CN108062586A - Marine main engine associated member state monitoring method and system based on decline contribution degree - Google Patents
Marine main engine associated member state monitoring method and system based on decline contribution degree Download PDFInfo
- Publication number
- CN108062586A CN108062586A CN201711235510.7A CN201711235510A CN108062586A CN 108062586 A CN108062586 A CN 108062586A CN 201711235510 A CN201711235510 A CN 201711235510A CN 108062586 A CN108062586 A CN 108062586A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- economic
- mover
- host
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000012544 monitoring process Methods 0.000 title claims abstract description 46
- 230000007423 decrease Effects 0.000 title claims abstract description 21
- 239000000446 fuel Substances 0.000 claims abstract description 87
- 230000036541 health Effects 0.000 claims abstract description 80
- 238000004458 analytical method Methods 0.000 claims abstract description 57
- 238000012549 training Methods 0.000 claims abstract description 26
- 230000002159 abnormal effect Effects 0.000 claims abstract description 19
- 238000012423 maintenance Methods 0.000 claims abstract description 18
- 238000011234 economic evaluation Methods 0.000 claims abstract description 11
- 238000011835 investigation Methods 0.000 claims abstract description 9
- 230000003862 health status Effects 0.000 claims abstract description 5
- 239000013598 vector Substances 0.000 claims description 74
- 238000005070 sampling Methods 0.000 claims description 41
- 230000008569 process Effects 0.000 claims description 29
- 230000015556 catabolic process Effects 0.000 claims description 20
- 238000006731 degradation reaction Methods 0.000 claims description 20
- 238000011156 evaluation Methods 0.000 claims description 20
- 238000012360 testing method Methods 0.000 claims description 19
- 238000010276 construction Methods 0.000 claims description 13
- 238000013139 quantization Methods 0.000 claims description 13
- 238000001816 cooling Methods 0.000 claims description 12
- 238000012163 sequencing technique Methods 0.000 claims description 7
- 210000002569 neuron Anatomy 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 description 7
- 238000005259 measurement Methods 0.000 description 6
- 230000002000 scavenging effect Effects 0.000 description 5
- 238000012795 verification Methods 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 3
- MWUXSHHQAYIFBG-UHFFFAOYSA-N nitrogen oxide Inorganic materials O=[N] MWUXSHHQAYIFBG-UHFFFAOYSA-N 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 238000000692 Student's t-test Methods 0.000 description 2
- 239000000498 cooling water Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 239000003921 oil Substances 0.000 description 2
- 238000012353 t test Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 239000002283 diesel fuel Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000000556 factor analysis Methods 0.000 description 1
- 239000000295 fuel oil Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000011155 quantitative monitoring Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Geometry (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Medicines Containing Antibodies Or Antigens For Use As Internal Diagnostic Agents (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention relates to a kind of marine main engine associated member state monitoring methods and system based on decline contribution degree, the method selects input data of the steady working condition data as off-line training in history health sample, builds host economic evaluation baseline and the economic sex-health baseline of associated member；Economy correlation monitoring data are obtained, using the host economic evaluation baseline on-line analysis host economy state of structure, when economy state is normal, are terminated；When economy abnormal state, abnormal alarm is sent, and carries out the analysis of host associated member health status；Obtain the health status result of each component and the order of each parts for maintenance investigation.This method and system are from the health status monitoring for being carried out at the same time host economy condition adjudgement and host associated member, to effectively instructing the maintaining of host and associated member, recover the economic performance of host, exclude associated member security risk or exception/fault and reduce fuel cost, great realistic meaning.
Description
Technical Field
The invention relates to the field of economic performance evaluation and analysis of ship hosts, in particular to a method and a system for monitoring states of ship host related parts based on degradation contribution degrees.
Background
The ship is used as a transportation tool with large transportation volume in transportation, and the operation cost of the ship is 40-60% of fuel consumption, wherein the ship main engine is used as the power 'heart' of the ship, and the fuel consumption of the ship main engine is usually more than 90% of the fuel consumption of the whole ship; taking an ocean ten thousand ton wheel as an example, the fuel consumption of a host computer reaches more than 20-30 tons every time the ocean ten thousand ton wheel sails for one day, which greatly influences the ship operation cost of shipping enterprises, and particularly in the continuous and low-priced market environment of the current shipping market, how to reduce cost and improve efficiency becomes the key point of concern of users of each shipping enterprise; meanwhile, the fuel consumption is closely related to the emission of pollutants, and excessive fuel consumption will inevitably result in the increase of the emission of nitrogen oxides, and influence the marine atmospheric environment in the transportation sea area.
Therefore, how to effectively control the fuel consumption cost of the host, accurately evaluate the economic performance of the host, control the fuel consumption and emission of the host and the like becomes a problem which is extremely concerned and needs to be solved by shipping users at present. For the marine main engine, economic influence factors are numerous, if the economic evaluation result (namely whether the main engine is normally consumed or not and whether the economic operation condition is met) can be directly and quickly obtained, and the influence reason can be quickly found out when the economy is abnormal, abnormal related parts can be positioned, and a countermeasure can be given, so that the economic performance of the main engine can be timely recovered, economic and safety hidden dangers or abnormalities/faults can be eliminated, the fuel consumption is reduced, the fuel cost is saved, and the environmental protection performance of the main engine operation is improved.
In the current technical scheme, a quantitative monitoring method for the state of the host computer related component, which is measured in real time according to the running process of the ship, is not available.
Disclosure of Invention
In view of the foregoing analysis, the present invention aims to provide a method and a system for monitoring a state of a ship host related component based on a degradation contribution degree, so as to solve the problem that the prior art cannot monitor the health state of the host related component in the ship operation process in real time.
The purpose of the invention is mainly realized by the following technical scheme:
on one hand, the method for monitoring the state of the ship host computer related component based on the decline contribution degree comprises the following steps:
selecting stable working condition data in a historical health sample as input data of offline training, constructing a host economic evaluation baseline according to a host rotating speed and a fuel consumption rate measured value in the input data, and constructing an economic association part health baseline by using an SOM (sequence of model) method according to the host rotating speed and economic association part state parameters;
adopting a multi-parameter monitoring system to monitor data, utilizing the established host economy evaluation baseline to analyze the economy state of the host for the monitored data, and ending when the economy state is normal; when the economic state is abnormal, an abnormal alarm is sent out, and the health state of the host computer related component is analyzed; the host associated component health status analysis comprising: analyzing the monitoring data by using the health baseline of the economic relevance component to obtain the degradation contribution degree of the host economic relevance component degradation to the host economic abnormity;
and sequencing the degradation contribution degrees, positioning the components in the front sequence as abnormal components, and performing maintenance and investigation on the corresponding components.
