cause the risks of oil seal oil leakage, oil pan oil leakage, main oil passage oil leakage, oil filter oil leakage and the like. Therefore, the causes of failure of the plain bearing system generally include bearing overheating, poor heat dissipation of the lubricating oil, water temperature of the cooling water, lubricating oil pressure, etc., and in the present embodiment, as shown in fig. 2, the operating state parameters of the plain bearing system include temperature parameters and pressure parameters; the temperature parameters include a first bearing temperature parameter of the drive end sliding bearing (i.e. the temperature of the bearing 1 in fig. 2), and a second bearing temperature parameter of the non-drive end sliding bearing (i.e. the temperature of the bearing 2 in fig. 2), and also include the temperature of lubricating oil and the temperature of cooling water; the pressure parameters include a first lubrication pressure parameter for the non-driven plain bearing (i.e., bearing 1 lubrication pressure in fig. 2) and a second lubrication pressure parameter for the non-driven plain bearing (i.e., bearing 2 lubrication pressure in fig. 2).
The gearbox is responsible for transmitting power to the propeller and provides a reduction ratio, and because the speed of the internal combustion engine is too high and the torque is small, the gearbox is required to reduce the speed and increase the torque, which is not suitable for ship propulsion; the gearbox mainly comprises a transmission shaft, a gear and a shell, the main performance indexes of the gearbox are transmission efficiency, water resistance coefficient and reliability and durability, before the gearbox fails, a vibration signal of the gearbox body is gradually increased, the fault judgment can be realized by installing a vibration acceleration sensor on the gearbox body to test the vibration signal, and in the embodiment, the running state parameters of the gearbox system comprise first acceleration parameters, and the first acceleration parameters comprise accelerations acquired at a plurality of measuring points on the gearbox; in this embodiment, the first acceleration parameter is vibration acceleration, and acceleration indexes (i.e., vibration acceleration) of 8 measuring points on the gearbox, such as acceleration of measuring point 1 to acceleration of measuring point 8 in fig. 3, detect vibration signals of the box body by using an acceleration sensor, implement real-time monitoring of the gearbox by time domain analysis, and use the acceleration as an operation state parameter thereof.
The rolling bearing system is one of important parts of the ship power system, and the working state of the rolling bearing system directly determines the performance and the running condition of the ship power system. In engineering practice, a minor failure of the rolling bearing system may lead to a shutdown of the production line or damage to the equipment, resulting in serious economic losses. Therefore, the working state of the rolling bearing system is monitored in real time, faults are found in time, and a reliable maintenance strategy is formulated, so that the method has important significance; the running state parameters of the rolling bearing system comprise second acceleration parameters, the second acceleration parameters comprise acceleration acquired at each rolling bearing, and the second acceleration parameters are vibration acceleration; in this embodiment, acceleration at 4 rolling bearings is collected, and vibration signals of the rolling bearings belong to high-frequency signals, so that an acceleration sensor is used for picking up signals.
After acquiring the operation state parameters of each subsystem in the ship power system, as shown in fig. 3, the method further includes:
establishing a parameter group based on a time sequence for each subsystem, wherein the parameter group comprises the time sequence and the numerical value of each running state parameter corresponding to the time sequence; the parameter groups comprise a sliding bearing system parameter group, a gear box system parameter group and a rolling bearing system parameter group, and the sliding bearing system parameter group, the gear box system parameter group and the rolling bearing system parameter group form a multi-source data group.
Step S12, carrying out interpolation processing on the missing values of the running state parameters in the parameter group based on a Lagrange interpolation method; the approximation corresponding to the missing values of the operational state parameters of the Lagrangian interpolation is calculated according to the following formula:
in the above, x 1 ,x 2 ,…,x l Representing a time sequence, i e (1, 2, …, l), j e (1, 2, …, l), and j+.i; y is 1 ,y 2 ,…y i ,…,y l Values representing the respective operating state parameters in the respective parameter groups, L i (x) Representing i sample point pairs (x 1 ,y 1 ),(x 2 ,y 2 ),(x 3 ,y 3 )...(x l ,y l ) Is the interpolation basis function of f l (x) And carrying in the point corresponding to the missing function value in order to obtain the l Lagrangian interpolation function by using the interpolation basis function, and obtaining the approximate value of the missing value corresponding to the point to perform interpolation filling.
