Abstract
Shape memory alloys (SMAs) are able to compensate any undergoing plastic deformations and return to their memorized shape. Such a behavior persuades industrialists to use them for different engineering applications, as smart actuators and sensors. Because of their vast applications, it is crucial to engineers to develop effective identification tools capable of simulating the behavior of SMAs. However, SMA actuators have complex and hysteric behavior that in turn obstructs the modeling process. The motivation behind the current study emanates in the pursuit of developing efficient prediction tools for effective modeling of SMA actuators. Actually, after several experiments and software simulations, the authors develop a hybrid intelligent tool which takes advantage of the self-organizing Pareto based evolutionary algorithm (SOPEA) and simultaneous recurrent neural network (SRNN), as a black-box model, to automatically identify the behavior of SMA. SOPEA is a multiobjective evolutionary algorithm which is based on the concepts of survival of the fittest, non-dominated sorting and information recruitment. The information recruitment is guaranteed by applying an un-supervised neuro computing technique, i.e. adaptive self organizing map (ASOM) with conscience mechanism. ASOM is an un-supervised network that assists SOPEA to recognize the non-dominated patterns and produce further non-dominated solutions. Together with the structure of SOPEA, the authors follow a comprehensive preference-based strategy to exploit the desired regions in the Pareto front. This occurs through introducing deliberate reference points. The outcome method is applied to the design of SRNN for modeling the SMA actuator. It is demonstrated that the designed optimization tool can show acceptable performance for the present case study within the imposed computational budget. Besides, through a rigorous experimental procedure, it is indicated that by applying an efficient artificial system, the behavior of SMA can be identified without any specific knowledge of the physical conditions and governing equations.
Similar content being viewed by others
References
Barzegari MM, Dardel M, Fathi A, Ghadimi M (2012) Aeroelastic characteristics of cantilever wing with embedded shape memory alloys. Acta Astronaut 79:189–202
Mani G, Feldman MD, Patel D, Agraval CM (2007) Coronary stents: a materials perspective. Biomaterials 28:1689–1710
Lanteigne E, Jnifene A (2008) An experimental study on a SMA driven pressurized hyper-redundant manipulator. J Intell Mater Syst Struct 19:1067–1075
Bellini A, Colli M, Dragoni E (2009) Mechatronic design of a shape memory alloy actuator for automotive tumble flaps: a case study. IEEE Trans Ind Electron 56:2644–2656
Tai NT, Ahn KW (2012) A hysteresis functional link artificial neural network for identification and model predictive control of SMA actuator. J Process Control 22:766–777
Arai K, Aramaki S, Yanagisawa K (1994) Continuous system modeling of shape memory alloy for control analysis. In: Proceeding of 6th international symposium on micro machine human science, pp 97–99
Lagoudas DC, Entchev PB (2004) Modeling of transformation-induced plasticity and its effect on the behavior of porous shape memory alloys. Part I: constitutive model for fully dense SMAs. Mech Mater 36:865–892
Langelaar M, Keulen FV (2008) Modeling of shape memory alloy shells for design optimization. Comput Struct 86:955–963
Gauthuer JY, Hubert A, Abadie J, Chaillet N, Lexcellent C (2008) Nonlinear Hamiltonian modeling of magnetic shape memory alloy based actuators. Sens Actuators A 141:536–547
Christ D, Reese S (2006) Finite-element modeling of shape memory alloys: a comparison between small-strain and large-strain formulations. Mater Sci Eng A 481–482:343–346
Liang C, Rogers A (1990) One dimensional thermo mechanical constitutive relations for shape memory materials. ASME J Intell Mater Syst Struct 1:207–234
