Available online at www.sciencedirect.com
ScienceDirect
Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available Energy Procedia 00 (2017) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia
Energy (2017) 000–000 354–359 EnergyProcedia Procedia138 00 (2017) www.elsevier.com/locate/procedia
2017 International Conference on Alternative Energy in Developing Countries and Emerging Economies 2017 AEDCEE, 25‐26 May 2017, Bangkok, Thailand
Theof 15thThin International on Districtfor Heating and Cooling Study LayerSymposium Drying Model Cassava Pulp a b Charmongkolpradit , Ratinun Luampon * AssessingSuparerk the feasibility of using the heat demand-outdoor Department of Mechanical Engineering, Faculty of Engineering, Rajamangala University of Technology Isan Khonkaen Campus temperature function for a long-term district heat demand forecast 150, Sri Chant Rd., Naimuang, Muang, Khonkaen, 40000, Thailand a a
bb
Department of Mechanical Engineering, Faculty of Engineering and Architecture, Rajamangala University of Technology Isan a,b,c 744, Suranarai a Rd., Naimuang, aMuang, Nakhonratchasima, b c 30000, Thailand c
I. Andrić
*, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrière , O. Le Corre
a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France c Abstract Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France
This thesis article was purposes to study on the thin layer drying of cassava pulp using the Model BINDER Incubator BD 53 dryer. In experimental the cassava pulp sample with initial moisture content at 80 % w.b. was drying at temperature 50, 60 and 70Abstract C and collected mass of cassava pulp at interval 20 minute until mass not change. The experimental data was analyzed to evaluate the thin layer drying models and guideline for developed a new empirical drying equation to be suitable for predicting the moisture ratio networks of cassavaare pulp. From experimental, was of decreased increased dryingfor time and drying District heating commonly addressed inthe themoisture literatureratio as one the mostwith effective solutions decreasing the time was decreased with increased temperature. Thesystems results analysis of thin layer drying models, the Midilli model greenhouse gas emissions from thedrying building sector. These require high investments which are returned through thewas heat 2 good predicted with experimental data.and It had coefficient of determination (R2)demand and rootinmean square could error (RMSE) sales.agreement Due to the changed climate conditions building renovation policies, heat the future decrease, better than another model. The newly developed equation had high accuracy with experimental data and had correlation values prolonging the investment return period. R22 = 0.9938 and RMSE = 0.0264 for drying at temperature 50 C - 70 C. The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand The districtPublished of Alvalade, locatedLtd. in Lisbon (Portugal), was used as a case study. The district is consisted of 665 ©forecast. 2017 The Authors. by Elsevier Published by ©buildings 2017 The Authors. Elsevierperiod Ltd. and typology. Three weather scenarios (low, medium, high) and three district vary in both construction Peer-reviewthat under responsibility of the scientific committee of the 2017 International Conference on Alternative Energy in Peer-review under responsibility of the Organizing Committee of 2017 AEDCEE. Drenovation eveloping Countries Economies. scenarios and wereEmerging developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were compared with results from a dynamic heat demand model, previously developed and validated by the authors. Keywords: Thin layer drying models; Moisture ratio; Cassava pulp The results showed that when only weather change is considered, the margin of error could be acceptable for some applications (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and improve the accuracy of heat demand estimations. © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and * Corresponding author. Tel.: +66-840-307-600. Cooling. E-mail address:
[email protected].
