Adaptive neuro-fuzzy approach for ducted tidal turbine performance estimation

Renewable and Sustainable Energy Reviews 59 (2016) 1111–1116

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

Adaptive neuro-fuzzy approach for ducted tidal turbine performance estimation Obrad Anicic n, Srdjan Jovic Faculty of Technical Sciences, Kosovska Mitrovica, University of Priština, 38 220 Kosovska Mitrovica, Kneza Milosa 7, Serbia

art ic l e i nf o

a b s t r a c t

Article history: Received 14 March 2015 Received in revised form 10 December 2015 Accepted 13 January 2016

The potential of marine power to produce electricity has been exploited recently. One of the ways of producing electricity is by using tidal current turbines. In this study, the hydrodynamic performance of a novel type of ducted tidal turbine was investigated. Tidal energy has become a large contender of traditional fossil fuel energy, particularly with the successful operation of multi-megawatt sized tidal turbines. Hence, quality of produced energy becomes an important problem in tidal energy conversion plants. Several control techniques have been applied to improve the quality of power generated from tidal turbines. In this study, the adaptive neuro-fuzzy inference system (ANFIS) is designed and adapted to estimate power coefficient value of the ducted tidal turbines. The back propagation learning algorithm is used for training this network. This intelligent controller is implemented using Matlab/Simulink and the performances are investigated. The simulation results presented in this paper show the effectiveness of the developed method. & 2016 Elsevier Ltd. All rights reserved.

Keywords: Ducted tidal turbine Power coefficient Neuro-fuzzy ANFIS

Contents 1. 2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1111 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1112 2.1. Ducted tidal turbine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1112 2.2. Adaptive neuro-fuzzy inference system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1113 3. Results and discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1114 4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1114 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1115

1. Introduction The population of the world is increasing with an alarming rate. This trend demands more resources for establishment of mankind. In the past, non-renewable energy resources were the major source of power production. The wide use of fossil fuels has led to various diseases such as skin cancer, respiratory diseases [1,2]. In this regard, alternative renewable resources are required to fulfill the growing demand of energy. Renewable sources such as biomass, sunlight, wind, ocean energy are the natural resources which are available in abundance and can be renewed in measurable time period [3–6]. n

Corresponding author. E-mail address: [email protected] (O. Anicic).

http://dx.doi.org/10.1016/j.rser.2016.01.031 1364-0321/& 2016 Elsevier Ltd. All rights reserved.

Tidal barrages were widely used for harnessing ocean energy [7,8]. The principle of tidal barrage was similar to dams. The drawbacks for both technologies are the same. In early 1950s the first tidal turbine was used. It gained significant attention in late 1970s when the proof of concept of SeaGen was given [9,10]. The horizontal axis tidal turbines were designed to harness kinetic energy from tidal currents caused by gravitational interaction among the sun, the moon and the earth [11]. Horizontal tidal turbine were used due to the higher efficiency compared to the vertical counterparts. However, recently vertical axis tidal turbines gained more attention due to the same principle of operation with wind turbines which are in the advanced stage of implementation and usage [12,13]. The invention of tidal turbine brought a new era in ocean energy.

