Euler’s First-Order Explicit Method–Peridynamic Differential Operator for Solving Population Balance Equations of the Crystallization Process

Using Euler’s first-order explicit(EE)method and the peridynamic differential operator(PDDO)to discretize the time and internal crystal-size derivatives,respectively,the Euler’s first-order explicit method–peridynamic differential operator(EE–PDDO)was obtained for solving the one-dimensional population balance equation in crystallization.Four different conditions during crystallization were studied:size-independent growth,sizedependent growth in a batch process,nucleation and size-independent growth,and nucleation and size-dependent growth in a continuous process.The high accuracy of the EE–PDDO method was confirmed by comparing it with the numerical results obtained using the second-order upwind and HR-van methods.The method is characterized by non-oscillation and high accuracy,especially in the discontinuous and sharp crystal size distribution.The stability of the EE–PDDO method,choice of weight function in the PDDO method,and optimal time step are also discussed.

An analytical solution for the population balance equation using a moment method

Brownian coagulation is the most important inter-particle mechanism affecting the size distribution of aerosols. Analytical solutions to the governing population balance equation （PBE） remain a challenging issue. In this work, we develop an analytical model to solve the PBE under Brownian coagulation based on the Taylor-expansion method of moments. The proposed model has a clear advantage over conventional asymptotic models in both precision and efficiency. We first analyze the geometric standard deviation （GSD） of aerosol size distribution. The new model is then implemented to determine two analytic solu- tions, one with a varying GSD and the other with a constant GSD, The varying solution traces the evolution of the size distribution, whereas the constant case admits a decoupled solution for the zero and second moments, Both solutions are confirmed to have the same precision as the highly reliable numerical model, implemented by the fourth-order Runge-Kutta algorithm, and the analytic model requires significantly less computational time than the numerical approach. Our results suggest that the proposed model has great potential to replace the existing numerical model, and is thus recommended for the study of physical aerosol characteristics, especially for rapid predictions of haze formation and evolution,

Multi-head neural networks for simulating particle breakage dynamics

The breakage of brittle particulate materials into smaller particles under compressive or impact loads can be modelled as an instantiation of the population balance integro-differential equation.In this paper,the emerging computational science paradigm of physics-informed neural networks is studied for the first time for solving both linear and nonlinear variants of the governing dynamics.Unlike conventional methods,the proposed neural network provides rapid simulations of arbitrarily high resolution in particle size,predicting values on arbitrarily fine grids without the need for model retraining.The network is assigned a simple multi-head architecture tailored to uphold monotonicity of the modelled cumulative distribution function over particle sizes.The method is theoretically analyzed and validated against analytical results before being applied to real-world data of a batch grinding mill.The agreement between laboratory data and numerical simulation encourages the use of physics-informed neural nets for optimal planning and control of industrial comminution processes.

MULTI-MONTE-CARLO METHOD FOR GENERAL DYNAMIC EQUATION CONSIDERING PARTICLE COAGULATION

Monte-Carlo （MC） method is widely adopted to take into account general dynamic equation （GDE） for particle coagulation, however popular MC method has high computation cost and statistical fatigue. A new Multi-Monte-Carlo （MMC） method, which has characteristics of time-driven MC method, constant number method and constant volume method, was promoted to solve GDE for coagulation. Firstly MMC method was described in details, including the introduction of weighted fictitious particle, the scheme of MMC method, the setting of time step, the judgment of the occurrence of coagulation event, the choice of coagulation partner and the consequential treatment of coagulation event. Secondly MMC method was validated by five special coagulation cases in which analytical solutions exist. The good agreement between the simulation results of MMC method and analytical solutions shows MMC method conserves high computation precision and has low computation cost. Lastly the different influence of different kinds of coagulation kernel on the process of coagulation was analyzed： constant coagulation kernel and Brownian coagulation kernel in continuum regime affect small particles much more than linear and quadratic coagulation kernel,whereas affect big particles much less than linear and quadratic coagulation kernel.

Bubble size modeling approach for the simulation of bubble columns

The constant bubble size modeling approach(CBSM)and variable bubble size modeling approach(VBSM)are frequently employed in Eulerian–Eulerian simulation of bubble columns.However,the accuracy of CBSM is limited while the computational efficiency of VBSM needs to be improved.This work aims to develop method for bubble size modeling which has high computational efficiency and accuracy in the simulation of bubble columns.The distribution of bubble sizes is represented by a series of discrete points,and the percentage of bubbles with various sizes at gas inlet is determined by the results of computational fluid dynamics(CFD)–population balance model(PBM)simulations,whereas the influence of bubble coalescence and breakup is neglected.The simulated results of a 0.15 m diameter bubble column suggest that the developed method has high computational speed and can achieve similar accuracy as CFD–PBM modeling.Furthermore,the convergence issues caused by solving population balance equations are addressed.

Simulation of gas-solid adsorption process considering particle-size distribution

The particle-size distribution of adsorbents usually plays an important role on the adsorption performance. In this study, population balance equation(PBE) is utilized in the simulation of an adsorption process to model the time-dependent adsorption amount distribution on adsorbent particles of a certain size distribution. Different adsorption kinetics model can be used to build the adsorption rate function in PBE according to specific adsorption processes. Two adsorption processes, including formaldehyde on activated carbon and CO_(2)/N_(2)/CH_(4) mixture on 4A zeolite are simulated as case studies, and the effect of particle-size distribution of adsorbent is analyzed. The simulation results proved that the influence of particle-size distribution is significant. The proposed model can help consider the influence of particlesize distribution of adsorbents on adsorption processes to improve the prediction accuracy of the performance of adsorbents.

Estimation of secondary nucleation kinetics of benzoic acid in batch crystallizer

The nucleation and growth kinetics of benzoic acid were determined in a population balance model,describing the seeded batch antisolvent crystallization process.The process analytical technologies(PATs)were utilized to record the evolution of chord length distributions(CLDs)in solid phase together with the concentration decay in liquid phase,which provided essential experimental information for parameter estimation.The model was solved using standard method of moments based on the moments calculated from CLDs and solute concentration.A developed model,incorporating the nucleation and crystal growth as functions of both supersaturation and solvent composition,has been constructed by fitting the zeroth moment of particles and concentration trends.The determined kinetic parameters were consequently validated against a new experiment with a different flow rate,indicating that the developed model predicted crystallization process reasonably well.This work illustrates the strategy in construct a population balance model for further simulation,model-based optimization and control studies of benzoic acid in antisolvent crystallization.

Solution of Smoluchowski coagulation equation for Brownian motion with TEMOM

The particle number density in the Smoluchowski coagulation equation usually cannot be solved as a whole,and it can be decomposed into the following two functions by similarity transformation:one is a function of time(the particle k-th moments),and the other is a function of dimensionless volume(self-preserving size distribution).In this paper,a simple iterative direct numerical simulation(iDNS)is proposed to obtain the similarity solution of the Smoluchowski coagulation equation for Brownian motion from the asymptotic solution of the k-th order moment,which has been solved with the Taylor-series expansion method of moment(TEMOM)in our previous work.The convergence and accuracy of the numerical method are first verified by comparison with previous results about Brownian coagulation in the literature,and then the method is extended to the field of Brownian agglomeration over the entire size range.The results show that the difference between the lognormal function and the self-preserving size distribution is significant.Moreover,the thermodynamic constraint of the algebraic mean volume is also investigated.In short,the asymptotic solution of the TEMOM and the self-preserving size distribution form a one-to-one mapping relationship;thus,a complete method to solve the Smoluchowski coagulation equation asymptotically is established.