CN111190349A - Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment - Google Patents
Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment Download PDFInfo
- Publication number
- CN111190349A CN111190349A CN201911398668.5A CN201911398668A CN111190349A CN 111190349 A CN111190349 A CN 111190349A CN 201911398668 A CN201911398668 A CN 201911398668A CN 111190349 A CN111190349 A CN 111190349A
- Authority
- CN
- China
- Prior art keywords
- equipment
- model
- monitoring
- state
- ship
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Abstract
The invention provides a method, a system and a medium for monitoring the state and diagnosing the fault of equipment in a marine engine room, which comprise the following steps: step 1: segmenting the ship model; step 2: selecting characteristic parameters of each sub-model, and extracting training samples; and step 3: training the training samples; and 4, step 4: and carrying out real-time state monitoring and fault diagnosis. The invention greatly reduces the threshold for constructing equipment state monitoring, and provides clear information guidance for subsequent fault diagnosis and assistant decision; the invention can meet the requirements of monitoring and evaluating the state of the ship equipment by the intelligent ship and the unmanned ship at present, and improve the safety of the ship.
Description
Technical Field
The invention relates to the technical field of ships, in particular to a method, a system and a medium for state monitoring and fault diagnosis of ship cabin equipment. In particular to a ship cabin equipment state monitoring and fault diagnosis method combining a self-organizing mapping network and a support vector machine.
Background
The complex and various and degree of automation of ship cabin equipment is higher and higher, the daily maintenance and troubleshooting of equipment are more and more difficult, the cost of equipment maintenance is higher and higher, and along with the release of China classification society's "intelligent ship standard" and the more and more of actual shipment quantity, the intelligent trend of ship cabin equipment is more and more obvious, what's more and more follows this, it is massive data, how to make full use of current data carries out initiative operation and maintenance service to key equipment in the cabin, and then effectively support and promote the intellectuality and the operation and maintenance management level of boats and ships, have become the key research direction in the intelligent field of boats and ships. At present, the algorithm for monitoring the equipment state and diagnosing the fault by utilizing big data machine learning has the accuracy which depends on the quality of a sample and the quantity of a fault sample, only normal data is needed at first in practical application, abnormal data is less, so that a model cannot be established, the accuracy rate is not enough in practical operation, and the diagnosis result has no interpretability, so that the fault location is not accurate enough.
The method can gradually enrich the fault samples by combining the machine learning model with the statistical data under the condition that only a few fault samples exist, and can gradually increase the confidence coefficient of the output result of the fault sample model in the algorithm decision along with the increase of the fault samples, so that the iteration and the optimization of the model are realized. Finally, deviation quantification is carried out on all parameters through a statistical model after a fault is monitored, and influence factors are sequenced according to the normalized distance, so that accurate fault feature positioning is realized.
In the process of monitoring the state of the key equipment in the cabin of the ship, the health state of the equipment in the healthy state is good at the initial stage of operation of the mechanical equipment, at the moment, the characteristic parameters at the moment can be considered to be normal state data samples, the abnormal state data samples are few, the normal and reverse samples are generally needed by a conventional machine learning algorithm, many pieces of equipment often need to be considered to simulate abnormal data through simulation or bench test data when a model is built, the data are often inaccurate, and the difference between the working condition change and the individual and use occasions of the equipment is not considered.
The invention is based on deep research on the incidence relation among all subsystems, equipment and equipment of the key equipment (such as a diesel engine) of a marine engine room, carries out hierarchical division, carries out monitoring characteristic parameter extraction and establishes a state monitoring model of the equipment aiming at all the projects, subsystems and equipment, and further establishes a complete monitoring model of the system-level running state of the marine equipment. The invention utilizes the multidimensional vector under each typical working condition of equipment operating for a period of time to carry out model training, and the trained model can be directly called and can realize the automatic diagnosis of the fault/failure state of the current equipment by a computer. The model can use the working condition data as characteristic parameters, a multi-dimensional characteristic vector generated by operation as a test set, calculate the current state of the model in real time, and when the abnormal state is monitored, calculate the Fisher of each parameter and a sample in a normal state to judge and obtain key parameters causing relevant faults/failures of equipment, thereby finding out the early potential faults of the equipment under different working conditions. And a multi-model conditional series-parallel operation mechanism is adopted to improve the online diagnosis efficiency and accuracy of equipment state monitoring and fault diagnosis in the current intelligent cabin system. The intelligent ship and unmanned ship state monitoring system meets the requirements of state monitoring and fault diagnosis of ship equipment by an existing intelligent ship and an existing unmanned ship.