1112

O. Anicic, S. Jovic / Renewable and Sustainable Energy Reviews 59 (2016) 1111–1116

Numerical methods are being widely used because of its better accuracy when compared with empirical equations. These techniques require multiple environment variables for predicting the performance of tidal turbines. Nevertheless, the accuracy of numerical method in prediction of the output variables highly relies on the input data. Numerical methods involve high computation resource and multiple exogenous parameters. In addition, these numerical models involve large number of meteorological and oceanographic data. Therefore, there is need for putting forth new method for estimating performance of tidal current turbine, which is simple yet shows better accuracy than traditional approaches. Recent research works were proposed using simpler method for predicting power coefficient of tidal current turbine using soft computing methods. One of the methods for predicting the performance is by using the machine modeling. In [14], the firefly algorithm is used to predict the wind turbine generation. In 2013 Tan et. al designed the optimized power generation by renewable energy by using firefly algorithm [15]. Long et al. also predicted the sea level by using the same approach [16]. The concept of ducted turbines has been studied for decades with no commercially successful designs to date. The first in-depth study of ducted turbines was done in the context of wind power by Lilley and Rainbird [17], who developed analytical models based on onedimensional momentum theory and potential flow methods in the 1950s. This study suggested that a reasonable duct could provide at least a 65 per cent increase in power over an ideal unshrouded turbine with the same rotor diameter. Attempts to develop ducted wind turbines have been unsuccessful for a number of reasons, the most important of which is arguably the immense loading on the duct in storm conditions or in yawed flows. There is renewed interest in ducted turbines in the context of tidal power generation since the direction and magnitude of tidal flows are quite predictable and tidal turbines would not be subject to such extreme storm loads as wind turbines. There are several prototypes for vertical axis tidal turbines; however, the dominant design and focus of this article is the ducted horizontal axis tidal turbine concept. This requires special considerations in blade design to ensure that the turbine can operate on both ebb and flow tides. Analytical models have been developed to characterize the performance of ducted turbines by Lilley and Rainbird [17], Foreman et al. [18], Lawn [19], van Bussel [20], and Jamieson [21]. However, all these models require empirical parameters to capture the effects of flow separation, base pressure, and viscous loss. At present, there is little experimental or numerical data to support a fundamental understanding of how these factors vary with changes to duct geometry. The models presented by Jamieson and van Bussel are based on a modified version of the standard actuator disc momentum analysis [22]. The tidal systems are non-linear power sources that need accurate on-line identification on the optimal operating point [23,24]. Also, the power from tidal varies depending on the environmental factors such as the fluctuation of wind velocity. Aiming at optimizing such systems to ensure optimal functioning of the unit, new techniques are used today such as the fuzzy logic (FL) [25], artificial neural network (ANN) [26] and neuro-fuzzy [27]. Artificial neural networks are flexible modeling tools with capabilities of learning the mathematical mapping between input and output variables of nonlinear systems. One of the most powerful types of neural network system is adaptive neuro-fuzzy inference system (ANFIS) [28–53]. ANFIS shows very good learning and prediction capabilities, which makes it an efficient tool to deal with encountered uncertainties in any system. ANFIS, as a hybrid intelligent system enhances the ability to automatically learn and adapt. The key goal of this investigation is to establish an ANFIS for estimation of the tidal turbine pressure coefficient C pc . An attempt is made to retrieve correlation between pressure coefficient C pc in regard to thrust coefficient and outer diffuser surface angle of the

tidal turbine. That system should be able to forecast the pressure coefficient in regards to the main turbine parameters. Fuzzy Inference System (FIS) is the main core of ANFIS. FIS is based on expertise expressed in terms of ‘IF–THEN’ rules and can thus be employed to predict the behavior of many uncertain systems. The advantage of FIS is that it does not require knowledge of the underlying physical process as a precondition for its application. Thus ANFIS integrates the fuzzy inference system with a back-propagation learning algorithm of neural network. An ANFIS model will be establish in this study to predict the tidal turbine pressure coefficient in relation to the two main turbine parameters. The experimental training and checking data for the ANFIS network are obtained from analytical analysis of the tidal turbine power output. The basic idea behind the soft computing methodology is to collect input/output data pairs and to learn the proposed network from these data. The ANFIS is one of the methods to organize the fuzzy inference system with given input/output data pairs. This technique gives fuzzy logic the capability to adapt the membership function parameters that best allow the associated fuzzy inference system to track the given input/output data. In Section 2, ducted tidal turbine pressure coefficient is explained in detail. The main principle of the adaptive neuro-fuzzy inference system (ANFIS) is presented in Section 3. Section 3 also presents the ANFIS model of the tidal turbine pressure coefficient estimation. Section 4 summarizes the results and provides the synthesis of measurement data. Finally, Section 5 offers some concluding remarks and future-work directions.