Keywords: Heat demand; Forecast; Climate change 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Organizing Committee of 2017 AEDCEE.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of the 2017 International Conference on Alternative Energy in Developing Countries and Emerging Economies. 10.1016/j.egypro.2017.10.138
2
Suparerk Charmongkolpradit et al. / Energy Procedia 138 (2017) 354–359 Suparerk Charmongkolpradit et al./ Energy Procedia 00 (2017) 000–000
355
1. Introduction The cassava or tapioca which is known biologically as “Manihot esculenta crantz”, is one of the important economics crops in Thailand. Thailand is the first major cassava exporter. Even though the major produce cassava of the world are Nigeria, Brazil, Indonesia, Thailand and Congo [1]. Thailand cassava productions during 2013 to 2014 were yields 28.745 million tons of fresh roots with 2.982 million rai of cultivated area [2]. Fresh root from farm would be produce for chip and pellets of 10 to 12 million tons and for starch is around 12 to 15 million tons [3]. However, production of cassava starch results in formation of 15 to 20% of the original processed root dry weight basis as solid waste. This waste is well known as “pulp”. Cassava pulp is retaining a high amount of carbohydrates and fiber. It can be used in biomass, ethanol, animal feeds and paper pulp. On the other hand, cassava pulp is still remains high moisture content (75-80% w/w), it may be causing environmental problems that including a strong and offensive putrefaction odor and local water contamination [4]. If those pulps could be dried that would help to solve exists problem. Simple dehumidifier of cassava pulps is sun drying but it may have some problem as required large area, relies on the weather condition and poor product quality. However, this study aims to solve this problem by implement the drying technology for dehumidified cassava pulp. More information on properly design and analysis that affected the vital parameters such as: equilibrium moisture content, moisture ratio, drying rate, density, specific heat, etc. is needed for studies. So, the main objective of this article was study the thin layer drying model for predicted the moisture ratio of cassava pulp. This study would not only reduce the environmental problem of cassava pulp from the industry but also add value for cassava crops [5] and drying data could describe the drying characteristic of cassava pulp. Nomenclature M M0 Me MR MRexp,i MRpre,i N n
Moisture content (w.b.) Initial moisture content (w.b.) Equilibrium moisture content (w.b.) Moisture ratio (decimal) The ith experimentally observed moisture ratio The ith predicted moisture ratio The number of observations The number of constant in thin layer model
2. Thin layer drying model The moisture ratio of cassava pulp during the thin layer drying experiments was using the following equation:
MR
M Me M0 Me
(1)
For the thin layer drying models in Table 1 were tested to select the best model for describing the drying curve of cassava pulp. The non-linear regression was performed using computer program. The coefficient of determination (R2) was the primary criterion for selecting the best equation to describe the drying curve [6]. In addition, the root mean square error analysis (RMSE) was used to determine the best fit. This parameter can be calculated as equation 2.
1 RMSE N
N
MR i 1
pre ,i
MRexp,i
2
1/ 2
(2)
Where MRexp,i is the ith experimentally observed moisture ratio, MRpre,i the ith predicted moisture ratio, N the number of observations and n is the number of constant in each thin layer model.
356
Suparerk Charmongkolpradit et al. / Energy Procedia 00 (2017) Suparerk Charmongkolpradit et al. / Energy Procedia 138000–000 (2017) 354–359
Table 1. List of thin layer models Name 1 .Lewis 2 .Page 3 .Henderson and Pabis 4 .Wang and Singh 5. Modified Henderson and Pabis 6 .Logarithmic 7 .Midilli
Model MR =exp(-kt) MR =exp(-ktn) MR =a exp(-kt) MR =1 +at +bt2 MR = a exp(-kt) + b exp(-gt) + c exp(-ht) MR =a exp(-kt) + b MR =exp(-ktn) + bt
3
References Bruce [7] Page [8] Henderson and Pabis [9] Wang and Singh [10] Henderson and Pabis [9] Togrul and Pehlivan [11] Midilli et al .[12]
The effects of some parameters related to the product or drying condition such as: sizing, drying air temperature, relative humidity, etc. Modelling the drying behavior of different agricultural products often requires the statistical methods of regression and correlation analysis. Linear and non-linear regression method are important tools to find the relationships between different variables those for which no established empirical relationship exists. In this study, the relationships of the drying constants in function of drying air temperature for the best suitable model were also determined by the non-linear regression technique for predicted moisture ratio of cassava pulp. 3. Experimental In this experiment, cassava pulp was drying with the Model BINDER Incubator BD 53 dryer. The cassava pulp sample with initial moisture content at 80 % w.b. was drying at temperature 50, 60 and 70 C and collected mass of cassava pulp at interval 20 minute until no more mass changes ( 0.01 g). 4. Experiment results From experiment, mass of cassava was collected for determined the moisture ratio by equation 1. It then used with the semi-empirical dynamic model and empirical drying equation for further predicts the moisture ratio. Non-linear regression method was using for analysis various parameters of thin layer equations which run on computer program and consideration in term of coefficient of determination (R2) and root mean square error (RMSE). Drying of cassava pulp started with an initial moisture content around 80 % w.b. and continued until no more mass changes. In this study, the variations of moisture ratio with drying time were considered. Additionally, empirical thin layer drying curve equation of cassava pulp was determined by applying the thin layer drying models in Table 1. The moisture ratio of cassava pulp at different drying temperature was shown in Fig. 1. From drying curve, these is an inverse relation between drying air temperature and drying time as moisture ratio was decreased with increased drying time and the drying time was decreased with increased drying temperature. 1.0
Experiment 50 °C Experiment 60 °C
Moisture ratio (MR)
0.8
Experiment 70 °C
0.6
0.4
0.2
0.0
0
50
100 150 Time (min)
200
250
300
Fig. 1. Moisture ratio of cassava pulp at different drying temperature.
