A method for detecting outliers in hydrological time series based on arima-svr

Abstract

The invention discloses an ARIMA-SVR-based hydrological time series abnormal-value detection method. The method includes the following steps: firstly, acquiring hydrological time series data, and carrying out ARIMA fitting on a hydrological time series, wherein an ARIMA model can well fit a linear part of the data, but a fitting effect is poorer when a nonlinear part exists in the data; then obtaining residual errors of an ARIMA part, and using a method of 10-fold cross-validation to find out optimum combination of gamma, cost and a kernel function of SVR; and finally, adding fitting values oftwo parts together to obtain final prediction values, obtaining a confidence interval, of which confidence is p, by solving, comparing the prediction values with the confidence interval, and determining that values outside the confidence interval are abnormal values. The invention provides, for practitioners related to water conservancy, a method of finding abnormal values in hydrological time series, and detection of the abnormal values in the hydrological time series has important guiding significance for work of flood control, drought relief and the like.

Description

technical field [0001] The present invention relates to a model construction method for outlier detection of hydrological time series based on the field of statistical learning and machine learning. value is checked. Background technique [0002] Outlier detection is an important part of hydrological data mining. The change of water level is affected by seasons and other abrupt factors, and it also contains noise factors. The detection ability of a single model is limited, and the traditional ARIMA model is not accurate enough to predict nonlinear time series. High, the support vector regression structure is complex, and it is easy to cause the problem of "overfitting". Individual optimization of these methods does not overcome the limitations of individual methods. Therefore, the outlier detection method combined with multiple models has become the direction of time series outlier detection. Contents of the invention [0003] Purpose of the invention: In view of the di...

AI Technical Summary

Problems solved by technology

[0003] Purpose of the invention: In view of the shortcomings of the existing time series outlier methods that are not pertinent, relatively low in sensitivity and specificity, according to the fluctuation characteristics of the hydrological time series, the method of combining the ARIMA model and SVR is used to analyze the anomaly of the hydrological time series. Detection, Improving Sensitivity and Specificity of Anomaly Detection in Hydrological Time Series

Images