On the basis of the above embodiment, as a preferred implementation manner, noise data is screened out and removed based on a 3 sigma criterion; the 3 sigma criterion is as follows:
in the above-mentioned method, the step of,
the arithmetic average value of all running state parameters is calculated, and n is the number of initial sample points to be denoised; sigma is standard deviation, R n If the residual is greater than 3σ, Z is noise data, and is eliminated, and if the residual is equal to or less than 3σ, Z (k) is a value of an operation state parameter at each time of the time series, k e (1, 2, …, n) is retained.
And interpolating and filling the missing values after removing the noise data based on the Lagrange interpolation method.
And step S13, performing feature selection on each running state parameter in the parameter group based on a Pearson correlation coefficient method, and extracting state evaluation feature parameters.
The larger the absolute value of the Pearson correlation coefficient is, the higher the correlation degree is, in a sliding bearing system parameter group, the index types are inconsistent, the system state is accurately represented, the bearing temperature of the sliding bearing at the driving end is taken as a reference temperature, a first Pearson correlation coefficient of the residual temperature parameter and the reference temperature is determined, and the temperature parameter of the first Pearson correlation coefficient lower than a preset first correlation coefficient threshold value is removed to obtain a state evaluation characteristic parameter corresponding to the sliding bearing system parameter group; and in the gearbox system parameter group and the rolling bearing system parameter group, the index types are acceleration signals, one acceleration is randomly selected as a reference acceleration for eliminating the redundancy of data, the second Pearson correlation coefficient between the residual acceleration in the gearbox system parameter group and the rolling bearing system parameter group and the reference acceleration is determined, and the acceleration of which the second Pearson correlation coefficient is higher than a second correlation coefficient threshold value is removed.
Based on the characteristic selection, the state evaluation characteristic parameters are preliminarily selected, and because the acceleration sampling frequency in the gear box system parameter group and the rolling bearing system parameter group is high, the change is fast, the data is complex, the time domain characteristic parameters of the gear box system parameter group and the rolling bearing system parameter group are required to be further extracted, and the Fourier change is adopted to extract the frequency domain characteristic parameters of the gear box system parameter group and the rolling bearing system parameter group, so that the state evaluation characteristic parameters corresponding to the gear box system parameter group and the rolling bearing system parameter group are obtained, and the parameter group comprising the state evaluation parameters is correspondingly formed.
On the basis of the above embodiment, as a preferred implementation manner, the extracted time domain and frequency domain characteristic parameters are as follows:
x max ＝max(|x(n)|)
x p-p ＝x max -x min
wherein x (N) is represented as a discretized signal after acceleration sampling, N is the number of sampling points, K is the number of spectral lines in a spectrum corresponding to x (N), Y (K) is the amplitude corresponding to the kth spectral line, and x max 、
x rms 、x p-p Mf represents maximum, absolute mean, root mean square, peak-to-peak, mean frequency, respectively.
Based on the above embodiments, as a preferred embodiment, the parameter matrix of each parameter group after feature selection and preliminary feature extraction is as follows:
where n represents the number of samples and p represents the number of state characteristic parameters initially selected.
And carrying out standardization processing and secondary feature extraction processing on the state evaluation feature parameters, and determining the weight of each state evaluation feature parameter in the parameter group.
In this embodiment, by using a fusion method of multi-state parameter features, filling and denoising data missing values are performed by using a data cleaning operation, and multi-source data redundant information is removed by using a feature selection and feature extraction operation, so that the defect that the existing ship power system depends on a single parameter threshold alarm method can be overcome, the degradation process before the failure of the ship power system can be identified in advance, and the state change of the ship power system during operation can be accurately estimated.
