**Application of Novel Thermal Technology in Foods Processing**

Editors

**Indrawati Oey Sze Ying Leong**

MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin

*Editors* Indrawati Oey University of Otago New Zealand

Sze Ying Leong University of Otago New Zealand

*Editorial Office* MDPI St. Alban-Anlage 66 4052 Basel, Switzerland

This is a reprint of articles from the Special Issue published online in the open access journal *Foods* (ISSN 2304-8158) (available at: https://www.mdpi.com/journal/foods/special issues/Application Novel Thermal Technology Foods Processing).

For citation purposes, cite each article independently as indicated on the article page online and as indicated below:

LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. *Journal Name* **Year**, *Volume Number*, Page Range.

**ISBN 978-3-0365-3002-4 (Hbk) ISBN 978-3-0365-3003-1 (PDF)**

© 2022 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications.

The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND.

## **Contents**



#### **Marcos Andr ´es Maza, Juan Manuel Mart´ınez, Guillermo Cebri ´an, Ana Cristina S ´anchez-Gimeno, Alejandra Camargo, Ignacio Alvarez and Javier Raso ´**


### **About the Editors**

**Indrawati Oey** is the head of the Food Science Department at the University of Otago, New Zealand and the principal investigator at the Riddet Institute. She received her MSc (1996) and PhD (2000) from Katholieke Universiteit Leuven (KULeuven), Belgium. Before joining the University of Otago as a professor of food science in 2009, she worked at the Research Foundation—Flanders (Belgium). Her research expertise is postharvest innovation and novel food processing technologies including advanced heating, high hydrostatic pressure, and pulsed electric fields that create food products with unique health and sensory characteristics in a sustainable way. She co-ordinated the training and career development for the EU-funded project "Novel processing methods for high-quality and safe foods—NovelQ". She is a fellow of the International Academic of Food Science and Technology (IAFoST) and the New Zealand Institute of Food Science and Technology (NZIFST). Professor Oey is an executive board member of Ag@Otago and a member of Food Waste Innovation, which measures food waste, develops reduction strategies, applies innovative technology, and works to modify producer and consumer behaviour.

**Sze Ying Leong** is a research fellow at the Department of Food Science of University of Otago, New Zealand. She received her BSc (2011) and PhD (2016) from the University of Otago. Currently, she is also an affiliate researcher with the Riddet Institute, a New Zealand Centre of Research Excellence with a focus on the effects of food processing (including emerging processing technologies such as high-pressure processing and pulsed electric fields) on plant-based foods and their resulting health benefits upon human digestion. Dr. Leong has recently completed a six-year programme (Food Industry Enabling Technology (FIET)) on food processing technology development, validation, and commercialisation, funded by the NZ Ministry of Business, Innovation, and Employment. She is co-author of 36 peer-reviewed research articles published in high-impact factor journals and has contributed 8 book chapters. Dr. Leong is a regular reviewer for many reputed journals in food science and technology. She is an active member of the New Zealand Institute of Food Science and Technology, the Institute of Food Technologies (Non-Thermal Processing Division), and the International Society for Electroporation-Based Technologies and Treatments and is a member of the Early Career Scientists Section of the International Union of Food Science and Technology.

### *Editorial* **Application of Novel Thermal Technology in Foods Processing**

**Sze Ying Leong 1,2 and Indrawati Oey 1,2,\***


Advanced and novel thermal technologies, such as ohmic heating and dielectric heating (e.g., microwave heating and radio frequency heating), have been developed to improve the effectiveness of heat processing whilst warranting food safety and eliminating undesirable impacts on the organoleptic and nutritional properties of foods. Novel thermal technologies rely on heat generation directly inside foods, which has implications for improving the overall energy efficiency of the heating process itself. The use of novel thermal technologies is dependent on the complexity and inherent properties of the food materials of interest (e.g., thermal conductivity, electrical resistance, water content, pH, rheological properties, food porosity, and presence of particulates). Moreover, there is a need to address the combined use of thermal processing with emerging technologies such as pulsed electric fields, high hydrostatic pressure and ultrasound to complement the conventional thermal processing of fluid or solid foods.

This Special Issue provides readers with an overview of the latest applications of various novel technologies in food processing. A total of eight cutting-edge original research papers and one comprehensive review paper discussing novel processing technologies from the perspectives of food safety, sustainability, process engineering, (bio)chemical changes, health, nutrition, sensory issues, and consumers are covered in this Special Issue.

Drying is a conventional thermal processing technique that is very effective in prolonging the shelf life of a food product by reducing microbial activities while facilitating its transportation and storage by decreasing the product weight and volume. The long drying time and decline in the product quality with drying duration has driven an urgent need to resolve these issues. Two approaches have recently been proposed: (i) the application of pretreatments such as microwave (MW) or ultrasound (US) on raw material prior to drying [1], and (ii) the development of either a hybrid drying process involving convective-infrared (IF) [1] or dielectric drying involving MW and radio frequency (RF) [2]. Using turnip slices as a case study, the work of Taghinezhad et al. [1] explored several independent variables such as pretreatments applied to the raw material prior to drying (MW [360 W for 2.5 min], US [30 ◦C for 10 min] and blanching [90 ◦C for 2 min], the temperature of the drying air (50, 60, and 70 ◦C) and the thickness of the materials (2, 4, and 6 mm) on the response variables including the quality indices (color difference and shrinkage) and drying factors (drying time, effective moisture diffusivity coefficient, specific energy consumption (SEC), energy efficiency and dryer efficiency) using a hybrid convective-IF dryer. The response surface method was used to optimize the drying process and the response variables were predicted by the adaptive neuro-fuzzy inference system model. The results indicated that an increase in the dryer temperature and a decline in the thickness of turnip slices can enhance the evaporation rate, which will decrease the drying time (from 40 to 20 min), SEC (from 168.98 to 21.57 MJ/kg), color difference (from 50.59 to 15.38) and shrinkage (from 67.84% to 24.28%) while increasing the effective moisture diffusivity coefficient (from 1.007 to 8.11 × <sup>10</sup>−<sup>9</sup> <sup>m</sup>2/s), energy efficiency (from 0.89% to 15.23%) and dryer efficiency (from 2.11% to 21.2%). Compared to US and blanching, MW pretreatment increased the energy and drying efficiency, while the variations in the color and shrinkage of products were

**Citation:** Leong, S.Y.; Oey, I. Application of Novel Thermal Technology in Foods Processing. *Foods* **2022**, *11*, 125. https://doi.org/ 10.3390/foods11010125

