Research Article  Open Access
Ngoc Thoai Tran, ThanhPhong Dao, Thao NguyenTrang, CheNgoc Ha, "Prediction of Fatigue Life for a New 2DOF Compliant Mechanism by ClusteringBased ANFIS Approach", Mathematical Problems in Engineering, vol. 2021, Article ID 6672811, 14 pages, 2021. https://doi.org/10.1155/2021/6672811
Prediction of Fatigue Life for a New 2DOF Compliant Mechanism by ClusteringBased ANFIS Approach
Abstract
Twodegreeoffreedom (2DOF) compliant mechanism has some outstanding characteristics in accurate positioning systems. Studying the fatigue life for the 2DOF compliant mechanism is a meaningful task to ensure a long working. However, a study for fatigue life prediction of this mechanism has not been conducted so far. In this article, a method for fatigue life prediction of 2DOF compliant mechanism is developed for the first time. This method is the combining of the differential evolution algorithm and the adaptive neurofuzzy inference system (ANFIS) with subtractive clustering. The numerical results on two case studies consisting of material steel A36 and the material AL 6061T6 show that the accuracy of the proposed method is very high. Compared to the actual fatigue life, the root mean square error of the proposed method lies in the range [1.7, 3.97] cycles for Case 1 and [2.03, 10.38] cycles for Case 2. The statistical test also indicates that the proposed method outperforms the traditional method using triangular membership function, bellshape, and Gaussian membership function, with the significance level from 0.05 to 0.1. These results demonstrate the feasibility of the proposed approach in fatigue life prediction of 2DOF compliant mechanism.
1. Introduction
Twodegreeoffreedom (2DOF) compliant mechanism is a monolithic structure that has been widely utilized in ultrahigh precise engineering. Basically, the 2DOF compliant mechanism inherits outstanding characteristics such as minimal positioning error, free friction, reduced assemble, and easy machining [1]. Several applications of this mechanism can be found in the nanopositioning system [2], micropositioning system [3–9], space pointing mechanism [10], and so on. Although the 2DOF compliant mechanism proposes excellent advantages in precision engineering and manipulators, this mechanism still exists major drawbacks, including nonlinear kinematic behaviors and fatigue influences. Considering the first disadvantage, nonlinear kinematic behaviors can be modeled and analyzed by pseudorigidbody model [11], compliance matrix method [12], and pseudostatic model [13]. Regarding the second disadvantage, fatigue influences on the 2DOF compliant mechanism can result in a failure under dynamic working conditions.
In similarity to rigidlink mechanisms, fatigue life of the 2DOF complaint mechanism is a term indicating how long an object will last before it fails due to concentrated stresses. Evaluation of the fatigue life is a critically important task in design process of compliant mechanisms so as to guarantee the working safety and reliability. Practically, compliant mechanisms are often subjected to cyclic and reversed loads. Under varying stress cycles, fatigue phenomena can appear in compliant mechanisms. So, it is necessary to predict the fatigue life to avoid any damage. Nevertheless, modeling of the fatigue life for the 2DOF compliant mechanism is so complicated task because the 2DOF compliant mechanism is a complex structure that has an integration of kinematic behaviors and mechanical ones. Therefore, the problem of fatigue prediction has been interested to research recently to prevent undesired failures. In [14], the artificial neural networks (ANNs) were used to predict the fatigue life of polyamide6 polymer. The results showed that the ANNbased approach can achieve a better performance than conventional methods. The adaptive neurofuzzy inference system (ANFIS) was used to model the fatigue behavior of a multidirectional laminate under four different loading conditions [15]. In the above study, both grid partitioning and subtractive clustering were used to build the ANFIS models. Nevertheless, a method for setting subtractive clustering parameters has not been conducted. The research in [16] provided a review of fatigue life modeling, fatigue life prediction, and application to composite structure design. The research studies in [17, 18] proposed a method for solving the partial differential equations using the deep neural network (DNN). This approach was then utilized for many applications in the field of Computational Mechanics. Various soft computing methods for predicting the fatigue life of different material systems were also studied in [19–28]. However, it can be noted that a study predicting the fatigue life for 2DOF compliant mechanism has not been conducted so far.
To fulfill the mentioned research gap, this paper develops a method based on the adaptive neurofuzzy inference system (ANFIS) with subtractive clustering for fatigue life prediction of 2DOF compliant mechanism. Furthermore, for searching the optimal radius parameters of the clustering algorithm, the 10fold crossvalidation and the differential evolution (DE) algorithm are used. Particularly, the average mean square error of 10folds is evaluated and optimized using the DE. The proposed framework is applied to predict the fatigue life of two cases of 2DOF compliant mechanism consisting of the material steel A36 and the material AL 6061T6 based on their design variables.
The rest of the paper is organized as follows. The related works including the adaptive neurofuzzy inference system, the subtractive clustering, and the differential evolution algorithm are presented in Section 2. The proposed framework is presented in Section 3. The new method is applied to predict fatigue life for 2DOF compliant mechanism consisting of steel materials A36 and AL 6061T6. In the conclusion, the main contribution, the results, and a few further research directions are summarized and discussed.