2. Methodology 2.1. Ducted tidal turbine Blade is the most crucial part of the turbine. The water velocity of the tidal currents strikes the turbine blades and lead to the rotation of turbine. This converts the kinetic energy of the tidal currents into rotational energy for the generator and eventually into the electricity. The power produced by such turbines is given by P ¼ 0:5C p ρ A V 3

ð1Þ

where, P (W) is the power produced, Cp (dimensionless) is the coefficient of power, A (m2) is the flow contact and V (m/s) is the inlet velocity for the turbine. Therefore, the properties of the turbine blades have a major effect on the performance of the turbine. In this study, a collapsible blade with three foldable entities was taken. The lower most entity was 260 mm and the scaling factor of 0.95 was followed for each succeeding blade. The turbine has the capacity to be folded up to half of its installed size. Table 1 represents the parameters of the turbine studied. The analytical framework for this study is based on the ducted turbine model as presented by Lawn [19], which is similar to an earlier derivation by Foreman et al. [8]. The model is developed by analyzing the variation of pressure through the duct. From the free stream condition (p0, u0) the flow either expands or contracts approaching the rotor disc plane. The variation in pressure is related to the change in velocity by Bernoulli’s equation modified with an efficiency term which parameterizes viscous loss in the inlet section. The pressure change across the actuator disc is defined according to the standard definition of the thrust coefficient CT. The pressure difference between the far wake, defined where full expansion back to p ¼p0 has occurred, and the diffuser outlet is parameterized as a base pressure coefficient Cpc. This definition reflects the assumption of full pressure recovery in the wake to the

O. Anicic, S. Jovic / Renewable and Sustainable Energy Reviews 59 (2016) 1111–1116

free stream value, and is an idealization required for this analytical model. In reality, as turbine arrays begin to extract significant amounts of energy from the flow, the downstream pressure will be measurably reduced from the free stream value owing to the restricted domains of tidal channels. In this investigation the vibrations and oscillations of the tidal turbine is neglected in order to simplify the analysis [54,55]. To study the effects of changing geometric features of the duct, several duct designs were tested. The geometric features expected to impact the duct performance were the inner and outer diffuser surface angles αin, αout, as depicted in Fig. 1. In the above sections, the impact of various parameters is discussed. Table 2 describes the parameters used to predict the tidal turbine performance. 2.2. Adaptive neuro-fuzzy inference system ANFIS can be used for classification, approximation of highly nonlinear functions, on-line identification in discrete control system and to predict a chaotic time series. ANFIS can serve as a basis for constructing a set of fuzzy ‘if–then’ rules with appropriate membership function to generate the stipulated input–output pairs. The membership functions are tuned to the input–output data. ANFIS is about taking an FIS system and tuning it with a back propagation algorithm based on the collection of input–output data. The basic structure of a FIS consists of three conceptual components: a rule base, which contains a selection of fuzzy rules; a database, which defines the membership functions (MFs) used in the fuzzy rules; and a reasoning mechanism, which performs the inference procedure upon the rules and the given facts to derive a reasonable output or conclusion. These intelligent systems combine knowledge, technique and methodologies from various sources. They possess human-like expertise within a specific domain – adapt themselves and learn to do better in changing environments. In ANFIS, neural networks recognize patterns, and help adaptation to environments. ANFIS is tuned with a back propagation algorithm based on the collection of input-output data. Here the training data is obtained by analytical expression for the tidal turbine pressure coefficient. One half of the data are used for training while the other half is used for checking and validation of the model. With a proper training scheme and fine filtered data-

sets, ANFIS is capable to estimate tidal turbine pressure coefficient quite accurately since it learns from the training data. This measurement-free architecture also makes it immediately available for operation once they are trained. There were three membership functions on each input. In this study bell-shaped membership functions were chosen with maximum equal to 1 and minimum equal to 0. Fuzzy logic toolbox in MATLAB was used for the entire process of training and evaluation of fuzzy inference system. Fig. 2 shows an ANFIS structure for two inputs. In this work, the first-order Sugeno model with two inputs and fuzzy IF–THEN rules of Takagi and Sugeno’s type is used: if x is A andy is Cthen f 1 ¼ p1 x þ q1 y þr 1