Received: 1 November 2021 Accepted: 28 December 2021 Published: 5 January 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the lowest in the US pretreatment. In another drying example using pecan nut kernels, the work of Zhang et al. [2] considered the dielectric properties (DPs) of raw material, an important factor in the design of effective MW and RF drying processes. The DPs of raw materials were investigated over frequencies ranging from 10 to 3000 MHz at moisture contents between 10 to 30% in a temperature range of 5–65 ◦C at varying salt strengths (mild, medium, and heavy). It was found that the dielectric constant (ε ) and loss factor (ε) of the kernels decreased significantly with increasing frequency in the RF band, but decreased gradually in the measured MW band. The moisture content and temperature increase greatly contributed to the increase in the ε and ε of samples, and ε increased sharply with increasing salt strength. Quadratic polynomial models were established to simulate DPs as functions of temperature and moisture content at four frequencies (27, 40, 915, and 2450 MHz), with R2 > 0.94. The average penetration depth of electromagnetic energy into the pecan kernels in the RF band was greater than that in the MW band (238.17 vs. 15.23 cm).

MW pasteurization and sterilization is an emerging thermal technology that combines preheating with hot water and MW energy (915 MHz) to achieve sterilization within a shorter time. However, a well-reported challenge with the MW processing of food is nonuniform heating, which could lead to the formation of cold spots in the processed products and is consequently unable to achieve the targeted sterilisation efficiency. Two research works from Soni et al. [3,4] addressed these food safety issues using novel validation method and tools. In their first study [3], novel spore pouches were developed using mashed potato as a food model inoculated with either *Geobacillus stearothermophilus* or *Clostridium sporogenes* spores to evaluate the sterilization efficiency of Coaxially induced MW pasteurization and sterilization (CiMPAS). The spore pouches were placed at predetermined specific locations, especially cold spots, in each food tray before being processed using two regimes (R-121 and R-65), which consisted of 121 ◦C and 65 ◦C at 12 and 22 kW, respectively, followed by recovery and enumeration of the surviving spores. To identify cold spots or the inoculation location, mashed potato was spiked with Maillard precursors and processed through CiMPAS, followed by measurement of lightness (*L\**) values. Inactivation equivalent to of 1–2 and >6 Log CFU/g for *G. stearothermophilus* and *C. sporogenes* spores, respectively, was obtained on the cold spots using R-121 (total processing time of 64.2 min). Inactivation of <1 and 2–3 Log CFU/g for *G. stearothermophilus* and *C. sporogenes* spores, respectively, on the cold spots was obtained using R-65 (total processing time of 68.3 min), whereas inactivation of 1–3 Log CFU/g of *C. sporogenes* spores was obtained on the sides of the tray. The results were reproducible across three processing replicates for each regime, and inactivation at the specific locations was clearly distinguishable. For their second study [4], hyperspectral imaging (HSI) was used to identify cold spots in CiMPAS-treated mashed potato and directly compared to the results of color changes due to the Maillard reaction after MW-induced sterilization. To visualize the HIS spectra of each tray in comparison with the control sample (raw mashed potato), the mean spectrum (i.e., mean of region of interest) of each tray, as well as the control sample, was extracted and then fed to the fitted principal component analysis model. The HIS results coincided with those post hoc analysis of the average reflectance values. Despite the presence of a visual difference in browning, the *L\** values were not significantly different to detect a cold spot among a range of 12 processed samples. At the same time, HSI could identify the colder trays among the 12 samples from one batch of microwave sterilization.

Apart from MW processing, the use of ohmic heating is another alternative technology to sterilize and pasteurize heat sensitive food products that can provide better thermal uniformity, a high heating rate and energy efficiency. The work of Joe et al. [5] took a novel approach developing an ohmic–vacuum combination (OH-VC) heating system to process a multiphase type of senior-friendly food. Changes in the physical and electrical properties of senior-friendly model foods were investigated depending on the experimental conditions such as vacuum pressure intensity and vacuum pretreatment time. Numerical simulations based on the experimental conditions were performed using COMSOL Multiphysics software. The OH-VC heating method with agitation reduced the heating time of the model food, and non-uniform temperature distribution in model food was successfully resolved due to the effects of vacuum and agitation. Furthermore, difference was found in the hardness of solid particles depending on the vacuum treatment time and intensity after the heating treatment.

In recent years, the use of pulsed electric field (PEF) technology has gained in popularity, particularly in the potato industry for the production of French fries and potato crisps, to "condition" the raw material (potato tubers) prior to subsequent processing (i.e., cutting, blanching and frying). However, the influence of PEF pretreatment on the frying process and related chemical reactions for food materials is still not fully understood. PEF treatment of plant tissue causes structural modifications, which are likely to influence heat, mass and momentum transfers, as well as alter the rate of chemical reactions during the frying process. Detailed insights into the frying process in terms of heat, mass (water and oil) and momentum transfers are outlined in a comprehensive review article by Xu et al. [6], in conjunction with the development of the Maillard reaction and starch gelatinization during frying. These changes occur during frying, and consequently impact oil uptake, moisture content, colour, texture and the amount of contaminants generated in fried foods, as well as the fried oil. Therefore, the effects of PEF pretreatment on these properties across a variety of fried plant-based foods are summarised in the review article. The different mathematical models used to potentially describe the influence of PEF on the frying process of plant-based foods and predict the quality parameters of fried foods produced from PEF-treated plant materials are also addressed in the review article.