Patent document CN103760871B (application number: 201410039061.9) discloses a diagnosis system and a diagnosis method for a ship state, the diagnosis system includes a ship-side data monitoring system and a shore-side cloud diagnosis system, the ship-side data monitoring system includes an engine room data acquisition base station, an industrial personal computer, a ship-side server and a ship-side multi-information fusion analysis module, and the shore-side cloud diagnosis system includes a ship data server cluster, a data algorithm server and a user client. The invention processes data into two stages: the first stage performs routine analysis on the real-time data, and has small calculated amount and high real-time performance; and the second stage sends the ship-side data to a shore-side cloud diagnosis system through ship-shore communication for further complex analysis and state trend prediction, and the method is more accurate in analysis, can realize a prediction function, but is slightly poor in real-time performance.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a method, a system and a medium for monitoring the state and diagnosing the fault of equipment in a marine engine room.
The invention provides a method for monitoring the state of equipment in a marine engine room and diagnosing faults of the equipment, which comprises the following steps:
step 1: segmenting the ship model;
step 2: selecting characteristic parameters of each sub-model, and extracting training samples;
and step 3: training the training samples;
and 4, step 4: and carrying out real-time state monitoring and fault diagnosis.
Preferably, the step 2 includes: extracting normal samples and extracting abnormal samples.
Preferably, the step 3 comprises: and carrying out SOM model training and SVM model training.
Preferably, the normal samples are extracted for training the self-organizing network mapping network and the support vector machine model.
Preferably, the fault samples are extracted for training the self-organizing network mapping network and the support vector machine model.
Preferably, the plurality of models are subjected to series-parallel operation under set conditions, so that the operation efficiency is improved, and the operation state of the equipment is preliminarily output.
Preferably, the step 4 comprises: under the condition that the equipment state is monitored to be abnormal, when the model is judged to belong to the abnormal class, calculating the discrimination distance between the normal sample and the sample after normalization processing of the abnormal sample;
the larger the discrimination distance is, the more serious the characteristic parameter deviates from the normal sample, and finally the running state of the equipment is output.
Preferably, the abnormal result is confirmed by a statistical method and then is put into the abnormal sample, the abnormal sample is gradually enriched, and the automatic updating of the abnormal sample is realized.
The invention provides a ship cabin equipment state monitoring and fault diagnosis system, which comprises:
module M1: segmenting the ship model;
module M2: selecting characteristic parameters of each sub-model, and extracting training samples;
module M3: training the training samples;
module M4: and carrying out real-time state monitoring and fault diagnosis.
Compared with the prior art, the invention has the following beneficial effects:
1. the establishment of the model only needs a small amount of abnormal data, the weight of the diagnosis result of the whole model is dynamically adjusted along with the gradual increase of the abnormal data, so that the threshold for establishing the equipment state monitoring is greatly reduced, the model can be dynamically updated and iterated, and when the model monitors that the health state is abnormal, the fault characteristic parameter can be found out by judging the distance, so that definite information guidance can be provided for subsequent fault diagnosis and auxiliary decision;
2. the invention can meet the requirements of monitoring and evaluating the state of the ship equipment by the intelligent ship and the unmanned ship at present, and improve the safety of the ship.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a management hierarchy chart of a key equipment model of a power type cabin;
FIG. 2 is a flow chart of the method of the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
The specific implementation method of the invention is as follows:
the method comprises the following steps: carrying out model segmentation:
when the state monitoring and fault diagnosis of a main power system of a ship engine room are solved, firstly, all subsystems, equipment and equipment under the main power system are hierarchically divided: and partitioning the system level model by adopting a modular partitioning method. As shown in FIG. 1, a management hierarchy chart of a key equipment model of a power type cabin is divided.
Step two: selecting characteristic parameters of each part (subsystem):
taking a diesel supercharger as an example to explain a selection method of characteristic parameters of each part (subsystem), wherein the diesel supercharger is taken as a core part of a diesel engine, and through mechanism analysis, the selected characteristic parameters comprise 6 characteristic parameters of supercharger lubricating oil inlet pressure, supercharger lubricating oil outlet temperature, supercharger inlet exhaust temperature, supercharger outlet exhaust temperature, supercharger rotating speed and supercharger vibration effective value. The rotating speed of the supercharger reflects the working condition of the supercharger, the effective value of the vibration of the supercharger can reflect the bearing fault of the supercharger and the dynamic balance change of a rotor of the supercharger, and the lubricating condition of the supercharger, the inlet exhaust temperature of the supercharger and the outlet exhaust temperature of the supercharger can reflect the working efficiency of the supercharger.