On the basis of the scheme, the invention is further improved as follows:
further, the economy-related components include a cylinder system, a cooling system, and a supercharger system.
Further, the monitoring data are data consistent with the input data type of the off-line training, and comprise the rotating speed of a host, the fuel consumption, the state parameter of the cylinder system, the state parameter of the cooling system and the state parameter of the supercharger system.
Further, constructing an economic-relevant component health baseline comprises the steps of:
selecting system state parameters of each related component related to economy, and constructing an economy related component observation vector of the nth sampleUsing corpus as input vector of networkWherein,sampling time range for stable running process of ship in navigationThe sampling time set used for the SOM network training, s (n) is the rotation speed, x1(n),x2(n),…,xm(n) state parameters of m associated component systems relating to economy;
a jth weight vector W based on the input vectorj(n) obtaining a corpus weight vector Wj(ii) a Wherein,
k is the number of neurons trained by the SOM network;
calculating the Euclidean distance between the input vector and the weight vector, and finding out the point with the minimum distance as the best matching node Wc；
By usingAdjusting the weight to obtain an adjusted corpus weight vector, wherein t is an iteration step, α is a learning rate,to optimally match the node WcCenter winning neighbor with jth neuron and best matched node WcA topological distance function between;
and carrying out iterative training of the SOM network, finishing the training when convergence occurs, and taking the trained SOM network as a health baseline of the economic association part.
Further, the method also comprises the step of testing the trained SOM network, and if the test fails, the training is carried out again until the test is passed.
Further, the step of testing the trained SOM network includes:
by sampling time ranges during stationary operationIs different fromSet of sampling instantsAn economic relevance component observation vector at the nth sample point as a test process time horizon
The full set of input vectors corresponding to the input trained SOM network is
Obtaining an estimate of the SOM network output by a trained SOM networkThe economic relevance component at the nth sample point estimates the vector:
calculating the health baseline deviation vector of the economic relevance part corresponding to the test data:
when economic relevance part health baseline deviation vector deltatAnd when the deviation mean value of each parameter does not exceed the threshold value, the test is considered to be passed.
Further, the method for analyzing the host economic state of the online monitoring data by using the constructed host economic evaluation baseline comprises the following steps:
two-dimensional observation vector based on fuel economy correlationCalculating the mean statistic of fuel efficiency for the same rotation speed, thereby obtaining the sampling time range in the smooth operation processFuel efficiency statistics related to rotational speedWherein n issThe number of different rotating speed values;
will observe the vectorOf different rotational speed values si,i＝1,...,nsInputting a host fuel economy evaluation baseline to obtain a corresponding host fuel economy evaluation baseline estimation value
Calculating the deviation of the input data corresponding to the evaluation baseline of the fuel economy of the host
If all the deviations are smaller than the threshold value, the economic state is normal, and the process is finished;
and if any value in the deviation is larger than the threshold value, the economic state is abnormal, and the health state analysis of the host computer related component is carried out.
Further, the online monitoring data are analyzed by utilizing the health baseline of the economic relevance component to obtain the degradation contribution degree of the host economic relevance component degradation to the host economic abnormity, and the method comprises the following steps:
selecting a smooth running process sampling time rangeSystem state parameters of all relevant parts related to economy form an economy relevant part observation vector of the l sampling pointHealth baseline input vector of economic association part with complete set as SOM network construction
After the SOM network is input, each parameter in the observation vector is calculated to the nearest best matching node WcThe distance value between them, noted as the minimum quantization difference MQE; the minimum quantization difference of the economy-related components for the ith sample point is:
obtaining the recession contribution degrees C (l) of all parts except the rotating speed from the MQE value,
is composed of at least oneInner mean value obtaining decay contribution of integral parts
The beneficial effects of the invention are as follows:
meanwhile, the evaluation of the economy of the host and the analysis of the health state of the related parts are carried out: calculating the fuel consumption rate of the ship in the actual operation process through the fuel consumption and the operation state of the host, and directly evaluating the fuel economy of the host; finding all the relevant parts related to the host computer economy, calculating the contribution degree of the decline of each economy relevant part to the host computer economy abnormity through a data-driven means, and obtaining corresponding economy influence factors and influence degrees quantitatively, thereby effectively guiding the maintenance investigation and maintenance of the host computer and the relevant parts, and having practical significance for recovering the economic performance of the host computer, eliminating potential safety hazards or abnormity/faults and reducing the fuel cost.
In another aspect, a ship host computer associated component state monitoring system based on a degradation contribution degree is provided, including:
the data selection module is configured to select stable working condition data in the historical health samples as input data of offline training;
a baseline construction module configured to perform the construction of a host economic assessment baseline and an economic-related component health baseline;
an online data acquisition module configured to perform online acquisition of economic-related monitoring data;
an economic status analysis module configured to perform analysis of the host economic status using the constructed host economic assessment baseline;
the judgment selection module is configured to execute the selection to be finished when the economy state is judged to be normal; when the economic state is judged to be abnormal, analyzing the health state of the host computer related component;
the system comprises a main engine related component health state analysis module, a main engine related component health state analysis module and a main engine economic related component health state analysis module, wherein the main engine related component health state analysis module is configured to obtain the decline contribution degree of the main engine economic related component by utilizing an SOM method based on an economic related component health baseline;
and the maintenance sequence determining module is configured to execute sequencing on the decline contribution degrees to obtain a health state result of each component, and obtain the sequence of maintenance and investigation of each component according to the result.
Further, the host computer related component health state analysis module is configured to execute the following specific steps:
selecting a smooth running process sampling time rangeSystem state parameters of all relevant parts related to economy form an economy relevant part observation vector of the l sampling pointHealth baseline input vector of economic association part with complete set as SOM network constructions represents the rotational speed, x1,x2,…,xmRepresenting state parameters of m associated component systems related to economy; l represents a sampling point;
after the SOM network is input, each parameter in the observation vector is calculated to the nearest best matching node WcThe distance value between them, noted as the minimum quantization difference MQE; the minimum quantization difference of the economy-related components for the ith sample point is:
obtaining the recession contribution degrees C (l) of all parts except the rotating speed from the MQE value,
is composed of at least oneInner mean value obtaining decay contribution of integral parts
The system of the invention has the same principle as the method, so the system also has the technical effect corresponding to the method.
In the invention, the technical schemes can be combined with each other to realize more preferable combination schemes. Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The drawings are only for purposes of illustrating particular embodiments and are not to be construed as limiting the invention, wherein like reference numerals are used to designate like parts throughout.
