Modelling solid-fluid multiphase flows are crucial to many applications such as fluidized beds, pyrolysis and gasification, catalytic cracking etc. Accurate modelling of the fluid-particle forces is essential for lab-scale and industry-scale simulations. Fluid-particle system solutions can be obtained using various techniques including the macro-scale TFM (Two fluid model), the meso-scale CFD-DEM (CFD - Discrete Element Method) and the micro-scale PRS (Particle Resolved Simulation method). As the simulation scale decreases, accuracy increases but with an exponential increase in computational time. Since fluid forces have a large impact on the dynamics of the system, this study trains deep learning models using micro-scale PRS data to predict drag forces on ellipsoidal particle suspensions to be applied to meso-scale and macro-scale models. Two different deep learning methodologies are employed, multi-layer perceptrons (MLP) and 3D convolutional neural networks (CNNs). The former trains on the mean characteristics of the suspension including the Reynolds number of the mean flow, the solid fraction of the suspension, particle shape or aspect ratio and inclination to the mean flow direction, while the latter trains on the 3D spatial characterization of the immediate neighborhood of each particle in addition to the data provided to the MLP. The trained models are analyzed and compared on their ability to predict three different drag force values, the suspension mean drag which is the mean drag for all the particles in a given suspension, the mean orientation drag which is the mean drag of all particles at specific orientations to the mean flow, and finally the individual particle drag. Additionally, the trained models are also compared on their ability to test on data sets that are excluded/hidden during the training phase. For instance, the deep learning models are trained on drag force data at only a few values of Reynolds numbers and tested on an unseen value of Reynolds numbers. The ability of the trained models to perform extrapolations over Reynolds number, solid fraction, and particle shape to predict drag forces is presented. The results show that the CNN performs significantly better compared to the MLP in terms of predicting both suspensions mean drag force and also mean orientation drag force, except a particular case of extrapolation where the MLP does better. With regards to predicting drag force on individual particles in the suspension the CNN performs very well when extrapolated to unseen cases and experiments and performs reasonably well when extrapolating to unseen Reynolds numbers and solid fractions. / M.S. / Multiphase solid-fluid flows are ubiquitous in various industries like pharmaceuticals (tablet coating), agriculture (grain drying, grain conveying), mining (oar roasting, mineral conveying), energy (gasification). Accurate and time-efficient computational simulations are crucial in developing and designing systems dealing with multiphase flows. Particle drag force calculations are very important in modeling solid-fluid multiphase flows. Current simulation methods used in the industry such as two-fluid models (TFM) and CFD-Discrete Element Methods (CFD-DEM) suffer from uncertain drag force modeling because these simulations do not resolve the flow field around a particle. Particle Resolved Simulations (PRS) on the other hand completely resolve the fluid flow around a particle and predict very accurate drag force values. This requires a very fine mesh simulation, thus making PRS simulations many orders more computationally expensive compared to the CFD-DEM simulations. This work aims at using deep learning or artificial intelligence-based methods to improve the drag calculation accuracy of the CFD-DEM simulations by learning from the data generated by PRS simulations. Two different deep learning models have been used, the Multi-Layer Perceptrons(MLP) and Convolutional Neural Networks(CNN). The deep learning models are trained to predict the drag forces given a particle's aspect ratio, the solid fraction of the suspension it is present in, and the Reynolds number of the mean flow field in the suspension. Along with the former information the CNN, owing their ability to learn spatial data better is additionally provided with a 3D image of particles' immediate neighborhood. The trained models are analyzed on their ability to predict drag forces at three different fidelities, the suspension mean drag force, the orientation mean drag, and the individual particle drag. Additionally, the trained models are compared on their abilities to predict unseen datasets. For instance, the models would be trained on particles of an aspect ratio of 10 and 5 and tested on their ability to predict drags of particles of aspect ratio 2.5. The results show that the CNN performs significantly better compared to the MLP in terms of predicting both suspension mean drag force and also mean orientation drag force, except a particular case of extrapolation where the MLP does better. With regards to predicting drag force on individual particles in the suspension, the CNN performs very well when extrapolated to unseen cases and experiments and performs reasonably well when extrapolating to unseen Reynolds numbers and solid fractions.