Step three: extracting training samples of each sub-model:
after the characteristic parameters are selected, training samples of each sub-model need to be extracted for training. The characteristic parameters first need to extract a training sample that can cover the full operating condition. Firstly, extracting original data according to the time label, and rejecting a sample with poor data quality on the time label. Secondly, randomly sampling from the data samples extracted for the first time to visualize the sampled data, ensuring the approximate balance of the samples, such as unbalance, extracting the training samples again, and storing the extracted samples in a CSV format for model training. And similarly, extracting a test sample of the model.
1) Extraction of a normal sample:
and (4) carrying out statistical screening on the data by adopting a threshold value set by an expert, and reserving the vector with normal characteristic parameters on the time mark to preliminarily classify the vector into a normal sample.
2) Extracting an abnormal sample:
and (3) carrying out statistical screening on the data by adopting a threshold value set by an expert, and retaining a vector of any one abnormal characteristic parameter of the time mark to preliminarily classify the data into normal samples.
Step four: constructing a sub-model and optimizing parameters:
and reading the data sample file extracted in the last step, and carrying out supercharger state model training.
1) Training of the SOM model:
SOM, proposed in 1981 by TeuvoKohonen, professor of helsinki university, finland, has now become the most widely used method for ad hoc neural networks, in which the WTA (winnertakeall) competition mechanism reflects the most fundamental feature of ad hoc learning. After the input vector is input into the network, the network randomly selects a sample X in the input vector, and then calculates the distance between the sample X and the weight vectors of all SOMs. The best matching cell BMU is the cell whose weight vector is closest to X. After the BMU is determined, the weight vector of the BMU and its topological relation neighbor cells are updated in time. After learning is complete, the weight vectors are classified according to their distance in the input space. After model training is complete, a recommended quantization error MQE is generated that characterizes the distance between the BMU and the input data. The state of the equipment and the degradation degree of the performance can be described qualitatively and quantitatively through the setting and the changing trend of the MQE value.
The SOM parameters mainly include initialization domain radius (initialradius), initial learning rate (initialerlearning), maximum iteration number, confidence coefficient, and the like. The parameters of the model can be used for generating one or more groups of hyper-parameters by adopting a grid method, the model trained by each group of parameters evaluates the precision and the operation efficiency of the model under different parameters by utilizing cross validation, and the hyper-parameter optimization is carried out on the basis.
2) Training of the SVM model:
before model training, the following hyper-parameters of the model need to be initialized: nu (penalty parameter), kernel function (using RBF), gamma coefficient of kernel function, etc.
The penalty parameter nu is an important parameter influencing the state monitoring of the model, namely the tolerance of errors. The larger nu is, the less tolerable errors are, the easier overfitting is, and the smaller nu is, the easier underfitting is. nu is too large or too small, which leads to poor generalization ability of the model.
gamma is a parameter of the kernel function itself after the RBF function is selected as the kernel. The distribution of the data after being mapped to a new feature space is determined implicitly, the larger the gamma is, the fewer the support vectors are, and the smaller the gamma value is, the more the support vectors are. The number of support vectors affects the speed of training and prediction.
The parameter optimization of the model is to seek the model trained by each pair of parameters in a two-dimensional parameter matrix consisting of nu and gamma, verify the model by using a test sample, and count the accuracy of the same batch of samples under different parameters. In order to ensure the accuracy of the result, a plurality of test sets can be used for verification, and then the optimal nu and gamma can be selected.
Step five: the multi-model conditional series-parallel operation mechanism carries out real-time state monitoring and fault diagnosis:
after the optimal parameters of the model are selected, the optimal parameters can be used for training a monitoring model of the equipment state of a certain piece of equipment, and the model is input as a multi-dimensional vector of the characteristic parameters of the current equipment. Because the efficiency of the SOM algorithm is better than that of the SVM, the feature vector is firstly calculated by using the SOMN, if the calculation result is normal, the next cycle is entered, if the calculation result is abnormal, the SVM calculation is further performed by using the SVMN, and in the program operation, the SOMN and the SVMN and the SOME and the SVME are in series relation. The SOMN and the SOME are in parallel relation, and the weight of the output result of the SOMN and the SOME is gradually adjusted through accumulation of fault samples. And finally giving out comprehensive judgment of normality or abnormality of the equipment through a judgment function f (x).
