Outlier data identification method and system in catalytic cracking device data

Abstract

The embodiment of the invention provides an outlier data identification method and system in catalytic cracking device data. The method comprises the following steps that: according to preset sampling time, obtaining the original attribute value of catalytic cracking production data, and obtaining the time series set of the original attribute value; using a sliding time window to reconstruct the time series set, and obtaining the sub-series set of the original attribute value, wherein the sub-series set comprises a plurality of sub series; obtaining the interactive weight vector of the sub-series set, and constructing a weighted series set according to the interactive weight vector and the sub-series set, wherein the weighted series set comprises a plurality of weighted sub series; and obtaining the local outlier factor of each weighted sub series in the weighted series set, and identifying the outlier data in the catalytic cracking device data according to the local outlier factor. The system is used for executing the above method. In the embodiment of the invention, outlier data identification accuracy in the catalytic cracking device data and the reliability of the catalytic cracking device data are improved.

Description

technical field [0001] The embodiments of the present invention relate to the field of petrochemical technology, and in particular to a method and system for identifying outlier data in catalytic cracking unit data. Background technique [0002] Heavy oil catalytic cracking plays an important role in my country's oil refining industry. For refineries that require reduced production energy consumption, material consumption, and abnormal shutdown, it is the most economical and convenient to ensure the "stable, long-term, full, and optimal" operation of the device through production operation optimization. s Choice. In the optimization of refining and chemical production operations, in addition to mechanism modeling, data modeling is also an important means. Use the data of daily operation, or arrange certain tests to obtain the test data of the process, and then use statistical methods or machine learning methods to build models, that is, data modeling. The data model has high...

AI Technical Summary

Problems solved by technology

The identification of outlier data in catalytic cracking unit data belongs to the research field of time series outlier detection. The difficulty lies in: the sampling data itself has volatility, and normal fluctuations cannot be treated as outliers. What needs to be paid attention to is these normal fluctuations Exception data other than

At present, in the field of engineering, statistical methods are mainly used to identify outlier data. This method is simple to calculate, but it needs to know or infer the distribution of data in advance. However, the data in reality is often far from the ideal mathematical distribution. Moreover, such methods are difficult to identify high-dimensional and periodic time series data

In the existing technology, time series outlier data identification methods are generally used, and some methods directly apply non-sequence data mining methods to time series data, ignoring periodicity and volatility, and have a high misjudgment rate

Images