2. Related Works
2.1. Fuzzy Set
Fuzzy set, which was introduced by La [29], is a wellknown theory for modeling data uncertainty. A fuzzy set A on universe X is a set defined by the membership function which is a mapping from X into the interval [0, 1]. If , x completely belongs to the fuzzy set A. If , x does not belong to the fuzzy set. If , x is a partial member of the fuzzy set A. There are several ways to define the membership function . Some wellknown membership functions are the trapezoidal membership function, the triangular membership function, and the Gaussian membership function.
2.2. Fuzzy Inference System
The fuzzy inference system (FIS) [30] is a system that uses fuzzy set theory to map inputs to output (independent variables to dependent variable in the case of regression). There are two main types of FIS including the Mamdani and the Sugeno systems. This paper concentrates on the Sugeno system which builds a separate subregression model corresponding to each rule and calculates the final output as the weightmean of all subregression outputs. For example, let X be a script variable that can be mapped to two corresponding fuzzy sets A_{1} and A_{2}, Y be a script variable that can be mapped to two corresponding fuzzy sets B_{1} and B_{2}, and Z be the output or dependent variable. Z can be computed via the Sugeno fuzzy logic inference as follows: If X is A_{1} and Y is B_{1}, then with the degree of applicability is If X is A_{2} and Y is B_{2}, then with the degree of applicability is If X is A_{1} and Y is B_{2}, then with the degree of applicability is Finally, the output Z is computed using the formula
In the above formulas, is the component predicted value of fuzzy rule i, the predicted value can be considered as the weightmean of all component predicted values , and are userdefined. A disadvantage of the Sugeno FIS is that expert knowledge is required to determine the fuzzy set parameters and the coefficients p, q, and r; therefore, the FIS is a nonoptimal model and lacks learning ability. To improve the performance of FIS, [31] presented a hybrid model called ANFIS that combines the Sugeno FIS and the artificial neural networks. ANFIS can take advantage of the fuzzy reasoning of FIS and the learning capability of ANN. A brief review of ANFIS is presented as follows.
2.3. Adaptive NeuroFuzzy Inference System
Soft computing refers to an approach that attempts to model the behavior of a complex system so that an approximate solution of this system can be provided. By the approximation, soft computing not only can provide a relatively good result but also can significantly reduce the computational cost. Therefore, this approach has many applications in engineering problems, such as civil engineering [32–34], mining engineering [35–37], and agricultural and biological engineering [38]. The adaptive neurofuzzy inference system (ANFIS) proposed by [31] can take advantage of the fuzzy reasoning of FIS and the learning capability of ANN. Therefore, ANFIS has been successfully performed in classification and prediction tasks and has become one of the most wellknown soft computing methods so far. A simple illustration of an ANFIS model is presented in Figure 1.
It can be seen from Figure 1 that ANFIS consists of five layers that perform different functions. Let be the output of the ith node of the lth layer. Assuming that the variable X can be partitioned into two fuzzy sets, A_{1} and A_{2}, whereas the variable Y can be partitioned into two fuzzy sets B_{1} and B_{2}. In Layer 1, each node calculates the membership function and , . In other words, for and for . Some wellknown membership functions are, namely, Gaussian membership function , triangular membership function , and generalized bell membership function , where a, b, and c are called as premise parameters. Each node in Layer 2 is the AND operator that calculates the joint membership degree of in A_{j} and B_{k}, j, k = 1,2. Finding the product of individual membership degrees is a popular method modeling the AND operator; therefore, , j, k = 1,2. Another operator for calculating the joint membership degree is defined as . In fact, both T_{P} and T_{M} are some specific examples of a tnorm which is used to generalize the conjunction in fuzzy logic. For all a, b in [0, 1], ; that is, the logical strength represented by T_{P} is stronger than T_{M}. For more details of tnorms, readers can refer to [39]. As shown in Figure 1, node 1 in Layer 2 calculates the membership degree of X is A_{1} and Y is B_{1}, whereas node 2 in Layer 2 calculates the membership degree of X is A_{2} and Y is B_{2}. Certainly, more combinations between A_{i} and B_{k}, for example, X is A_{1} and Y is B_{2}, can be handled. For the sake of convenience, only two nodes are illustrated in Figure 1. Let be the ith rule membership or firing strength obtained in Layer 2. Layer 3 is utilized for the normalization purpose. In this layer, the output of node i is the ratio of the ith rule’s firing strength to the sum of all rules’ firing strengths. For example, from ANFIS architecture in Figure 1, it can be implied that , . Layer 4 calculates the reasoning result for each fuzzy rule using the formula where and are called consequent parameters. Finally, Layer 5 calculates the overall output using the sum of all , that is, .
ANFIS can optimize all parameters using a hybrid learning method. In the forward pass, the leastsquares method is utilized to identify the consequent parameters in Layer 4. In the backward pass, the gradient descent method is utilized to identify the premise parameters in Layer 1. Therefore, the ANFIS inherits both the fuzzy reasoning of FIS and the learning abilities of ANN.