Step six: and (3) calculating a discrimination distance when the model is abnormal:
if the real-time state monitoring model detects the current equipment fault, the judgment distances of all the parameters are calculated respectively:
assume a sample set: x ═ X1，X2，X3…,Xn，
Class omega 1 (positive)Common sample set): x ═ X11，X12，X13…,X1n
ω 2 class (outlier sample set): x ═ X21，X22，X23…,X2n
The solution for the discrimination distance D is as follows:
wherein m is1A certain parameter mean vector of ω 1 class, m2Mean value of a certain parameter of the omega 2 class, s1Variance, s, of a certain parameter of the ω 1 class2The variance of some parameter of the ω 2 class.
And then, the discrimination of each characteristic parameter is sequenced to obtain each characteristic parameter, and the larger the discrimination distance D of the parameter is, the larger the degree of deviation of the parameter from the normal sample is.
Step seven: constructing a system level model by using the sub-models:
and (4) constructing multi-dimensional vector models of different subsystems and components by using the second step to the sixth step, and then constructing a system-level state monitoring model by using testability modeling software, so as to perform complex equipment or system-level reasoning.
The invention is applied to the ship diesel engine state monitoring project, and can diagnose the fuel oil, the lubricating oil, the cooling water, the exhaust system, the air intake system, the supercharger, the cylinder unit and the like of the ship.
As shown in fig. 2, is a flow chart of the method of the present invention, including: the specific implementation process and steps of the invention are as follows:
the method comprises the following steps: carrying out model segmentation;
step two: selecting characteristic parameters of each part (subsystem);
step three: extracting training samples of each sub-model;
1) extracting a normal sample;
2) extracting an abnormal sample;
step four: constructing a sub-model and optimizing parameters;
1) training a SOM model;
2) training an SVM model;
step five: the multi-model conditional series-parallel operation mechanism carries out real-time state monitoring and fault diagnosis;
step six: calculating a discrimination distance when the model is abnormal;
step seven: constructing a system level model by utilizing the sub-models;
and (4) constructing multi-dimensional vector models of different subsystems and components by using the second step to the sixth step, and then constructing a system-level state monitoring model by using testability modeling software, so as to perform complex equipment or system-level reasoning.
Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A ship cabin equipment state monitoring and fault diagnosis method is characterized by comprising the following steps:
step 1: segmenting the ship model;
step 2: selecting characteristic parameters of each sub-model, and extracting training samples;
and step 3: training the training samples;
and 4, step 4: and carrying out real-time state monitoring and fault diagnosis.
2. The marine vessel cabin equipment state monitoring and fault diagnosis method according to claim 1, wherein the step 2 includes: extracting normal samples and extracting abnormal samples.
3. The marine vessel cabin equipment state monitoring and fault diagnosis method according to claim 1, wherein the step 3 includes: and carrying out SOM model training and SVM model training.
4. The method for monitoring the state of equipment in the marine engine room and diagnosing the fault according to claim 2, wherein a normal sample is extracted to train the self-organizing network mapping network and the support vector machine model.
5. The method for monitoring the state of equipment and diagnosing the fault in the cabin of the ship according to claim 2, wherein a fault sample is extracted to train a self-organizing network mapping network and a support vector machine model.
6. The method for monitoring the state of the equipment in the marine engine room and diagnosing the fault of the equipment according to claim 1, wherein a plurality of models are connected in series and in parallel under a set condition, so that the operation efficiency is improved, and the operation state of the equipment is preliminarily output.
7. The marine vessel cabin equipment state monitoring and fault diagnosis method according to claim 1, wherein the step 4 includes: under the condition that the equipment state is monitored to be abnormal, when the model is judged to belong to the abnormal class, calculating the discrimination distance between the normal sample and the sample after normalization processing of the abnormal sample;
the larger the discrimination distance is, the more serious the characteristic parameter deviates from the normal sample, and finally the running state of the equipment is output.
8. The method for monitoring the state of and diagnosing faults of equipment in the marine engine room according to claim 1, wherein the abnormal result is confirmed by a statistical method and then is put into the abnormal sample, so that the abnormal sample is gradually filled, and the automatic updating of the abnormal sample is realized.
