Abstract
In this work, a hybrid method based on neural network and particle swarm optimization (PSO) was applied to literature data to develop and validate a model that can predict with precision the solubility of binary systems (CO2 + solid drugs). ANN was used for modeling the nonlinear process. The PSO was used for two purposes: replacing the standard backpropagation in training the ANN and optimizing the process. The training and validation strategy has been focused on the use of a validation agreement vector, determined from linear regression analysis of the predicted versus experimental outputs, as an indication of the predictive ability of the neural network model. Statistical analysis of the predictability of the optimized neural network model trained with trainpso algorithm shows excellent agreement with experimental data. Furthermore, the comparison in terms of average relative deviation (AARD%) between the predicted results for each binary for the whole temperature and pressure range and results predicted by density-based models and a set of equations of state shows that the ANN–PSO model with optimized configuration, five neurons in input and hidden layers and one neuron in output layer (5-5-1) correlates far better the solubility of the solid drugs in scCO2. A control strategy was also developed by using the inverse artificial neural network method. The sensitivity analysis showed that all studied inputs have strong effect on the solubility and allowed the estimation of some solid properties from the solubility data with good accuracy without need to the group contribution methods.
Similar content being viewed by others
Abbreviations
- AARD:
-
Average absolute relative deviation
- ACO:
-
Ant colony optimization
- ANN:
-
Artificial neural network
- ANNi:
-
Inverse artificial neural network
- b:
-
Bias of artificial neural model
- br:
-
Bayesian regularization
- I:
-
Relative importance
- GA:
-
Genetic algorithm
- lm:
-
Levenberg–Marquardt
- MT:
-
Mendez-Santiago and Teja model
- QLF:
-
Quasi-chemical lattice fluid
- T:
-
Equilibrium temperature (K)
- Tc:
-
Critical temperature (K)
- P:
-
Pressure (MPa)
- Pc:
-
Critical pressure (MPa)
- PR:
-
Peng–Robinson
- PSO:
-
Particle swarm optimization
- R:
-
Regression coefficient
- SAFT:
-
Statistical associating fluid theory
- SRK:
-
Soave–Redlich–Kwong
- V:
-
Output layer–hidden layer synaptic weights of artificial neural model
- w:
-
Input layer–hidden layer synaptic weights of artificial neural model
- ω:
-
Acentric factor
- y 2 :
-
Solubility of solid drugs
- calc:
-
Calculated property
- exp:
-
Experimental property
- h:
-
Hidden
- o:
-
Output
- 2:
-
Solute (solid)
- c:
-
Critical property
- i:
-
Weight (i)
References
Tabernero A, Vieira de Melo SA, Mammucari R, Martín del Valle EM, Foster NR (2014) Modelling solubility of solid active principle ingredients in sc-CO2 with and without cosolvents: a comparative assessment of semi empirical models based on Chrastil’s equation and its modifications. J Supercri Fluids 93:91–102
Ayyildiz M, Çetinkaya K (2015) Comparison of four different heuristic optimization algorithms for the inverse kinematics solution of a real 4-DOF serial robot manipulator. Neural Comput Appl. doi:10.1007/s00521-015-1898-8
Duan S, Dong Z, Xiaofang H, Wang L, Li H (2015) Small-world Hopfield neural networks with weight salience priority and memristor synapses for digit recognition. J Neural Comput Appl. doi:10.1007/s00521-015-1899-7
Sakabe J, Uchida H, Shimoyama Y (2015) Mixed solid phase model using equation of state based on hole-theory for solubility prediction of pharmaceutical compound in supercritical CO2. J Supercrit Fluids 100:26–33
Mohanraja M, Jayarajb S, Muraleedharan C (2012) Applications of artificial neural networks for refrigeration, air-conditioning and heat pump systems—a review. Renew Sustain Energy Rev 16(2):1340–1358
Kalogirou SA (2001) Artificial neural networks in renewable energy systems applications: a review. Renew Sustain Energy Rev 5(4):373–401
Ryan A, El Sayed M, Andrew SG, Tjong J, Saeid H (2015) Artificial neural network training utilizing the smooth variable structure filter estimation strategy. Neural Comput Appl. doi:10.1007/s00521-015-1875-2
