Antecedent Moisture - Data-Based Approaches

Data-Based Approaches

An alternative approach for modeling antecedent moisture is to start from measurements of the behavior of the system and the external influences (inputs to the system) and try to determine a mathematical relation between them without going into the details of what is actually happening inside the system. This approach is called system identification. System identification is applied in several fields beyond engineering, ranging from economics to astronomy and comes under other names such as inverse modeling, time series analysis, and empirical physical modeling. System identification is a general term to describe mathematical tools and algorithms that build dynamical models from measured data. A dynamical model in this context is a mathematical description of the dynamic behavior of a system or process. A so-called white-box model based on first principles, e.g. a model for a physical process from the Newton equations, in many cases will be overly complex and possibly even impossible to obtain in reasonable time due to the complex nature of many systems and processes.

Data-based approaches based on system identification, such as the i3D antecedent moisture model, have been applied to hydrologic modeling for simulating antecedent moisture effects on wet weather events in sanitary collection systems. This modeling approach differs from traditional techniques because it is based on system identification and is guided by system observations (i.e. data) and mathematical routines are used to generate the correct model structure, rather than physically based first principles. This is in contrast to assuming that the correct model is known beforehand, as is typically the case for modeling within civil engineering. This technique allows information within the observations to guide the modeling algorithms so that only the relevant and observed dynamics are present in the model structure. The resulting models are not black box, but are grey box models that have parameters and structure that tie directly to physical understanding and interpretation.