Elahinia MH, Ahmadian M (2005) An enhanced SMA phenomenological model: the shortcomings of the existing models. Smart Mater Struct 14:1297–1308
Barforoushi SD, Fathi A, Danai S (2011) Experimental model of shape memory alloy actuators using modified Prandtl-Ishlinskii model. U.P.B. Sci Bull Ser B 73:255–266
Ma N, Song G, Lee HJ (2004) Position control of shape memory alloy actuators with internal electrical resistance feedback using neural networks. Smart Mater Struct 13:777–783
Song G, Chaudhry V, Batur C (2003) Precision tracking control of shape memory alloy actuators using neural networks and a sliding-mode based robust controller. Smart Mater Struct 12:223–231
Pirge G, Hacioglu A, Ermis M, Altintas S (2009) Determination of the compositions of NiMnGa magnetic shape memory alloys using hybrid evolutionary algorithms. Comput Mater Sci 45:189–193
Asua E, Etxebarria V, Garcia-Arribas A (2008) Neural network-based micropositioning control of smart shape memory alloy actuators. Eng Appl Artif Intell 21:796–804
Ahn KK, Kha NB (2007) Internal model control for shape memory alloy actuators using fuzzy based Preisach model. Sens Actuators A 136:730–741
Fathi A, Mozaffari A (2014) Modeling a shape memory alloy actuators using an evolvable recursive black-box and hybrid heuristic algorithms inspired based on the annual migration of salmons in nature. Appl Soft Comput 14:229–251
Mozaffari A, Fathi A, Khajepour A, Toyserkani E (2013) Optimal design of laser solid freeform fabrication system and real-time prediction of melt pool geometry using intelligent evolutionary algorithms. Appl Soft Comput 13:1505–1519
Gacto MJ, Alcala R, Herrera F (2012) A multi-objective evolutionary algorithm for an effective tuning of fuzzy logic controllers in heating, ventilating and air conditioning systems. Appl Intell. doi:10.1007/s10489-010-0264-x
Forouraghi B (2000) A genetic algorithm for multiobjective robust design. Appl Intell 12:151–161
Blecic I, Cecchini A, Trunfion GA (2007) A decision support tool coupling a causal model and a multi-objective genetic algorithm. Appl Intell 26:125–137
Deb K, Gupta H (2006) Introducing robustness in multi-objective optimization. Evol Comput 14:463–494
Atashkari K, Nariman-Zadeh N, Pilechi A, Jamali A, Yao X (2005) Thermodynamic Pareto optimization of turbojet engines using multi-objective genetic algorithms. Int J Therm Sci 44:1061–1071
Furtuna R, Curteanu S, Leon F (2012) Multi-objective optimization of a stacked neural network using an evolutionary hyper-heuristic. Appl Soft Comput 12:133–144
Guillen A, Rubio G, Toda I, Rivera A, Pomares H, Rojas I (2010) Applying multiobjective RBFNNs optimization and feature selection to a mineral reduction problem. Expert Syst Appl 37:4050–4057
Fernandez JC, Hervas C, Martinez-Estudillo FJ, Gutierrez PA (2011) Memetic Pareto evolutionary artificial neural networks to determine growth/no-growth in predictive microbiology. Appl Soft Comput 11:534–550
Delgado M, Pegalajar MC (2005) A multiobjective genetic algorithm for obtaining the optimal size of a recurrent neural network for grammatical inference. Pattern Recogn 38:1444–1456
Mouret JB, Doncieux S (2012) Encouraging behavioral diversity in evolutionary robotics: An empirical study. Evol Comput 20:91–133
Blanco A, Delgado M, Pegalajar MC (2001) A real-coded genetic algorithm for training recurrent neural networks. Neural Netw 14:93–105
Juang CF, Chang PH (2011) Recurrent fuzzy system design using elite-guided continuous ant colony optimization. Appl Soft Comput 11:2687–2697
Mozaffari A, Gorji-Bandpy M, Samadian P, Rastgar R, Rezania Kolai A (2013) Comprehensive preference optimization of an irreversible thermal engine using Pareto based mutable smart bee algorithm and generalized regression neural network. Swarm Evol Comput 9:90–103
Quek C, Pasquier M, Kumar N (2008) A novel recurrent neural network-based prediction system for option trading and hedging. Appl Intell 29:138–151
Holldobler S, Kalinke Y, Storr HS (1999) Approximating the semantics of logic programs by recurrent neural networks. Appl Intell 11:45–58