The remaining problem of ANFIS is how to identify the number of fuzzy sets for each variable. The conventional ANFIS uses the grid partitioning method that requires a given number of fuzzy sets defined by expert knowledge. Furthermore, this method causes the dimensional curse when applied to largescale data. For example, given an input of 12 variables, if each variable can be partitioned into two fuzzy sets, the grid partitioning method results in 2^{12} = 4096 rules. This number is very large for any practical learning method. Therefore, the conventional ANFIS with the grid partitioning method might be not completely objective and might be infeasible when dealing with largescale data. An alternative method is using ANFIS with subtractive clustering technique, which is presented in the following section.
2.4. Subtractive Clustering
Cluster analysis is to discover the underlying structure of a dataset by partitioning the data into groups such that each obtained cluster (group) is a set of similar data points [40]. In other words, all the elements in a cluster share the same characteristic behavior of the system. Therefore, each cluster’s center can be considered as the basis of a rule that represents a system behavior. Because clustering can create a concise representation of the system behavior, it can reduce the number of fuzzy rules or prevent rule explosion. Let be N ddimensional normalized data points. Subtractive clustering [41], which is often applied to ANFIS, is summarized in the following steps: Step 1: normalize the data set, set k = 1, and calculate the likelihood that each data point is a cluster center, using where P_{i} is the likelihood of point x_{i} and is the neighborhood radius in dimension m. Step 2: select the data point with the highest likelihood as the kth cluster’s center. Step 3: update the likelihood that each data point is a cluster center, using formula where and denote the kth cluster center and its likelihood, respectively. is a constant. Using formula (2), the closer the distance between x_{i} and is, the more reduction of its likelihood is. Therefore, all points with an updated P_{i} of 0 are assigned to cluster k and are not considered in the next steps. Step 4: repeat Steps 2 and 3 until all points are assigned to their clusters.
In the above algorithm, the value of r_{m} has an important effect on the clustering result. When r_{m} tends to zero, the algorithm tends to result in N single clusters, each of which has only one element. In contrast, when r_{m} tends to one, the algorithm tends to result in a single cluster of N elements. It is not easy to determine a suitable value for r_{m} that can work well for all case studies. Therefore, an additional framework for optimizing the set of r_{m} is developed. The developed framework is based on the differential evolution algorithm, which will be presented in the next section.
2.5. Differential Evolution Algorithm
The differential evolution algorithm, DE, which is a wellknown global search method based on population, is designed to solve both continuous and discrete problems [42]. To clarify the used notation, this article refers to the minimization of the objective function , where x is a vector of d variables in the decision space and x_{l} and x_{u} are the vectors of lower bound and upper bound, respectively. The DE seeks an optimal solution through generations (iterations). In each generation, the DE evolves a population P of size NP, . Each element in this set or each feasible solution is called a chromosome, which is a vector of d variables, socalled genes. The four major operators of the DE algorithm, which include initialization, mutation, crossover, and selection, are briefly summarized as follows.
2.5.1. Initialization
In this phase, an initial population of NP chromosomes is generated through a random sampling technique. Specifically, each individual is represented as a chromosome containing N genes and is generated bywhere , NP is the size of the population, and and are the lower and upper limits of , respectively [43].
2.5.2. Mutation
In this phase, each individual x_{i} in the current population generates a mutant vector using some mutation operators listed as follows: rand/1: rand/2: best/1: best/2: currenttobest/1: where r_{1}, r_{2}, r_{3}, r_{4}, and r_{5} are mutually exclusive integers randomly chosen in {1,2, …, NP}, F is the scale factor randomly selected in the range [0, 2], and x is the best individual in the current population.
2.5.3. Crossover
After completing mutation, each target vector x_{i} produces a trial vector u_{i} by substituting some components of the vector x_{i} by some components of the mutant vector v_{i} through the following binomial crossover operation [43]:where i∈{1,2, …, NP}, j∈{1,2, …, d}, and CR is the crossover control parameter chosen in [0,1].
2.5.4. Selection
In this phase, each trial vector u_{i} created after the crossover phase will be evaluated and compared with the target vector x_{i} to choose a better individual for the next generation.
2.6. Evaluating an ANFIS Model
Let y_{i} and be the ith actual and predictive data points, respectively. Some criteria for evaluating the ANFIS model are presented as follows: Mean absolute error: Mean squared error: Mean absolute percentage error:
2.7. KFold CrossValidation Method
Kfold crossvalidation is a typical process being often incorporated into the training process for avoiding the overfitting problem [44–46]. In this procedure, the original training data set is randomly divided into k subsets. k−1 subsets are used as new training data sets to develop an ANFIS model. The remainder is used for evaluating the model using the criteria presented above. This process is repeated k times until every subset is used once for evaluating. Finally, the overall performance is calculated by taking the average of k individual measures. The ANFIS with the lowest average of MAE, MSE, and MAPE can be considered as the most suitable model. Additionally, it can be noted that the test data set is not used in the training and crossvalidation process but is only used in the final validation when the optimized ANFIS has been already built. An illustration for kfold crossvalidation process is presented in Figure 2 where k is set as three and is the ratio of the test set compared to the data set.
