CESSIPy: A Python open-source module for stochastic system identification in civil engineering

Abstract

This paper presents CESSIPy, a Python open-source module for identifying modal properties of a vibrating structure from output-only measurements. The identified properties are natural frequencies, damping ratios and modal shapes. System inputs are assumed to be broad band stationary Gaussian processes. Five system identification methods are available: (1) Basic Frequency Domain (BFD), (2) Enhanced Frequency Domain Decomposition (EFDD), (3) Covariance-Driven Stochastic Subspace Identification (SSI COV), (4) Data-Driven Stochastic Subspace Identification (SSI DATA), and (5) Instrumental Variable (IV). All functions have been validated through numerical simulations. The software is fully documented with a user guide, in-code descriptions, and script examples to demonstrate code usage.

Keywords

• System identification
• Operational modal analysis
• Structural dynamics
• Enhanced frequency-domain decomposition
• Peak-picking
• Stochastic subspace identification