CN112083355A - Ship cabin equipment health management and fault prediction system and method - Google Patents
Ship cabin equipment health management and fault prediction system and method Download PDFInfo
- Publication number
- CN112083355A CN112083355A CN202010938464.2A CN202010938464A CN112083355A CN 112083355 A CN112083355 A CN 112083355A CN 202010938464 A CN202010938464 A CN 202010938464A CN 112083355 A CN112083355 A CN 112083355A
- Authority
- CN
- China
- Prior art keywords
- fault
- equipment
- maintenance
- red
- sensors
- 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
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/56—Testing of electric apparatus
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/005—Testing of complete machines, e.g. washing-machines or mobile phones
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The utility model provides a boats and ships cabin equipment health management and failure prediction system, including control management equipment, communication equipment, data collection equipment and on-the-spot collection equipment, control management equipment includes the host computer, communication equipment includes the switch, data collection equipment includes data acquisition card, on-the-spot collection equipment includes a plurality of temperature sensor that are used for installing on boats and ships cabin equipment, voltage sensor, current sensor, pressure sensor and flow sensor, temperature sensor, voltage sensor, current sensor, pressure sensor and flow sensor all are connected with data acquisition card, data acquisition card passes through the switch and is connected with the host computer. The system can realize the capabilities of fault diagnosis, predictive maintenance, knowledge base management and the like of the typical equipment of the ship cabin, thereby distinguishing the resource waste caused by manual regular maintenance, reducing the maintenance cost, improving the automation and intelligent level of the operation and maintenance of the cabin, and further improving the operation reliability and the safety of the equipment of the ship cabin.
Description
Technical Field
The invention relates to the technical field of ship cabin equipment management, in particular to a ship cabin equipment health management and fault prediction system, and particularly relates to a fault prediction method of the ship cabin equipment health management and fault prediction system.
Background
The ship cabin equipment health management technology mainly takes ship cabin equipment as a research object to analyze the reliability of the cabin equipment, and utilizes the collected and processed data to perform data modeling on cabin auxiliary equipment, so that the health management functions of fault diagnosis, fault prediction, predictive maintenance and the like of the equipment are realized, the running reliability of the cabin auxiliary equipment is improved, and the stable running of the auxiliary equipment is guaranteed. At present, the research and development of the health management technology of the ship cabin equipment are laggard, and the development requirements of intelligent ships cannot be met; research on a ship cabin equipment health management system and a fault prediction method is needed to meet the requirements for improving the overall level and competitiveness of ocean engineering and intelligent ships, improve the maintenance level and use efficiency of ship electromechanical equipment, form a long-term effective data base, and improve the quality and service capacity of ship core electromechanical equipment.
For the research on the health management of the auxiliary engine of the cabin of the ship, the cabin monitoring and alarming system is always in the leading position abroad, is a multifunctional management system based on a network platform, has the characteristics of intellectualization, digitalization, networking, integration and the like, and can realize the functions of fault diagnosis, real-time data display, equipment state monitoring, fault alarming and the like. Ship electrical product manufacturers such as SIEMENS (SIEMENS) in germany, ABB group in switzerland, Kongsberg in norway, schneider electrical, JRCS in japan, etc. have technically mature system products. China has accumulated certain research on equipment level and subsystem level of intelligent ships, and the research on the system level is gradually developed. However, the automation degree of ship equipment is not uniform, most of equipment still stays in simple control or only realizes state monitoring, intelligent decision and self-adaptive intelligent health equipment is lacked, and the problems that a large amount of equipment lacks effective and uniform information comprehensive analysis and processing technology and the like exist; although some devices can realize monitoring functions, only intelligent monitoring, fault isolation and remote servo closed-loop control can be realized, the learning capability and the self-adaptive capability of the devices are lacked, and the health management functions of cabin equipment such as fault prediction, fault self-repair and the like are difficult to realize.
Disclosure of Invention
The invention aims to solve the technical problem of providing a ship cabin equipment health management and fault prediction system which can realize the capabilities of fault diagnosis, predictive maintenance, knowledge base management and the like of typical equipment of a ship cabin, distinguishes resource waste caused by manual regular maintenance, reduces maintenance cost, improves the automation and intelligence level of cabin operation and maintenance, and further improves the operation reliability and safety of the ship cabin equipment.
Another technical problem to be solved by the present invention is to provide a method for health management and fault prediction of the above ship cabin equipment health management and fault prediction system.