3. The Proposed Method
This section develops a method for fatigue life prediction of 2DOF compliant mechanism. The ANFIS model which to be estimated has the form Y = f(X_{1}, X_{2},…, X_{d}) where X_{1}, X_{2},…, X_{d} are d design variables and Y is the dependent variable. Using the subtractive clustering technique, the model performance relies on the choice of radius parameters r_{i} (0< = r_{i} < = 1, i = 1, 2, …, d + 1). Therefore, the 10fold crossvalidation is used and the differential evolution algorithm is utilized for optimizing the mean square error (MSE) average over 10folds. Because the DE defines a solution by a chromosome, each solution needs to be first encoded. The encoding phase is summarized in Figure 3. As shown in Figure 3, a possible solution is defined as a chromosome. Each chromosome contains d + 1 genes representing the corresponding radii of the design variables and the dependent variables. The MSE, whose values may be different for different chromosomes, can be considered as an implicit objective function of r_{i}.
After the encoding phase is completed, chromosomes or solutions can be handled with mutation, crossover, and selection operators. Through each iteration i, only a fixed number of solutions can be selected for the next iteration. After a given number of iterations, maxiter, a chromosome that provides the smallest MSE average over 10folds can be considered as the best solution and can be used as the parameters of the ANFIS. The overall process of the proposed method is outlined in Figure 4.
4. Numerical Example
4.1. The Data Set
Figure 5 gives a design scheme of a new 2DOF compliant mechanism. This mechanism includes a shuttle table which is located at the middle position. The shuttle table can move in the xdirection when a piezostack actuator (PSAX) applies a force to an end of the table. Besides, when a force comes from the PSAY, the shuttle table is moved in the ydirection. The overall movement of the shuttle table is based on the elastic deflection of flexure hinges, including flexure hinge 1 (thickness T_{1} and length V_{1}), flexure hinge 2 (thickness T_{2} and length H), and flexure hinge 3 (length V_{2}). Overall dimensions of the mechanism are given, as in Figures 5(a) and 5(b). The width of the mechanism is b of 10 mm (see Figure 5(c)). It is noted that the motions of flexure hinges are repeated due to these hinges are worked in an elastic area. Therefore, the mechanism is easy to subject a fatigue failure. Predicting fatigue life for the mechanism is critically important to prevent failure and ensure a safety operation.
During the design phase, manufacturing material affects the fatigue life of the 2DOF mechanism because each material has a specific limitation of strength. This section applies the proposed method for predicting the fatigue life of the new 2DOF compliant mechanism in two cases: material steel A36 and material AL 6061T6. The physical properties, chemical components of steel A36, and Al 6061T6 are shown in Table 1, whereas the alternative stresses of the two materials are shown in Table 2 [47, 48]. In both cases, a load, F, of 20 N is applied the shuttle table along the xaxis and fixed supports are located by screws at three holes, as depicted in Figure 5.


After boundary conditions are applied for the model, the model is meshed. A 3D model of the 2DoF mechanism is built for finite element analysis, as shown in Figure 6. However, uncertainties and errors are critically important for any data. Because the datasets of the fatigue life are determined through finite element simulations in ANSYS software, the numerical datasets are strongly influenced by some controllable parameters. Datasets are therefore also affected by those parameters. First of all, a few main parameters of the meshing model, such as number of elements, number of nodes, and type of elements, are determined to permit a good analysis. Adjusting any parameter also leads to a difference result. So, before implementing the simulations, a meshing quality is evaluated to ensure a good predicting accuracy. The mechanism is meshed with 55376 nodes and 24884 elements. The type of tennode tetrahedral elements is utilized for the meshing model. Meshing quality is evaluated, as shown in Figure 7. It is noted that a better meshing quality ensures a more simulation accuracy. The results indicated that the meshing distribution is relatively good with a metric average value of 0.2546 and a standard deviation of 0.1891.
Datasets of the fatigue life are then determined by simulations in ANSYS software R19 and are summarized in Table 3. It can be noted in Table 3 that the range of input variables is set by the user, but the range of output variable is based on the simulated results. Furthermore, the natural log transformation is applied to the output variable for producing smaller values. The data sets can be also downloaded at https://sites.google.com/tdtu.edu.vn/nguyentrangthao/home.