In agreement with the review article by Xu et al. [6], the work of Gholamibozanjani et al. [7] conducted a timely investigation on the use of suitable heat and mass transfer models to predict temperature distribution during potato frying after pre-treatment with PEF. Meanwhile, the work of Abduh et al. [8] reported the kinetics of colour development during the frying of four potato cultivars pre-treated with PEF, in which the kinetic result can later aid in the optimisation of frying conditions for deep-fried potato industries where PEF technology is implemented. Based on an unsteady-state heat conduction, a mathematical model was developed by Gholamibozanjani et al. [7] to describe the simultaneous heat and moisture transfer during potato frying. For the first time, the equation was solved using both enthalpy and Variable Space Network methods, based on a moving interface defined by the boiling temperature of water in a potato disc during frying. Two separate regions of the potato disc, namely fried (crust) and unfried (core), were considered heat transfer domains. A variable boiling temperature of the water in potato discs was required as an input parameter for the model. This is because water is continuously evaporated during frying, resulting in an increase in the soluble solid concentration of the potato sample. The application of PEF pretreatment prior to frying had no significant effect on the measured moisture content, thermal conductivity or frying time compared to potatoes that did not receive PEF pretreatment. However, a PEF pretreatment at 1.1 kV/cm and 56 kJ/kg reduced the temperature variation in the experimentally measured potato center by up to 30%. On the other hand, the effect of PEF (1 kV/cm; 50 and 150 kJ/kg) followed by blanching (3 min., 100 ◦C) on the colour development (*L\**) of potato slices during frying was studied on a kinetic basis by Abduh et al. [8]. Four potato cultivars, 'Crop77', 'Moonlight', 'Nadine', and 'Russet Burbank', with different glucose and amino acid contents, were used. The implementation of PEF and blanching as a sequential pre-treatment prior to frying was found to be effective in improving the lightness of the fried products for all potato cultivars. PEF pre-treatment did not change the kinetics of *L\** reduction during frying (150–190 ◦C), which followed first-order reaction kinetics. The estimated reaction rate constant (*k*) and activation energy (*Ea* based on Arrhenius equation) for non-PEF and PEF-treated samples were cultivar-dependent. The estimated *Ea* values during the frying of PEF-treated 'Russet Burbank' and 'Crop77' were significantly lower (up to 30%) than their non-PEF counterparts, indicating that the change in the *k* value of *L\** became less temperature-dependent during frying.

The use of PEF can be also extended to red winemaking to achieve a similar outcome as the pre-fermentation heating (or thermovinification) of red grapes, which is a common practice in commercial wineries in Europe. PEF treatment applied to red grapes before the maceration-fermentation stage allows for a reduction in the contact time of grape skins, and helps to obtain wines with a higher polyphenolic content without involving heat in the vinification process. The work of Maza et al. [9] monitored the evolution of the polyphenolic compounds and sensory properties of wines obtained from Grenache grapes, either untreated or treated with PEF, during the course of either bottle aging or oak aging followed by bottle aging. Immediately prior to aging in bottles or in barrels, enological parameters that depend on phenolic extraction during skin maceration were higher when grapes were treated with PEF. In terms of color intensity, phenolic families, and individual phenols, the wine obtained with grapes treated by PEF followed a similar evolution to untreated control wine in the course of aging. Sensory analysis revealed that the application of a PEF treatment resulted in wines that are sensorially different, where panelists preferred wines obtained from grapes treated with PEF. Physicochemical and sensory analyses showed that grapes treated with PEF are suitable for obtaining wines that require aging in bottles or in oak barrels.

In summary, all the papers published in this Special Issue highlighted a large portion of the research activities in the field of advanced thermal processing, aiming to improve the efficiency of food processing, ensure food safety, enhance product quality and reduce food waste. The development of novel thermal-processing technologies and an exploration of the combined use of nonthermal processing to complement thermal processing will remain a very active research area in the coming decades. Future studies should consider the feasibility and applicability of these technologies or processing approaches for industrialscale environments.

**Funding:** This research received no external funding.

**Acknowledgments:** S.Y.L. and I.O. would like to thank all the contributors that published their works in this Special Issue, as well as the reviewers that have critically evaluated all submissions to ensure all works are published at their highest scientific quality. S.Y.L. and I.O. would also like to thank the publisher, MDPI, and the editorial staff of *Foods* for their professional support as well as for their invitation to edit this Special Issue.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Optimization and Prediction of the Drying and Quality of Turnip Slices by Convective-Infrared Dryer under Various Pretreatments by RSM and ANFIS Methods**

**Ebrahim Taghinezhad 1,\*, Mohammad Kaveh <sup>2</sup> and Antoni Szumny <sup>3</sup>**


**Abstract:** Drying can prolong the shelf life of a product by reducing microbial activities while facilitating its transportation and storage by decreasing the product weight and volume. The quality factors of the drying process are among the important issues in the drying of food and agricultural products. In this study, the effects of several independent variables such as the temperature of the drying air (50, 60, and 70 ◦C) and the thickness of the samples (2, 4, and 6 mm) were studied on the response variables including the quality indices (color difference and shrinkage) and drying factors (drying time, effective moisture diffusivity coefficient, specific energy consumption (*SEC*), energy efficiency and dryer efficiency) of the turnip slices dried by a hybrid convective-infrared (HCIR) dryer. Before drying, the samples were treated by three pretreatments: microwave (360 W for 2.5 min), ultrasonic (at 30 ◦C for 10 min) and blanching (at 90 ◦C for 2 min). The statistical analyses of the data and optimization of the drying process were achieved by the response surface method (RSM) and the response variables were predicted by the adaptive neuro-fuzzy inference system (ANFIS) model. The results indicated that an increase in the dryer temperature and a decline in the thickness of the sample can enhance the evaporation rate of the samples which will decrease the drying time (40–20 min), *SEC* (from 168.98 to 21.57 MJ/kg), color difference (from 50.59 to 15.38) and shrinkage (from 67.84% to 24.28%) while increasing the effective moisture diffusivity coefficient (from 1.007 <sup>×</sup> <sup>10</sup>−<sup>9</sup> to 8.11 <sup>×</sup> <sup>10</sup>−<sup>9</sup> <sup>m</sup>2/s), energy efficiency (from 0.89% to 15.23%) and dryer efficiency (from 2.11% to 21.2%). Compared to ultrasonic and blanching, microwave pretreatment increased the energy and drying efficiency; while the variations in the color and shrinkage were the lowest in the ultrasonic pretreatment. The optimal condition involved the temperature of 70 ◦C and sample thickness of 2 mm with the desirability above 0.89. The ANFIS model also managed to predict the response variables with *R*<sup>2</sup> > 0.96.

**Keywords:** blanching; drying; efficiency; energy; microwave; ultrasound

#### **1. Introduction**

The turnip has been long used in the human diet due to its high vitamin and mineral contents. Its use dates back to the prehistoric era. The turnip is cultivated in Europe and Iran, especially in cold regions [1]. Recently, the turnip has attracted the attention of consumers due to its high antioxidant content and anti-inflammatory, anti-diabetes, and anticancer features, in addition to its glucosinolates, flavonoids, and phenylpropanoid contents [2].