Parameter

Description for tidal turbine

Blade type Hydrofoil Number of blades Blade length Radius of the turbine Rotational speed Collapsible entities for each blade Foldable capacity Ratio of scaling

Foldable NACA 0025 Three 710 mm (260 mm, 236 mm, 214 mm) 450 mm 60 rpm Three Half of the installed size 0.90

ð2Þ

The first layer consists of input variables membership functions (MFs), input 1 and input 2. This layer just supplies the input values to the next layer. The inputs are thrust coefficient and outer diffuser surface angle of the tidal turbine. In the first layer every node is an adaptive node with a node function O ¼ μAB ðxÞ and O ¼ μCD ðxÞ where μAB ðxÞ and μCD ðxÞ are MFs. In this study, bell-shaped MFs with maximum equal to 1 and minimum equal to 0 is chosen. The second layer (membership layer) checks for the weights of each MFs. It receives the input values from the 1st layer and acts as MFs to represent the fuzzy sets of the respective input variables. Every node in the second layer is non-adaptive and this layer multiplies the incoming signals and sends the product out like wi ¼ μAB ðxÞ  μCD ðyÞ. Each node output represents the firing strength of a rule. Table 2 Input variables in terms of the definition. Inputs

Parameters description

Parameters characterization

Input 1 Thrust coefficient CT Input 2 Outer diffuser surface angle of the tidal αout turbine.

Layer 1 Layer 2

A x

Table 1 Parameters of the tidal turbine.

1113

Layer 3

Layer 4 x w1 *

w1

B

w 1 * f1 f

C

*

w 2 f2

w2 *

w2

y

Layer 5

y

D

x

y

Fig. 2. ANFIS structure.

Training data 0.9 0.8 0.7

Cpc

0.6 0.5 0.4 0.3 0.2 0.1 0 0

10

20

30

Data samples Fig. 1. Duct geometric attributes used by the regression-based model.

Fig. 3. ANFIS training data.

40

50

60

1114

O. Anicic, S. Jovic / Renewable and Sustainable Energy Reviews 59 (2016) 1111–1116

Fig. 4. ANFIS structure.

fifth layer is not adaptive and this node computes the overall output as the summation of all incoming signals P X wf O4i ¼ wi xf ¼ Pi i ð3Þ i wi i The hybrid learning algorithms were applied to identify the parameters in the ANFIS architectures. In the forward pass of the hybrid learning algorithm, functional signals go forward until Layer 4 and the consequent parameters are indentified by the least squares estimate. In the backward pass, the error rates propagate backwards and the premise parameters are updated by the gradient descent.

3. Results and discussion Fig. 5. ANFIS predicted relationship for tidal turbine pressure coefficient.

The third layer is called the rule layer. Each node (each neuron) in this layer performs the pre-condition matching of the fuzzy rules, i.e. they compute the activation level of each rule, the number of layers being equal to the number of fuzzy rules. Each node of these layers calculates the weights which are normalized. The third layer is also non-adaptive and every node calculates the ratio of the rule’s firing strength to the sum of all rules' firing strengths like wi ¼ w1 wþi w2 , i ¼ 1; 2: The outputs of this layer are called normalized firing strengths. The fourth layer is called the defuzzification layer and it provides the output values resulting from the inference of rules. Every node in the fourth layer is an adaptive node with node function
 O4i ¼ wi xf ¼ wi ðpi x þ qi y þ r i Þ where pi ; qi ; r is the parameter set and in this layer is referred to as consequent parameters. The fifth layer is called the output layer which sums up all the inputs coming from the fourth layer and transforms the fuzzy classification results into a crisp (binary). The output represents estimated tidal turbine pressure coefficient. The single node in the

