CN-115374961-B - Ship power operation and maintenance system and operation and maintenance method

CN115374961BCN 115374961 BCN115374961 BCN 115374961BCN-115374961-B

Abstract

The invention discloses a ship power operation and maintenance system which is characterized by comprising a parameter setting module, a real-time data acquisition module, a fault analysis module, a ship history big data module, a fault tree optimization module and a data terminal module, wherein the fault analysis module is respectively connected with the parameter setting module, the real-time data acquisition module, the fault tree optimization module and the data terminal module, the ship history big data module is connected with the fault tree optimization module, the fault analysis module analyzes a ship fault through parameters set by the parameter setting module and a fault tree model constructed by the ship history big data module, and the fault tree optimization module optimizes the fault tree model constructed by the fault analysis module through historical data of the ship history big data module. According to the invention, through carrying out correlation coefficient analysis on the attributes, merging the attributes with close relations, increasing the operation speed, reducing the branching and judging times and improving the efficiency.

Inventors

YANG YIFEI
WU YANG
HE ZUJUN
SU ZHEN
QI LIANG
YE SHUXIA
YUAN WEI

Assignees

江苏科技大学

Dates

Publication Date: 20260505
Application Date: 20220711

Claims (3)

1. The ship power operation and maintenance method is characterized by comprising the following steps of: step 1, acquiring historical fault data, and performing data preprocessing to obtain a training set and a testing set which comprise diesel engine state attributes; Step 2, randomly extracting the training set sample and the testing set sample with a put back to form a plurality of sub-training sets and a plurality of sub-testing sets, and filling the missing diesel engine state attribute of the plurality of sub-training sets and performing continuous discrete processing on the diesel engine state attribute of the plurality of sub-training sets; Step 3, merging similar attributes in the diesel engine state attributes of the sub training set; step 4, determining the maximum depth critical value of the fault tree according to the relativity of each attribute in the diesel engine state attribute, constructing the fault tree by an information gain method according to the maximum depth critical value and the training set, When the difference of the information gain before and after splitting does not meet the preset condition, the splitting factor item in the information gain rate is repeatedly optimized according to the difference of the information gain before and after splitting until the preset condition is met, and splitting is carried out to form a new branch; When the split factor term in the information gain rate is optimized, the acquisition formula of the split factor term is as follows: Wherein S k represents the number of sample data of the current kth attribute, S is the number of data set samples, beta k is a probability average coefficient, Wherein p k is the current attribute probability, p max is the attribute probability maximum value, and p min is the attribute probability minimum value; When the split factor item in the information gain rate is repeatedly optimized according to the difference between the information gain before and after splitting, the acquisition formula of the split factor item is as follows: Wherein S k represents the number of sample data of the current kth attribute, S is the number of data set samples, beta k is a probability average coefficient, Wherein p k is the current attribute probability, p max is the maximum value of the attribute probability, and p min is the minimum value of the attribute probability, and alpha is the difference between the information gain value before splitting and the information gain value after splitting; Step 5, pruning the constructed fault tree, and testing the fault tree by using a test set; And 6, inputting the real-time diesel engine state into a fault tree to diagnose the fault.
2. The ship power operation and maintenance method according to claim 1, wherein the determining method of the maximum depth critical value in the step 4 comprises: when the number of the current attributes is less than 5, the maximum depth critical value limitation is not carried out, and when the number of the current attributes is more than or equal to 5, the fault tree critical value is Where m represents the number of attributes.
3. The ship power operation and maintenance method according to claim 1, wherein the preset condition is that a difference value between an information gain value before splitting and an information gain value after splitting is smaller than a preset value, and the preset value is obtained by determining the preset value through a square value r 2 of a pearson correlation attribute coefficient r, and the value range is between (0 and 1).

Description

Ship power operation and maintenance system and operation and maintenance method Technical Field The invention belongs to the technical field of ship operation and maintenance, and particularly relates to a ship power operation and maintenance system and an operation and maintenance method. Background As the most important component of the ship machinery, the ship power system equipment is a combination of main machinery for various energy generation, transmission and consumption. With scientific and industrial technology as support, modern ships are increasingly large, and ship power systems are also developed towards large, integrated, complicated and intelligent directions. Modern ship power systems begin to apply more and more high-tech technologies, and the system scale structure and the technical complexity are increased increasingly, so that faults are easy to occur in the operation process. The complexity and difficulty of the technology and equipment required by fault maintenance are increased sharply, the maintenance cost is also greatly increased, and the current operation and maintenance service cannot meet the increasing operation and maintenance requirements of modern ships. Poor maintenance and failure handling are a very important cause of marine accidents. In the management of engineering ships, especially large transport ships, the daily dynamic management of the ships is more needed, the information such as the dynamic state of the ships and the running state of mechanical equipment is accurately mastered, but the operation is inevitably wrong only by manual maintenance, so that enterprises or using units are at risk of suffering economic loss, and a lot of inconvenience is brought to the operation and management of the ships. In view of the above problems, establishing a ship power operation and maintenance system and an operation and maintenance method for guaranteeing remote operation and maintenance management of the ship power system, guaranteeing spare part management and supply, and checking the operation condition of ship power equipment, maintenance management and the like has become a key to be solved in the modernization of ship operation and maintenance guarantee. Disclosure of Invention Aiming at the defects in the prior art, the invention provides a ship power operation and maintenance system and an operation and maintenance method, which are used for solving the technical problems that the operation and maintenance condition of a ship cannot be accurately mastered and the accurate operation and maintenance method cannot be obtained in the prior art. The invention provides a ship power operation and maintenance system, which comprises a parameter setting module, a real-time data acquisition module, a fault analysis module, a ship history big data module, a fault tree optimization module and a data terminal module, wherein the parameter setting module is used for setting the parameter of the ship; The fault analysis module is respectively connected with the parameter setting module, the real-time data acquisition module, the fault tree optimization module and the data terminal module, and the ship history big data module is connected with the fault tree optimization module; The fault analysis module is used for analyzing the ship faults according to the parameters set by the parameter setting module and the fault tree model constructed by the ship history big data module and the real-time data acquired by the real-time data acquisition module; The data terminal module generates an alarm signal and an operation and maintenance decision according to the ship fault analysis result; And the fault tree optimization module optimizes the fault tree model constructed by the fault analysis module through the historical data of the ship historical big data module. The parameter setting module is used for setting equipment basic parameters and shore-based parameters, wherein the equipment basic parameters comprise a system knowledge base, an equipment fault base and design data which are provided by equipment manufacturers, and the shore-based parameters comprise ship power system use scene data, reliability data and maintenance data which are provided by shore-based. Further, the fault analysis module builds a fault tree model through an S4000P standard to analyze ship faults. Further, the fault tree optimization module builds a fault tree model for reference through the historical data of the ship historical big data module, compares the fault model built by the fault analysis module with the fault tree model for reference, and optimizes the fault model by adding and/or deleting branches. The technical scheme of the invention is a ship power operation and maintenance method, which comprises the following steps: step 1, acquiring historical fault data, and performing data preprocessing to obtain a training set and a testing set which comprise diesel engine state attributes; Step 2, ran