| Soybean is an important grain and oil crop in China.The production,consumption,and actual storage for soybeans in China are among the highest in the world.However,preservation of soybeans for long-term storage has always been a challenge due to infestation by various insects.The pest problems can make soybean pollution through associated microbial spoilage caused by their metabolic activities and contaminations,resulting in serious decline of soybean quality and great economic loss.Radio frequency(RF)technology has long been proposed as a potential alternative to chemical fumigation and applied for control of different insects for various agricultural products.However,the non-uniformity in RF heating is still a major problem in developing effective RF heat treatments for pest control and other applications.Computer simulation and mathematical modeling may serve as valuable tools to acquire deep insights on the heating uniformity of products and offer opportunity to clearly understand RF interactions with food components without the necessity of extensive experiments.Therefore,a coupled electromagnetic and thermal model was established with finite element-based commercial software,COMSOL Multiphysics for better understanding the unclear RF heating mechanism and causes of non-uniform heating.A systematic analysis of factors that influence the uneven RF power distribution within the RF cavity and treated products was studied from the aspects of selective heating,heating rate and heating uniformity based on both the experimental and simulated methods.The effective measures were proposed according to the simulation results to improve the heating uniformity of soybeans.Soybeans packed in a rectangular container were used to determine the heating uniformity and validate the simulation model using a 27.12 MHz,6 kW RF system.The validated model was used to optimize the treatment parameters,such as proposing some effective methods to adapt and interven the power domain on the basis of the features of RF power distribution,exploring the best heating protocol and parameters for RF heating and providing an optimal solution to improve the heating uniformity.The main results were described as follows:(1)The fundamental based thermal and electromagnetic models for RF heating were established based on the coupled Maxwell’s equations of electromagnetics with heat transfer equations.The physical and dielectric properties of bulk soybeans were measured in terms of the mathematical equations that describe the electromagnetic energy and heat transfer flux.The dielectric properties of soybean decreased with increasing frequency,and the specific heat capacity and thermal conductivity increased with the increase of temperature and moisture content.On this basis,the transient theoretical model for RF heating of soybeans was established.(2)To study the effect of soybean size on model precision,a single layer of spherical particles was developed using two different formulations of the particles-an approach that considers the geometry of individual particles and a simplified approach that considers the effective properties of an air-particle mixture.The spherical particles,prepared from soybean powder,represented dry granular foods.Model predictions were validated using point and spatial temperature measurements.Results showed that consideration of an individual particle was more appropriate for larger size particles(diameter > 1 cm)because of the temperature difference between the individual particles was larger and should not be ignored.For smaller particles(diameter ≤ 1 cm),consideration of effective properties of an air-particle mixture was adequate as the particle’s internal temperature range was much smaller(~3 °C)due to lower electric field and also very uniform temperature distribution over the cross-sections.(3)A homogenous porous medium model was established based on the different dielectric constant and loss factor values of 3 kg soybeans and insect larvae.Indianmeal moth larvae were selected as the target insect for experimental validation of the simulation results.Results from computer simulation and experimental methods both showed that the mean temperature differences between insects and soybeans at the top,middle,and bottom layers were 5.9,6.6,and 6.2 oC,respectively.This demonstrated the selective RF heating characteristics between pests and soybeans.Simulated results also showed that when the insect was placed on the top surface center,horizontally placed in the host medium,and large insect size would cause relatively slow heating rate and low average temperature.(4)For better understanding the complex RF heating process,analysis of the main factors that influence the heating rate was necessary to help us learn how process parameters evolve in space and time during heating.Therefore,a 3D numerical model based on the relationship between heating rate and model parameters was constructed and validated by experimental measurements.Simulation results confirmed that the heat capacity and density of sample affected the RF heating rate in similar trend.The RF heating rate had a logarithmic linear relationship with the electrode gap.It had higher heating rates for top surface or bottom surfaces of the samples,which was close to the electrode.The RF heating rate was significantly influenced by top electrode voltage,dielectric properties and moisture content of the sample.The RF heating rate changed 74 % when changing the dielectric properties of soybean by ± 20 %.The equal loss factor to the dielectric constant provided the fast RF heating rate.(5)The difference in heating rate greatly affects the heating uniformity of soybeans,so a heating uniformity evaluation model was developed and validated by measuring the transient temperature profiles of three layers of soybean.The validated model was used to predict the effects of electrode configuration,sample composition,and special dielectric materials around samples on the behaviour of RF heating uniformity.When the top electrode area was changed into 1.32 times of the sample size,the maximum temperature of middle layers was reduced from 71 oC to 57 oC,and the uniformity index was reduced from 0.14 to 0.07.On the middle layers,the difference between maximum and minimum temperatures was found to be reduced from 24 to 15 oC,and the uniformity index was reduced from 0.14 to 0.06 using similar dielectric constant material,which implies a better heating uniformity.When the sample was placed in the middle of two electrodes,the uniformity index was reduced from 0.14 to 0.12.The above results showed that these measures were useful in the optimization process of RF heating uniformity,which influenced the RF energy distributions significantly.(6)To study the influence of different surrounding materials on the energy distribution of RF field,an experimentally validated simulation model was developed to investigate the effect of applying a surrounding container(polypropylene and polystyrene)on RF heating uniformity improvement in soybeans.The temperature uniformity was greatly improved by placing soybean samples in the polystyrene container other than the polypropylene one.Using the container thickness of 8 cm,combined with corner radius of 8 cm significantly improved the RF heating uniformity as compared to a single method.The maximum temperature difference in the middle layer decreased from 20 °C to 5 °C,and the uniformity index was reduced from 0.122 to 0.085.Results generated above showed that this method can significantly improve the uniformity of RF power distribution,which is feasible for industrial applications.(7)In order to increase the practicability of RF heating technology,a new method for optimizing the heating uniformity of soybeans was established for food processing industry.To quantify effects of dielectric properties(DPs)and densities of a surrounding container and treated food products on RF heating uniformity,a simulation model was established and experiments were performed to validate the developed model.Results developed a linear relationship for dielectric constant and density of surrounding container and treated products with high coefficient of determination.The optimum RF heating uniformity of food products could be achieved when the loss factor values of surrounding container lied between 0.01-0.1 % of the sample’s,and dielectric constant/density consistent with the developed correlation equations.Therefore,this provides a solid foundation data and theoretical support for improving RF heating uniformity and designing an effective industrial-scale RF treatment protocols in disinfestations of agriculture products. |
PDF Full Text Request