Step S14, based on a multi-index comprehensive evaluation method and a preset index interval, dividing the state evaluation characteristic parameters in each parameter group into a forward index, a moderate index and a reverse index; and carrying out standardization processing based on the corresponding strategy. And carrying out standardization processing based on system evaluation indexes on the parameter matrix of each parameter group so as to improve the accuracy of subsequent state evaluation. The system evaluation index is mainly divided into three types, one is a forward index, that is, the larger the index value is, if the index value is larger than a preset first threshold value, the better the working state is, the reverse index is, namely, the smaller the index value is, if the index value is smaller than a preset second threshold value, the better the working state is; and the other is a moderate index, and the parameter value is within a certain preset range, so that the system health state can be adversely affected by too small parameter values.
If the state evaluation characteristic parameter is judged to be a forward index, and the actual measurement value of the running state parameter is not more than an average value; or if the state evaluation characteristic parameter is judged to be a reverse index, and the actual measurement value of the running state parameter is larger than the average value; the running state parameter after the standard deviation processing is determined to be 0.
If the state evaluation characteristic parameter is judged to be a forward index, and the actual measurement value of the running state parameter is larger than an average value; or if the state evaluation characteristic parameter is judged to be a moderate index; or if the state evaluation characteristic parameter is judged to be a reverse index, and the actual measurement value of the running state parameter is not more than the average value; the running state parameters after standard deviation processing are determined as follows: the difference between the measured value and the average value is divided by the standard deviation of the operating state parameter.
On the basis of the above examples, as a preferred embodiment, the forward index is normalized by the following formula:
for the moderation index, it is normalized using the following formula:
for the reverse index, the following formula is adopted for normalization processing:
in the above, x' ij Is the normalized value, x, of the jth operating state parameter ij As an actual measurement value of the operating state parameter,
for the average value of the j-th operating state parameter, < >>
Is the standard deviation of the jth operating state parameter.
The data matrix parameter matrix of the k index group after normalization becomes:
and S15, carrying out secondary feature extraction on the parameter group based on a principal component analysis method, and carrying out secondary feature extraction operation by adopting PCA principal component analysis to reduce the feature dimension due to excessive state evaluation feature parameters in the gear box parameter group and the rolling bearing parameter group.
Wherein x is 1 ,x 2 ,...,x p As the original parameter variable, y 1 ,y 2 ,...,y m (m is less than or equal to p) is a main variable after dimension reduction, a mp Coefficient values representing original parameter variables in the main variables after dimension reduction; y is i And y is j (i.noteq.j; j=1, 2, …, m) independent of each other, lambda i Representing the contribution degree of principal component i, CPV representing the ratio of each principal component variable to the total variable, namely the contribution rate of the principal component to the total variance of the sample, m being the number of principal component variables selected according to the contribution rate, the number m of principal component variables being selected according to the contribution rate, the embodiment taking CPV of 90% or more as the standard, alpha i The weight of each principal component is represented.
On the basis of the above embodiments, as a preferred implementation manner, the initial weight of each state evaluation feature parameter in the parameter group is calculated based on an entropy method, and the following formula is adopted:
K＝1/ln(n)
In the above, X' k The data matrix after normalization and PCA feature dimension reduction is used as the kth parameter group, n is the number of samples, m is the number of finally selected state evaluation feature parameters, namely principal component variables selected in the steps, and x' ij Is X' k The ith row and the jth column elements of P ij Evaluating the contribution degree of the ith sample under the characteristic parameters for the jth state, and if P ij Taking P if =0 ij ·lnP ij Is set to be 0, the number of the components is set to be 0,
evaluating the entropy of the characteristic parameters for the individual states of the kth parameter group, < >>
Is X' k Objective weight of the j-th index in the (b).
And correcting the initial weight based on a weighted average method, wherein the initial weight is as shown in the following formula:
in the method, in the process of the invention,
representing the final objective weight of the state-estimated characteristic parameter of the kth parameter group,
and->
The objective weights calculated by principal component analysis and entropy method are respectively represented, w represents the comprehensive weight vector, and m represents the number of finally selected state evaluation characteristic parameters, namely the number of principal component variables.