The technical problem to be solved by the present invention is achieved by the following technical means. The invention relates to a health management and fault prediction system for ship cabin equipment, which comprises control management equipment, communication equipment, data collection equipment and field collection equipment, wherein the control management equipment comprises an upper computer, the communication equipment comprises a switch, the data collection equipment comprises a data collection card, the field collection equipment comprises a plurality of temperature sensors, voltage sensors, current sensors, pressure sensors and flow sensors which are used for being installed on the ship cabin equipment, the temperature sensors, the voltage sensors, the current sensors, the pressure sensors and the flow sensors are all connected with the data collection card, and the data collection card is connected with the upper computer through the switch.
The technical problem to be solved by the invention can be further realized by the following technical scheme that for the ship cabin equipment health management and fault prediction system, the upper computer is a computer or a programmable controller.
The technical problem to be solved by the present invention can be further solved by the following technical solution, wherein for the above-mentioned ship cabin equipment health management and fault prediction system, the temperature sensor is selected from one or more of NTC thermistor, platinum RTD, thermocouple and semiconductor.
The technical problem to be solved by the invention can be further realized by the following technical scheme that for the ship cabin equipment health management and fault prediction system, a ship cabin equipment health management and fault prediction method comprises
(1) Equipment reliability and failure mode analysis:
by using a fault tree method, reliability analysis and equipment failure mode analysis are carried out on ship cabin equipment, the fault characteristics of the equipment are obtained, and reasonable maintenance strategies are formulated;
(2) signal acquisition and preprocessing:
the method comprises the steps that various sensors are arranged to communicate with equipment, various information of the running state of the equipment is collected, the information is preprocessed to obtain parameters capable of reflecting the pressure, the temperature and the flow of the running state of the equipment, the running state data of the equipment is obtained, and data support is provided for real-time state monitoring, fault diagnosis and alarming and early warning;
(3) online fault diagnosis
Acquiring real-time monitoring equipment working state data on line, and performing real-time online diagnosis on a fault by using a fault model and an expert system in combination with a reserved fault mode and a failure criterion in the system; when the abnormality exists, the system fault is diagnosed on line by combining various health historical state data, working states and maintenance historical records of the equipment and integrating an intelligent diagnosis method, so that codes, characteristics, reasons for generation and positions of the faults are obtained, and timely and accurate data support is provided for maintenance;
(4) fault prediction
The method comprises the steps of evaluating the current performance state of a fault and forecasting the early stage of the fault by utilizing predicted fault characteristic parameters based on a fault prediction model of a state set sequence, and effectively realizing early stage identification of equipment faults by improving a data mining algorithm so as to effectively reduce economic and personal losses caused by faults which are high in concealment and difficult to find;
(5) predictive maintenance
According to the online monitoring information and the judgment result of the health state, the early detection of abnormal symptoms of equipment by maintenance personnel is facilitated, and corresponding maintenance suggestions and maintenance instruction manuals are provided;
(6) operation and maintenance knowledge management
The management and the update of the model base information, the knowledge base information and the maintenance optimization information are realized, and the management of the fault name, the solution method and the maintenance suggestion in the operation and maintenance knowledge base is provided by storing the fault type, the fault algorithm model and the maintenance suggestion measure; and secondly, continuously optimizing a fault model and perfecting an operation and maintenance knowledge base.
The technical problem to be solved by the invention can be further realized by the following technical scheme, and for the ship cabin equipment health management and fault prediction method, the method adopts a particle calculation processing mode, effectively solves the complex problem by selecting the most appropriate particle layer, processes the mining of fuzzy information and mass data, deletes redundant knowledge and unnecessary attribute characteristics, reduces a data table, reduces the number of fault characteristic attributes, and reduces the scale and complexity of a rule knowledge base, and the specific process is as follows:
reducing the decision table by adopting a relative granularity attribute reduction algorithm, wherein the input of the relative granularity attribute reduction algorithm is that the decision table S is (U, C, U, D, V, f), C is a conditional attribute set, D is a decision attribute set, and the output reduction result is represented by RED; the reduction method comprises the following specific steps:
(1) merging the same rules in the decision table to make RED equal to phi;
(2) for each one of ciC/RED as the element of the attribute, CiImportance sig (c) of reduction result RED to decision set Di，RED，D)＝GD(D|RED)-GD(D|RED∪ci)；
(3) All sigs (c) calculated in step (2)iD, D) selecting the attribute c corresponding to the maximum valueiAs c iskIf a plurality of attributes meet the condition, selecting the first attribute as ck；
(4) If sig (c)kRED, D) > 0, then RED ═ RED ≧ CkAnd continuing to step (2) to circularly calculate; if sig (c)kIf RED, D)' is 0, the loop is ended, and the reduced result RED is output.
The technical problem to be solved by the invention can be further realized by the following technical scheme that for the ship cabin equipment health management and fault prediction method, various sensors are arranged in a ship cabin, fault characteristics are extracted by using various signal processing methods, and a fault model and an expert system starvation method are used for carrying out real-time online diagnosis on the fault, so that a worker can find the fault of the equipment in time and take corresponding maintenance measures.