4
Suparerk Charmongkolpradit et al. / Energy Procedia 138 (2017) 354–359 Suparerk Charmongkolpradit et al./ Energy Procedia 00 (2017) 000–000
357
From non-linear regression analysis carried on seven thin layer models. The statistical results applied to those models by taking into consideration all temperature was shown in Table 2. The best model to describing the thin layer drying characteristics of cassava pulp was chosen as highest R2 and lowest RMSE. From Table 2, the Midilli model was good agreement predicted with experimental data. It had coefficient of determination (R2) and root mean square error (RMSE) better than another model for drying temperature range between 50 C - 70 C. After that, the drying constant k and coefficients b, n of the Midilli model was calculated by using non-linear regression technique for each drying temperature. The results of non-linear regression analyse at drying temperature 50, 60 and 70 C were shown in Table 3. Table 2. Correlation statistical analysis of thin layer models. Model Lewis Page Handerson & Pabis Wang & Singh Modified Handerson & Pabis Logarithmic Midilli
Correlation R2 RMSE R2 RMSE R2 RMSE R2 RMSE R2 RMSE R2 RMSE R2 RMSE
Table 3. Values of the drying constant and coefficient of Midilli model. Model Drying condition Midilli 50 C 60 C 70 C
50 C 0.9305 0.1127 0.9816 0.0487 0.9163 0.1018 0.9840 0.0432 0.9164 0.1019 0.9843 0.0423 0.9849 0.0422 b -3.810x10-3 -1.330x10-3 -1.002x10-3
60 C 0.9655 0.0765 0.9881 0.0374 0.9583 0.0707 0.9982 0.0149 0.9578 0.0710 0.9979 0.0149 0.9983 0.0146 k 1.131x10-8 3.677x10-3 4.703x10-3
70 C 0.9727 0.0702 0.9893 0.0353 0.9670 0.0647 0.9959 0.0214 0.9665 0.0648 0.9955 0.0224 0.9959 0.0213 n 2.376x10-7 1.101 1.158
In order to take into account the effect of drying temperature on the constant and coefficients of the Midilli model, namely k, n and b (Table 1), the regression analysis was evaluated the relations between these parameters and drying temperature .Thus, the regression equations of these parameters against drying temperature T (C) and the accepted model was shown in equation 3. For New Midilli model (equation 3), it had the correlation values with experimental data as R2 = 0.9915 and RMSE = 0.0309. The moisture ratio predicted of New-Midiili model and experimental was shown in Fig. 2. 1.0
Moisture ratio (MR)
Experiment 50 °C Experiment 60 °C
0.8
Experiment 70 °C New-Midilli 50 °C New-Midilli 60 °C
0.6
New-Midilli 70 °C
0.4
0.2
0.0
0
50
100 150 Time (min)
200
250
300
Fig. 2. Moisture ratio predicted of New-Midilli model and experimental at different drying temperature.
Suparerk Charmongkolpradit et al. / Energy Procedia 138 (2017) 354–359 Suparerk Charmongkolpradit et al. / Energy Procedia 00 (2017) 000–000
358
5
New Midilli model: MR =exp(-ktn) + bt
(3)
where the drying constant k and coefficient b, n are given by
k 1.325 105 T 2 1.826 103 T 0.05815 1.22121T 61.06 n T 48.91 5 2 b 1.076 10 T 1.432 103 T 48.49 103
From analysis results of seven models were guideline for developed a new empirical drying equation to be suitable for predicting the moisture ratio of cassava pulp. The curve fitting method was used to build the new empirical drying equation. The new empirical drying equation was shown at equation 4, drying constant and coefficient were shown in Table 4. Table 4. Correlation values, drying constant and coefficient of new empirical drying equation. New empirical drying equation Temperature Parameter a = 1.518 b = 1.011x10-4 50 °C k = 1.745 n = -0.2016 a = 1.592 b = 2.796x10-3 n MR = a exp - kt +b 60 °C k = 1.099 n = -0.6006 a = 1.308 b = 4.545x10-3 70 °C k = 1.124 n = 0.3077
R2
RMSE
0.9894
0.0347
0.9981
0.0141
0.9959
0.0212
1.0 Experiment 50 °C Experiment 60 °C
Moisture ratio (MR)
0.8
Experiment 70 °C New Eq.50 °C New Eq.60 °C
0.6
New Eq.70 °C
0.4
0.2
0.0
0
50
100 150 Time (min)
200
250
300
Fig. 3. Moisture ratio predicted of the new empirical drying equation and experimental at different drying temperature.