During the drying process, the moisture content of the product will be declined by heat and simultaneous mass transfer between the surroundings and sample surfaces. This

**Citation:** Taghinezhad, E.; Kaveh, M.; Szumny, A. Optimization and Prediction of the Drying and Quality of Turnip Slices by Convective-Infrared Dryer under Various Pretreatments by RSM and ANFIS Methods. *Foods* **2021**, *10*, 284. https://doi.org/10.3390/ foods10020284

Academic Editor: Sze Ying Leong Received: 25 December 2020 Accepted: 18 January 2021 Published: 31 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

process can be used as one of the important storage methods to prolong the shelf life of the product, reduce its transportation costs and minimize its packing requirements [3]. Drying can also prevent the spoilage and wastes of the crops after their harvest [4].

Among various industrial commercial dryers, convective dryers have found extensive applications in diverse industries including food and agriculture. This method, however, suffers from serious problems such as long processing time, low efficiency, high energy consumption rate, and declining quality of the product [5]. To resolve these issues, novel technologies such as hybrid dryers with the use of pretreatments can be employed [6].

The infrared method is one of the recent approaches used in the drying of food products, this method often applied in combination with convective methods and its goal is to accelerate the process of drying, reduce the energy consumption and improve the quality of the final product [7]; for instance, a hybrid convective-infrared (HCIR) dryer was used to dry blackberry [7,8] and potato [9]. Today, various pretreatments have been employed to reduce the drying time and improve the quality of the crops. Using these pretreatments, it is possible to reduce some of the unwanted variations such as textural and color changes [10]. So far, various pretreatments have been employed in the drying industry. Ultrasound and blanching pretreatments were used in a hybrid microwave-convective dryer to dry parsley leaves [11]. In another study, osmotic and ultrasound pretreatment were employed for drying strawberries under convective drying [12]. Ethanol and ultrasound were used as a pretreatment to dry potatoes using an infrared (IR) drying approach [13], citric acid and blanching were used to dry cauliflower using the convective dryer [10]. Other studies have been conducted by various dryers using different pretreatments to dry diverse crops for instance, blackberry [14], raspberries [15], Mirabelle plum [16], cranberry snacks [17], carrot discs [5], and cabbage [18]. These studies indicated that the use of these pretreatments can increase the effective moisture diffusion coefficient while reducing the drying time and specific energy consumption (*SEC*); hence improving the quality of the dried products. To the best of our knowledge, no study has addressed the influence of various pretreatments on the drying process of turnip slices using a hybrid convective-infrared (HCIR) dryer.

The relationship between the independent and dependent variables of the drying process is of crucial significance. Although some of the numerical methods have managed to some extent to resolve the complexity of the non-linear behavior. Due to the limitations of these methods, researchers have focused on other statistical methods such as adaptive neuro-fuzzy inference system (ANFIS) and response surface methods [19]. The neural-fuzzy deductive systems simultaneously exploit the merits of the artificial neural network and fuzzy logic. This method can be used to approximate the non-linear relationship between the inputs and outputs and has shown bright capabilities in the training, construction, and classification stages [20]. The response surface method (RSM) is a series of mathematical and statistical methods to model and analyze problems in which the response variable is under the influence of several independent variables. This method is aimed to optimize the response variables [21]. In RSM optimization, input variables are defined as the independent ones and their influence on the response (dependent) variable is explored. Numerous researchers have used RSM and ANFIS to model and optimize the quality and drying process of various crops including okra [22], quince [23], yacon [24], lavender leaves [25], and rough rice [26] for RSM and blackberries [8], almond [20], and yam slices [27] using the ANFIS method.

Regarding the importance of turnip in the human diet, its storage at high quality is of crucial significance. Previous studies have shown that no work has addressed the use of RSM and ANFIS to optimize and predict the quality and drying process of the turnip slices using HCIR dryers. In this regard, the aim of the present study is to model and optimize the effects of independent variables (slice thickness, temperature) on the dependent variables (drying time, effective moisture diffusion coefficient, *SEC*, energy efficiency, drying efficiency, shrinkage, and color) in drying turnip slices. In this research, turnip slices at the thicknesses of 2, 4, and 6 mm were dried by an HCIR dryer after various pretreatments (microwave, ultrasound, and blanching).

#### **2. Materials and Methods**

#### *2.1. Turnip Preparation*

Fresh turnips were provided from ParsAbad City, Ardebil province (Iran). The samples were kept in a refrigerator at + 4 ◦C. Prior to the experiments, the turnip samples were left at room temperature for 1 h. The initial moisture of the samples was determined at 10.23% (d.b.) using an oven (Memmert company, UFB50 model, Schwabach Germany) at 70 ◦C for 24 h.

#### *2.2. Pretreatments*

The pretreatments were carried out on the turnip samples before the drying process as follows:

#### 2.2.1. Blanching

For blanching pretreatment, a warm water bath (Memmert, WNB 14, Schwabach, Germany) was used. This bath had a maximum temperature of 120 ◦C and an accuracy of ±0.1 ◦C. The samples were placed in a warm bath at 90 ◦C for 2 min [17].

#### 2.2.2. Ultrasound Pre-Treatment

Ultrasound pretreatment was carried out using an ultrasonic bath (Parsonic, 7500s, Tehran, Iran) at the frequency of 28 kHz and power of 70 W regarding the constant frequency of the bath, turnip samples were immersed in distilled water at 30 ◦C and exposed to ultrasound waves for 10 min [28].

#### 2.2.3. Microwave Pre-Treatment

A domestic microwave oven (Panasonic NN-C2002W, Tokyo, Japan) at the frequency of 50 Hz and maximum heating power of 1000 W (with the capability of tuning the power at 90, 180, 360, 600, and 900 W) was employed for microwave pretreatment of the samples, the pretreatment was carried out at the power of 360 W for 2.5 min [29].

#### *2.3. Hybrid Convective-Infrared (HCIR) Dryer*

After pretreatments, the drying process was conducted using an HCIR dryer (GC 400 model, company Grouc, Tehran, Iran). This dryer includes two Infrared (IR) lamps (Philips model, Flemish, Belgium) working at the power of 500 W which are installed at the upper part of the drying chamber at the height of 30 cm the dryer has a centrifuge blower to blow hot air parallel to the substrate. To create the input air, a centrifuge fan equipped with an inverter (Vincker VSD2, ABB Co., Taipei, Taiwan) was employed. The input air speed was set at 1 m/s. The samples were placed on a meshed container on a digital balance (AND, GF-6000, A&D Company Ltd., Tokyo, Japan) at the accuracy of 0.01 g placed beneath the channel. The turnips were cut into 2, 4 and 6 mm thick pieces and pretreated using blanching, microwave and ultrasonic methods. The samples were then dried with a HCIR dryer at three temperature levels (50, 60 and 70 ◦C). Before the tests, the dryer was operated for 15 min to reach a constant temperature and air speed. In each test, one layer of 40 g turnip slice was placed on the dryer tray. During the drying process, the mean temperature and air humidity were 20 ± 4 ◦C and 15 ± 5%, respectively.