The technical problem to be solved by the invention can be further realized by the following technical scheme that for the ship cabin equipment health management and fault prediction method, the SDG fault prediction method based on bidirectional reasoning is adopted, firstly, all possible fault sources are searched by utilizing reverse reasoning, and then, forward reasoning is carried out on the possible fault sources in sequence; if a certain fault can completely or maximally explain the states of the abnormal variable nodes, the fault is a prediction result with the highest reliability.
The technical problem to be solved by the invention can be further solved by the following technical scheme that for the ship cabin equipment health management and fault prediction method, the ship cabin equipment for health management and fault prediction comprises a cargo oil pump, a marine diesel engine and a fuel supply unit.
Compared with the prior art, the invention has the advantages and technical effects that:
1. the invention improves the data quality and the performance of model diagnosis by a data preprocessing technology based on a rough set:
the invention is based on rough set theory to preprocess the original data before establishing machine learning model training, and under the condition of ensuring the decision ability to be unchanged, the redundant knowledge and unnecessary attribute characteristics are deleted, the fault model is reduced, the number of fault characteristic attributes is reduced, the scale and complexity of a rule knowledge base are reduced, and the data quality and the diagnostic performance of the model are improved;
2. according to the invention, the fault prediction precision of the ship cabin equipment is improved by a fault prediction technology based on the state set sequence:
in order to realize early discovery and prediction accuracy of the equipment fault of the ship cabin, the invention introduces a fault prediction technology based on a state set sequence, judges the possible future fault by combining the running states, structural characteristics and historical data of the equipment such as vibration signals, pressure, temperature and the like through an established SDG model, and predicts the nature, type, degree, reason and position of the fault. When a fault is predicted, a strategy can be maintained rapidly and accurately according to a prediction result, and a reasonable later-stage operation scheme of the equipment is provided by combining the operation environment of the equipment; meanwhile, relevant information is transmitted to an operator in time; the invention can effectively realize early identification of cabin equipment faults and reduce economic and personal losses caused by faults which are high in concealment and difficult to detect.
Drawings
FIG. 1 is a block diagram schematically illustrating an arrangement of the present invention;
FIG. 2 is a schematic diagram of one embodiment of the present invention;
FIG. 3 is a schematic diagram of fault model training of the present invention;
FIG. 4 is a schematic diagram of an SDG fault prediction model of the present invention;
FIG. 5 is a schematic diagram of operation and maintenance knowledge management according to the present invention;
FIG. 6 is a schematic diagram of the online fault diagnosis of the present invention;
FIG. 7 is a flow chart of SDG fault prediction in accordance with the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-7, a ship cabin equipment health management and fault prediction system, the system comprises a control management device, a communication device, a data collection device and a field collection device, wherein the control management device comprises an upper computer, the communication device comprises a switch, the data collection device comprises a data collection card, the field collection device comprises a plurality of temperature sensors, voltage sensors, current sensors, pressure sensors and flow sensors, the temperature sensors, the voltage sensors, the current sensors, the pressure sensors and the flow sensors are used for being installed on the ship cabin equipment, and the data collection card is connected with the upper computer through the switch. According to the requirements of a ship cabin health system, the research depth of a ship information system architecture is combined, the ship cabin equipment health management system information system architecture is researched aiming at the problems of multiple equipment types, large acquired data quantity and the like of a ship cabin, and the research is mainly carried out from a field layer, a data layer, a management layer and a service layer; data acquisition research on ship cabin equipment, such as diesel engines, cargo oil pumps, fuel supply units and the like, is carried out on a field layer; a centralized management and control platform is arranged to realize monitoring, fault alarming, prediction and control of the operation state of cabin equipment; the field acquisition and control modules are distributed on the field according to the arrangement positions of the cabin equipment, and can also be used as regional monitoring relay equipment to solve the problem of large signal attenuation caused by too long distance between the management and control equipment and a regional monitoring box; in addition, a certain number of temperature, pressure and flow sensors are arranged in the pipeline and the cabin, and the running state of the system is measured in real time. The health management of the ship cabin equipment mainly starts research from cabin equipment state acquisition, data acquisition is carried out on the equipment state through a sensor, and the control equipment controls the equipment at the bottom layer through a communication network; the research of data sharing and transmission is carried out, and data information is classified and stored by a multi-database integration technology; health management research such as top-level data analysis, decision, maintenance and the like is carried out, and guarantee is provided for normal operation of cabin equipment; meanwhile, the research of visual management and control is developed, and fault information, monitoring data of equipment, health management conditions and the like are displayed on a display platform in a centralized mode.