In this paper ANFIS training data were extracted using computational fluid dynamics according to Section 2.1. The training data is shown in Fig. 3. The ANFIS network has six bell-shaped membership functions for each input separately. Minimal training error of the neural network for the used membership functions was 0.00001471. It is not appropriate to further increase the number of the membership functions since there are too many parameters. ANFIS structure for two inputs and six membership functions for each input are shown in Fig. 4.

4. Conclusion The impact of the variation in the thrust coefficient and outer diffuser surface angle of the tidal turbine on the performance of the tidal turbine is investigated in the paper. As the parameter for measuring performance of the tidal turbine pressure coefficient C pc was used. A systematic approach to achiev the tidal turbine pressure coefficient by means of ANFIS strategy was investigated. The main advantage of designing the ANFIS coordination scheme is to estimate tidal turbine pressure coefficient as the main turbine parameter.

O. Anicic, S. Jovic / Renewable and Sustainable Energy Reviews 59 (2016) 1111–1116

Simulations were run in MATLAB and the results were observed on the corresponding output blocks (Fig. 5). The main advantages of the ANFIS scheme are: computationally efficient, well-adaptable with optimization and adaptive techniques. The developed strategy is not only simple, but also reliable and may be easy to implement in real time applications using some interfacing cards like the dSPACE, data acquisition cards, NI cards, etc. for control of various parameters. This can also be combined with expert systems and rough sets for other applications. ANFIS can also be used with systems handling more complex parameters. Another advantage of ANFIS is its speed of operation, which is much faster than in other control strategies; the tedious task of training membership functions is done in ANFIS. One of the most important features of the proposed ANFIS network is identification and estimation of the optimal tidal turbine pressure coefficient.

