Monoclonal antibodies (MAbs) have extensive biomedical application and are produced in mammalian cell bioreactor at a variety of scales, with... Show moreMonoclonal antibodies (MAbs) have extensive biomedical application and are produced in mammalian cell bioreactor at a variety of scales, with glucose and glutamine being the principle carbon and nitrogen sources required for cellular metabolism. Fed-batch operation has certain inherent advantage over batch culture for MAbs production. Design, optimization, scale-up, and control of bioreactor used for MAbs production requires reliable predictive empirical or mechanistic models for key cellular activity. The models used in prior studies have largely been the first principle based models (FPMs), although data-driven models are receiving increasing attention due to their certain inherent benefits and increasing information available about biological processes. The simpler and much less rigid structure of data-driven models facilitates frequent updating of parameters and prediction of process trajectories, increasing their utility in representation, monitoring and control of these processes.Multi-sampling rate recursive time series models are developed for representation of a mammalian cell culture, with process variables that are determinants of the performance of the culture being measured at different sampling frequencies. For this reason, a composite of an adaptive autoregressive moving average with exogenous input (ARMAX) model, a dual-rate adaptive autoregressive exogenous input (DR-ARX) model and a irregular-rate adaptive autoregressive exogenous input (IR-ARX) model is used. Appropriate parameter constraints have been imposed in parameter estimation algorithms and stability of these has been examined and is ensured. The data required of estimation of parameters are generated from simulation experiments using a well tested first principle-based kinetic model (FPM) considering random variations in manipulated inputs, kinetic parameters in the FPM, and measurement error for outputs. Glucose and glutamine being determinants of mammalian cell metabolism, their supply rates are considered to be inputs. The predictive ability of the data-driven model is examined and demonstrated over a broad range of prediction horizon. Show less