Kaikhah K, Garlick R (2000) Variable hidden layer sizing in Elman recurrent neuro-evolution. Appl Intell 12:193–205
Luitel Q, Venayagamoorthy GK (2010) Quantum inspired PSO for the optimization of simultaneous recurrent neural networks as MIMO learning systems. Neural Netw 23:583–586
Jaeger H (2002) A tutorial on training recurrent neural networks, covering BPPT, RTRL, EKF and the “echo state network” approach. GMD report 159, German National Research Center for Information Technology, pp 1–48
Pavlovic I, Ciric I, Djekic P, Nikolic V, Pavlovic R, Cojbasic Z, Radenkovic G (2013) Rheological model optimization using advanced evolutionary computation for the analysis of the influence of recycled rubber on rubber blend dynamical behaviour. Meccanica 48:2467–2477
Fathi A, Mozaffari A (2012) Vector optimization of laser solid freeform fabrication system using a hierarchical mutable smart bee-fuzzy inference system and hybrid NSGA-II/self-organizing map. J Intell Manuf. doi:10.1007/s10845-012-0718-6
Mozaffari A, Gorji-Bandpy M, Samadian P, Mohammadrezaei SN (2012) Analyzing, controlling and optimizing Damavand power plant’s operating parameters using a synchronous parallel shuffling self-organized Pareto strategy and neural network: a survey. Proc IMechE Part A J Power Energy 226:848–866
Rahmani-Monfared K, Fathi A, Mozaffari A, Rabiee SM (2012) Application of self-learning evolutionary algorithm for optimal design of a porous polymethylmethacrylate scaffold fabricated by laser drilling process. Proc IMechE Part E J Process Mech Eng 227:211–224
Billings SA, Voon WSF (1986) Correlation based model validity tests for non-linear models. Int J Control 44:235–244
Wickramasinghe UK, Li X (2008) Integrating user preferences with particle swarms for multi-objective optimization. Genetic and evolutionary computation conference (GECCO), 12–16 July, Portland, OR. ACM Press, NewYork, pp 745–752
Carrese R, Winarto H, Li X, Sobester A, Ebenezer S (2012) A comprehensive preference-based optimization framework with application to high-lift aerodynamic design. Eng Optim 44:1209–1227
Sinha A, Saxena DK, Deb K, Tiwaru A (2013) Using objective reduction and interactive procedure to handle many-objective optimization problems. Appl Soft Comput 13:415–427
Lakshmi K, Rao RM (2013) Multi-objective optimal design of laminated composite skirt using hybrid NSGA. Meccanica 48:1431–1450
Mirmahdi SH, Norouzi M (2013) On the comparative optimal analysis and synthesis of four-bar function generating mechanism using different heuristic methods. Meccanica 48:1995–2006
Kohonen T (2013) Essentials of the self-organizing map. Neural Netw 37:52–65
Author information
Authors and Affiliations
Corresponding author
Appendix: Self organizing Pareto based evolutionary algorithm
Appendix: Self organizing Pareto based evolutionary algorithm
In this section, the authors provide a succinct description of the main operators of the SOPEA algorithm. These operators can be divided into evolutionary operators (i.e. recombination, mutation, selection and the interaction of artificial bees) [20], an SOM operator, an external archive and sharing factor.
1.1 Adaptive SOM operator
An adaptive SOM with a gradient learning rule and concise mechanism is used to learn the characteristics of non-dominated solutions. Considering the non-dominated solutions as the input data, the SOM topology can emulate the non-dominated characteristics and augment the intensification of the Pareto front. The adaptive SOM proposed in [49] is an unsupervised neural network that automatically adapts the learning rate and neighborhood function of the neuron weights. One of the major applications of SOM is to minimize all of the distances between any input vector (non-dominated solutions in our case study) and the synaptic neuron weight vector. This is done by providing a suitable topological ordering for the input distribution. Figure 23 shows a schematic illustration of a typical SOM topology.