References [1] Kampa M, Castanas E. Human health effects of air pollution. Environ Pollut 2008;151(2):362–7. [2] Brook RD, et al. Air pollution and cardiovascular disease A statement for healthcare professionals from the expert panel on population and prevention science of the American Heart Association. Circulation 2004;109(21):2655–71. [3] Bridgwater A. The technical and economic feasibility of biomass gasification for power generation. Fuel 1995;74(5):631–53. [4] Karki R, Hu P, Billinton R. A simplified wind power generation model for reliability evaluation. Energy Convers IEEE Trans 2006;21(2):533–40. [5] Shaheen SE, Ginley DS, Jabbour GE. Organic-based photovoltaics: toward lowcost power generation. MRS Bulletin 2005;30(01):10–9. [6] Halamay DA, Brekken TKA, Simmons A, McArthur S. Reserve requirement impacts of large-scale integration of wind, solar, and ocean wave power generation. Sustain Energy, IEEE Trans 2011;2(3):321–8. [7] Xia J, Falconer RA, Lin B. Hydrodynamic impact of a tidal barrage in the Severn Estuary, UK. Renew Energy 2010;35(7):1455–68. [8] Frid Chris, Andonegi Eider, Depestele Jochen, Judd Adrian, Rihan Dominic, Rogers. Stuart I. The environmental interactions of tidal and wave energy generation devices. Environ Impact Assess Rev 2012;32(1):133–9. [9] Fraenkel P. Practical tidal turbine design considerations: a review of technical alternatives and key design decisions leading to the development of the SeaGen 1.2 MW tidal turbine. In: Proceedings of the fluid machinery groupocean power fluid machinery seminar. 2010. [10] Fraenkel P. Marine current turbines: pioneering the development of marine kinetic energy converters. Proc Inst Mech Eng Part A: J Power Energy 2007;221(2):159–69. [11] Cartwright DE, Melchior P. Tides: a scientific history, vol.7. Cambridge Univ. Press.; 1999., 1 edition (August 28, 2000), ISBN-10: 521797462, ISBN-13: 9780521797467. [12] Bhutta Muhammad Mahmood Aslam, Hayat Nasir, Farooq Ahmed Uzair, Ali Zain, Jamil Sh Rehan, Hussain Zahid. Vertical axis wind turbine–a review of various configurations and design techniques. Renew Sustain Energy Rev 2012;16(4):1926–39. [13] Yang B, Lawn C. Fluid dynamic performance of a vertical axis turbine for tidal currents. Renew Energy 2011;36(12):3355–66. [14] Shuhui Li, Wunsch DC, O'Hair EA, Giesselmann MG. Using neural networks to estimate wind turbine power generation. Energy Convers IEEE Trans 2001;16 (3):276–82. [15] Tan Wen-Shan, Hassan Mohammad Yusri, Majid Md Shah, Rahman Hasimah Abdul. Optimal distributed renewable generation planning: a review of different approaches. Renew Sustain Energy Rev 2013;18:626–45. [16] Long, NC and P Meesad. Meta-heuristic algorithms applied to the optimization of type-1 and type 2 TSK fuzzy logic systems for sea water level prediction. In: Proceedings of the IEEE Sixth International Workshop on Computational Intelligence & Applications (IWCIA). 2013. [17] Lilley GM, Rainbird WJ. Rep. 102: A preliminary report on the design and performance of ducted windmills. College of Aeronautics, Cranfield, 1956. Available as Tech. Rep. C/T 119. UK: The Electrical Research Association, Leatherhead; 1957. [18] Foreman KM, Gilbert B, Oman RA. Diffuser augmentation of wind turbines. Sol Energy 1978;20:305–11. [19] Lawn CJ. Optimization of the power output from ducted turbines. Proc IMechE Part A: J Power Energy 2003;217:107–17. [20] van Bussel GJ. The science of making more torque from wind: Diffuser experiments and theory revisited. J Phys Conf Ser 2007;75:1–12. [21] Jamieson P. Generalized limits for energy extraction in a linear constant velocity flow field. Wind Energy 2008;11:445–57. [22] Burton T, Sharpe D, Jenkins N, Bossanyi E. Wind energy handbook. Chichester: John Wiley & Sons Ltd.; 2001. [23] No TS, Kim J-E, Moon JH, Kim SJ. Modeling, control, and simulation of dual rotor wind turbine generator system. Renew Energy 2009;34:2124–32.