5. Results and Discussion
The performance of the proposed method is compared (subtractive clustering + DE) with four different ANFIS structures consisting of grid partitioning + triangular membership function (Model 1), grid partitioning + bellshaped membership function (Model 2), grid partitioning + Gaussian membership function (Model 3), and subtractive clustering + random range of influence r (Model 4). The considered problems have five design variables and one output variable. For the traditional ANFIS with the grid partitioning method, the number of fuzzy is set as 2, for each variable. Based on the above information, the parameters of Models 1, 2, and 3 can be determined. Particularly, the number of nodes, the number of linear parameters, and the number of fuzzy rules are, respectively, 92, 192, and 32, for all the three models; the number of nonlinear parameters is 30 for Model 1 and 2 and is 20 for Model 3. For Model 4 and the proposed method, the above parameters do not need to be prior determined but depend upon the value of radius r, which are randomly chosen in Model 4 and are optimized using DE in the proposed method. For the proposed method, the convergence behavior depends upon the hyperparameters of DE, such as mutation rate F, crossover rate CR, and stop criteria. Nevertheless, a comprehensive investigation for all above parameters is inefficient in practice. In this paper, firstly, some hyperparameters of DE are selected based on the investigations in [43, 49, 50]. Specifically, the size of population NP is 20, the mutation rate F is randomly chosen in the range [0.4, 1], crossover rate CR is randomly chosen in the range [0.7, 1], and the tolerance Tol is 10−6. Finally, the number of maximum iterations maxiter is investigated.
Figure 8 illustrates the convergence behavior of DE regarding the iterations, for some cases. It can be seen that the value of objective function sharply decreases at some first iterations. After 10–15 iterations, the value of the objective function still decreases but is not significant. In this study, the number of maximum iteration maxiter is set as 20 to trade off the quality of objective function and the computational cost. In other words, to design an efficient algorithm, a relative optimal solution rather than an exact optimal solution is chosen.
Let be the ratio of the test set compared to the data set. For each material, the performance of the proposed method for different values of will be evaluated. For each , the training set is first divided into kfolds. Applying the DE algorithm to training data, the optimal ANFIS that minimizes the average MSE (AMSE) over the kfolds can be found. Table 4 shows the optimal results for different values of and materials.

Based on the obtained optimal parameters, the respective ANFIS networks can be built. The obtained models are next evaluated on the test data set. Their predictions are, respectively, shown in Figures 9 and 10, for some typical cases. It can be observed from Figures 9 and 10 that the three ANFIS models with the grid partitioning method result in underfitting in some instances. Meanwhile, most of instances are quite well predicted using the two ANFIS models with the subtractive clustering method. Compared to the subtractive clustering with random values of r, the subtractive clustering with DE has a slightly better prediction (9/16 and 9/14 samples for Case 1 and Case 2, respectively).
The average MSE of the five comparative ANFIS models for different values is presented in Figures 11 and 12 (two cases of materials). It can be seen that the average MSE in terms of the natural logarithm of number cycles of the proposed method lies in the range [0.28, 1.92] for Case 1 and [0.50, 5.52] for Case 2; that is, the average RMSEs in terms of the number of cycles of the proposed method lie in the range [1.7, 3.97] for Case 1 and [2.03, 10.38] for Case 2. These results show that the accuracy of the proposed method is very high. Besides, it can be observed that the effect of the ratio of test data on the performances of two models with subtractive clustering is not so strong. As shown in Figures 11 and 12, in both cases, their average MSEs slightly fluctuate around small values even for high values of . Meanwhile, it can be observed that the average MSEs of other ANFIS models will increase with either too high or too small values.
The box plots in Figures 13 and 14 provide some statistical visualization of the performance of all methods for two kinds of materials. Among the three ANFIS models using the grid partitioning method, it can be seen from Figures 13 and 14 that the model with triangular membership function has a better result for steel A36 but not for Al 6061T6. Hence, the most suitable membership function cannot be determined. It might depend on which material is considered. These figures also demonstrate the superiority of the two ANFIS models using subtractive clustering over the other models. In comparison with the subtractive clustering with the random values of r, the subtractive clustering with DE has a slightly better performance in both cases. As shown in Figures 13 and 14, the median of the subtractive clustering with DE is lower than those of subtractive clustering with random values of r. Besides, the boxes, which stand for the interquartile range (from 25th percentile to 75th percentile), of the subtractive clustering with DE are also lower than those of subtractive clustering with random values of r, for both cases.
Although Figures 13 and 14 provide the first insight into the performance of all ANFIS models, it is more reliable if a statistical test could be conducted. For this purpose, the Wilcoxon signedrank test with a statistical significance value of 0.1 is used. The null hypothesis H0 is built as “there is no difference between the performance of the method A and the performance of the method B,” where A is the subtractive clustering with DE and B is a comparative method (grid partitioning + triangular membership function, grid partitioning + bellshaped membership function, grid partitioning + Gaussian membership function, and subtractive clustering + random range of influence r). According to Table 5, it can be concluded that the subtractive clustering with DE can improve the performance of ANFIS models using the grid partitioning method, with a significance level of 0.1. The difference between subtractive clustering with DE and the subtractive clustering with random r is not significant due to the following reasons. Firstly, the difference between the subtractive clustering with DE and the subtractive clustering with random r is not as large as the differences observed in comparison with other methods. Secondly, the sample size is not large enough for rejection of the null hypothesis H0. Therefore, the subtractive clustering with random r can be used for other cases in which the computational time is important. On the other hand, more experiments on other ratios of test data and other materials are required to provide a comprehensive comparison. The results show an efficient application of clusteringbased ANFIS in fatigue life prediction of 2DOF compliant mechanism.

