Mechanistic modeling has been repeatedly successfully applied in process development and control of protein chromatography. For each combination of adsorbate and adsorbent, the mechanistic models have to be calibrated. Some of the model parameters, such as system characteristics, can be determined reliably by applying well-established experimental methods, whereas others cannot be measured directly. In common practice of protein chromatography modeling, these parameters are identified by applying time-consuming methods such as frontal analysis combined with gradient experiments, curve-fitting, or combined Yamamoto approach. For new components in the chromatographic system, these traditional calibration approaches require to be conducted repeatedly. In the presented work, a novel method for the calibration of mechanistic models based on artificial neural network (ANN) modeling was applied. An in silico screening of possible model parameter combinations was performed to generate learning material for the ANN model. Once the ANN model was trained to recognize chromatograms and to respond with the corresponding model parameter set, it was used to calibrate the mechanistic model from measured chromatograms. The ANN model's capability of parameter estimation was tested by predicting gradient elution chromatograms. The time-consuming model parameter estimation process itself could be reduced down to milliseconds. The functionality of the method was successfully demonstrated in a study with the calibration of the transport-dispersive model (TDM) and the stoichiometric displacement model (SDM) for a protein mixture.

No abstract

No abstract

In this study, a novel hybrid metaheuristic model was developed to forecast the undrained soil shear (USS) property from cone penetration test (CPT) data (data from bore log sample from 70 different sites in Louisiana). This algorithm produced with the integration of grey wolf optimization (GWO) and multilayer perceptron neural network (MLP), named GWO - MLP, where different numbers of hidden layers were tested (1 to 4). The duty of optimization algorithm was to determine the optimal number of neurons in each hidden layer. To this objective, the system comprised five inputs entitled sleeve friction, cone tip persistence, liquid limit, plastic limitation, too much weight, and USS as outcome. The developed models for forecasting the USS of soil show the proposed best models have R2 at 0.9134 and 0.9236 in the training and predicting stage. Although the total ranking score of GWO-MLP2 and GWO-MLP4 is equal, the OBJ value shows that GWO-MLP4 has better performance than GWO-MLP2. In this case, considering the time of model running and a greater number of hidden layers suggests that GWO-MLP2 could be most appropriate. Therefore, the GWO-MLP3 model outperforms other GWO-MLP networks in the training and testing phase.