1115

[24] Rocha R. A sensorless control for a variable speed wind turbine operating at partial load. Renew Energy 2011;36:132–41. [25] Bououden S, Chadli M, Filali S, El Hajjaji A. Fuzzy model based multivariablepredictive control of a variable speed wind turbine: LMI approach. Renew Energy 2012;37:434–9. [26] Yilmaz AS, Özer Z. Pitch angle control in wind turbines above the rated wind speed by multi-layer perceptron and radial basis function neural networks. Expert Syst Appl 2009;36:9767–75. [27] Mohandes M, Rehman S, Rahman SM. Estimation of wind speed profile using adaptive neuro-fuzzy inference system(ANFIS). Appli. Energy 2011;88 (11):4024–32. [28] Jang J-SR. ANFIS: adaptive-network-based fuzzy inference systems. IEEE Trans Syst Man Cybern 1993;23:665–85. [29] Dalibor Petković, Mirna Issa, Nenad Pavlović D, Nenad Pavlović T, Lena Zentner. Adaptive neuro-fuzzy estimation of conductive silicone rubber mechanical properties. Expert Syst. Appli 2012;39(10):9477–82. [30] Dalibor Petković, Mirna Issa, Nenad Pavlović D, Lena Zentner, Žarko Ćojbašić. Adaptive neuro fuzzy controller for adaptive compliant robotic gripper. Expert Syst. Appli. 2012;39(18):13295–304. [31] Dalibor Petković, Nenad Pavlović D, Žarko Ćojbašić, Nenad Pavlović T. Adaptive neuro fuzzy estimation of underactuated robotic gripper contact forces. Expert Syst. Appli 2013;40(1):281–6. [32] Dalibor Petković, Žarko Ćojbašić. Adaptive neuro-fuzzy estimation of automatic nervous system parameters effect on heart rate variability. Neural Computing Appli 2012;21(8):2065–70. [33] Dalibor Petković, Nenad Pavlović D. Applications and Adaptive Neuro-Fuzzy Estimation of Conductive Silicone Rubber Properties. Strojarstvo: časopis za teoriju i praksu u strojarstvu 2013;54(3):197–203. [34] Dalibor Petković, Žarko Ćojbašić, Stevo Lukić. Adaptive neuro fuzzy selection of heart rate variability parameters affected by autonomic nervous system. Expert Syst. Appli 2013;40(11):4490–5. [35] Dalibor Petković, Žarko Ćojbašić, Vlasta Nikolić. Adaptive neuro-fuzzy approach for wind turbine power coefficient estimation. Renew. Sustain. Energy Rev 2013;28:191–5. [36] Dalibor Petković, Nenad Pavlović T, Shahaboddin Shamshirband, Miss Laiha Mat Kiah, Badrul Anuar Nor, Mohd Yamani Idna Idris. Adaptive neuro-fuzzy estimation of optimal lens system parameters. Opt Lasers Eng 2014;55:84–93. [37] Dalibor Petković, Žarko Ćojbašić, Vlasta Nikolić, Shahaboddin Shamshirband, Laiha Mat Kiah Miss, Badrul Anuar Nor, et al. Adaptive neuro-fuzzy maximal power extraction of wind turbine with continuously variable transmission. Energy 2014;64:868–74. [38] Dalibor Petković, Shahaboddin Shamshirband, Nenad Pavlović T, Badrul Anuar Nor, Miss Laiha Mat Kiah. Modulation transfer function estimation of optical lens system by adaptive neuro-fuzzy methodology. Opt Spectrosc 2014;117(1):121–31. http://dx.doi.org/10.7868/S0030403414070046. [39] Dalibor Petković, Shahaboddin Shamshirband, Badrul Anuar Nor, Md Nasir Mohd Hairul Nizam, Nenad Pavlović T, Shatirah Akib. Adaptive neuro-fuzzy prediction of modulation transfer function of optical lens system. Infrared Phys Techno 2014;65:54–60. http://dx.doi.org/10.7868/S0030403414070046. [40] Dalibor Petković, Shahaboddin Shamshirband, Žarko Ćojbašić, Vlasta Nikolić, Badrul Anuar Nor, Md Sabri Aznul Qalid, et al. Adaptive neuro-fuzzy estimation of building augmentation of wind turbine power. Computers Fluids 2014;97:188–94. [41] Dalibor Petković, Shahaboddin Shamshirband, Badrul Anuar Nor, Sareh Naji, Laiha Mat Kiah Miss, Abdullah Gani. Adaptive neuro-fuzzy evaluation of wind farm power production as function of wind speed and direction. Stoch Environ Res Risk Assess 2015;29(3):793–802. http://dx.doi.org/10.1007/ s00477-014-0901-8. [42] Dalibor Petković, Shahaboddin Shamshirband, Javed Iqbal, Badrul Anuar Nor, Nenad Pavlović D, Laiha Mat Kiah Miss. Adaptive neuro-fuzzy prediction of grasping object weight for passively compliant gripper. Appl Soft Computing 2014;22:424–31. [43] Dalibor Petković, Mirna Issa, Nenad Pavlović D, Lena Zentner, Ridzuan Daud Md Nor, Shahaboddin Shamshirband. Contact positions estimation of sensing structure using adaptive neuro-fuzzy inference system. Kybernetes 2014;43 (5):783–96. [44] Dalibor Petković, Shahaboddin Shamshirband, Nenad Pavlović D, Hadi Saboohi, Torki Altameem A, Abdullah Gani. Determining the joints most strained in an underactuated robotic finger by adaptive neuro-fuzzy methodology. Advances Eng Softw 2014;77:28–34. [45] Dalibor Petković. Adaptive neuro-fuzzy fusion of sensor data. Infrared Phys Technol 2014;67:222–8. http://dx.doi.org/10.1016/j.infrared.2014.07.031. [46] Dalibor Petković, Mirna Issa, Nenad Pavlović D, Lena Zentner. Potential of adaptive neuro-fuzzy inference system for contact positions detection of sensing structure. Measurement 2014;61:234–42. http://dx.doi.org/10.1016/j. measurement.2014.10.040. [47] Petković Dalibor, Shamshirband Shahaboddin, Anuar Nor Badrul, Sabri Aznul Qalid Md, Rahman Zulkanain Bin Abdul, Pavlović Nenad D. Input Displacement Neuro-fuzzy control and object recognition by compliant multi-fingered passively adaptive robotic gripper. J Intell Robotic Syst http://dx.doi.org/10. 1007/s10846-015-0182-6. [48] Dalibor Petković, Shahaboddin Shamshirband, Wen Tong Chong, Tamah AlShammari Eiman. Generalized adaptive neuro-fuzzy based method for wind speed distribution prediction. Flow Meas Instrum 2015;43:47–52. http://dx. doi.org/10.1016/j.flowmeasinst.2015.03.003.