In the embodiment, the comprehensive weighting method based on principal component analysis and entropy weighting summation ensures the accuracy of weight distribution, thereby effectively improving the accuracy of a subsequently constructed state reference model and further improving the accuracy of state evaluation.
And S2, dividing the operation working conditions of the ship power system into various types, extracting state evaluation characteristic parameters under each type of operation working conditions, and constructing a state evaluation reference model corresponding to the operation working conditions based on the state evaluation characteristic parameters.
In the embodiment, the operating condition characteristic parameters of the ship power system are selected, the historical operating condition data are subjected to operating condition division operation, the operating condition of the ship power system is divided by adopting a K-means algorithm, and the operating condition characteristic parameters comprise the motor rotating speed and the ambient temperature. And according to the collected historical data, primarily dividing the operation working conditions of the system into T-type working conditions.
And (3) performing expansion operation on the data of each working condition by adopting a Smote method, and according to the following formula:
wherein x is Synthesis In order to synthesize a new sample after the synthesis,
for sample point x i Is [0,1 ]]Is a random number. According toNew sample x synthesized as above Synthesis Is at sample point x i And sample dot->
And randomly selecting points on the connected line segments, and equalizing the number of the data set samples under each operation working condition through data expansion.
Based on the above embodiments, as a preferred implementation manner, a BP neural network may be used to identify the real-time operation condition of the ship power system. Assume a given training sample dataset { (x) (1) ，y (1) )，(x (2) ，y (2) )，...，(x (T) ，y (T) ) -wherein y (i) E {1,2,., T }, T is the number of operating condition classes, y can be calculated (i) The conditional probability of =t, t= {1,2,.. 1 ,α 2 ,...,α T 。
On the basis of the above embodiments, as a preferred implementation manner, the state evaluation reference model construction operation is performed based on the extended multi-source sensing data:
wherein K represents the number of gaussian distributions in the gaussian mixture model, K represents the sequential cycling of the K gaussian distributions from 1 to K; p (x|Θ) is the output of the state estimation reference model, ω k Is the weight of the kth Gaussian distribution, and
μ k and C k Respectively represent Gaussian probability density functions N k ＝(x|μ k ,C k ) Mean and covariance of (c). And x is a feature vector formed by evaluating feature parameters by using the selected states according to the operation conditions. Θ= { ω k ,μ k ,C k And represents the set of all parameters of the state evaluation reference model. N (N) k ＝(x|μ k ,C k ) Is the kth gaussian probability density function.
And S3, acquiring an operation condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the operation condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
Step S31, determining a first Mahalanobis distance d between the state evaluation feature vector at the current moment and each Gaussian distribution based on the state evaluation feature vector x at the current moment, the corresponding comprehensive weight vector w, the mean vector μ of the kth Gaussian distribution of the state evaluation reference model under the jth operation condition, and the covariance matrix C k (x)。
Weight coefficient omega based on Gaussian probability density function k The first Markov distance is used for determining a second Markov distance D between the state evaluation feature vector at the current moment and the state evaluation reference model under each operation condition j’ (x)。
In the above, D 1 (x)，D 2 (x)，…，D j' (x)，…D T (x) Respectively representing the second mahalanobis distance between the state evaluation feature vector at the current moment and the reference model of the operation condition j'; omega k Representing the probability of each Gaussian distribution, an
Based on the second mahalanobis distance and the probability alpha that the running condition of the ship power system at the current moment belongs to the condition category j j’ J' e { T }; and determining a third mahalanobis distance between the state evaluation feature vector at the current moment and a state evaluation reference model of T operating conditions, wherein T is the number of the operating condition categories.
Constructing a state evaluation index sa (t) based on the product of the third mahalanobis distance and the normal operation state constant value c of the ship power system:
sa(t)＝cD(x)
limiting the value range of sa (t) to be [0,1], wherein the closer the value of sa (t) is to 0, the worse the current state of the ship power system is; the closer to 1, the better the state of the system at the current time. s denotes the number of sliding windows, i.e. the number of state evaluation feature parameters based on the final selection selected in the embodiment, and SA (t) denotes the average state evaluation index within the sliding window.