In summary, the above experiments have demonstrated the potential benefit of ANFIS in fatigue life prediction of 2DOF compliant mechanism. Particularly, the two ANFIS models using subtractive clustering provide better performance compared to the three ANFIS models using the grid partitioning method. In comparison with the subtractive clustering with random r, the subtractive clustering using DE has slightly better performance, although the difference is not significant. Based on the obtained results, it can be expected that the proposed method can be wellapplied for other case studies (materials). The root mean square error of the fatigue life prediction is expected to be smaller than 100 cycles or smaller than 22 in terms of mean square error (MSE) of natural log transformation. Using this criterion, careful users can evaluate the suitability of the proposed method after obtained the training model to decide whether the proposed method can be applied or not.
6. Conclusions
This paper for the first time develops a method for fatigue life prediction of 2DOF compliant mechanism. The adaptive neurofuzzy inference system (ANFIS) is utilized for the prediction purpose. In addition to the conventional ANFIS model using the grid partitioning method, the ANFIS model using subtractive clustering is applied. Furthermore, the differential evolution algorithm is utilized for optimizing parameters of the clustering algorithm. Five ANFIS models are applied to two case studies consisting of the steel A36 and the AL 6061T6. The obtained results show that the average root mean square error in terms of the number of cycles lies in the range [1.7, 3.97] for Case 1 and [2.03, 10.38] for Case 2. Compared to the ANFIS models using the grid partitioning method, the proposed method achieves a better performance with a significance level from 0.05 to 0.1. Although the subtractive clustering using DE is slightly better than the subtractive clustering with random r, the respective difference has no statistical significance. The experiments also demonstrate the potential application of ANFIS in fatigue life prediction of 2DOF compliant mechanism. The results can be extended in predicting the fatigue life for other compliant mechanisms and related engineering applications. In future work, some physical prototypes can be fabricated to estimate the fatigue life for the proposed 2DOF compliant mechanism.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This research was funded by the Vietnam National Foundation for Science and Technology Development (NAFOSTED) under grant no. 107.012019.14.
References
 T.P. Dao and S.C. Huang, “Design and multiobjective optimization for a broad selfamplified 2DOF monolithic mechanism,” SadhanaAcademy Proceedings in Engineering Sciences, vol. 42, pp. 1527–1542, 2017. View at: Publisher Site  Google Scholar
 W.L. Zhu, Z. Zhu, Y. Shi, X. Wang, K. Guan, and B.F. Ju, “Design, modeling, analysis and testing of a novel piezoactuated XY compliant mechanism for large workspace nanopositioning,” Smart Materials and Structures, vol. 25, no. 11, Article ID 115033, 2016. View at: Publisher Site  Google Scholar
 Q. Xu, “New flexure parallelkinematic micropositioning system with large workspace,” IEEE Transactions on Robotics, vol. 28, no. 2, pp. 478–491, 2012. View at: Publisher Site  Google Scholar
 T.P. Dao and S.C. Huang, “Optimization of a two degrees of freedom compliant mechanism using Taguchi methodbased grey relational analysis,” Microsystem Technologies, vol. 23, pp. 4815–4830, 2017. View at: Publisher Site  Google Scholar
 S.C. Huang and T.P. Dao, “Multiobjective optimal design of a 2DOF flexurebased mechanism using hybrid approach of greytaguchi coupled response surface methodology and entropy measurement,” Arabian Journal for Science and Engineering, vol. 41, no. 12, 2016. View at: Publisher Site  Google Scholar
 S.C. Huang and T.P. Dao, “Design and computational optimization of a flexurebased XY positioning platform using FEAbased response surface methodology,” International Journal of Precision Engineering and Manufacturing, vol. 17, no. 8, 2016. View at: Publisher Site  Google Scholar
 B. Ding, Y. Li, X. Xiao, Y. Tang, and B. Li, “Design and analysis of a 3DOF planar micromanipulation stage with large rotational displacement for micromanipulation system,” Mechanical Sciences, vol. 8, no. 1, p. 117, 2017. View at: Publisher Site  Google Scholar
 B. Ding, Z. Yang, and Y. Li, “Design of flexurebased modular architecture micropositioning stage,” Microsystem Technologies, vol. 26, no. 9, pp. 2893–2901, 2020. View at: Publisher Site  Google Scholar
 B. Ding, Z.X. Yang, X. Xiao, and G. Zhang, “Design of reconfigurable planar micropositioning stages based on function modules,” IEEE Access, vol. 7, pp. 15102–15112, 2019. View at: Publisher Site  Google Scholar
 E. G. Merriam, J. E. Jones, S. P. Magleby, and L. L. Howell, “Monolithic 2 DOF fully compliant space pointing mechanism,” Mechanical Sciences, vol. 4, no. 2, pp. 381–390, 2013. View at: Publisher Site  Google Scholar
 W.L. Zhu, Z. Zhu, Y. Shi et al., “A novel piezoelectrically actuated 2DoF compliant micro/nanopositioning stage with multilevel amplification,” Review of Scientific Instruments, vol. 87, no. 10, Article ID 105006, 2016. View at: Publisher Site  Google Scholar
 N. Wang, Z. Zhang, X. Zhang, and C. Cui, “Optimization of a 2DOF micropositioning stage using corrugated flexure units,” Mechanism and Machine Theory, vol. 121, pp. 683–696, 2018. View at: Publisher Site  Google Scholar
 M. Ling, J. Cao, Z. Jiang, M. Zeng, and Q. Li, “Optimal design of a piezoactuated 2DOF millimeterrange monolithic flexure mechanism with a pseudostatic model,” Mechanical Systems and Signal Processing, vol. 115, pp. 120–131, 2019. View at: Publisher Site  Google Scholar
 J. Yang, G. Kang, Y. Liu, K. Chen, and Q. Kan, “Life prediction for ratedependent lowcycle fatigue of PA6 polymer considering ratchetting: semiempirical model and neural network based approach,” International Journal of Fatigue, vol. 136, Article ID 105619, 2020. View at: Publisher Site  Google Scholar
 A. P. Vassilopoulos and R. Bedi, “Adaptive neurofuzzy inference system in modelling fatigue life of multidirectional composite laminates,” Computational Materials Science, vol. 43, no. 4, pp. 1086–1093, 2008. View at: Publisher Site  Google Scholar
 A. P. Vassilopoulos, Fatigue Life Prediction of Composites and Composite Structures, Woodhead Publishing, Sawston, UK, 2019.
