SPRITE (Smart Preprocessing & Robust Integration Emulator)

SPRITE (Smart Preprocessing & Robust Integration Emulator) is an open-source suite of analytical and data-driven models that predict the performance of renewable carbon feedstock preprocessing units and system integration. The suite includes models such as the Population Balance Model (PBM), Enhanced Deep Neural Operator (DNO+), and Physics-Informed DNO+ (PIDNO+).

In order to achieve the desired material properties of granular biomass, such as particle size distribution (PSD), a milling process is necessary during preprocessing. The PSD of biomass is crucial in biofuel manufacturing, so accurately predicting it is vital in the design of preprocessing systems. While the PBM can rapidly predict the post-milling PSD of granular biomass after empirical calibration and validation, it does not consider moisture content as an input parameter.

To address this limitation, the code implements an enhanced deep learning model called the Deep Neural Operator (DNO+). This model not only retains the capabilities of the PBM in handling complex mapping functions but also incorporates additional factors that influence the system. By taking into account experimental conditions like feed moisture content and discharge screen size, the trained DNO+ model can effectively predict the PSD after milling.

Using this code, which includes these models, will help guide the selection of milling parameters to achieve the desired biomass PSD.

This software is open source and available at no cost.