The upper computer is a computer or a programmable controller. The computer is an industrial PC computer with wide temperature range, no fan and no hard disk, and has the characteristics of long service life, strong anti-interference capability and the like; the design is miniaturized, the installation is flexible, and the space is saved; all units are mutually independent, the operation reliability is ensured, and the maintenance is simple; the damping design enables the work to be more stable; the expansion groove is provided with a plurality of expansion grooves, so that the expansion capability is stronger; the programmable controller is used as a main body of the signal acquisition box, integrates control, information processing and communication functions into a single controller system, and provides a double-controller architecture to meet different tasks. The Ethernet and CAN bus interfaces with double interfaces are provided, high-density I/O acquisition interfaces are provided through cascade connection or daisy chain connection, functions such as analog input, analog output, digital input, digital output, thermocouple input, thermal resistance input and the like CAN be realized, and the requirements of information processing and communication are met.
The temperature sensor is selected from one or more of NTC thermistor, platinum RTD, thermocouple and semiconductor. When the hardware of the ship cabin typical equipment health system is designed in a model selection mode, a proper sensor needs to be selected according to the application requirement, and the cost is reduced to the maximum extent under the condition that the performance, the accuracy or the reliability are not influenced.
In particular, the primary consideration in selecting a temperature sensor is the temperature range. For example, for operating environments in excess of 1000 degrees celsius, thermocouples are often the only option; however, only a few applications involve such extreme temperatures. For most industrial, medical, automotive, consumer and general embedded systems, the typical operating temperature range is much narrower; when semiconductor-based components are used, the range is even more limited; for example, MCU used for commercial and consumer applications has a nominal temperature of 0 ℃ to 85 ℃; MCU for industrial applications can extend the range to-40 ℃ to 100 ℃, whereas automotive MCU needs to work in the temperature range of-40 ℃ to 125 ℃; thus, engineers may generally choose to use any standard type of temperature sensor.
Secondly, the shape and installation problems of the temperature sensor need to be considered; the temperature sensing assembly requires different packaging depending on the application configuration being measured; for example, semiconductor-based sensors cannot be immersed directly in hot oil; a low cost sensor may be an option for epoxy coated packaging. For higher temperature operation, the temperature sensor may be sealed in glass. This protects its temperature probe from other environmental factors, including liquids and debris. The sensor may also be placed in a stainless steel housing to improve stability; the higher the complexity required of the housing, the higher the cost of the sensor.
Sensors also come in a variety of shapes and sizes; selecting an appropriate sensor for an application may improve performance, responsiveness, and reliability; for example, all temperature sensors are subject to self-heating due to power passing through them; this self-heating can increase the ambient temperature around the sensor, thereby introducing errors and negatively impacting accuracy.
By using the NTC thermistor, the mass of the sensor can be increased so as to reduce errors caused by self-heating; even small-scale variations can have a large effect on reducing self-heating; for example, the volume/mass of the 3x3x3mm thermistor is greater than 3 times compared to the 2x2x2mm thermistor; only thermistors can achieve this flexibility. Semiconductor-based sensors are inherently fixed; since both RTDs and thermocouples are wire based, this limits the ability of engineers to adjust the mass to reduce self-heating errors.
A ship cabin equipment health management and fault prediction method can realize the capabilities of fault diagnosis, predictive maintenance, knowledge base management and the like of typical equipment of a ship cabin, and further distinguishes the resource waste caused by manual regular maintenance, reduces the maintenance cost, improves the automation and intelligent level of cabin operation and maintenance, and further improves the operation reliability and safety of the ship cabin equipment; the method mainly takes three types of core cabin equipment as research objects, wherein the three types of core cabin equipment comprise a cargo oil pump, a marine diesel engine and a fuel supply unit, and the specific contents are as follows:
(1) device reliability and failure mode analysis
By utilizing a fault tree method, reliability analysis and equipment failure mode analysis are carried out on three types of core cabin equipment (a cargo oil pump, a marine diesel engine and a fuel supply unit), the fault characteristics of the equipment are obtained, and reasonable maintenance strategies are formulated;
(2) signal acquisition and preprocessing
The method comprises the steps that various sensors are arranged to communicate with equipment, various information of the running state of the equipment is collected, the information is preprocessed to obtain parameters such as pressure, temperature and flow which can reflect the running state of the equipment, the running state data of the equipment is obtained, and data support is provided for real-time state monitoring, fault diagnosis and alarm and early warning;
(3) online fault diagnosis
And (3) acquiring real-time monitoring equipment working state data, and performing real-time online diagnosis on the fault by using a fault model and an expert system in combination with a reserved fault mode and a reserved failure criterion in the system. When the abnormality exists, the system fault is diagnosed online by combining information such as various health historical state data, working states, maintenance historical records and the like of the equipment and integrating an intelligent diagnosis method, so that codes, characteristics, reasons for generation and positions of the faults are obtained, and timely and accurate data support is provided for maintenance;