FIG. 1 is a flowchart of a method of example 1 of the present invention;
FIG. 2 is a schematic diagram illustrating a clustering principle of a self-organizing map according to embodiment 1 of the present invention;
FIG. 3 is a schematic diagram of a clustering method for self-organizing maps according to embodiment 1 of the present invention;
FIG. 4 is a flowchart showing the overall analysis method in example 1 of the present invention;
FIG. 5 is a baseline chart of economic evaluation of a host according to embodiment 2 of the present invention;
FIG. 6 is a baseline chart of host economy-related component health in accordance with example 2 of the present invention;
FIG. 7 shows the results of the economic analysis of the host according to embodiment 2 of the present invention;
fig. 8 is a graph illustrating the degree of degradation of the economic association unit of the host according to embodiment 2 of the present invention.
Detailed Description
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate preferred embodiments of the invention and together with the description, serve to explain the principles of the invention and not to limit the scope of the invention.
The economic indexes of the ship main engine are divided into direct characterization parameters and indirect characterization parameters.
1) The parameters are directly characterized by the fact that,
2) namely a direct economic key indicator. For marine main engines, the most critical indicator for evaluating the economic performance of their machines is the specific fuel consumption, also known as fuel efficiency, i.e. the amount of fuel consumed per hour per unit of available work, usually expressed as fuel consumption per kilowatt-hour. Therefore, the theoretical calculation formula of the fuel consumption rate of the main engine is as follows:
in the formula,the specific value of fuel oil consumption is unit g/kW.h;the unit is kg/h of fuel consumption per hour;is the effective power of the host machine, and has unit kW.
It should be noted that, in general, a factory bench test of a host computer gives a single value of fuel consumption rate under rated power of a diesel engine as a key economic performance index for evaluating the machine. However, when the main engine is actually operated after being shipped, the fuel consumption rate of the main engine changes along with the change of the real-time rotating speed, and the accuracy of the factory numerical value is poor. I.e. the fuel consumption g of the main engine during the voyageeIs a non-linear function of the speed of rotation s, which can be denoted as ge(s) and the fuel consumption and the diesel power are both related to the rotational speed.
2) And indirectly characterizing parameters, namely system state parameters of each associated component related to the economy. In the online monitoring system of the host computer, each related part system related to the economy comprises an air cylinder system, a cooling system (comprising a cylinder sleeve cooling system and an air cooler) and a supercharger system, and indirect characterization parameters, namely state monitoring parameters of the part systems comprise a plurality of temperature parameters of the air cylinder, temperature parameters of the cooling system and rotating speed parameters of the supercharger.
By combining the two economic index parameters, the observation vector X of the host monitoring parameter related to the economic performance is uniformly described as follows:
X(k)＝[s(k),ge(k),x1(k),x2(k),…,xm(k)]T(2)
wherein s (k) is the rotation speed of the kth monitoring sampling point, representing the current working condition of the host (i.e. the working condition of the host), ge(k) For monitoring the fuel efficiency, i.e. specific fuel consumption, of the sampling point k, which is directly related toThe measured rotation speed, fuel consumption and effective power are related, xi(k) I is 1.. m is the ith economy-related component state parameter, and m is the number of economy-related component state parameters. According to the measuring point conditions of different online monitoring systems, m can generally take the number of state parameters (analog quantities) of a cylinder system, a cooling system and a supercharger system.
Example 1
As shown in fig. 1, the present embodiment provides a method for analyzing the economy of a low-speed diesel engine, including the following steps:
step S1 off-line training, selecting historical health sample data as input data of the off-line training, and constructing an economic evaluation baseline of the low-speed diesel engine and a health baseline of an economic correlation component;
step S2, carrying out online analysis, carrying out economic state analysis according to the online monitoring data, and ending when the economic state is normal; when the economic state is abnormal, analyzing the health state;
analyzing the economic state, namely analyzing the fuel economy based on the economic evaluation baseline of the low-speed diesel engine to obtain a fuel economy state result of the low-speed diesel engine;
and analyzing the health state, namely analyzing the health state of the economy-related component of the low-speed diesel engine based on the health baseline of the economy-related component to obtain the health state result of the economy-related component.
Step S1 is to obtain stable working condition data by cleaning the economic relevant data of the historical faultless low-speed diesel engine, and further to construct the economic baseline of the low-speed diesel engine, including the fuel economy evaluation baseline of the low-speed diesel engine and the health baseline of the economic relevant parts.
Step S2 is based on the low-speed diesel engine economy analysis of the degradation contribution, that is: and selecting stable working condition data according to the economic relevant data of the low-speed diesel engine in a period of operation time, carrying out on-line fuel economy analysis based on the fuel evaluation baseline of the low-speed diesel engine, and further analyzing the recession contribution degree of each relevant module causing economic abnormity when the low-speed diesel engine is abnormal, thereby providing a data basis for eliminating the abnormity in the next step.
Step S1 includes the following specific steps:
s101, selecting historical health sample data as input data of offline training;
s102, constructing an economic evaluation baseline of the low-speed diesel engine;
s103, building an economic relevance component health baseline.
It should be noted that the configuration of the low-speed diesel engine economic baseline of the embodiment includes two parts: firstly, constructing a low-speed diesel engine fuel economy evaluation baseline for representing direct economy key indexes; and secondly, establishing an economic relevance component health baseline for representing indirect economic indexes of the system health state of each relevant component related to the economic property.
The reason for selecting and constructing the two economic baselines is that in order to improve the efficiency of economic analysis, firstly, the evaluation baselines of the fuel economy of the low-speed diesel engine are used for directly judging whether the economic efficiency is abnormal or not (only two-dimensional data is included, and the analysis result can be quickly obtained), if the economic efficiency is normal, further analysis is not needed, and if the economic efficiency is abnormal, the health decline analysis of economic-related parts is triggered, and the analysis conclusion is obtained to give a maintenance guidance suggestion. However, if only one baseline is constructed, which includes both the direct index and the indirect index, the calculation amount of the online analysis is increased, the analysis time is prolonged (the timeliness of the analysis is deteriorated), and unnecessary calculation resources are wasted.
Wherein, step S101 includes the following substeps:
and step S1011, data source acquisition.
Through boats and ships low-speed diesel engine monitoring system, set for historical time length T to find out all no trouble data in the time length, record as N healthy samples, the low-speed diesel engine monitoring parameter observation vector X that wherein the nth monitoring sampling point is relevant with the economic nature, can describe according to (2) as:
X(n)＝[s(n),ge(n),x1(n),x2(n),…,xm(n)]T,n＝1,...,N (3)
wherein the fuel efficiency g heree(n) measuring the fuel consumption G from a direct measurement at the nth monitor sampling pointe(N) and the effective power Ne(n) is obtained by calculation of the formula (1).
And step S1012, determining stable working conditions and obtaining stable working condition data related to economy.