Carrying out state evaluation on the ship power system based on the average state evaluation index in the sliding window, wherein if the average state evaluation index of a plurality of continuous sampling points (such as 8 sampling points) is judged to be larger than a first index threshold value or the state evaluation index of at least one sampling point is judged to be larger than a second index threshold value, judging that the ship power system is in an abnormal state; if the average state evaluation index of the continuous multiple sampling points is larger than the second index threshold or the state evaluation index of at least one sampling point is larger than the third index threshold, judging that the immediate shutdown processing is needed; wherein the first indicator threshold is less than the second indicator threshold, and the second indicator threshold is less than the third indicator threshold. In this embodiment, the first index threshold value is 1, the second index threshold value is 2, and the third index threshold value is 3.
The embodiment of the invention also provides a multisource sensing information fusion and expansion device based on the system state representation, and the multisource sensing information fusion and expansion method based on the system state representation in the embodiments comprises the following steps:
the multi-source data fusion module is used for obtaining the running state parameters of all subsystems in the ship power system, and carrying out data cleaning, feature extraction and principal component analysis on the running state parameters so as to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system;
The operating condition dividing module divides the operating condition of the ship power system into various types, extracts state evaluation characteristic parameters under each type of operating condition, and constructs a state evaluation reference model corresponding to the operating condition based on the state evaluation characteristic parameters;
the state evaluation module is used for acquiring the running condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the running condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
Based on the same conception, the embodiment of the present invention further provides a physical structure schematic diagram, as shown in fig. 4, where the server may include: processor 810, communication interface (Communications Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. The processor 810 may invoke logic instructions in the memory 830 to perform the steps of the multi-source sensory information fusion and augmentation method based on system state characterization as described in the various embodiments above. Examples include:
acquiring running state parameters of all subsystems in a ship power system, and performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system;
Dividing the operation working conditions of the ship power system into various types, extracting state evaluation characteristic parameters under each type of operation working conditions, and constructing a state evaluation reference model corresponding to the operation working conditions based on the state evaluation characteristic parameters;
acquiring an operation condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the operation condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Based on the same conception, the embodiments of the present invention also provide a non-transitory computer readable storage medium storing a computer program, where the computer program includes at least one piece of code executable by a master control device to control the master control device to implement the steps of the multi-source sensing information fusion and expansion method based on system state characterization according to the embodiments described above. Examples include:
acquiring running state parameters of all subsystems in a ship power system, and performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system;
dividing the operation working conditions of the ship power system into various types, extracting state evaluation characteristic parameters under each type of operation working conditions, and constructing a state evaluation reference model corresponding to the operation working conditions based on the state evaluation characteristic parameters;
acquiring an operation condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the operation condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
Based on the same technical concept, the embodiments of the present application also provide a computer program, which is used to implement the above-mentioned method embodiments when the computer program is executed by the master control device.
The program may be stored in whole or in part on a storage medium that is packaged with the processor, or in part or in whole on a memory that is not packaged with the processor.
Based on the same technical concept, the embodiment of the application also provides a processor, which is used for realizing the embodiment of the method. The processor may be a chip.
In summary, according to the multisource sensing information fusion and expansion method and system based on system state characterization provided by the embodiment of the invention, only the irreparable key fault factors such as the sliding bearing system parameter group, the gear box system parameter group and the rolling bearing system parameter group are used as the ship power system state evaluation index system, so that the running state of a ship power system can be better reflected, and the effective evaluation of the power system state can be realized; the method has the advantages that the defect that the existing ship power system depends on a single parameter threshold alarming method can be overcome, the degradation process before the failure of the ship power system can be recognized in advance, and the state change of the ship power system during operation can be accurately estimated; and meanwhile, the method adopts a comprehensive weighting method based on principal component analysis and entropy weighting summation to ensure the accuracy of weight distribution, thereby effectively improving the accuracy of a state reference model constructed subsequently, further improving the accuracy of state evaluation and further effectively improving the rationality of an evaluation result.
The embodiments of the present invention may be arbitrarily combined to achieve different technical effects.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk), etc.