Calero R, Stashenko E, Martínez JR, López-Giraldo J (2014) Formulation of a new generic density-based model for modeling solubility of polyphenols in supercritical carbon dioxide and ethanol. J Supercrit Fluids 85:116–122
Chin-Sheng K, I-Cheng Y (2015) Using neural networks to integrate structural analysis package and optimization package. Neural Comput Appl. doi:10.1007/s00521-015-1878-z
Chapman WG, Gubbins KE, Jackson G, Radosz M (1989) SAFT: equation of state solution model for associating fluid. Fluid Phase Equilib 52:31–38
Chapman WG, Gubbins KE, Jackson G, Radosz M (1990) New reference equation of state for associating liquids. Ind Eng Chem Res 29:1709–1721
Huang SH, Radosz M (1990) Equation of state for small, large, polydisperse, and associating molecules. Ind Eng Chem Res 29:2284–2294
Gross J, Sadowski G (2001) Perturbed-chain SAFT: an equation of state based on a perturbation theory for chain molecules. Ind Eng Chem Res 40:1244–1260
Gross J, Sadowski G (2000) Application of perturbation theory to a hard-chain reference fluid: an equation of state for square-well chains. Fluid Phase Equilib 168:183–199
Gross J, Vrabec J (2006) An equation-of-state contribution for polar components: dipolar molecules. AlChE J 52(3):1194–1204
Gross J, Sadowski G (2001) Perturbed-chain SAFT: an equation of state based on a perturbation theory for chain molecules. Ind Eng Chem Res 40(4):1244–1260
Gross J, Sadowski G (2002) Application of the perturbed-chain SAFT equation of state to associating systems. Ind Eng Chem Res 41(22):5510–5515
Selvi V, Umarani R (2010) Comparative analysis of ant colony and particle swarm optimization techniques. Int J Comput Appl 5(4):0975–8887
Simon ST, Macnaughton JS, Tomaske LD, Foster R (1993) Solubility of naproxen in supercritical carbon dioxide with and without cosolvents. Ind Eng Chem Res 23:1471–1481
Chen YM, Chen YP (2009) Measurements for the solid solubilities of antipyrine, 4-aminoantipyrine and 4-dimethylaminoantipyrine in supercritical carbon dioxide. Fluid Phase Equilib 282:82–87
Chen YM, Lin PC, Tang M, Chen YP (2010) Solid solubility of antilipemic agents and micronization of gemfibrozil in supercritical carbon dioxide. J Supercrit Fluids 52:175–182
Khimeche K, Alessi P, Kikicand I, Dahmani A (2007) Solubility of diamines in supercritical carbon dioxide: experimental determination and correlation. J Supercrit Fluids 41:10–19
Zhen H, Yu-hua G, Miao H, Teng LJ (2014) Solubility of progesterone in supercritical carbon dioxide and its micronization through RESS. Powder Technol 258:66–77
Li S, Varadarajan GS, Hartland S (1991) Solubilities of theobromine and caffeine in supercritical carbon dioxide: correlation with density-based models. Fluid Phase Equilib 68:263
Cortesi A, Kikic I, Alessi P, Turtoi G, Garnier S (1999) Effect of chemical structure on the solubility of antioxidants in supercritical carbon dioxide: experimental data and correlation. J Supercrit Fluids 14:139
Knez Ž, Markočič E, Leitgeb M, Primožič M, Hrnčič MK, Škerget M (2014) Industrial applications of supercritical fluids: a review. Energy 77:235–243
Beckman EJ (2004) Supercritical and near-critical CO2 in green chemical synthesis and processing. J Supercrit Fluids 28:121–191
Ashwini E, Devtalu V, Bari Manoj M, Barhate SD (2013) A review on: novel solubility enhancement technique hydrotropy. Indo Am J Pharm Res 3(6):4670–4679
Pfohl O, Petkov S, Brunner G (1998) High pressure fluid-phase equilibria containing supercritical fluids. In: 8th International conference on properties and phase equilibria for product and process design. Noordwijkerhout, Netherlands
Nannoolal Y, Rarey J, Ramjugernath D (2008) Estimation of pure component properties Part 3. Estimation of the vapour pressure of non-electrolyte organic compounds via contribution and group interactions. Fluid Phase Equilib 269:117–133
Marrero J, Gani R (2001) Group-contribution based estimation of pure component properties. Fluid Phase Equilib 183–184:183–208
