ISSN 1000-0526
CN 11-2282/P
CN 11-2282/P
Archive > Volume 41 Issue 11 > 2015,41(11):1398-1407. DOI:10.7519/j.issn.1000-0526.2015.11.010 Prev Next
Application of Inter Verification Sequence Alignment Model to Two Data Source Splicing of AWS Hourly Precipitation
- Article
- Figures
- Metrics
- Preview PDF
- Reference
- Related
- Cited by
- Materials
Abstract:
Nondefault inconsistent hourly precipitation data are an abnormal status in automatic weather station (AWS) observation, which can be often met as hourly precipitation data are transmitted and recorded in 2 sources. Three groups of related instances are listed first, and the direct reasons that deeply hidden are found out manually. To solve this problem, Inter VerificationSequence Alignment (IVSA) model in smaller time scale is raised in this article. When nondefault inconsistent data from same station appears at the same time, verification with smaller time scale data (minute precipitation) is made respectively. If both data cannot be proved wrong with inner verification step, then unit error possibility is added into sequence alignment method. Correlation credibility is calculated and more reliable data can be selected accordingly. After then, monthly data in May 2012 (1360 pairs of instances) are used to train the parameters, and the data (4017 pairs of instances) from 1 June to 31 July 2012 are input to verify the efficiency of applying IVSA in real time data environment, getting an accuracy of 99.65%. It is concluded that IVSA model can eliminate nondefault inconsistence in hourly precipitation data under running rules.
Keywords:
Project Supported:
Clc Number:
Mobile website
