A novel adaptive sampling based methodology for feasible region identification of compute intensive models using artificial neural network

Abstract

Identification of feasible region of operations in multivariate processes is a problem of interest in several fields. This is particularly challenging when the process model is black‐box in nature and/or is computationally expensive, as analytical solutions are not available and the number of possible model evaluations is limited. An efficient methodology is required to identify samples where the model is evaluated for developing a computationally efficient surrogate model. In this work, an artificial neural network based surrogate model is proposed which is integrated with a statistical‐based approach (Jack‐knifing) to estimate the variance of the surrogate model prediction. This allows implementation of an adaptive sampling approach where new samples are identified close to the feasible region boundary or in regions of high prediction uncertainty. The proposed approach performs better than a previously published kriging based method for different dimensionality case studies.

Prediction and design of cyclodextrin inclusion complexes formation via machine learning-based strategies

“…ANN, one of the parallel computational ML models, exhibits powerful capabilities in data regression and pattern recognition in chemical engineering research (Metta et al, 2020;Szoplik and Ciuksza, 2021). Because the model parameters in the learning algorithm are tuned from known measured inputs and outputs, ANNs can describe arbitrary functions and the actual problems hidden beneath.…”

Maximizing information from chemical engineering data sets: Applications to machine learning

“…Identifying feasible regions is a key component in flexibility analysis (Swaney and Grossmann, 1985;Grossmann et al, 2014), but the classical approaches assume that algebraic constraints containing some uncertain parameters are known (Halemane and Grossmann, 1983;Grossmann and Floudas, 1987). We refer the interested reader to Banerjee et al (2010), Rogers and Ierapetritou (2015), Ierapetritou (2017), andMetta et al (2020) for reviews covering approaches to estimate feasible regions and constraints based on surrogate functions. However, to efficiently model constraints with surrogate functions requires data samples in all regions of interests and a mixture of data points where the constraints are both satisfied and violated.…”

Multifeedstock and Multiproduct Process Design Using Neural Network Surrogate Flexibility Constraints

“…The ANN model with sigmoid activation function was used as the surrogate for different unit operation models based on Aspen Plus simulation . Adaptive sampling strategy has been implemented using both ANN and Kriging surrogate models to optimize computationally expensive processes. , Kim and Boukouvala extended Gaussian Process and ANN to mixed-integer surrogate models by using one-hot encoding for optimization . Rectified linear unit (ReLU) is another widely used ANN activation function that demonstrates the ability to capture the nonlinearity of the model with low model complexity .…”