Plumb A, Rowe RC, York P, Brown M (2005) Optimisation of the predictive ability of artificial neural network (ANN) models: a comparison of three ANN programs and four classes of training algorithm. Eur J Pharm Sci 25:395
Faúndez CA, Quiero FA, Valderrama JO (2010) Phase equilibrium modeling in ethanol + congener mixtures using an artificial neural network. Fluid Phase Equilib 292:29–35
Si-Moussa C, Hanini S, Derriche R, Bouhedda M, Bouzidi A (2008) Prediction of high-pressure vapor liquid equilibrium of six binary systems, carbon dioxide with six esters, using an artificial neural network model. Braz J Chem Eng 25(1):183–199
Battiti R (1996) Reactive search: towards self-tuning heuristics. In: Rayward-smith VJ, Osman IH, Reeves CR, Smith GD (eds) Modern heuristic search methods, chap 4. Wiley, New York, NY, pp 61–83
Kennedy J, Eberhart R, Shi Y (2001) Swarm intelligence. Morgan Kaufmann, San Francisco, CA
Pazuki GR, Dashtizadeh A, Taghikhane V, Ghotbi C (2006) A new two-parameter cubic equation of state for predicting phase behaviour of pure compounds and mixtures. Fluid Phase Equilib 242:19–28
Slokar YM, Zupanb J, Marechal AM (1999) The use of ANN for modeling of the H2O2/UV decoloration process: part I. Dyes Pigm 42:123–135
Garson GD (1991) Interpreting neural network connections weights. Al Expert: Miller Freeman, Inc, San Francisco, p 46
El Hamzaoui Y, Hernández JA, Silva-Martínez S, Bassam A, Álvarez A, Lizama-Bahena C (2011) Optimal performance of COD removal during aqueous treatment of alazine and gesaprim commercial herbicides by direct and inverse neural network. Desalination 277:325–337
Hernández JA, Rivera W, Colorado D, Moreno-Quintanar G (2012) Optimal COP prediction of a solar intermittent refrigeration system for ice production by means of direct and inverse artificial neural networks. Sol Energy 86:1108–1117
Laidi M, Hanini S (2013) Optimal solar COP prediction of a solar-assisted adsorption refrigeration system working with activated carbon/methanol as working pairs using direct and inverse artificial neural network. Int J Refrig 36:247–257
Hattab N, Motelica-Heino M (2014) Application of an inverse neural network model for the identification of optimal amendment to reduce copper toxicity in phytoremediated contaminated soils. J Geochem Explor 136:14–23
Mohanraj M, Jayaraj S, Muraleedharan C (2015) Applications of artificial neural networks for thermal analysis of heat exchangers—a review. Int J Therm Sci 90:150–172
Hernández JA (2009) Optimum operating conditions for heat and mass transfer in food stuffs drying by means of neural network inverse. Food Control 20:435–438
Coimbra P, Duarte CM, De Sousa HC (2006) Cubic equation-of-state correlation of the solubility of some anti-inflammatory drugs in supercritical carbon dioxide. Fluid Phase Equilib 239:188–199
Coimbra P, Fernandes D, Ferreira P, Gil MH, De Sousa HC (2008) Solubility of Irgacure 2959 photoinitiator in supercritical carbon dioxide: Experimental determination and correlation. J supercrit Fluids 45:277–281
Stassi A, Betteni R, Gazzaniga A, Giordano F, Schiraldi A (2000) Assessment of solubility of ketoprofen and vanillic acid in supercritical CO2 under dynamic condition. J Chem Eng Data 45:161–165
Housaindokht MR, Bozorgmehr MR (2008) Calculation of solubility of methimazole, phenazopyridine and propranolol in supercritical carbon dioxide. J Supercrit Fluids 43:390–397
Li JH, Huang Z, Wei JL, Xu L (2013) A new optimization method for parameter determination in modeling solid solubility in supercritical CO2. Fluid Phase Equilib 25:117–124
Burgos GI, Brennecke S, Stadtherr MA (2004) Solubility measurement and modeling of molecules of biological and pharmaceutical interest with supercritical CO2. Fluid Phase Equilib 220(1):57–69
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Abdallah el hadj, A., Laidi, M., Si-Moussa, C. et al. Novel approach for estimating solubility of solid drugs in supercritical carbon dioxide and critical properties using direct and inverse artificial neural network (ANN). Neural Comput & Applic 28, 87–99 (2017). https://doi.org/10.1007/s00521-015-2038-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-015-2038-1