 C. Anitescu, E. Atroshchenko, N. Alajlan, and T. Rabczuk, “Artificial neural network methods for the solution of second order boundary value problems,” Computers, Materials & Continua, vol. 59, no. 1, pp. 345–359, 2019. View at: Publisher Site  Google Scholar
 E. Samaniego, C. Anitescu, S. Goswami et al., “An energy approach to the solution of partial differential equations in computational mechanics via machine learning: concepts, implementation and applications,” Computer Methods in Applied Mechanics and Engineering, vol. 362, Article ID 112790, 2020. View at: Publisher Site  Google Scholar
 Q. Liu, W. Shi, and Z. Chen, “Rubber fatigue life prediction using a random forest method and nonlinear cumulative fatigue damage model,” Journal of Applied Polymer Science, vol. 137, no. 14, p. 48519, 2020. View at: Publisher Site  Google Scholar
 J. C. Stinville, W. C. Lenthe, M. P. Echlin, P. G. Callahan, D. Texier, and T. M. Pollock, “Microstructural statistics for fatigue crack initiation in polycrystalline nickelbase superalloys,” International Journal of Fracture, vol. 208, no. 12, pp. 221–240, 2017. View at: Publisher Site  Google Scholar
 K. V. Anderson and S. R. Daniewicz, “Statistical analysis of the influence of defects on fatigue life using a Gumbel distribution,” International Journal of Fatigue, vol. 112, pp. 78–83, 2018. View at: Publisher Site  Google Scholar
 T. Loutas, N. Eleftheroglou, and D. Zarouchas, “A datadriven probabilistic framework towards the insitu prognostics of fatigue life of composites based on acoustic emission data,” Composite Structures, vol. 161, pp. 522–529, 2017. View at: Publisher Site  Google Scholar
 D. Shi and H. Ma, “A bayesian inference method and its application in fatigue crack life prediction,” IEEE Access, vol. 7, pp. 118381–118387, 2019. View at: Publisher Site  Google Scholar
 W. Sai, G. B. Chai, and N. Srikanth, “Fatigue life prediction of GLARE composites using regression tree ensemblebased machine learning model,” Advanced Theory and Simulations, vol. 3, no. 6, Article ID 2000048, 2020. View at: Google Scholar
 D. Nowell and P. W. Nowell, “A machine learning approach to the prediction of fretting fatigue life,” Tribology International, vol. 141, Article ID 105913, 2020. View at: Publisher Site  Google Scholar
 M. Zhang, C.N. Sun, X. Zhang et al., “High cycle fatigue life prediction of laser additive manufactured stainless steel: a machine learning approach,” International Journal of Fatigue, vol. 128, Article ID 105194, 2019. View at: Publisher Site  Google Scholar
 Y. AlAssaf and H. El Kadi, “Fatigue life prediction of unidirectional glass fiber/epoxy composite laminae using neural networks,” Composite Structures, vol. 53, no. 1, pp. 65–71, 2001. View at: Publisher Site  Google Scholar
 J. A. Lee, D. P. Almond, and B. Harris, “The use of neural networks for the prediction of fatigue lives of composite materials,” Composites Part A: Applied Science and Manufacturing, vol. 30, no. 10, pp. 1159–1169, 1999. View at: Publisher Site  Google Scholar
 Z. La, “Fuzzy sets,” Information and Control, vol. 8, pp. 338–353, 1965. View at: Google Scholar
 T. Takagi and M. Sugeno, “Fuzzy identification of systems and its applications to modeling and control,” IEEE Transactions on Systems, Man, and Cybernetics, vol. 15, no. 1, pp. 116–132, 1985. View at: Publisher Site  Google Scholar
 J.S. R. Jang, “ANFIS: adaptivenetworkbased fuzzy inference system,” IEEE Transactions on Systems, Man, and Cybernetics, vol. 23, no. 3, pp. 665–685, 1993. View at: Publisher Site  Google Scholar
 D. DinhCong, S. PhamDuy, and T. NguyenThoi, “Damage detection of 2D frame structures using incomplete measurements by optimization procedure and model reduction,” Journal of Advanced Engineering and Computation, vol. 2, no. 3, pp. 164–173, 2018. View at: Publisher Site  Google Scholar
 H. Naderpour and M. Mirrashid, “Shear failure capacity prediction of concrete beamcolumn joints in terms of ANFIS and GMDH,” Practice Periodical on Structural Design and Construction, vol. 24, no. 2, Article ID 4019006, 2019. View at: Publisher Site  Google Scholar
