Open AccessSeawater Temperature Outlier Detection Strategy Based on Wavelet Analysis for Solving the Migration Problem of Fishery Companies
Author(s) -
Yuan Liu,
Zheyong Qiu,
Zhongyue Wang,
Jiaqi He
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1575/1/012061
Subject(s) - wavelet , seawater , outlier , wavelet transform , transformation (genetics) , discrete wavelet transform , anomaly detection , computer science , mathematics , econometrics , fishery , algorithm , geology , statistics , oceanography , data mining , artificial intelligence , biology , biochemistry , gene
As the global ocean temperature rises, many marine organisms begin to find new habitats. The problem of fish migration has a profound impact on the development of fisheries. This paper uses wavelet anomaly analysis method to find the abnormal value of seawater temperature, uses discrete wavelet decomposition and reconstruction to de-noise the time series, and predicts the age of the fishing company. Firstly, we conducted continuous wavelet analysis of seawater temperature data to obtain the relationship between seawater temperature change rate and time. We performed Fourier transform to obtain its transform function, and then reconstructed the sequence to obtain wavelet coefficients of seawater temperature transformation speed. After that, we used wavelet coefficients to conduct simulation screening to obtain all outliers. In this way, the time for bankruptcy under the best and worst conditions of the fishing company can be obtained. Although the model comprehensively considers many factors, in order to establish a universal model, idealizing many influencing factors has certain limitations. The predicted bankruptcy time may have a certain error from the actual.