#### *2.4. Moisture Ratio*

Moisture ratio of the hybrid convective-infrared (HCIR)-dried turnip slices was determined by Equation (1) [1]:

$$MR = \frac{M\_t - M\_c}{M\_b - M\_c} \tag{1}$$

#### *2.5. Effective Moisture Diffusivity*

Fick's law, Equation (2), can describe the moisture transport in the descending stage of the drying process [30]:

$$\frac{\partial X}{\partial t} = D\_{\varepsilon f f} \frac{\partial^2 X}{\partial x^2} \tag{2}$$

The second Fick's law is related to the mass diffusivity during the descending phase of the drying process, using appropriate boundary conditions, it is possible to solve the Fick's equation for various geometries. For a thin layer, the Fick equation can be solved by Equation (3) [31]:

$$MR = \frac{8}{\pi^2} \sum\_{n=1}^{\infty} \frac{1}{\left(2n+1\right)} \exp\left(\frac{-D\_{eff}\left(2n+1\right)^2 \pi^2 t}{4L^2}\right) \tag{3}$$

The effective diffusivity coefficient can be determined from the slope of Equation (4) [17]:

$$\ln(MR) = \ln(\frac{8}{\pi^2}) - \ln\left(\frac{-D\_{eff}\pi^2 t}{4L^2}\right) \tag{4}$$

Generally, the diffusivity coefficient can be determined by plotting the experimental data of ln(*MR*) versus the time. The slope of the obtained line can be substituted in Equation (5) to determine the diffusivity coefficient [28]:

$$K = \left(\frac{D\_{eff}\pi^2}{4L^2}\right) \tag{5}$$

#### *2.6. Specific Energy Consumption (SEC), Energy and Drying Efficiency*

After the drying tests, the drying curve and hence the drying time can be determined for each specific condition. The specific energy consumption of the drying process can be obtained by Equation (6) [32]:

$$SEC = \left(\frac{E\_t}{\mathcal{M}\_w}\right) \tag{6}$$

Energy efficiency can be also determined by Equation (7) [33]:

$$
\eta\_{\mathcal{C}} = \left(\frac{E\_{cvap}}{E\_t}\right) \times 100\tag{7}
$$

The HCIR dryer efficiency can be calculated by the following equation [34]:

$$
\eta\_d = \left(\frac{E\_{\text{evap}} + E\_{\text{heating}}}{E\_t}\right) \times 100\tag{8}
$$

#### *2.7. Shrinkage Measurement*

Shrinkage refers to the variations in the sample volume relative to its initial volume. This phenomenon can be assigned to the water removal from the cellular space and its substitution with the air. During the drying process, the shape and size of the product may also change. The alterations in the physical properties can finally result in some changes in the final texture (shrinkage) of the dried products. Shrinkage can be determined by [35]:

$$S\_d = (1 - \frac{V\_t}{V\_0}) \times 100\tag{9}$$

In which *Sa* shows the shrinkage percentage, *Vt* denotes the apparent volume of the dried sample (cm3) after the time of t and *V*<sup>0</sup> represents the volume of the raw samples (cm3). The apparent volume of the samples was measured by the toluene displacement method using a glass pycnometer (50 mL) in this method, the samples with determined weight were transferred into a semi-filled pycnometer containing toluene. The remaining volume of the pycnometer was then closely filled with the solvent and its weight was measured. The apparent volume of the samples (*V*) can be determined by the following equations [36]:

$$V = V\_f - \frac{M\_{sf}}{\rho\_s} \tag{10}$$

$$M\_{sf} = M\_{t+s} - M\_f - M \tag{11}$$

*2.8. Color Difference*

Color is a significant factor in the evaluation of food products and their marketability [37]. To evaluate the color of the samples, a color meter was used to measure various parameters including *L* (lightness), *a* (red-green), and *b* (yellow-blue). The total color difference of the samples was also determined by Equation (12) [5].

$$
\Delta E = \sqrt{(L - L\_0^\*)^2 + (a - a\_0^\*)^2 + (b - b\_0^\*)^2} \tag{12}
$$

#### *2.9. Response Surface Methodology (RSM)*

In this research, the influence of the independent variables (drying air temperature in three levels of 50, 60, and 70 and the sample thickness in three levels of 2, 4, and 6 mm) on the dependent variables (drying time (min), effective moisture diffusivity (m2/s), *SEC* (Mj/kg), energy efficiency (%), drying efficiency (%), shrinkage (%), and color difference) was evaluated for the samples pretreated by microwave, ultrasound, and blanching.

For the predicted responses, it was assumed that:

$$y\_k = f\_k(\varepsilon\_1, \varepsilon\_2, \varepsilon\_3) \tag{13}$$

In which *yk* is the predicted response and *ε*1, *ε*<sup>2</sup> and *ε*<sup>3</sup> denote the natural (independent) variables. The second-order response surface equations are also presented in Equation (14) [25]:

$$y\_k = \beta\_0 + \sum\_{j=1}^k \beta\_j \mathbf{x}\_j + \sum\_{j=1}^k \beta\_{j\bar{j}} \mathbf{x}\_j^2 + \sum\_{i$$

In the above equation, *β*0, *βj*, *βjj*, and *βij* are the regression coefficients. *xj* also denotes the coded input variables. Design-expert software was used for fitting the response surfaces and optimize the drying process through solving a multiple regression equation (Equation (14)) using historical data and RSM. The mathematical models of each response were assessed by multiple linear regression analysis. The statistical significance of the independent variables for the response variables was explored at the confidence level of 95% (*p* < 0.05). Only the significant variables were included in the proposed regression equation. Finally, the optimal point of the process was determined according to the boundary conditions and the target functions as shown in Table 1.