Within N healthy samples, hourly fuel consumption measurements G are selected that are stable on the fly (speed s > 0 (in RPM, i.e., revolutions per minute))e(N; s > 0), N ∈ { 1.., N } and a low speed diesel engine output power measurement Ne(N; s > 0), N is belonged to { 1.., N }, and the fuel consumption rate g related to the rotating speed at the corresponding moment is calculated according to the formula (1)e(n；s＞0),n∈{1,...,N}。
Further, in stable underway samples in the N health samples, the range of the underway smooth operation process is foundObtaining the fuel consumption rate in the smooth operation process
The check level α is set to 0.05, and the fuel consumption rate sample value g at each sampling time is calculated by a hypothesis test method of t-teste(N; s > 0), N ∈ { 1.., N }, test statistic T:
wherein,is the average number of samples, μ0Assuming the population mean of the samples, S is the sample standard deviation,is the sample volume at s > 0.
From this, each specific fuel consumption sample value g is determinede(N, s is more than 0), and whether N is in the range of {1,.., N } can be checked by the hypothesis of the t-test check level α being 0.05, so that sample values (namely fuel consumption rate singular values) which do not pass the check are removed, and the fuel consumption rate in the smooth operation process is obtainedWherein,and sampling a time range for the stable running process of the ship in the process of sailing.
At the same time, the measured value of the rotation speed in the corresponding moment is selectedAnd low-speed diesel engine monitoring parameter observation vectors related to economy at the nth stable working condition monitoring sampling point in corresponding timeAccording to (2) can be described as:
whereby stable condition data relevant to economy is obtained.
Step S102, establishing a fuel economy evaluation baseline, comprising:
step S1021. According to the historical rotating speed in the process of sailing stable operationAnd fuel consumption rate measurementEstablishing a two-dimensional measurement incidence relation;
step S1022, obtaining a rotation speed-fuel consumption rate evaluation model by using a nonlinear function fitting modeAs a baseline for low speed diesel fuel economy evaluation.
Step S103, establishing an economic relevance component health baseline, specifically:
the economy-associated component health baselines are preferably constructed using a network of SOMs (Self-organizing maps, or Self-organizing features maps, SOMs for short), the principles and method of which are schematically illustrated in FIGS. 2-3.
The SOM is a simplified version of a neural network, "self-organizing" in the sense that it can automatically sort and cluster input data through self-learning without inputting attribute information of the data; SOM is a typical "unsupervised learning method" that allows classification and similarity analysis even without data relationship and class information. In general, for a low-speed diesel engine which is just put into use in a ship, it is easy to obtain normal (no fault) historical economy-related data, and it is difficult to obtain economy abnormality data, i.e., it is difficult to obtain various tag data, and therefore, for the analysis of economy abnormality of such data, the unsupervised learning method of SOM is an excellent choice.
The steps for constructing an economy associated part health baseline using the SOM network are as follows:
step S1031, Input vector (Input vector): the observation in the formula (5)Vector quantityIn the method, system state parameters of all relevant parts related to the economy, including state monitoring parameters of a rotating speed, a cylinder system, a cooling system, a supercharger system and the like, are selected to form an economy relevant part observation vector of an nth sampling pointUsing corpus as input vector of networkWherein,sampling time range for stable running process of ship in navigationThe set of sampling instants used for SOM network training.
Step S1032, input initial Weight vector (Weight vector): the jth weight vector for the input vector is noted ask is the number of neurons, and the total set is the weight vector Wj. The initial weight may be set arbitrarily, and for simplicity, may be set to [0,1,0,1.,]Tand performing later iteration adjustment to obtain a weight vector at the convergence moment.
Step S1033, SOM network training process:
finding the best Matching node bmu (best Matching unit): calculating Euclidean distances between the input vector and the weight vector, and finding the point with the minimum distance, i.e.
Wherein,Wcis marked as the best matching node BMU.
Adjusting the weights to iterate: the weight adjustment is performed using the following learning rule:
where t is the iteration step, α is the learning rate,to optimally match the node BMU (W)c) The conventional gaussian function is chosen here as the topological distance function between the jth neuron and the BMU in the central winning neighborhood.
Iterative training is thus performed, ending the training when converging, and using the trained SOM network as an economic association component health baseline.
Preferably, the method further comprises the steps of S1034, an SOM network test process, i.e. validity verification of the baseline;
by sampling time ranges during stationary operationIs different fromSet of sampling instantsAs the time range of the test process, the input vector complete set corresponding to the input trained SOM network isI.e. the economic relevance component observation vector at the nth sample point
Obtaining an estimation value of the output of the SOM network through a trained SOM network (namely, an economic relevance component health baseline of the SOM network construction)I.e. the economic relevance component estimate vector at the nth sample point
Calculating the health baseline deviation vector of the economic relevance part corresponding to the test data:
when economic relevance part health baseline deviation vector deltatWhen the deviation mean value of each parameter does not exceed 5%, the validity verification of a base line is considered to pass; otherwise a retraining of the baseline is required.
The step S2 of online analysis specifically includes the following steps:
step S201, acquiring related data
And step S2011, monitoring the economic relevant data on line.
In the running process of a ship, a multi-parameter monitoring system is adopted to monitor parameters in real time, when online low-speed diesel engine economy analysis is required, low-speed diesel engine economy related data in a period before the node is input, and a low-speed diesel engine monitoring parameter observation vector X related to economy is obtained, which can be described as follows according to (3):
X(l)＝[s(l),ge(l),x1(l),x2(l),…,xm(l)]T,l′＞L,l＝l′-L,...,l′-1,l′(9)
wherein L' is the current time, and L is the time length.
Step S2012, stable operating condition data is selected.
Obtaining the fuel consumption rate in the smooth operation processWherein,for sampling the time range of the stationary operation process within the input time (i.e. the set of sampling moments of the stationary operation process), and at the same time, optionally taking out the measured values of the rotational speed within the corresponding momentsAnd low-speed diesel engine monitoring parameter observation vectors related to economy at the corresponding moment, namely the ith stable working condition monitoring sampling pointAccording to (2) can be described as:
thereby obtaining the input stable working condition data related to the economical efficiency in the period, wherein the fuel efficiency ge(l) Measuring fuel consumption G by monitoring direct measurements at sampling points during the first smooth rune(l) And effective power Ne(l) Obtained by calculation of formula (1).
