Bulk cargo ship equipment state detection method based on SVM

Abstract

The invention relates to a bulk cargo ship equipment state detection method based on SVM, and the method comprises the steps: S1, obtaining a bulk cargo ship equipment state total training data set, and randomly selecting an initial training data sample; S2, training an SVM model by adopting the training data sample to obtain a rough separation plane, finding out and removing training data sampleswhich are not support vectors, predicting the residual training data sample size according to the trained SVM model, and determining removal or reservation of the training data samples according to aboundary data decision function value so as to reconstruct a training data set; S3, incrementally selecting a ship equipment state training data sample from the reconstruction training data set, turning to the step S2, and gradually training an SVM model to screen out a support vector until the training data sample is finally reconstructed; and S4, performing global SVM training on the final reconstruction training data sample to obtain a classification hyperplane, and obtaining a ship equipment state detection result according to the classification hyperplane. According to the method, the problems of long consumed time and large occupied memory in the SVM training process are solved.

Description

technical field [0001] The invention relates to the technical field of big data processing, in particular to an SVM-based bulk cargo ship equipment state detection method. Background technique [0002] With the development of science and technology, for the shipbuilding industry, with the wide application of network technology and information technology, ship automation, control and navigation systems are developing towards distributed, network and intelligent systems. From the perspective of ship operation, with the improvement of ship automation and informatization, more and more data are accumulated on board. These massive data contain all information on the status of ship equipment, but for ship owners, this information is not Intuitive and understandable, and data-based ship management can help ship owners fully collect and utilize this information, refine and analyze it, and provide services such as assessment, forecast, management, and decision-making for the status o...

AI Technical Summary

Problems solved by technology

[0004] In view of the above analysis, the present invention aims to provide a SVM-based bulk cargo ship equipment status detection method to solve the problem that the existing SVM model training process takes a long time, takes up a large amount of hardware resources and memory, and cannot meet the needs of rapid processing of high-dimensional large-scale ship data. To detect problems with the equipment status of bulk cargo ships

Images