**Table 1.** Boundary conditions and the independent and dependent variables.




#### *2.10. Adaptive Neuro-Fuzzy Inference System (ANFIS)*

Compatible deductive neural-fuzzy systems combine the ANN and fuzzy logic concepts and employ a series of if-then fuzzy laws. In this study, neural-fuzzy modeling was achieved using Matlab software. To this end, a Sugeno system was employed and the desirable membership function was determined among various functions (triangular, trapezoidal, bell-shaped, Gaussian, Pi, type-II Gaussian, and sigmoid). Their membership degree was also obtained by trial and error. A combinational training algorithm (including error back propagations algorithm and minimum square error method) was employed to train and match with the fuzzy deductive system. This model was used to predict the drying time, effective moisture diffusivity coefficient, *SEC*, energy efficiency, drying efficiency, color, and shrinkage of the turnip samples dried under various pretreatment conditions. ANFIS inputs were the input air temperature and the sample thickness. In the present study, 75% of the data were used for training, and the remaining 25% were used for validation. The model evaluation and comparison was carried out by the determination coefficient (*R*2), and mean square root error (*MSE*).

$$R^2 = 1 - \frac{\sum\_{i=1}^{N} (S\_k - T\_k)^2}{\sum\_{i=1}^{N} (S\_k - T\_m)^2} \tag{15}$$

$$MSE = \frac{1}{N} \sum\_{i=1}^{N} (S\_K - T\_k)^2 \tag{16}$$

#### **3. Results and Discussion**

#### *3.1. Drying Time*

Table 2 shows the results obtained from the RSM method for predicting the drying time of the turnip slices based on the independent variables (drying air temperature, and slice thickness) for various pretreatments. The drying air temperature and slice thickness had a significant effect on the drying time for all three pretreatments (*p* < 0.05). The fitted models were linear and second-order polynomial equations. The positive and negative signs of the estimated regression in the equations indicated the significant direct and indirect effects on the response variable, respectively (*p* < 0.05).

Figure 1, depicts the effect of the drying temperature and slice thickness on the drying time of the turnip samples for the three studied pretreatments using an HCIR dryer. According to Figure 1b, the shortest drying time (40 min) was for the drying air temperature of 70 ◦C and thickness of 2 mm for the sample pretreated by microwave. The longest drying time (250 min) was also recorded for the drying air temperature of 50 ◦C for the control samples with the thickness of 6 mm (Figure 1a) the decline in the thickness and the rise in the temperature could enhance the thermal gradient within the turnip samples, hence raising the moisture evaporation rate. The microwave pretreatment also led to a high pressure difference between the center and the surface of the product and incremented the drying rate; this will enhance the mass transfer, hence shortening the drying time [38]. Similar results were reported by the other researchers using a convective dryer and various pretreatments for drying blackberry [39], apple [29], potatoes [31], and black mulberry [7]. According to Figure 1c,d), ultrasound pretreatment also caused a significant (*p* < 0.05) reduction in the drying time, as compared with the blanching pretreatment. The shortest drying time for the ultrasound (140 min) and blanching (170 min) pretreatments were observed in the sample with a thickness of 2 mm dried at the temperature of 70 ◦C. The ultrasound-induced cavitation can lead to the formation of a series of microchannels in the product which can decrease the boundary layer of the propagation and enhance the mass transfer; this will, in turn, facilitate the water removal from the product [37]. These results are in line with the previous reports. For apple [38] and rose flower [40] drying, the drying time was significantly decreased by ultrasound pretreatment as compared with the blanching pretreatment.


**Table 2.** Response surface method (RSM) modeling results for predicting the drying time under a hybrid convective-infrared (HCIR) dryer with various pretreatments.

A: Drying temperature (◦C); B: Thickness (mm). *R*2: determination coefficient and *CV*: Coefficient of variation.

**Figure 1.** Effect of the drying temperature and sample thickness on the drying time (min) of the turnip slices dried under an hybrid convective-infrared (HCIR) dryer with various pretreatments (**a**) control, (**b**) microwave, (**c**) ultrasound, and (**d**) blanching.

13

#### *3.2. Effective Moisture Diffusivity Coefficient (Deff)*

Table 3 lists *Deff* results for various pretreatments at the studied temperature and thicknesses. *R*<sup>2</sup> was larger than 0.6 indicating that the demonstrated models were the best models for predicting the value of *Deff*. According to Table 3, *Deff* showed a linear and significant variation in different pretreatments (*p* < 0.05).

**Table 3.** Response surface method (RSM) modeling for predicting effective moisture diffusivity coefficient (*Deff*) under a hybrid convective-infrared (HCIR) dryer with various pretreatments.


A: Drying temperature (◦C); B: Thickness (mm). *R*2: determination coefficient and *CV*: Coefficient of variation.

Figure 2 shows the influence of the air temperature and turnip thickness on *Deff* for an HCIR dryer with various pretreatments. The highest *Deff* value (8.11 × <sup>10</sup>−<sup>9</sup> m2/s) was observed for the microwave-pretreated samples dried at the temperature of 70 ◦C and thickness of 2 mm (Figure 2b); while the lowest *Deff* (1.007 × <sup>10</sup>−<sup>9</sup> m2/s) was recorded for the control samples with the thickness of 6 mm dried at 50 ◦C (Figure 2a). Other researchers reported the effective moisture diffusivity in the range of 5.47 × <sup>10</sup>−<sup>10</sup> to 4.82 × <sup>10</sup>−<sup>9</sup> m2/s [1,41]. Based on Figure 3, an increase in the input air temperature and a decline in the sample's thickness can raise *Deff*. At high temperatures, the free water of the sample can be evaporated rapidly, hence dramatically reducing the drying time and increasing *Deff*. The use of microwave pretreatment is also enhanced, compared to the other pretreatments. By polarizing the water molecules, the microwave increased the internal temperature of the product. Moreover, it destroyed the product texture and formed channels with larger diameters, thus preventing the surface from hardening, hence accelerating the free water evaporation. *Deff* will decrease as a result of a decline in the drying time [33]. Similar results were reported by other researchers for cranberry snacks [17], blackberry [30], and okra [42]. They declared that the use of different pretreatments can increase the moisture diffusivity coefficient compared to the control samples.

Based on Figure 2c,d, *Deff* was higher in the ultrasonic pretreatment as compared with the blanching as ultrasonic treatment could open capillary paths due to the dispersion of the surface species; giving rise to longer microscopic channels as a result of the deformation of the cell. Therefore, ultrasonic pretreatment can deform and destroy the cell walls and accelerate moisture evaporation [38]. These results are in line with the previous reports by other researchers [30,39].

**Figure 2.** Effect of the drying temperature and sample thickness on the effective moisture diffusivity coefficient (*Deff*) (m2/s) of the turnip slices dried under an HCIR dryer with various pretreatments (**a**) control, (**b**) microwave, (**c**) ultrasound, and (**d**) blanching.