From Table 4 the parameters a, b, k and n were use the non-linear regression technique to account for the function of drying temperature. The accepted drying constant and coefficients were as follows: MR =a exp(-ktn) + b where the drying constant k and coefficient a, b, n are given by
a 3.59 103 T 2 0.43825T 11.78
(4)
6
Suparerk Charmongkolpradit et al. / Energy Procedia 138 (2017) 354–359 Suparerk Charmongkolpradit et al./ Energy Procedia 00 (2017) 000–000
359
b 3.46 103 T 2 0.42045T 12.17 k 4.73 106 T 2 789.75 106 T 0.02756 n 3.355 103 T 2 0.4337T 15.04 The new empirical drying equation (equation 4) can predict moisture ratio of cassava pulp in range 50 - 70 °C. It had the correlation values with experimental data as R2 = 0.9938 and RMSE = 0.0264. The moisture ratio predicted of new empirical drying equation was shown in Fig. 3. 5. Conclusion The drying of cassava pulp by using the Model BINDER Incubator BD 53 dryer. In experimental the cassava pulp sample with initial moisture content at 80 % w.b. was drying at temperature 50, 60 and 70 C and collected mass of cassava pulp at interval 20 minute until mass was constant. The thin layer drying models were used to predicted moisture content of cassava pulp and were guideline for developed a new empirical drying equation to be suitable for predicting the moisture ratio of cassava pulp. From experimental, the moisture ratio was decreased with increased drying time and the drying time was decreased with increased drying temperature. The results analysis of thin layer drying models, the Midilli model was good agreement predicted with experimental data. From non-linear regression analysis to evaluated drying constant and coefficient, when the Midilli model was compare with experimental data it had coefficient of determination (R2) and root mean square error (RMSE) with 0.9915 and 0.0309, especially. From analysis results of seven models were guideline for developed a new empirical drying equation to be suitable for predicting the moisture ratio of cassava pulp. The curve fitting method was used to build the new empirical drying equation. The new empirical drying equation can predict moisture ratio of cassava pulp in range 50-70 °C drying temperature. It had the correlation values with experimental data as R2 = 0.9938 and RMSE = 0.0264. Acknowledgement The authors gratefully acknowledge Department of mechanical engineering, Rajamangala university of technology isan and Khon kaen campus for their support in this study under research projects. References [1] N. Poramacom, A. Ungsuratana, P. Ungsuratana, P. Supavitipattana. Cassava production, prices and related policy in Thailand, American International Journal of Contemporary Research. 2013; 3(5): p. 43-51. [2] North Eastern Tapioca Trade Association. Summary of the Survey for Production and Trade. Annual report; 2012. [3] National Science and Technology Development Agency, Ministry of Science and Technology Thailand. Strategic Planning Allience II: SPA II. [4] Djuma ali, N. Soewarno, Sumarno, D. Primarini, W. Sumaryono. Cassava pulp as a biofuel feedstock of an enzymatic hydrolysis process. MAKARA TEKNOLOGI 2011; 15(2): p. 183-192. [5] Pantipa Phowan, Paiboon Danvirutai. Hydrogen Production from cassava pulp hydrolysate by mixed seed cultures: Effects of initial pH. substracte and biomass concentrations. BIOMASS & BIO ENERGY 2014; 64: p. 1-10. [6] Guarte, RC .Modelling the drying behaviour of copra and development of a natural convection dryer for production of high quality copra in the Philippines .Ph.D .dissertation .Hohenheim University: Germany:Stuttgart; 1996. [7] Bruce DM. Exposed-layer barley drying, three models fitted to New data up to 150 C, Journal of Agricultural Engineering Research 1985; 32: p. 337-347. [8] Page GE. Factors influencing the maximum rate so of air drying shelled corn in thin layers, M. S. thesis . Department of Mechanical Engineering:Purdue University: Purdue:USA; 1949. [9] Henderson SM, Pabis S. Grain drying theory. II. Temperature effects on drying coefficients. Journal of Agricultural Engineering Research 1961; 6: p. 169-174. [10] Togrul IT, Pehlivan D. Mathematical modeling of solar Drying of apricots in thin layers. Journal of Food Engineering 2002; 55: p. 209-216. [11] Wang CY, Singh RP. Use of variable equilibrium moisture content in modeling rice drying. Transactions of American Society of Agricultural Engineers 1978; 11: p. 668-672. [12] Midilli A, Kucuk H, Yapar Z. A new model for single layer drying. Drying Technology 2002; 20(7): p. 1503–1513.