9. A ship cabin equipment state monitoring and fault diagnosis system is characterized by comprising:
module M1: segmenting the ship model;
module M2: selecting characteristic parameters of each sub-model, and extracting training samples;
module M3: training the training samples;
module M4: and carrying out real-time state monitoring and fault diagnosis.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911398668.5A CN111190349A (en) | 2019-12-30 | 2019-12-30 | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911398668.5A CN111190349A (en) | 2019-12-30 | 2019-12-30 | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111190349A true CN111190349A (en) | 2020-05-22 |
Family
ID=70707846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911398668.5A Pending CN111190349A (en) | 2019-12-30 | 2019-12-30 | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111190349A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113033011A (en) * | 2021-04-07 | 2021-06-25 | 大连海事大学 | Ship machinery health state assessment method and system |
| CN113257329A (en) * | 2021-04-12 | 2021-08-13 | 中国空间技术研究院 | Memory fault diagnosis method based on machine learning |
| CN113657664A (en) * | 2021-08-16 | 2021-11-16 | 大连海事大学 | System and method for predicting state of equipment in marine engine room |
| CN114137915A (en) * | 2021-11-18 | 2022-03-04 | 成都航天科工大数据研究院有限公司 | Fault diagnosis method for industrial equipment |
| CN114662710A (en) * | 2022-01-27 | 2022-06-24 | 武汉船用机械有限责任公司 | Method, device and equipment for determining maintenance requirement of marine engine room pump |
| CN116027724A (en) * | 2022-09-23 | 2023-04-28 | 河北东来工程技术服务有限公司 | Ship equipment risk monitoring method and system |
| CN116030552A (en) * | 2023-03-30 | 2023-04-28 | 中国船舶集团有限公司第七一九研究所 | Intelligent comprehensive display control method and system for ship |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105319071A (en) * | 2015-09-21 | 2016-02-10 | 天津大学 | Diesel engine fuel oil system fault diagnosis method based on least square support vector machine |
| CN108062586A (en) * | 2017-11-30 | 2018-05-22 | 中国船舶工业系统工程研究院 | Marine main engine associated member state monitoring method and system based on decline contribution degree |
| CN108491580A (en) * | 2018-02-26 | 2018-09-04 | 上海理工大学 | A kind of equipment fault diagnosis apparatus and system |
| CN108961468A (en) * | 2018-06-27 | 2018-12-07 | 大连海事大学 | A kind of ship power system method for diagnosing faults based on integrated study |
| CN109670584A (en) * | 2018-12-27 | 2019-04-23 | 张婧 | A kind of fault diagnosis method and system based on big data |
| CN110261122A (en) * | 2019-06-20 | 2019-09-20 | 大连理工大学 | A kind of boat diesel engine fault monitoring method based on piecemeal |
| CN110263846A (en) * | 2019-06-18 | 2019-09-20 | 华北电力大学 | The method for diagnosing faults for being excavated and being learnt based on fault data depth |
-
2019
- 2019-12-30 CN CN201911398668.5A patent/CN111190349A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105319071A (en) * | 2015-09-21 | 2016-02-10 | 天津大学 | Diesel engine fuel oil system fault diagnosis method based on least square support vector machine |
| CN108062586A (en) * | 2017-11-30 | 2018-05-22 | 中国船舶工业系统工程研究院 | Marine main engine associated member state monitoring method and system based on decline contribution degree |
| CN108491580A (en) * | 2018-02-26 | 2018-09-04 | 上海理工大学 | A kind of equipment fault diagnosis apparatus and system |
| CN108961468A (en) * | 2018-06-27 | 2018-12-07 | 大连海事大学 | A kind of ship power system method for diagnosing faults based on integrated study |
| CN109670584A (en) * | 2018-12-27 | 2019-04-23 | 张婧 | A kind of fault diagnosis method and system based on big data |
| CN110263846A (en) * | 2019-06-18 | 2019-09-20 | 华北电力大学 | The method for diagnosing faults for being excavated and being learnt based on fault data depth |
| CN110261122A (en) * | 2019-06-20 | 2019-09-20 | 大连理工大学 | A kind of boat diesel engine fault monitoring method based on piecemeal |
Non-Patent Citations (3)
| Title |
|---|
| 何艳 等: "基于SVM-SOM集成学习的故障诊断方法研究", 《自动化技术与应用》 * |
| 林新通 等: "支持向量机在船舶柴油机废气涡轮增压器故障诊断中的应用", 《上海海事大学学报》 * |
| 舒服华: "基于粗糙集与支持向量机的推土机故障诊断", 《筑路机械与施工机械化》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113033011A (en) * | 2021-04-07 | 2021-06-25 | 大连海事大学 | Ship machinery health state assessment method and system |