Those of ordinary skill in the art will appreciate that implementing all or part of the above-described method embodiments may be accomplished by a computer program to instruct related hardware, the program may be stored in a computer readable storage medium, and the program may include the above-described method embodiments when executed. And the aforementioned storage medium includes: ROM or random access memory RAM, magnetic or optical disk, etc.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The multi-source sensing information fusion and expansion method based on system state characterization is characterized by comprising the following steps of:
acquiring running state parameters of all subsystems in a ship power system, and performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system;
Dividing the operation working conditions of the ship power system into various types, extracting state evaluation characteristic parameters under each type of operation working conditions, and constructing a state evaluation reference model corresponding to the operation working conditions based on the state evaluation characteristic parameters;
acquiring an operation condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the operation condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
2. The system state characterization-based multi-source sensing information fusion and expansion method according to claim 1, wherein the operational state parameters of the sliding bearing system include temperature parameters and pressure parameters; the temperature parameters comprise a first bearing temperature parameter of the drive end sliding bearing and a second bearing temperature parameter of the non-drive end sliding bearing, and the running state parameters of the sliding bearing system further comprise lubricating oil temperature and cooling water temperature; the pressure parameters comprise a first lubricating oil pressure parameter of the non-driving sliding bearing and a second lubricating oil pressure parameter of the non-driving sliding bearing;
the operational state parameters of the gearbox system include a first acceleration parameter including acceleration collected at a plurality of points on the gearbox;
The operating state parameters of the rolling bearing system include a second acceleration parameter including an acceleration collected at each rolling bearing;
after the operation state parameters of all subsystems in the ship power system are obtained, the method further comprises the following steps:
establishing a parameter group based on a time sequence for each subsystem, wherein the parameter group comprises the time sequence and the numerical value of each running state parameter corresponding to the time sequence; the parameter sets include a plain bearing system parameter set, a gearbox system parameter set, and a rolling bearing system parameter set.
3. The method for fusing and expanding multi-source sensing information based on system state characterization according to claim 2, wherein the data cleaning, feature extraction and feature analysis are performed on the running state parameters to obtain state evaluation feature parameters, and specifically comprising:
interpolation processing is carried out on the missing values in the parameter group based on a Lagrange interpolation method, noise data are screened out and removed based on a 3 sigma criterion, and interpolation filling is carried out on the missing values after the noise data are removed based on the Lagrange interpolation method;
performing feature selection on each running state parameter in the parameter group based on a Pearson correlation coefficient method, and extracting state evaluation feature parameters;
And carrying out standardization processing and secondary feature extraction processing on the state evaluation feature parameters, and determining the weight of each state evaluation feature parameter in the parameter group.
4. The method for fusing and expanding multi-source sensing information based on system state characterization according to claim 3, wherein the method for extracting state evaluation feature parameters based on Pearson correlation coefficient method performs feature selection on each operation state parameter in the parameter group, specifically comprises:
in a sliding bearing system parameter group, taking the bearing temperature of a driving end sliding bearing as a reference temperature, determining a first Pearson correlation coefficient of a residual temperature parameter and the reference temperature, and eliminating a temperature parameter of which the first Pearson correlation coefficient is lower than a preset first correlation coefficient threshold value to obtain a state evaluation characteristic parameter corresponding to the sliding bearing system parameter group;
randomly selecting one acceleration from the gear box system parameter group and the rolling bearing system parameter group as a reference acceleration, determining second Pearson correlation coefficients of residual accelerations in the gear box system parameter group and the rolling bearing system parameter group and the reference acceleration, and eliminating accelerations of which the second Pearson correlation coefficients are higher than a second correlation coefficient threshold value; and extracting time domain characteristic parameters and frequency domain characteristic parameters of each acceleration in the gear box system and the rolling bearing system to obtain state evaluation characteristic parameters corresponding to the gear box system parameter group and the rolling bearing system parameter group.