1116

O. Anicic, S. Jovic / Renewable and Sustainable Energy Reviews 59 (2016) 1111–1116

[49] Dalibor Petković, Milan Gocić, Slavisa Trajković, Shahaboddin Shamshirband, Shervin Motamedi, Roslan Hashim, et al. Determination of the most influential weather parameters on reference evapotranspiration by adaptive neuro-fuzzy methodology. Comp Electron Agric 2015;114:277–84. http://dx. doi.org/10.1016/j.compag.2015.04.012. [50] Dalibor Petković. Adaptive neuro-fuzzy approach for estimation of wind speed distribution. Electr Power Energy Syst 2015;73:389–92. http://dx.doi. org/10.1016/j.ijepes.2015.05.039. [51] Dalibor Petković, Shahaboddin Shamshirband, Almas Abbasi, Kourosh Kiani, Tamah Al-Shammari Eiman. Prediction of contact forces of underactuated finger by adaptive neuro fuzzy approach. Mech Syst Signal Process 2015;64– 65:520–7. http://dx.doi.org/10.1016/j.ymssp.2015.03.013. [52] Dalibor Petković. Adaptive Neuro-Fuzzy optimization of the net present value and internal rate of return of a wind farm project under wake effect, JCC. Bus. Econ Res 2015;8:11–28. http://dx.doi.org/10.7835/jcc-berj-2015-0102.

[53] Jović S, Raičević V. Vibro-impact forced oscillations of a heavy mass particle along a rough circle excited by a single-frequency force. ISSN: 0001-5970. Acta Mech. 2012;223(6):1153–68. http://dx.doi.org/10.1007/s00707-012-0623-2v. [54] Jović S, Raičević V, Garić LJ, Vibro-impact system based on forced oscillations of heavy mass particle along a rough parabolic line, Mathematical Problems in Engineering, Hindawi Publishing Corporation, Vol. 2012, Article ID 846390, [17 p] http://dx.doi.org/10.1155/2012/846390, ISSN: 1024-123x, (Online First), 2012. [55] Jović S., Raičević V., The phase portrait of the vibro-impact dynamics of two mass particle motions along rough circle, 2012, ISSN 1392–1207, Kaunas University of Technology, Scintific Journal "MECHANIKA", 2012 Nr 18(6), Article ID-No 526, p. 657–665.