| CN113033011B (en) * | 2021-04-07 | 2023-06-06 | 大连海事大学 | Ship mechanical health state assessment method and system |
| CN113257329A (en) * | 2021-04-12 | 2021-08-13 | 中国空间技术研究院 | Memory fault diagnosis method based on machine learning |
| CN113657664A (en) * | 2021-08-16 | 2021-11-16 | 大连海事大学 | System and method for predicting state of equipment in marine engine room |
| CN114137915A (en) * | 2021-11-18 | 2022-03-04 | 成都航天科工大数据研究院有限公司 | Fault diagnosis method for industrial equipment |
| CN114662710A (en) * | 2022-01-27 | 2022-06-24 | 武汉船用机械有限责任公司 | Method, device and equipment for determining maintenance requirement of marine engine room pump |
| CN116027724A (en) * | 2022-09-23 | 2023-04-28 | 河北东来工程技术服务有限公司 | Ship equipment risk monitoring method and system |
| CN116027724B (en) * | 2022-09-23 | 2024-01-12 | 河北东来工程技术服务有限公司 | Ship equipment risk monitoring method and system |
| CN116030552A (en) * | 2023-03-30 | 2023-04-28 | 中国船舶集团有限公司第七一九研究所 | Intelligent comprehensive display control method and system for ship |
| CN116030552B (en) * | 2023-03-30 | 2023-07-25 | 中国船舶集团有限公司第七一九研究所 | Intelligent comprehensive display control method and system for ship |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111190349A (en) | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment | |
| CN115270956B (en) | Continuous learning-based cross-equipment incremental bearing fault diagnosis method | |
| CN112418277A (en) | Method, system, medium, and apparatus for predicting remaining life of rotating machine component | |
| CN111680875B (en) | Unmanned aerial vehicle state risk fuzzy comprehensive evaluation method based on probability baseline model | |
| CN114813105A (en) | Gear box fault early warning method and system based on working condition similarity evaluation | |
| CN110570012B (en) | Storm-based power plant production equipment fault early warning method and system | |
| CN116448419A (en) | Zero sample bearing fault diagnosis method based on depth model high-dimensional parameter multi-target efficient optimization | |
| CN114676742A (en) | Power grid abnormal electricity utilization detection method based on attention mechanism and residual error network | |
| CN115419558A (en) | Wind turbine generator running state evaluation method and device | |
| Shi et al. | Health index synthetization and remaining useful life estimation for turbofan engines based on run-to-failure datasets | |
| CN115822887A (en) | Performance evaluation and energy efficiency diagnosis method and system of wind turbine generator | |
| Wang et al. | A remaining useful life prediction model based on hybrid long-short sequences for engines | |
| CN112241606A (en) | Cooperative decision-making method for operation and maintenance of ship intelligent equipment based on CPS decision-making module | |
| CN112421631A (en) | New energy consumption capacity assessment method and system | |
| CN111273125A (en) | RST-CNN-based power cable channel fault diagnosis method | |
| CN115526258A (en) | Power system transient stability evaluation method based on Spearman correlation coefficient feature extraction | |
| Ahsan et al. | Developing computational intelligence for smart qualification testing of electronic products | |
| CN107527093B (en) | Wind turbine generator running state diagnosis method and device | |
| Jiang et al. | A SVDD and K-means based early warning method for dual-rotor equipment under time-varying operating conditions | |
| CN113689072A (en) | Offshore wind turbine generator running state evaluation method based on Markov chain | |
| CN105426665B (en) | Method is determined based on the DYNAMIC RELIABILITY of status monitoring | |
| CN115598459A (en) | Power failure prediction method for 10kV feeder line fault of power distribution network | |
| CN111461565A (en) | Power supply side power generation performance evaluation method under power regulation | |
| Yuan | Unsupervised Machine Condition Monitoring Using Segmental Hidden Markov Models. | |
| Zhao et al. | Remaining useful life prediction method based on convolutional neural network and long short-term memory neural network |
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: 20200522 |
|
| RJ01 | Rejection of invention patent application after publication |