**Figure 3.** Effect of the drying temperature and sample thickness on the specific energy consumption (*SEC*, MJ/kg) of an HCIR dryer with various pretreatments (**a**) control, (**b**) microwave, (**c**) ultrasound, and (**d**) blanching.

#### *3.3. Specific Energy Consumption (SEC)*

Table 4 shows the modeling results for *SEC* of drying turnip slices at various temperatures and sample thicknesses and pretreatments. Based on this table, the linear variables of air temperature and sample thickness could significantly (*p* < 0.05) affect *SEC* in different pretreatments. *R*<sup>2</sup> was larger than 0.84, indicating the suitability of this linear model for predicting the value of *SEC*. It must be noted that only the coefficients with significant (*p* < 0.05) impact on *SEC* are included in the equation.

**Table 4.** Modeling results by the use of RSM for prediction of specific energy consumption (*SEC*) under an HCIR dryer with various pretreatments.


A: Drying temperature (◦C); B: Thickness (mm).

Figure 3, depicts the effects of the temperature of the drying air and the sample thickness on the value of *SEC* for various pretreatments. The highest *SEC* (168.98 MJ/kg) was related to the control samples with the thickness of 6 mm dried at 50 ◦C (Figure 3a); while the lowest *SEC* (21.57 kJ/kg) was observed for the microwave-pretreated samples with the thickness of 2 mm dried at 70 ◦C (Figure 3b). Similar results were reported by the other researchers in drying black mulberry [7], blackberry [39], and apple [34] using convective dryer under different pretreatments. They indicated that microwave-treated and control samples had the lowest and highest *SEC* values, respectively. In the current study, microwave pretreatment declined the *SEC* compared to the other two pretreatments. Using microwave pretreatments, the destruction in the texture of the product will be enhanced which will elevate the moisture removal rate; hence declining the *SEC* value [32]. Compared to blanching pretreatment, ultrasonic pretreatment led to lower *SEC* values (Figure 3c,d). Food products such as turnip will form a hard layer on their surface following the moisture removal which may decelerate the evaporation. Ultrasonic pretreatment prevents the formation of this layer, hence increasing the moisture removal rate, shortening the drying time, and hence reducing the *SEC* value [20]. Similar results were reported for drying parsley leaves by a microwave-convective dryer [11] and blackberry by an HCIR dryer [43]; as they showed that ultrasound pretreatment can result in lower *SEC* values, compared to the blanching pretreatment.

#### *3.4. Energy (ηe) and Dryer (ηd) Efficiency*

Table 5 lists the results obtained by modeling the effects of drying air temperature and sample thickness on the energy and dryer efficiency using an HCIR dryer with different pretreatments. Under all the studied conditions, *R*<sup>2</sup> was above 0.89 for the energy efficiency and above 0.8 for the dryer efficiency indicating that these models can predict the energy and dryer efficiencies well. Under the ultrasound pretreatment, the influence of the input air temperature and sample thickness was significant (*p* < 0.05) through a secondorder equation; while for the other pretreatment, these effects were linear and significant (*p* < 0.05). The variations in the dryer efficiency followed a second-order equation for the microwave and control samples; whereas the other pretreatments showed linear significant variation trends (*p* < 0.05).


**Table 5.** RSM modeling of the energy and dryer efficiencies under an HCIR dryer with different pretreatments.

A: Drying temperature (◦C); B: Thickness (mm). Energy (*ηe*) and dryer (*ηd*) efficiency.

A comparison of Figures 4 and 5 indicated that the elevation of the temperature enhanced the energy and dryer efficiencies. Temperature can augment the rate of moisture removal and hence decline the drying time; therefore, both the efficiencies will show ascending trends with temperature enhancement. With an increase in the sample thickness, the energy and dryer efficiencies declined as the drying time was increased. On the other hand, comparing the studied pretreatments showed that the highest efficiencies can be achieved using the microwave pretreatment (Figures 4b and 5b); while the control samples exhibited the lowest efficiencies (Figures 4a and 5a). Microwave pretreatment destroyed the product texture and accelerated moisture removal. Results have shown that an increase in the drying temperature and a decline in the thickness of the sample can improve both energy and dryer efficiencies. As shown in Figure 4c,d, ultrasonic and blanching pretreatments enhanced the destruction in the product texture, hence no hard layer will be formed during the drying process, and therefore the product will be dried faster. Energy efficiency varied from 0.89% to 6.48% for the controls, 5.99% to 15.23% for the microwave pretreatment, 4.88% to 9.87% for the ultrasound pretreatment, and 1.90% to 7.77% for the blanching pretreatment. The dryer efficiency of the control, microwave, ultrasound, and blanching pretreatments, varied in 2.11–9.11%, 3.45–9.99%, 5.77–11.54%, and 8.74–21.4%, respectively. Other researchers have also shown that various pretreatments can enhance the energy and drying efficiencies [34].

**Figure 4.** Effect of the drying temperature and sample thickness on the energy efficiency (%) of an HCIR dryer with various pretreatments (**a**) control, (**b**) microwave, (**c**) ultrasound, and (**d**) blanching.

**Figure 5.** Effect of the drying temperature and sample thickness on the drying efficiency (%) of an HCIR dryer with various pretreatments (**a**) control, (**b**) microwave, (**c**) ultrasound, and (**d**) blanching.

#### *3.5. Shrinkage*

Table 6 lists the coefficients of the equations obtained by the fitted models for the shrinkage parameter. The air temperature and sample thickness could significantly affect the shrinkage of the samples (*p* < 0.05). Table 6 also shows *R*2, adj-*R*2, Pre-*R*2, and *CV* values. Regarding high *R*<sup>2</sup> values (above 0.97), the presented model is the best one for predicting the shrinkage level of the samples.

**Table 6.** RSM modeling for predicting shrinkage of the turnip samples under an HCIR dryer with different pretreatments.


A: Drying temperature (◦C); B: Thickness (mm).

Figure 6, shows the influence of the drying air temperature and sample thickness on the shrinkage of the samples pretreated by different methods. As seen, the highest shrinkage can be observed in the control samples while the ultrasound-pretreated samples exhibited the lowest shrinkage (Figure 6a). A comparison of the pretreatments indicated that blanching led to the highest shrinkage as the intercellular water was replaced by air which led to stress in the cell structure, hence the texture failed in maintaining its structure (Figure 6d). As a result, the extracellular structure will collapse resulting in higher shrinkage [29].