The SOM network with concise mechanism uses the following updating rule:
where t is the sub-generation of SOM network and n represents the generation number of SOPEA. y j (t)is a controlling parameter that leads the weight vectors toward the non-dominated solutions. In other words, if the fitness of input value f R , which is a non-dominated solution, is lower than \( f_{{W_{j} (n)}} \) then y j (t) = 1, and consequently, the neuron center moves toward the non-dominated solution (the input of network); otherwise y j (t) = 0 and the neuron center do not approach the solution. The mathematical expression of the abovementioned descriptions can be given as:
where w n+1 j (t) refers to the updated weight vector and w j (t) is the old weight vector. ‖R n i − w n j ‖ represents the distance between the input vectors where R n i is the i th non-dominate solution in the n th generation. The learning rate which is a descending function is defined as follows:
The learning rate parameter h j (0) should be initialized with a value close to unity. α can take any arbitrary value between 0 and 1. s f is a descending constant and should be set based on the condition of problem. In this paper, s f is set to be 1,000. The function f(.) should be designed in a fashion that the following criterion is satisfied appropriately:
f(z) can be set as:
The scaling value sl for an n dimensional input is adjusted using the following equations:
where i represents the number of variable in each solution. \( E_{k}^{i} \left( 0 \right) \) is initialized with some small random values. Figure 24a depicts the concept of neurons updating mechanism in the SOM grid.
Concise mechanism is applied to revive the dead units (weights) in SOM neuron center. A schematic illustration of the procedure is visualized in Figure 24b. The dead unit is a term that refers to the weights with trivial chance of learning and adaption during the progress. Reviving of these units is often called concise mechanism. Here, a simple mechanism is used for tuning of the bias of nodes (neuron):
1.2 External archive
An external archive is used to collect all non-dominated solutions. The main provocation of devising an external archive is to save a historical record of the non-dominated solutions detected over an optimization procedure. The archive interacts with the solutions of SOPEA in a real-time fashion to produce additional non-dominated solutions and eliminate the inferior solutions archived previously. A solution can be saved in the external archive if it complies any of the following criteria:
-
1.
The candidate solution dominates any existing solution in the archive,
-
2.
There is no stored solution in the archive,
-
3.
The archive is full, but the candidate solution is non-dominated and enhances the diversity of the Pareto front significantly.
The prominent asset of using an external archive lies in a reduction of the computational complexity, especially, when the algorithm captures a higher number of the non-dominated solutions. This is because its task is just retaining the obtained non-dominated solutions. Furthermore, using an external archive provides some elitism by preserving non-dominated solutions. The feedback of applying such a policy illustrates that the speed of algorithm does not decline significantly when working with a higher number of the non-dominated solutions.
1.3 Sharing factor
Sharing factor (ξ) is a threshold value that determines the amount of solutions that should be inserted into each of the optimizing operators (i.e. bee and evolutionary operators). In this regard, SOPEA can be considered as ensemble of two co-evolutionary operators in which the sharing factor is ‘trigger of ensemble’. The value of sharing factor (ξ) is confined within the range of unity [0, 1]. Due to the characteristics of an optimization problem, it can be either constant or self-adaptive. The mathematical formulation of sharing process is implemented as:
where \( N_{{P_{evolutionary} }} \) represents the matrix of solutions allocated to the evolutionary phase,\( N_{{P_{bee - inspired} }} \) represents the number of solutions in the bee-inspired phase, \( P_{s } \) is the number SOPEA solutions and ξ is the sharing factor. As it can be inferred, the proposed formulation is a strategy for shuffling all of the solutions between two evolutionary phases.
Rights and permissions
About this article
Cite this article
Mozaffari, A., Fathi, A. & Lashgarian Azad, N. Preferred design of recurrent neural network architecture using a multiobjective evolutionary algorithm with un-supervised information recruitment: a paradigm for modeling shape memory alloy actuators. Meccanica 49, 1297–1326 (2014). https://doi.org/10.1007/s11012-014-9894-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11012-014-9894-0