Particle group combustion presents a strong temporal and spatial inhomogeneity owing to the complicated interphase interactions. Based on the data set from the fictitious domain method, the recurrent fully connected and convolutional parallel neural network (R-FC&CNN) architecture and its two comparable simplified models, that is, the recurrent fully connected neural network (R-FCNN) and the recurrent convolutional neural network (R-CNN) architectures, were constructed for predicting the gas−solid momentum exchange coefficient, β (kg•s −1 • m −3 ), average combustion rate per unit surface area of particles, r A / c (kg• s −1 •m −2 ), and comprehensive NaCl release parameter, γ, selectively. A time sequence of average particle temperature, T̅ (K), and particle volume fraction, ε, which can be extended in the matrix form, were constructed as the features selectively according to their correlation with the target physical quantity. The average relative error, δ̅ , and coefficient of determination, R 2 , were used as the evaluators. Through final testing, in the mild combustion domain, the R-CNN and R-FCNN models with simple structures showed good performance for β(δ̅ = 0.13, R 2 = 0.8) and r A / c (δ̅ = 0.04, R 2 = 0.84), respectively, while in the severe combustion domain, the R-FC&CNN model, with more complete features and functional structure, performed the best (for β, δ̅ = 0.12, R 2 = 0.85; for r A / c , δ̅ = 0.05, R 2 = 0.92; and for γ, = = R 0.05, 0.96 2). A fine-tuning and interpolated prediction method was developed to investigate further the model's expansibility. Both ensured acceptable performance on a new similar problem. In summary, the feasibility of physical meaningoriented machine learning--based model(s) for predicting the combustion characteristics of nonuniformly distributed particles was confirmed.