The shrinkage increased by increasing the temperature and sample thickness. An increment in the drying temperature enhanced the thermal gradient between the product and the environment, promoting the moisture migration from the internal layers to the sliced layers; this will cause a moisture gradient between the surface and internal layers and hence augment the shrinkage [18]. By drying mushrooms [44] and barley seeds [35] at various temperatures, other researchers also showed an increase in the shrinkage by the temperature elevation. The reason for the increased shrinkage in thicker samples can be explained as follows: a rise in the sample thickness will reduce the water release of the cell and hence decline the stress applied to the cell by the liquid. Such a decline in the stress will enhance the textural shrinkage. The shrinkage of the turnip samples pretreated by microwave (Figure 6b), ultrasound (Figure 6c), and blanching (Figure 6d) method, as well as the controls, varied from 24.28–46.67%, 19.28–42.49%, 26.20–52.21%, and 36.36–67.84%, respectively.

**Figure 6.** Effect of the drying temperature and sample thickness on the shrinkage (%) of the turnip slices dried under an HCIR dryer with various pretreatments (**a**) control, (**b**) microwave, (**c**) ultrasound, and (**d**) blanching.

#### *3.6. Color Difference (*Δ*E)*

As presented in Table 7, the drying air temperature and sample thickness linearly and significantly altered Δ*E* of the dried turnip (*p* < 0.059). *R*2, adj-*R*2, and Pre-*R*<sup>2</sup> values of Δ*E* index were above 0.97, above 0.96, and above 0.92. Therefore, the presented equations can well fit the experimental data.

**Table 7.** RSM modeling for predicting color difference (Δ*E*) of the samples dried under an HCIR dryer with different pretreatments.


A: Drying temperature (◦C); B: Thickness (mm).

Figure 7 shows the effects of the drying temperature and sample thickness on Δ*E* of the turnip samples dried by an HCIR dryer for different pretreatments. An increase in the temperature and sample thickness enhanced the Δ*E* value since an increase in these two factors implies drying at higher temperatures which will result in browning reactions and an increase of the brunt areas on the sample surface [11]. Similar results were reported on the variations of Δ*E* during drying different products such as almond kernel [45] and cabbage [18].

**Figure 7.** Effect of the drying temperature and sample thickness on the color difference of the turnip slices dried under an HCIR dryer with various pretreatments (**a**) control, (**b**) microwave, (**c**) ultrasound, and (**d**) blanching.

The highest color variation (Δ*E*) was 50.59 and observed in the control samples with the thickness of 6 mm dried at 70 ◦C (Figure 7a); while the lowest color difference (11.12) was for the ultrasound-treated samples with the thickness of 2 mm dried at 50 ◦C (Figure 7c). The results indicated that the color indices were closer to the fresh samples when the products were thinner and dried at lower temperatures. According to Figure 7, the studied pretreatments caused some color variations. Similar results were also reported for other agricultural products such as mushrooms [44], star anise [46], cranberry snacks [17], and blackberry [39].

#### *3.7. Optimization*

Table 8 lists the optimized values of the independent and response variables along with their desirability function based on the desirability index. The optimal independent variables were drying temperature of 70 ◦C and thickness of 2 mm for all the pretreatments and control samples (accuracy over 0.89). Under this optimal condition, the response variables such as drying time (11.33 min), *SEC* (59.31 MJ/kg), shrinkage (54.87%) and color variation (40.83) were minimized while, *Deff* (2.15 × <sup>10</sup>−<sup>9</sup> <sup>m</sup>2/s) energy efficiency (6.64%) and dryer efficiency (9.13) showed their maximal levels. Other researchers also used the RSM method to optimize the drying process of various crops including apricots [45], lavender leaves [25], sunflower seeds [21], and pistachio [47].


**Table 8.** Optimization of the response parameters for turnip drying under an HCIR dryer with different pretreatments by RSM.

#### *3.8. ANIFIS*

Table 9 presents the results obtained by the ANFIS model to predict the drying time, *Deff*, *SEC*, energy and dryer efficiencies, shrinkage, and color variation of the dried turnip samples using an HCIR dryer. To measure the performance of the model, developed equations and two statistical functions, root mean square error (RMSE) and determination coefficient (*R*2), were used. In this table the lowest RMSE and highest *R*<sup>2</sup> are presented. According to Table 7, *R*<sup>2</sup> of prediction of drying time, *Deff*, *SEC*, energy efficiency, dryer efficiency, shrinkage, and color were 0.9965, 0.989, 0.000, 0.9993, 0.9989, and 0.9990, respectively (other pretreatments are shown in Table 9). According to Table 9, it can be concluded that the ANFIS model offered higher accuracy for all the studied parameters as compared with the RSM model. By drying almonds [20] and blackberry [30], the researchers have shown that the ANFIS model can successfully predict the drying properties of the products.

**Table 9.** Prediction of the response parameters for turnip drying under an HCIR dryer with different pretreatments by ANFIS.


#### **4. Conclusions**

In this study, drying time, *Deff*, *SEC*, energy efficiency, drying efficiency, color, and shrinkage of the turnip samples dried by an HCIR dryer were evaluated under various pretreatments (microwave, ultrasonic, and blanching). The following results were obtained:


the process; while the ANFIS method did not have this capability. ANFIS, however, showed better performance in predicting the dependent variables.

This study provides an in-depth understanding of the drying kinetics, and energy consumption, energy efficiency and quality properties (shrinkage and color) of HCIR drying process with four pretreatments, which will be helpful for the selection of pretreatment methods in the turnip industry.

**Author Contributions:** Conceptualization, E.T. and M.K.; methodology, E.T. and M.K.; validation, E.T. and M.K.; formal analysis, E.T.; investigation, E.T. and M.K.; resources, E.T. and A.S.; data curation, E.T. and M.K.; writing—original draft preparation, E.T. and A.S.; writing—review and editing, E.T., A.S.; and M.K.; visualization, E.T. and A.S.; funding acquisition, E.T. and A.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the office of vice chancellor for research at Mohaghegh Ardabili University (by a research project with Number: 99.d.9.8954 and date 2020.08.15), and Wrocław University of Environmental and Life Sciences.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data for this research will available.

**Acknowledgments:** The authors are highly thankful to Department of Agricultural Technology Engineering, Mohaghegh Ardabili University, Ardabil, Iran for providing facilities to conduct this research work.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**



#### **References**