5. The method for fusing and expanding multi-source sensing information based on system state characterization according to claim 3, wherein the step of performing normalization processing and secondary feature extraction processing on the state evaluation feature parameters and determining weights of all operation state parameters in the parameter group specifically comprises the steps of:
based on a multi-index comprehensive evaluation method and a preset index interval, classifying the state evaluation characteristic parameters in each parameter group into a forward index, a moderate index and a reverse index; carrying out standardization processing based on the corresponding strategy;
if the state evaluation characteristic parameter is judged to be a forward index, and the actual measurement value of the running state parameter is not more than an average value; or if the state evaluation characteristic parameter is judged to be a reverse index, and the actual measurement value of the running state parameter is larger than the average value; determining that the running state parameter after standard deviation processing is 0;
if the state evaluation characteristic parameter is judged to be a forward index, and the actual measurement value of the running state parameter is larger than an average value; or if the state evaluation characteristic parameter is judged to be a moderate index; or if the state evaluation characteristic parameter is judged to be a reverse index, and the actual measurement value of the running state parameter is not more than the average value; the running state parameters after standard deviation processing are determined as follows: the difference between the measured value and the average value is divided by the standard deviation of the running state parameter;
And carrying out secondary feature extraction on the parameter group based on a principal component analysis method, calculating initial weights of all state evaluation feature parameters in the parameter group based on an entropy method, and correcting the initial weights based on a weighted average method.
6. The multi-source sensing information fusion and expansion method based on system state characterization according to claim 3, wherein the constructing a state evaluation reference model corresponding to an operation condition based on the state evaluation characteristic parameter specifically comprises:
acquiring a set of state evaluation characteristic parameters corresponding to the current class of operation conditions, and forming a state evaluation characteristic vector based on the state evaluation characteristic parameters;
determining the weight of each state evaluation characteristic parameter in the state evaluation characteristic vector, and modeling the distribution of the state evaluation characteristic parameters as a Gaussian probability density function;
and constructing a state evaluation reference model of the current class of operation conditions based on the weight of each state evaluation characteristic parameter and the corresponding Gaussian probability density function.
7. The multi-source sensing information fusion and expansion method based on system state characterization according to claim 6, wherein the method is characterized in that a state evaluation index is constructed based on the running condition and the mahalanobis distance of the state evaluation reference model, and the state evaluation is performed on the ship power system based on the state evaluation index, and specifically comprises the following steps:
Based on the state evaluation feature vector at the current moment, the corresponding weight, the mean value vector of the Gaussian distribution of the state evaluation reference model under each operation condition, and the covariance matrix, determining the first Markov distance between the state evaluation feature vector at the current moment and each Gaussian distribution;
determining a second mahalanobis distance between the state evaluation feature vector at the current moment and the state evaluation reference model under each operation condition based on the weight coefficient of the Gaussian probability density function and the first mahalanobis distance;
determining a third mahalanobis distance between a state evaluation feature vector at the current moment and a state evaluation reference model of T operating conditions based on the second mahalanobis distance and the probability that the operating conditions of the ship power system at the current moment belong to each category, wherein T is the number of the operating condition categories;
constructing a state evaluation index based on the product of the third mahalanobis distance and a normal running state constant value of the ship power system;
carrying out state evaluation on the ship power system based on the average state evaluation index in the sliding window, wherein if the average state evaluation index of the continuous multiple sampling points is judged to be larger than a first index threshold value or the state evaluation index of at least one sampling point is judged to be larger than a second index threshold value, the abnormal state is judged; if the average state evaluation index of the continuous multiple sampling points is larger than the second index threshold or the state evaluation index of at least one sampling point is larger than the third index threshold, judging that the immediate shutdown processing is needed; wherein the first indicator threshold is less than the second indicator threshold, and the second indicator threshold is less than the third indicator threshold.