(4) fault prediction
The method comprises the steps of evaluating the current performance state of a fault and forecasting the early stage of the fault by utilizing predicted fault characteristic parameters based on a fault prediction model of a state set sequence, and effectively realizing early stage identification of equipment faults by improving a data mining algorithm so as to effectively reduce economic and personal losses caused by faults which are high in concealment and difficult to find;
(5) predictive maintenance
According to the online monitoring information and the judgment result of the health state, the online monitoring system helps maintenance personnel to find abnormal symptoms of equipment at an early stage, provides corresponding maintenance suggestions and maintenance instruction manuals, finds out the cause of the fault as soon as possible, and predicts the influence of the fault, thereby realizing planned and targeted maintenance according to the circumstances and solving safety problems and hidden dangers in time;
(6) operation and maintenance knowledge management
And the operation and maintenance knowledge management realizes the management and the update of contents such as model base information, knowledge base information, maintenance optimization information and the like. The system provides management of information such as fault names, solution methods, maintenance suggestions and the like in an operation and maintenance knowledge base by storing fault types, fault algorithm models and maintenance suggestion measures; in addition, the system has a learning function, continuously optimizes a fault model and perfects an operation and maintenance knowledge base.
The invention of this application lies in:
1. data preprocessing technology based on rough set
In general, an original data set acquired by a sensor contains a large amount of noise and has problems of data loss, abnormality and the like, so that the original data needs to be preprocessed before a machine learning model is established for training to improve the data quality and the performance of the model; the rough set theory is a new important mathematical tool for processing fuzzy and inaccurate problems; the method adopts a particle calculation processing mode, effectively solves the complex problem by selecting the most appropriate particle layer, and processes the mining of fuzzy information and mass data. Redundant knowledge and unnecessary attribute characteristics are deleted, a data table is reduced, the number of fault characteristic attributes is reduced, the scale and complexity of a rule knowledge base are reduced, and the data processing efficiency is improved;
reducing the decision table by adopting a relative granularity attribute reduction algorithm, wherein the input of the relative granularity attribute reduction algorithm is that the decision table S is (U, C, U, D, V, f), C is a conditional attribute set, D is a decision attribute set, and the output reduction result is represented by RED; the reduction method comprises the following specific steps:
(1) merging the same rules in the decision table to make RED equal to phi;
(2) for each one of ciC/RED as the element of the attribute, CiImportance sig (c) of reduction result RED to decision set Di，RED，D)＝GD(D|RED)-GD(D|RED∪ci)；
(3) All sigs (c) calculated in step (2)iD, D) selecting the attribute c corresponding to the maximum valueiAs c iskIf a plurality of attributes meet the condition, selecting the first attribute as ck；
(4) If sig (c)kRED, D) > 0, then RED ═ RED ≧ CkAnd continuing to step (2) to circularly calculate; if sig (c)kIf RED, D)' is 0, the loop is ended, and the reduced result RED is output.
The method is an incremental reduction algorithm, and comprises the steps of selecting irreducible attributes and sequentially adding the irreducible attributes into a null set by calculating the importance of each attribute on RED to a decision set D until the importance of the remaining attributes is 0, and finishing adding, so that the minimum reduction result of a decision table is obtained.
2. Real-time online fault diagnosis technology
The ship cabin equipment is numerous, the arrangement is dispersed, the position is secret, frequent inspection is not easy to carry out, and the equipment abnormality is difficult to find in time. On the basis of full investigation, a real-time online fault diagnosis technology is provided, so that the working personnel can find the fault and hidden danger of the equipment in time, maintain and repair the equipment in a targeted manner, and reduce unnecessary repair, thereby prolonging the repair period and reducing the production cost;
various sensors are installed in a ship cabin system, fault characteristics are extracted by using various signal processing methods, and real-time online diagnosis is performed on the faults by using methods such as a fault model and an expert system, so that a worker can find the faults of equipment in time and take corresponding maintenance measures.
3. Fault prediction technique based on state set sequence
The method comprises the steps that a fault prediction model based on a state set sequence carries out estimation on the current performance state and early prediction on faults by utilizing predicted fault characteristic parameters, and early recognition on auxiliary equipment faults is more effectively realized by improving a data mining algorithm, so that the method is strong in effective reduction; the SDG is a qualitative mathematical model, all nodes are connected according to a certain rule by utilizing a directional branch to form a network topological graph, the influence relation between process variables and the deduction process of a fault can be described in an intuitive and simple graphical mode, a large amount of potential information is contained, the mechanism and the relation of a system are expressed, and the prediction of the fault is effectively realized;
the SDG fault prediction method based on the bidirectional reasoning comprises the steps of firstly, searching all possible fault sources by utilizing the reverse reasoning, and then sequentially carrying out the forward reasoning on the possible fault sources; if a certain fault can completely or maximally explain the states of the abnormal variable nodes, the fault is a prediction result with the highest reliability.