Step S202, performing fuel economy analysis based on the low-speed diesel engine fuel economy evaluation baseline, and comprising the following steps of:
step S2021 input data statistics:
two-dimensional observation vector based on fuel economy correlationCalculating the average value statistic of the fuel efficiency aiming at the same rotating speed,thereby obtaining a smooth running process sampling time rangeFuel efficiency statistics related to rotational speedWherein n issThe number of different rotating speed values;
step S2022 baseline data estimation:
will observe the vectorOf different rotational speed values si,i＝1,...,nsBaseline for evaluating fuel economy of low-speed diesel engineNamely a 'rotating speed-fuel consumption rate' evaluation model, and a corresponding fuel economy evaluation baseline estimation value of the low-speed diesel engine is obtained
Step S2023 baseline deviation calculation:
calculating the deviation of the input data corresponding to the fuel economy evaluation baseline of the low-speed diesel engine
Confidence of 0.05 (5% fluctuation above and below the baseline estimate) was used as a standard:
step S2024 judges selection:
if deviation Deltag(si)≤5％,i＝1,...,nsIf the economic state is normal, ending;
if deviation Deltag(si),i＝1,...,nsAny value of (a) is > 5%,then an economic state abnormity early warning is triggered, and the step S203 is carried out to analyze the health state of the economic association part of the low-speed diesel engine based on the decline contribution degree.
Step S203, analyzing the health state of the low-speed diesel engine economy-related component based on the decline contribution degree, and the method comprises the following steps:
optionally, the present embodiment selects an SOM method of "self-organizing map-minimum quantization difference (SOM-MQE)" to calculate the degradation contribution degree of each associated component:
step S2031, selecting the sampling time range in the steady operation processSystem state parameters (including rotation speed, cylinder system, cooling system, supercharger system and other state monitoring parameters, and kept consistent with input vector of off-line training) of all relevant parts related to economy form economy relevant part observation vector of the first sampling pointHealth baseline input vector of economic association part with complete set as SOM network construction
Step S2032, calculating each parameter in the observation vector to the nearest best matching node BMU (W) after inputting the SOM networkc) The distance between the two is expressed as the minimum quantization difference (MQE), i.e. the minimum quantization difference of the economy-related component of the ith sample point is:
step S2033, obtaining the decline contribution degrees C (l) of all the components except the rotating speed from the MQE value, namely:
the decay contribution of the whole components isInner mean value obtained, recorded as
Step S204: and positioning the problem component, and performing maintenance and troubleshooting.
Namely: according to the degree of contribution of decay of each componentDrawing a decline contribution degree distribution graph of each part, sequencing the decline contribution degrees of each part from high to low, and sequentially giving a sequence for carrying out maintenance and investigation of the parts according to the high and low sequence of the decline contribution degrees of each part, thereby giving an economic analysis guidance suggestion.
FIG. 4 is a flowchart illustrating the detailed steps of the present embodiment.
In this embodiment:
(1) economic assessment and impact factor analysis were performed from two perspectives: firstly, from a direct angle, the fuel consumption rate (namely fuel efficiency) is one of important indexes for judging the performance of the diesel engine, and is directly related to the economy, emission index and reliability of the diesel engine, namely the fuel consumption rate in the actual running process of a ship is calculated through the fuel consumption and running state of a host, and the fuel economy of the host is directly evaluated; and secondly, from an indirect angle, finding all the related components related to the economy of the host, calculating the contribution degree of the decline of each economy related component to the economy abnormity of the host by a data driving means, and quantitatively obtaining corresponding economy influence factors and influence degrees, thereby having practical significance for effectively guiding the maintenance of the host, recovering the economic performance of the host, eliminating potential safety hazards or abnormity/fault and reducing the fuel cost.
(2) By analyzing and monitoring multi-parameter data related to economy, economy abnormity is timely and accurately found, and corresponding economy influence factors and influence degrees are quantitatively analyzed, so that the maintenance of the host is effectively guided, the economic performance of the host is recovered, potential safety hazards or abnormity/faults are eliminated, and the fuel cost is reduced.
Example 2
The method is adopted to carry out real ship monitoring on a host (brand is MAN, model is 5S60ME) which is put into a certain ten-thousand-ton bulk cargo ship for use, and the economic related monitoring data of the diesel engine comprise typical parameters such as operation state, fuel consumption (direct economic parameters), a host cylinder system (including cylinder exhaust temperature, cylinder piston cooling oil outlet temperature, cylinder jacket cooling water outlet temperature, cylinder scavenging box fire temperature), a supercharger system (supercharger rotating speed), an air cooler system (scavenging temperature difference between the front and the rear of an air cooler) and the like.
A first part: off-line training results
According to the off-line training procedure of the embodiment 1, the economic relevant data of the faultless smooth operation in the ships 2016/8/15-2016/9/15, including the data of the rotating speed, the fuel efficiency, the No.1-5 cylinder exhaust temperature, the No.1-5 cylinder piston cooling oil outlet temperature, the No.1-5 cylinder scavenging box fire temperature, the No.1-5 cylinder jacket cooling water outlet temperature, the scavenging temperature difference between the front and the rear of the air cooler and the supercharger rotating speed, and the main engine fuel economy evaluation baseline and the economic relevant component health baseline are obtained; the baseline graphs of the model are shown in FIGS. 5-6.
The verification error of the trained SOM network does not exceed 5%, and the requirement of health baseline verification of the economical related parts can be met for online analysis.
A second part: on-line analysis of results
The economic-related data for smooth operation within the vessels 2016/9/16-2016/10/15 was entered according to the example 1 on-line analysis process.
The results shown in fig. 7 are obtained by analyzing the fuel economy, the fuel economy baseline deviation exceeding 5% is obtained, the economy abnormity early warning is triggered, the health state analysis is carried out, the degradation contribution degree distribution condition and the sequencing result are obtained by calculation, the degradation degree of each relevant component parameter is shown in fig. 8, the degradation degree of each relevant component parameter is sequentially displayed from top to bottom, the corresponding degradation degree of each relevant component can be represented, and the fault component can be effectively positioned for guiding the specific maintenance and investigation sequence.
Therefore, maintenance and inspection suggestions are given according to the sequence, namely, the air cooler, the cylinder piston, the supercharger, the scavenging box, the cylinder sleeve and the cylinder exhaust temperature of the main engine are inspected in sequence.
The present example well verifies the validity of the method.
Example 3
The embodiment provides a host computer economy analysis system, which comprises an offline training module and an online analysis module;
the data selection module is configured to select stable working condition data in the historical health samples as input data of offline training;
a baseline construction module configured to perform the construction of a host economic assessment baseline and an economic-related component health baseline;
an online data acquisition module configured to perform online acquisition of economic-related monitoring data;
an economic status analysis module configured to perform analysis of the host economic status using the constructed host economic assessment baseline;
the judgment selection module is configured to execute the selection to be finished when the economy state is judged to be normal; when the economic state is judged to be abnormal, analyzing the health state of the host computer related component;
the system comprises a main engine related component health state analysis module, a main engine related component health state analysis module and a main engine economic related component health state analysis module, wherein the main engine related component health state analysis module is configured to obtain the decline contribution degree of the main engine economic related component by utilizing an SOM method based on an economic related component health baseline;
and the maintenance sequence determining module is configured to execute sequencing on the decline contribution degrees to obtain a health state result of each component, and obtain the sequence of maintenance and investigation of each component according to the result.
