Affiliation:
1. Federal Research Center for Information and Computational Technologies
Abstract
The article presents the IntvalPy library which implements interval computations in Python. Unlike other existing interval libraries, IntvalPy allows one to work with both classical interval arithmetic and complete Kaucher interval arithmetic. In addition, the library was developed taking into account the IEEE 1788-2015 standard for interval arithmetic on digital computers, which guarantees high accuracy of the results and compatibility with the other existing software products. The top-level functionality of the IntvalPy library implements state-of-the-art methods for recognizing and estimating solution sets for interval linear systems of equations, computing their formal solutions, and visualizing solution sets for interval equations and systems of equations. As an example of the library application, we solve the practically important problem of estimating the parameters of the electrochemical process of the formation of loose metal powder precipitates. Additionally, numerical computation was carried out, as well as qualitative comparisons with other interval libraries, in order to demonstrate the functionality and optimality of implemented interval classes.
Publisher
Novosibirsk State University (NSU)
Reference25 articles.
1. Androsov A. S., Shary S. P. The IntvalPy library [Online]. URL: https://github.com/AndrosovAS/intvalpy.
2. Kearfott R. B., Nakao M., Neumaier A., Rump S. M. et al. Standardized notation in interval analysis. Reliable Computing, 2010, vol. 15, no. 1, pp. 70-13.
3. Rump S. M. Verification methods: Rigorous results using floating-point arithmetic. Acta Numerica, 2010, vol. 19, pp. 287-449.
4. Hansen E., Walster G. W. Global Optimization Using Interval Analysis: Revised And Expanded CRC Press; 2nd edition (December 19, 2003). 2003, 728 p.
5. Sharaya I. A. Boundary intervals and visualization of AE-solution sets for interval systems of linear equations. Reliable Computing, 2012, pp. 435-467.