8. The utility model provides a multisource sensing information fuses and expansion device based on system state characterization which characterized in that includes:
the system comprises a multisource data fusion module, a state evaluation module and a state evaluation module, wherein the multisource data fusion module is used for acquiring operation state parameters of all subsystems in a ship power system, and performing data cleaning, feature extraction and principal component analysis on the operation state parameters to obtain state evaluation feature parameters; the subsystem includes a plain bearing system, a gearbox system, and a rolling bearing system;
the operating condition dividing module divides the operating condition of the ship power system into various types, extracts state evaluation characteristic parameters under each type of operating condition, and constructs a state evaluation reference model corresponding to the operating condition based on the state evaluation characteristic parameters;
the state evaluation module is used for acquiring the running condition and a state evaluation reference model at the current moment, constructing a state evaluation index based on the running condition and the mahalanobis distance of the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor performs the steps of the system state characterization based multisource sensory information fusion and augmentation method of any one of claims 1 to 7 when the program is executed.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the multisource sensory information fusion and expansion method based on system state characterization according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310344544.9A CN116383615A (en) | 2023-03-29 | 2023-03-29 | Multisource sensing information fusion and expansion method based on system state characterization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310344544.9A CN116383615A (en) | 2023-03-29 | 2023-03-29 | Multisource sensing information fusion and expansion method based on system state characterization |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116383615A true CN116383615A (en) | 2023-07-04 |
Family
ID=86967018
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310344544.9A Pending CN116383615A (en) | 2023-03-29 | 2023-03-29 | Multisource sensing information fusion and expansion method based on system state characterization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116383615A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117828306A (en) * | 2024-03-01 | 2024-04-05 | 青岛哈尔滨工程大学创新发展中心 | Data sample expansion method and system based on ship motion frequency spectrum characteristics |
-
2023
- 2023-03-29 CN CN202310344544.9A patent/CN116383615A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117828306A (en) * | 2024-03-01 | 2024-04-05 | 青岛哈尔滨工程大学创新发展中心 | Data sample expansion method and system based on ship motion frequency spectrum characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111596604B (en) | Intelligent fault diagnosis and self-healing control system and method for engineering equipment based on digital twinning | |
| CN112016251B (en) | Nuclear power device fault diagnosis method and system | |
| CN108022058B (en) | Wind turbine state reliability assessment method | |
| CN110378042B (en) | Wind turbine generator gearbox oil temperature anomaly detection method and system based on SCADA data | |
| CN111415070A (en) | Wind turbine generator gearbox oil temperature over-temperature fault early warning method based on SCADA data | |
| CN112834211A (en) | Fault early warning method for transmission system of wind turbine generator | |
| CN116383615A (en) | Multisource sensing information fusion and expansion method based on system state characterization | |
| CN112580267A (en) | Aero-engine surge prediction method based on multi-branch feature fusion network | |
| CN110570013B (en) | Single-station online wave period data prediction diagnosis method | |
| Wang et al. | Fault diagnosis of centrifugal pump using symptom parameters in frequency domain | |
| CN116311594B (en) | Ship subsystem state analysis method, device and storage medium | |
| Zhang et al. | Remaining Useful Life Prediction of Rolling Bearings Using Electrostatic Monitoring Based on Two‐Stage Information Fusion Stochastic Filtering | |
| CN113112000B (en) | Construction method and application of multilayer convolutional neural network model | |
| CN110969185A (en) | Equipment abnormal state detection method based on data reconstruction | |
| CN114091525A (en) | Rolling bearing degradation trend prediction method | |
| CN112418306A (en) | Gas turbine compressor fault early warning method based on LSTM-SVM | |
| Ding et al. | Gear fault diagnosis based on VMD sample entropy and discrete hopfield neural network | |
| CN115982663A (en) | Fault early warning system of new forms of energy fan equipment | |
| CN117520960A (en) | Bearing fault diagnosis method based on XGBoost and feature combination | |
| CN112308391A (en) | Real-time monitoring and anomaly detection method for equipment state based on neural network | |
| Shan et al. | Semisupervised fault diagnosis of gearbox using weighted graph-based label propagation and virtual adversarial training | |
| CN112326241A (en) | Nuclear power main pump bearing fault early warning method based on fusion degradation index | |
| CN117034001A (en) | Wind turbine generator system fault prediction method and device and electronic equipment | |
| CN117093938A (en) | Fan bearing fault detection method and system based on deep learning | |
| CN116862284A (en) | Marine diesel engine health state assessment method based on SOM network and rule fusion |
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 |