A host associated component health state analysis module configured to perform the specific steps of:
selecting a smooth running process sampling time rangeSystem state parameters of all relevant parts related to economy form an economy relevant part observation vector of the l sampling pointHealth baseline input vector of economic association part with complete set as SOM network constructions represents the rotational speed, x1,x2,…,xmRepresenting state parameters of m associated component systems related to economy; l represents a sampling point;
after the SOM network is input, each parameter in the observation vector is calculated to the nearest best matching node WcThe distance value between them, noted as the minimum quantization difference MQE; the minimum quantization difference of the economy-related components for the ith sample point is:
obtaining the recession contribution degrees C (l) of all parts except the rotating speed from the MQE value,
is composed of at least oneInner mean value obtaining decay contribution of integral parts
Since the system of the present invention has the same principle as the method, the system also has the technical effects corresponding to the analysis method, and the related parts can be referred to each other, and the detailed description of the embodiment is not repeated.
Those skilled in the art will appreciate that all or part of the flow of the method implementing the above embodiments may be implemented by a computer program, which is stored in a computer readable storage medium, to instruct related hardware. The computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention.
Claims (10)
1. A ship host computer associated component state monitoring method based on decline contribution degree is characterized by comprising the following steps:
selecting stable working condition data in a historical health sample as input data of offline training, constructing a host economic evaluation baseline according to a host rotating speed and a fuel consumption rate measured value in the input data, and constructing an economic association part health baseline by using an SOM (sequence of model) method according to the host rotating speed and economic association part state parameters;
adopting a multi-parameter monitoring system to monitor data, utilizing the established host economy evaluation baseline to analyze the economy state of the host for the monitored data, and ending when the economy state is normal; when the economic state is abnormal, an abnormal alarm is sent out, and the health state of the host computer related component is analyzed; the host associated component health status analysis comprising: analyzing the monitoring data by using the health baseline of the economic relevance component to obtain the degradation contribution degree of the host economic relevance component degradation to the host economic abnormity;
and sequencing the degradation contribution degrees, positioning the components in the front sequence as abnormal components, and performing maintenance and investigation on the corresponding components.
2. The method of claim 1, wherein the economy-related components include a cylinder system, a cooling system, and a supercharger system.
3. The method of claim 2, wherein the monitored data is data consistent with an input data type of offline training, including a host speed, a fuel consumption amount, a cylinder system state parameter, a cooling system state parameter, and a supercharger system state parameter.
4. The method of claim 3, wherein constructing an economic-related component health baseline comprises the steps of:
selecting system state parameters of each related component related to economy, and constructing an economy related component observation vector of the nth sampleUsing corpus as input vector of networkWherein,for vessels in the voyageSampling time range in steady operation processThe sampling time set used for the SOM network training, s (n) is the rotation speed, x1(n),x2(n),…,xm(n) state parameters of m associated component systems relating to economy;
a jth weight vector W based on the input vectorj(n) obtaining a corpus weight vector Wj(ii) a Wherein,
k is the number of neurons trained by the SOM network;
calculating the Euclidean distance between the input vector and the weight vector, and finding out the point with the minimum distance as the best matching node Wc；
By usingAdjusting the weight to obtain an adjusted corpus weight vector, wherein t is an iteration step, α is a learning rate,to optimally match the node WcCenter winning neighbor with jth neuron and best matched node WcA topological distance function between;
and carrying out iterative training of the SOM network, finishing the training when convergence occurs, and taking the trained SOM network as a health baseline of the economic association part.
5. The method of claim 4, further comprising the step of testing the trained SOM network, and if the test fails, retraining until the test is passed.
6. The method of analyzing mainframe economy of claim 5 wherein the step of testing the trained SOM network comprises:
by sampling time ranges during stationary operationIs different fromSet of sampling instantsAn economic relevance component observation vector at the nth sample point as a test process time horizon
The full set of input vectors corresponding to the input trained SOM network is
Obtaining an estimate of the SOM network output by a trained SOM networkThe economic relevance component at the nth sample point estimates the vector:
<mrow> <msub> <mover> <mi>X</mi> <mo>^</mo> </mover> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mover> <mi>s</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mn>2</mn> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>,</mo> <mo>...</mo> <mo>,</mo> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>n</mi> <mo>&Element;</mo> <msub> <mover> <mi>T</mi> <mo>&OverBar;</mo> </mover> <mrow> <mi>t</mi> <mi>e</mi> <mi>s</mi> <mi>t</mi> </mrow> </msub> <mo>;</mo> </mrow>
calculating the health baseline deviation vector of the economic relevance part corresponding to the test data:
<mrow> <msub> <mi>&Delta;</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <mo>|</mo> <mfrac> <mrow> <mi>s</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>-</mo> <mover> <mi>s</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mover> <mi>s</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>|</mo> <mo>&times;</mo> <mn>100</mn> <mi>%</mi> <mo>,</mo> <mo>|</mo> <mfrac> <mrow> <msub> <mi>x</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>|</mo> <mo>&times;</mo> <mn>100</mn> <mi>%</mi> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mo>|</mo> <mfrac> <mrow> <msub> <mi>x</mi> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>|</mo> <mo>&times;</mo> <mn>100</mn> <mi>%</mi> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>n</mi> <mo>&Element;</mo> <msub> <mover> <mi>T</mi> <mo>&OverBar;</mo> </mover> <mrow> <mi>t</mi> <mi>e</mi> <mi>s</mi> <mi>t</mi> </mrow> </msub> </mrow>
when economic relevance part health baseline deviation vector deltatAnd when the deviation mean value of each parameter does not exceed the threshold value, the test is considered to be passed.
7. The method according to one of claims 1 to 6,
the method for analyzing the host economic state of the online monitoring data by using the constructed host economic evaluation baseline comprises the following steps of:
two-dimensional observation vector based on fuel economy correlationCalculating the mean statistic of fuel efficiency for the same rotation speed, thereby obtaining the sampling time range in the smooth operation processFuel efficiency statistics related to rotational speedWherein n issThe number of different rotating speed values;
will observe the vectorOf different rotational speed values si,i＝1,...,nsInputting a host fuel economy evaluation baseline to obtain a corresponding host fuel economy evaluation baseline estimation value
Calculating the deviation of the input data corresponding to the evaluation baseline of the fuel economy of the host
If all the deviations are smaller than the threshold value, the economic state is normal, and the process is finished;
and if any value in the deviation is larger than the threshold value, the economic state is abnormal, and the health state analysis of the host computer related component is carried out.