The inventive principle of the present application:
according to the construction requirements of the ship cabin equipment health management system, the research depth of a ship information system architecture is combined, the research of a cabin equipment health management system is developed aiming at the problems of multiple cabin equipment types, large data volume, low intelligent level and the like, and the research is mainly carried out according to the equipment health management technical architecture and the equipment general technical requirements; under the guidance of 'universal technical requirements', three types of core cabin equipment with high value, high added value, strong fault hazard and high maintenance cost are selected for system development and research; by combining advanced technologies such as a data preprocessing technology based on a rough set, a real-time online fault diagnosis technology, a fault prediction based on a state set sequence and the like, functions of online monitoring, fault diagnosis, fault prediction, predictive maintenance and the like of cabin equipment can be effectively improved, the operation and maintenance capacity of a ship cabin is enhanced, the safe operation level of an auxiliary engine is improved, the operation cost of an enterprise is reduced, and the health management level of the cabin is improved; the popularization and application of the project can improve the digitization, networking, intellectualization, safety operation and maintenance integration capability of the ship cabin core equipment, and help enterprises to realize the development goals of quality improvement, efficiency improvement, quality and quantity preservation.
The application has the advantages that:
1. the invention improves the data quality and the performance of model diagnosis by a data preprocessing technology based on a rough set:
the invention is based on rough set theory to preprocess the original data before establishing machine learning model training, and under the condition of ensuring the decision ability to be unchanged, the redundant knowledge and unnecessary attribute characteristics are deleted, the fault model is reduced, the number of fault characteristic attributes is reduced, the scale and complexity of a rule knowledge base are reduced, and the data quality and the diagnostic performance of the model are improved;
2. according to the invention, the fault prediction precision of the ship cabin equipment is improved by a fault prediction technology based on the state set sequence:
in order to realize early discovery and prediction accuracy of the equipment fault of the ship cabin, the invention introduces a fault prediction technology based on a state set sequence, judges the possible future fault by combining the running states, structural characteristics and historical data of the equipment such as vibration signals, pressure, temperature and the like through an established SDG model, and predicts the nature, type, degree, reason and position of the fault. When the fault is predicted, the strategy can be maintained rapidly and accurately according to the prediction result, and a reasonable later-period operation scheme of the equipment is provided by combining the operation environment of the equipment. Meanwhile, relevant information is transmitted to an operator in time; the invention can effectively realize early identification of cabin equipment faults and reduce economic and personal losses caused by faults which are high in concealment and difficult to detect.
Claims (8)
1. A ship cabin equipment health management and fault prediction system is characterized in that: the system comprises a control management device, a communication device, a data collection device and a field collection device, wherein the control management device comprises an upper computer, the communication device comprises a switch, the data collection device comprises a data collection card, the field collection device comprises a plurality of temperature sensors, voltage sensors, current sensors, pressure sensors and flow sensors, the temperature sensors, the voltage sensors, the current sensors, the pressure sensors and the flow sensors are used for being installed on ship cabin equipment, the temperature sensors, the voltage sensors, the current sensors, the pressure sensors and the flow sensors are all connected with the data collection card, and the data collection card is connected with the upper computer through the switch.
2. The ship cabin equipment health management and failure prediction system of claim 1, wherein: the upper computer is a computer or a programmable controller.
3. The ship cabin equipment health management and failure prediction system of claim 1, wherein: the temperature sensor is selected from one or more of NTC thermistor, platinum RTD, thermocouple and semiconductor.