 H. Naderpour and M. Mirrashid, “Moment capacity estimation of spirally reinforced concrete columns using ANFIS,” Complex & Intelligent Systems, vol. 6, no. 1, pp. 97–107, 2020. View at: Publisher Site  Google Scholar
 R. Mikaeil, S. S. Haghshenas, Y. Ozcelik, and H. H. Gharehgheshlagh, “Performance evaluation of adaptive neurofuzzy inference system and group method of data handlingtype neural network for estimating wear rate of diamond wire saw,” Geotechnical and Geological Engineering, vol. 36, no. 6, pp. 3779–3791, 2018. View at: Publisher Site  Google Scholar
 A. R. Dormishi, M. Ataei, R. Khaloo Kakaie, R. Mikaeil, and S. Shaffiee Haghshenas, “Performance evaluation of gang saw using hybrid ANFISDE and hybrid ANFISPSO algorithms,” International Journal of Surface Mining Reclamation and Environment, vol. 10, no. 2, pp. 543–557, 2019. View at: Google Scholar
 A. Aryafar, R. Mikaeil, S. S. Haghshenas, and S. S. Haghshenas, “Application of metaheuristic algorithms to optimal clustering of sawing machine vibration,” Measurement, vol. 124, pp. 20–31, 2018. View at: Publisher Site  Google Scholar
 Y. Huang, Y. Lan, S. J. Thomson, A. Fang, W. C. Hoffmann, and R. E. Lacey, “Development of soft computing and applications in agricultural and biological engineering,” Computers and Electronics in Agriculture, vol. 71, no. 2, pp. 107–127, 2010. View at: Publisher Site  Google Scholar
 R. Babuška, Fuzzy Modeling for Control, vol. 12, Springer Science & Business Media, Berlin, Germany, 2012.
 T. VoVan and T. NguyenTrang, “Similar coefficient for cluster of probability density functions,” Communications in StatisticsTheory and Methods, vol. 47, no. 8, pp. 1792–1811, 2018. View at: Publisher Site  Google Scholar
 S. L. Chiu, “Fuzzy model identification based on cluster estimation,” Journal of Intelligent and Fuzzy Systems, vol. 2, no. 3, pp. 267–278, 1994. View at: Publisher Site  Google Scholar
 R. Storn and K. Price, “Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces,” Journal of Global Optimization, vol. 11, no. 4, pp. 341–359, 1997. View at: Publisher Site  Google Scholar
 N.T. Thao, “An improved fuzzy time series forecasting model using the differential evolution algorithm,” Journal of Intelligent & Fuzzy Systems, vol. 36, no. 2, pp. 1727–1741, 2019. View at: Publisher Site  Google Scholar
 G. J. McLachlan, K.A. Do, and C. Ambroise, Analyzing Microarray Gene Expression Data, vol. 422, John Wiley & Sons, Hoboken, NJ, USA, 2005.
 J. Zhou, X. Li, and H. S. Mitri, “Classification of rockburst in underground projects: comparison of ten supervised learning methods,” Journal of Computing in Civil Engineering, vol. 30, no. 5, Article ID 4016003, 2016. View at: Publisher Site  Google Scholar
 G. Vanwinckelen and H. Blockeel, “On estimating model accuracy with repeated crossvalidation,” in Benelearn 2012: Proceedings of the 21st BelgianDutch Conference on Machine Learning, pp. 39–44, Ghent, Belgium, March 2012. View at: Google Scholar
 T. L. Boyer and E. Howard, Metals Handbook, American Society for Metals, Materials Park, OH, USA, 1985.
 P. D. Harvey, Engineering Properties of Steel, ASM International, Cleveland, OH, USA, 1982.
 V. HoHuu, T. NguyenThoi, T. VoDuy, and T. NguyenTrang, “An adaptive elitist differential evolution for optimization of truss structures with discrete design variables,” Computers & Structures, vol. 165, pp. 59–75, 2016. View at: Publisher Site  Google Scholar
 T. NguyenTrang, T. NguyenThoi, T. TruongKhac, A. T. PhamChau, and H. Ao, “An efficient hybrid optimization approach using adaptive elitist differential evolution and spherical quadratic steepest descent and its application for clustering,” Scientific Programming, vol. 2019, Article ID 7151574, 2019. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2021 Ngoc Thoai Tran et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.