8. The method for analyzing the host economy of claim 7, wherein the online monitoring data is analyzed by using the health baseline of the economy-related component to obtain the degradation contribution degree of the degradation of the economy-related component of the host to the economic abnormity of the host, and the method comprises the following steps:
selecting a smooth running process sampling time rangeSystem state parameters of all relevant parts related to economy form an economy relevant part observation vector of the l sampling pointHealth baseline input vector of economic association part with complete set as SOM network construction
Calculating each parameter in observation vector to the nearest after inputting SOM networkBest matching node W ofcThe distance value between them, noted as the minimum quantization difference MQE; the minimum quantization difference of the economy-related components for the ith sample point is:
<mfenced open = "" close = ""> <mtable> <mtr> <mtd> <mrow> <mi>M</mi> <mi>Q</mi> <mi>E</mi> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <msub> <mi>MQE</mi> <mi>s</mi> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>1</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>2</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mi>m</mi> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <mo>|</mo> <mi>s</mi> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>s</mi> </mrow> </msub> <mo>|</mo> <mo>,</mo> <mo>|</mo> <msub> <mi>x</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>1</mn> </msub> </mrow> </msub> <mo>|</mo> <mo>,</mo> <mo>|</mo> <msub> <mi>x</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>2</mn> </msub> </mrow> </msub> <mo>|</mo> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mo>|</mo> <msub> <mi>x</mi> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mi>m</mi> </msub> </mrow> </msub> <mo>|</mo> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> </mtable> </mfenced>
obtaining the recession contribution degrees C (l) of all parts except the rotating speed from the MQE value,
<mfenced open = "" close = ""> <mtable> <mtr> <mtd> <mrow> <mi>C</mi> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <msub> <mi>C</mi> <msub> <mi>x</mi> <mn>1</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>C</mi> <msub> <mi>x</mi> <mn>2</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>C</mi> <msub> <mi>x</mi> <mi>m</mi> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>1</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>/</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>1</mn> </msub> </mrow> </msub> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>2</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>/</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>2</mn> </msub> </mrow> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mi>m</mi> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>/</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mi>m</mi> </msub> </mrow> </msub> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> </mtable> </mfenced>
is composed of at least oneInner mean value obtaining decay contribution of integral parts
9. A ship host computer associated component state monitoring system based on decline contribution degree is characterized by comprising:
the data selection module is configured to select stable working condition data in the historical health samples as input data of offline training;
a baseline construction module configured to perform the construction of a host economic assessment baseline and an economic-related component health baseline;
an online data acquisition module configured to perform online acquisition of economic-related monitoring data;
an economic status analysis module configured to perform analysis of the host economic status using the constructed host economic assessment baseline;
the judgment selection module is configured to execute the selection to be finished when the economy state is judged to be normal; when the economic state is judged to be abnormal, analyzing the health state of the host computer related component;
the system comprises a main engine related component health state analysis module, a main engine related component health state analysis module and a main engine economic related component health state analysis module, wherein the main engine related component health state analysis module is configured to obtain the decline contribution degree of the main engine economic related component by utilizing an SOM method based on an economic related component health baseline;
and the maintenance sequence determining module is configured to execute sequencing on the decline contribution degrees to obtain a health state result of each component, and obtain the sequence of maintenance and investigation of each component according to the result.
10. The system of claim 9, wherein: a host associated component health state analysis module configured to perform the specific steps of:
selecting a smooth running process sampling time rangeSystem state parameters of all relevant parts related to economy form an economy relevant part observation vector of the l sampling pointHealth baseline input vector of economic association part with complete set as SOM network constructions represents the rotational speed, x1,x2,…,xmRepresenting state parameters of m associated component systems related to economy; l represents a sampling point;
after the SOM network is input, each parameter in the observation vector is calculated to the nearest best matching node WcThe distance value between them, noted as the minimum quantization difference MQE; the minimum quantization difference of the economy-related components for the ith sample point is:
<mfenced open = "" close = ""> <mtable> <mtr> <mtd> <mrow> <mi>M</mi> <mi>Q</mi> <mi>E</mi> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <msub> <mi>MQE</mi> <mi>s</mi> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>1</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>2</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mi>m</mi> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <mo>|</mo> <mi>s</mi> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>s</mi> </mrow> </msub> <mo>|</mo> <mo>,</mo> <mo>|</mo> <msub> <mi>x</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>1</mn> </msub> </mrow> </msub> <mo>|</mo> <mo>,</mo> <mo>|</mo> <msub> <mi>x</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>2</mn> </msub> </mrow> </msub> <mo>|</mo> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mo>|</mo> <msub> <mi>x</mi> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mi>m</mi> </msub> </mrow> </msub> <mo>|</mo> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> </mtable> </mfenced>
obtaining the recession contribution degrees C (l) of all parts except the rotating speed from the MQE value,
<mfenced open = "" close = ""> <mtable> <mtr> <mtd> <mrow> <mi>C</mi> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <msub> <mi>C</mi> <msub> <mi>x</mi> <mn>1</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>C</mi> <msub> <mi>x</mi> <mn>2</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>C</mi> <msub> <mi>x</mi> <mi>m</mi> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>=</mo> <msup> <mrow> <mo>&lsqb;</mo> <mrow> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>1</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>/</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>1</mn> </msub> </mrow> </msub> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mn>2</mn> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>/</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mn>2</mn> </msub> </mrow> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>MQE</mi> <msub> <mi>x</mi> <mi>m</mi> </msub> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>/</mo> <msub> <mi>W</mi> <mrow> <mi>c</mi> <mo>,</mo> <msub> <mi>x</mi> <mi>m</mi> </msub> </mrow> </msub> </mrow> <mo>&rsqb;</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> <mi>l</mi> <mo>&Element;</mo> <mover> <mi>L</mi> <mo>&OverBar;</mo> </mover> </mrow> </mtd> </mtr> </mtable> </mfenced>