4. A ship cabin equipment health management and fault prediction method is characterized by comprising the following steps: the method using the ship cabin equipment health management and fault prediction system of any one of claims 1-3, the method comprising:
(1) equipment reliability and failure mode analysis:
by using a fault tree method, reliability analysis and equipment failure mode analysis are carried out on ship cabin equipment, the fault characteristics of the equipment are obtained, and reasonable maintenance strategies are formulated;
(2) signal acquisition and preprocessing:
the method comprises the steps that various sensors are arranged to communicate with equipment, various information of the running state of the equipment is collected, the information is preprocessed to obtain parameters capable of reflecting the pressure, the temperature and the flow of the running state of the equipment, the running state data of the equipment is obtained, and data support is provided for real-time state monitoring, fault diagnosis and alarming and early warning;
(3) online fault diagnosis
Acquiring real-time monitoring equipment working state data on line, and performing real-time online diagnosis on a fault by using a fault model and an expert system in combination with a reserved fault mode and a failure criterion in the system; when the abnormality exists, the system fault is diagnosed on line by combining various health historical state data, working states and maintenance historical records of the equipment and integrating an intelligent diagnosis method, so that codes, characteristics, reasons for generation and positions of the faults are obtained, and timely and accurate data support is provided for maintenance;
(4) fault prediction
The method comprises the steps of evaluating the current performance state of a fault and forecasting the early stage of the fault by utilizing predicted fault characteristic parameters based on a fault prediction model of a state set sequence, and effectively realizing early stage identification of equipment faults by improving a data mining algorithm so as to effectively reduce economic and personal losses caused by faults which are high in concealment and difficult to find;
(5) predictive maintenance
According to the online monitoring information and the judgment result of the health state, the early detection of abnormal symptoms of equipment by maintenance personnel is facilitated, and corresponding maintenance suggestions and maintenance instruction manuals are provided;
(6) operation and maintenance knowledge management
The management and the update of the model base information, the knowledge base information and the maintenance optimization information are realized, and the management of the fault name, the solution method and the maintenance suggestion in the operation and maintenance knowledge base is provided by storing the fault type, the fault algorithm model and the maintenance suggestion measure; and secondly, continuously optimizing a fault model and perfecting an operation and maintenance knowledge base.
5. The method for health management and fault prediction of marine vessel cabin equipment of claim 4, wherein: the method adopts a particle calculation processing mode, effectively solves the complex problem by selecting the most appropriate particle layer, processes the mining of fuzzy information and mass data, deletes redundant knowledge and unnecessary attribute characteristics, reduces a data table, reduces the number of fault characteristic attributes, and reduces the scale and complexity of a rule knowledge base, and the specific process comprises the following steps:
reducing the decision table by adopting a relative granularity attribute reduction algorithm, wherein the input of the relative granularity attribute reduction algorithm is that the decision table S is (U, C, U, D, V, f), C is a conditional attribute set, D is a decision attribute set, and the output reduction result is represented by RED; the reduction method comprises the following specific steps:
(1) merging the same rules in the decision table to make RED equal to phi;
(2) for each one of ciC/RED as the element of the attribute, CiImportance sig (c) of reduction result RED to decision set Di,RED,D)＝GD(D|RED)-GD(D|RED∪ci)；
(3) All sigs (c) calculated in step (2)iD, D) selecting the attribute c corresponding to the maximum valueiAs c iskIf a plurality of attributes meet the condition, selecting the first attribute as ck；
(4) If sig (c)kRED, D) > 0, then RED ═ RED ≧ CkAnd continuing to step (2) to circularly calculate; if sig (c)kIf RED, D)' is 0, the loop is ended, and the reduced result RED is output.
6. The method for health management and fault prediction of marine vessel cabin equipment of claim 4, wherein: the method comprises the steps of installing various sensors in a ship cabin, extracting fault characteristics by using various signal processing methods, and carrying out real-time online diagnosis on the faults by using a fault model and an expert system starvation method, so that working personnel can find the faults of equipment in time and take corresponding maintenance measures.
7. The method for health management and fault prediction of marine vessel cabin equipment of claim 4, wherein: firstly, searching all possible fault sources by utilizing reverse reasoning, and then sequentially carrying out forward reasoning on the possible fault sources; if a certain fault can completely or maximally explain the states of the abnormal variable nodes, the fault is a prediction result with the highest reliability.
8. The method for health management and fault prediction of marine vessel cabin equipment of claim 4, wherein: the ship cabin equipment for health management and fault prediction by the method comprises a cargo oil pump, a marine diesel engine and a fuel supply unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010938464.2A CN112083355A (en) | 2020-09-09 | 2020-09-09 | Ship cabin equipment health management and fault prediction system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010938464.2A CN112083355A (en) | 2020-09-09 | 2020-09-09 | Ship cabin equipment health management and fault prediction system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112083355A true CN112083355A (en) | 2020-12-15 |
Family
ID=73731570
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010938464.2A Pending CN112083355A (en) | 2020-09-09 | 2020-09-09 | Ship cabin equipment health management and fault prediction system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112083355A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112858872A (en) * | 2020-12-31 | 2021-05-28 | 华中光电技术研究所（中国船舶重工集团公司第七一七研究所） | Circuit board health management circuit, device, control method and circuit board health manager |
| CN113189918A (en) * | 2021-05-25 | 2021-07-30 | 上海海事大学 | Ship anti-pollution equipment and safety equipment online monitoring system and method |
| CN113311280A (en) * | 2021-07-30 | 2021-08-27 | 中国人民解放军海军工程大学 | Health grading monitoring device for complex electromechanical system |
| CN113538874A (en) * | 2021-07-05 | 2021-10-22 | 南京盛航海运股份有限公司 | Ship running state detection early warning method, system, equipment and storage medium |
| CN115453236A (en) * | 2022-08-24 | 2022-12-09 | 大连海事大学 | Fault diagnosis and health assessment method for ship wind wing power system |
| CN116976865A (en) * | 2023-09-22 | 2023-10-31 | 中国船舶集团有限公司第七一九研究所 | Ship maintenance device allocation management system based on big data analysis |
-
2020
- 2020-09-09 CN CN202010938464.2A patent/CN112083355A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112858872A (en) * | 2020-12-31 | 2021-05-28 | 华中光电技术研究所（中国船舶重工集团公司第七一七研究所） | Circuit board health management circuit, device, control method and circuit board health manager |
| CN113189918A (en) * | 2021-05-25 | 2021-07-30 | 上海海事大学 | Ship anti-pollution equipment and safety equipment online monitoring system and method |
| CN113538874A (en) * | 2021-07-05 | 2021-10-22 | 南京盛航海运股份有限公司 | Ship running state detection early warning method, system, equipment and storage medium |
| CN113311280A (en) * | 2021-07-30 | 2021-08-27 | 中国人民解放军海军工程大学 | Health grading monitoring device for complex electromechanical system |
| CN113311280B (en) * | 2021-07-30 | 2021-12-28 | 中国人民解放军海军工程大学 | Health grading monitoring device for complex electromechanical system |
| CN115453236A (en) * | 2022-08-24 | 2022-12-09 | 大连海事大学 | Fault diagnosis and health assessment method for ship wind wing power system |
| CN116976865A (en) * | 2023-09-22 | 2023-10-31 | 中国船舶集团有限公司第七一九研究所 | Ship maintenance device allocation management system based on big data analysis |
| CN116976865B (en) * | 2023-09-22 | 2023-12-19 | 中国船舶集团有限公司第七一九研究所 | Ship maintenance device allocation management system based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112083355A (en) | Ship cabin equipment health management and fault prediction system and method | |
| Liu et al. | Fault detection, diagnostics, and prognostics: software agent solutions | |
| RU2313815C2 (en) | Device and method for controlling technical plant, which contains a set of systems, in particular, electric power plant | |
| CN106368816B (en) | A kind of online method for detecting abnormality of marine low speed diesel engine based on baseline offset | |
| KR102073912B1 (en) | Method and system for diagnostic rules for heavy duty gas turbines | |
| JP5203984B2 (en) | Method and apparatus for analyzing and evaluating measurement data of a measurement system | |
| CN110397476B (en) | Online monitoring and analyzing system and method for oil products of steam turbine for nuclear power | |
| KR20160073945A (en) | System and method for managing wind plant | |
| EP1242923A2 (en) | A process for the monitoring and diagnostics of data from a remote asset | |
| CN110738331A (en) | intelligent marine engine room system | |
| CN102141808A (en) | Embedded type fault pre-diagnosis system and method for steam turbine generator unit | |
| CN105988457A (en) | Ship or offshore mobile platform watertight equipment monitoring and fault diagnosis system | |
| Garcia | Improving heat exchanger supervision using neural networks and rule based techniques | |
| WO2023063887A2 (en) | Method and apparatus for predicting state of wind turbine blade, and device and storage medium therefor | |
| CN112254972A (en) | Excavator oil temperature early warning method and device, server and excavator | |
| Green et al. | NILM dashboard: Actionable feedback for condition-based maintenance | |
| Nugraha et al. | The Auxiliary Engine Lubricating Oil Pressure Monitoring System Based on Modbus Communication | |
| CN114648212A (en) | Cloud computing-based ship equipment performance intelligent management system and method | |
| CN212483819U (en) | Ship cabin equipment health management and fault prediction system | |
| Breikin et al. | Dynamic modelling for condition monitoring of gas turbines: Genetic algorithms approach | |
| RU121944U1 (en) | AUTOMATED SYSTEM FOR IDENTIFYING FAULT SENSORS AMONG SENSORS INTENDED FOR CONTROL OF TECHNOLOGICAL PROCESSES | |
| Fernandes et al. | An overview on degradation modelling for service cost estimation | |
| Tao et al. | A state and fault prediction method based on RBF neural networks | |
| Theilliol et al. | Sensor fault diagnosis based on energy balance evaluation: Application to a metal processing | |
| Papa et al. | Cyber physical system based proactive collaborative maintenance |
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 |