is composed of at least oneInner mean value obtaining decay contribution of integral parts
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711235510.7A CN108062586B (en) | 2017-11-30 | 2017-11-30 | Ship host machine associated component state monitoring method and system based on decline contribution degree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711235510.7A CN108062586B (en) | 2017-11-30 | 2017-11-30 | Ship host machine associated component state monitoring method and system based on decline contribution degree |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108062586A true CN108062586A (en) | 2018-05-22 |
| CN108062586B CN108062586B (en) | 2020-03-27 |
Family
ID=62135911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711235510.7A Active CN108062586B (en) | 2017-11-30 | 2017-11-30 | Ship host machine associated component state monitoring method and system based on decline contribution degree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108062586B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109255201A (en) * | 2018-10-24 | 2019-01-22 | 哈工大机器人（山东）智能装备研究院 | A kind of ball screw assembly, health evaluating method based on SOM-MQE |
| CN110502390A (en) * | 2019-07-08 | 2019-11-26 | 中国地质大学（武汉） | A kind of colleges and universities' cloud computing center automation operation management system |
| CN110696990A (en) * | 2019-11-11 | 2020-01-17 | 中国船舶工业系统工程研究院 | Ship generator component influence identification method and system based on data driving |
| CN111190349A (en) * | 2019-12-30 | 2020-05-22 | 中国船舶重工集团公司第七一一研究所 | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment |
| CN112378651A (en) * | 2020-12-08 | 2021-02-19 | 中国船舶工业系统工程研究院 | Data-driven-based equipment dynamic reliability assessment method |
| CN113379223A (en) * | 2021-06-04 | 2021-09-10 | 江苏科技大学 | Ship-borne spare part multi-level configuration method for ship main engine based on fault correlation model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880170B (en) * | 2012-10-08 | 2015-03-25 | 南京航空航天大学 | System failure early warning method based on baseline model and Bayesian factor |
| CN105737922A (en) * | 2016-01-26 | 2016-07-06 | 中国船舶工业系统工程研究院 | Method and device for early warning on fuel consumption rate of low-speed diesel engine of ship |
| CN106368816A (en) * | 2016-10-27 | 2017-02-01 | 中国船舶工业系统工程研究院 | Method for online abnormity detection of low-speed diesel engine of ship based on baseline deviation |
| CN106777554A (en) * | 2016-11-29 | 2017-05-31 | 哈尔滨工业大学（威海） | Aerial engine air passage cell cube health status evaluation method based on state baseline |
-
2017
- 2017-11-30 CN CN201711235510.7A patent/CN108062586B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880170B (en) * | 2012-10-08 | 2015-03-25 | 南京航空航天大学 | System failure early warning method based on baseline model and Bayesian factor |
| CN105737922A (en) * | 2016-01-26 | 2016-07-06 | 中国船舶工业系统工程研究院 | Method and device for early warning on fuel consumption rate of low-speed diesel engine of ship |
| CN106368816A (en) * | 2016-10-27 | 2017-02-01 | 中国船舶工业系统工程研究院 | Method for online abnormity detection of low-speed diesel engine of ship based on baseline deviation |
| CN106777554A (en) * | 2016-11-29 | 2017-05-31 | 哈尔滨工业大学（威海） | Aerial engine air passage cell cube health status evaluation method based on state baseline |
Non-Patent Citations (2)
| Title |
|---|
| HOUMAN HANACHI等: "A Physics-Based Modeling Approach for Performance Monitoring in Gas Turbine Engines", 《IEEE TRANSACTIONS ON RELIABILITY》 * |
| 马剑等: "船舶主推进系统故障预测与健康管理设计", 《南京航空航天大学学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109255201A (en) * | 2018-10-24 | 2019-01-22 | 哈工大机器人（山东）智能装备研究院 | A kind of ball screw assembly, health evaluating method based on SOM-MQE |
| CN109255201B (en) * | 2018-10-24 | 2023-07-14 | 哈工大机器人（山东）智能装备研究院 | SOM-MQE-based ball screw pair health assessment method |
| CN110502390A (en) * | 2019-07-08 | 2019-11-26 | 中国地质大学（武汉） | A kind of colleges and universities' cloud computing center automation operation management system |
| CN110502390B (en) * | 2019-07-08 | 2021-06-01 | 中国地质大学（武汉） | Automatic operation and maintenance management system of colleges and universities cloud computing center |
| CN110696990A (en) * | 2019-11-11 | 2020-01-17 | 中国船舶工业系统工程研究院 | Ship generator component influence identification method and system based on data driving |
| CN111190349A (en) * | 2019-12-30 | 2020-05-22 | 中国船舶重工集团公司第七一一研究所 | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment |
| CN112378651A (en) * | 2020-12-08 | 2021-02-19 | 中国船舶工业系统工程研究院 | Data-driven-based equipment dynamic reliability assessment method |
| CN113379223A (en) * | 2021-06-04 | 2021-09-10 | 江苏科技大学 | Ship-borne spare part multi-level configuration method for ship main engine based on fault correlation model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108062586B (en) | 2020-03-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108062586B (en) | Ship host machine associated component state monitoring method and system based on decline contribution degree | |
| CN108062618B (en) | Low-speed diesel engine economy analysis method and system based on double baselines | |
| Cipollini et al. | Condition-based maintenance of naval propulsion systems with supervised data analysis | |
| Cipollini et al. | Condition-based maintenance of naval propulsion systems: Data analysis with minimal feedback | |
| CN106368816B (en) | A kind of online method for detecting abnormality of marine low speed diesel engine based on baseline offset | |
| Basurko et al. | Condition-based maintenance for medium speed diesel engines used in vessels in operation | |
| Wang et al. | Research on the fault monitoring method of marine diesel engines based on the manifold learning and isolation forest | |
| Coraddu et al. | A novelty detection approach to diagnosing hull and propeller fouling | |
| Ellefsen et al. | Online fault detection in autonomous ferries: Using fault-type independent spectral anomaly detection | |
| KR20210124229A (en) | Methods and systems for reducing marine fuel consumption | |
| CN113847950A (en) | Intelligent ship equipment state monitoring system based on cloud computing and information interaction method | |
| Coraddu et al. | Vessels fuel consumption: A data analytics perspective to sustainability | |
| Bui et al. | Advanced data analytics for ship performance monitoring under localized operational conditions | |
| Vorkapić et al. | Machine learning methods in monitoring operating behaviour of marine two-stroke diesel engine | |
| CN115659263A (en) | Ship control behavior risk assessment system and assessment method based on big data | |
| Karatuğ et al. | Design of a decision support system to achieve condition-based maintenance in ship machinery systems | |
| CN115392782A (en) | Method and system for monitoring and diagnosing health state of process system of nuclear power plant | |
| Zeng et al. | A data-driven intelligent energy efficiency management system for ships | |
| Fan et al. | Comprehensive evaluation of machine learning models for predicting ship energy consumption based on onboard sensor data | |
| Gupta et al. | Big data analytics as a tool to monitor hydrodynamic performance of a ship | |
| JP6846896B2 (en) | Analysis of ship propulsion performance | |
| Feng et al. | Comparison of SOM and PCA-SOM in fault diagnosis of ground-testing bed | |
| Golovan et al. | System of Water Vehicle Power Plant Remote Condition Monitoring | |
| Wei et al. | Multi-sensor monitoring based on-line diesel engine anomaly detection with baseline deviation | |
| Arias Chao | Combining deep learning and physics-based performance models for diagnostics and prognostics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |