Taoyuan, Taiwan
Taoyuan, Taiwan
SEARCH FILTERS
Time filter
Source Type

Lin J.,National Taipei University of Technology | Lian R.-J.,VaNung University
IEEE Transactions on Industrial Electronics | Year: 2011

A self-organizing fuzzy controller (SOFC) has been proposed to control engineering applications. During the control process, the SOFC continually updates the learning strategy in the form of fuzzy rules, beginning with empty fuzzy rules. This eliminates the problem of finding appropriate membership functions and fuzzy rules for the design of a fuzzy logic controller. It is, however, arduous to select appropriate parameters (learning rate and weighting distribution) in the SOFC for control engineering applications. To solve the problem caused by the SOFC, this study developed a hybrid self-organizing fuzzy and radial basis-function neural-network controller (HSFRBNC). The HSFRBNC uses a radial basis-function neural-network (RBFN) to regulate in real time these parameters of the SOFC, so as to gain optimal values, thereby overcoming the problem of the SOFC application. To confirm the applicability of the proposed HSFRBNC, the HSFRBNC was applied in manipulating an active suspension system. Then, its control performance was evaluated. Simulation results demonstrated that the HSFRBNC offers better control performance than the SOFC in improving the service life of the suspension system and the ride comfort of a car. © 2011 IEEE.


Lian R.-J.,VaNung University
Information Sciences | Year: 2012

A self-organizing fuzzy controller (SOFC) under system control has online learning capabilities; nevertheless, the SOFC may excessively modify its fuzzy rules when its learning rate and weighting distribution are inappropriately selected. This results in oscillatory phenomena in the system during the control process. Moreover, the SOFC is mainly used to manipulate single-input single-output systems. When it is used to handle robotic systems, which are multiple-input multiple-output systems, the dynamic coupling effects between degrees of freedom (DOF) in the robotic systems are difficult to eliminate. To eliminate these problems, this study developed a grey-prediction self-organizing fuzzy controller (GPSOFC) for robotic systems. The GPSOFC introduces a grey-prediction algorithm into the SOFC to pre-correct fuzzy rules to reasonable ones for the control of robotic systems. This solves the problem caused by the inappropriate selection of parameters in the SOFC and compensates for the dynamic coupling effects between the DOFs in the robotic systems. To evaluate the feasibility of the proposed method, this study used the GPSOFC to manipulate a 6-DOF robot to determine its control performance. The GPSOFC yielded better control performance than the SOFC for robotic motion control, as shown in experimental results. © 2012 Elsevier Inc. All rights reserved.


Lian R.-J.,VaNung University
Expert Systems with Applications | Year: 2012

It is difficult to determine the stability of a self-organizing fuzzy controller (SOFC). Therefore, a system controlled using the SOFC cannot guarantee the stability of the system during the control process. To eliminate the problem caused by the SOFC, this study developed an enhanced self-organizing fuzzy sliding-mode controller (EASFSC) for robotic systems. Instead of using the system's output error and its error change, the EASFSC uses a sliding surface and its differentiation as the input variables of a fuzzy logic controller (FLC) in the SOFC. Using the fuzzy operation, these variables generate a control input that ensures the stability of the system. The proposed method also employs an adaptive law to modify the fuzzy consequent parameter of the FLC in the SOFC to improve the control performance of the system. The stability of the EASFSC has been proven using the Lyapunov stability theorem. Simulation results of a two-link robotic manipulator application verified that the EASFSC provides superior control performance as compared with the SOFC. © 2011 Elsevier Ltd. All rights reserved.


Liu S.-T.,VaNung University
Expert Systems with Applications | Year: 2011

Conventional portfolio optimization models have an assumption that the future condition of stock market can be accurately predicted by historical data. However, no matter how accurate the past data is, this premise will not exist in the financial market due to the high volatility of market environment. This paper discusses the fuzzy portfolio optimization problem where the asset returns are represented by fuzzy data. A mean-absolute deviation risk function model and Zadeh's extension principle are utilized for the solution method of portfolio optimization problem with fuzzy returns. Since the parameters are fuzzy numbers, the gain of return is a fuzzy number as well. A pair of two-level mathematical programs is formulated to calculate the upper bound and lower bound of the return of the portfolio optimization problem. Based on the duality theorem and by applying the variable transformation technique, the pair of two-level mathematical programs is transformed into a pair of ordinary one-level linear programs so they can be manipulated. It is found that the calculated results conform to an essential idea in finance and economics that the greater the amount of risk that an investor is willing to take on, the greater the potential return. An example, which utilizes the data from Taiwan stock exchange corporation, illustrates the whole idea on fuzzy portfolio optimization problem. © 2011 Elsevier Ltd. All rights reserved.


Lian R.-J.,VaNung University
IEEE Transactions on Industrial Electronics | Year: 2011

A self-organizing fuzzy controller (SOFC) has been developed to control complicated and nonlinear systems. However, it is arduous to choose an appropriate learning rate and a suitable weighting distribution of the SOFC to achieve satisfactory performance for system control. Furthermore, the SOFC is mainly used to control single-input single-output systems. When the SOFC is applied to manipulating a robotic system, which is an example of multiple-input multiple-output systems, it is difficult to eliminate the dynamic coupling effects between the degrees of freedom (DOFs) of the robotic system. To address the problems, this study developed a self-organizing fuzzy radial basis-function neural-network (RBFN) controller (SFRBNC) for robotic systems. The SFRBNC uses an RBFN to regulate in real time these parameters of the SOFC to optimal values, thereby solving the problem faced when the SOFC is applied. The RBFN has coupling weighting regulation ability, so it can eliminate the dynamic coupling effects between the DOFs for robotic system control. From the experimental results of the 6-DOF robot tests, the SFRBNC demonstrated better control performance than the SOFC. © 2006 IEEE.


Liu S.-T.,VaNung University
Expert Systems with Applications | Year: 2011

Financial holding companies in Taiwan play an important role in the process of economic development. Facing the financial globalization and market liberalization, competition between financial holding companies is growing. In data envelopment analysis (DEA) studies, the efficiency measurement can have a two-stage structure. A common approach to the two-stage problem is to apply a standard DEA model separately in each stage; such approaches treat the stages in a two-stage process as operating independently of one another. Different from previous studies, this paper takes the series relationship of the two individual stages into account in measuring the profitability and marketability efficiencies of the Taiwan financial holding companies. It is found that the overall efficiencies of all financial holding companies are inefficient and the low efficiency score of the whole process is mainly due to the low efficiency score of the marketability process. Decomposing the overall efficiency into the component efficiencies helps a company identify the stage that causes inefficiency. © 2010 Elsevier Ltd. All rights reserved.


Liu S.-T.,VaNung University
Computers and Industrial Engineering | Year: 2012

The integration of production and marketing planning is crucial in practice for increasing a firm's profit. However, the conventional inventory models determine the selling price and demand quantity for a retailer's maximal profit with exactly known parameters. When the demand quantity, unit cost, and production rate are represented as fuzzy numbers, the profit calculated from the model should be fuzzy as well. Unlike previous studies, this paper develops a solution method to find the fuzzy profit of the integrated production and marketing planning problem when the demand quantity, unit cost, and production rate are represented as fuzzy numbers. Based on Zadeh's extension principle, we transform the problem into a pair of two-level mathematical programming models to calculate the lower bound and upper bound of the fuzzy profit. According to the duality theorem of geometric programming technique, the two-level mathematical program is transformed into the one-level conventional geometric program to solve. At a specific α-level, we can derive the global optimum solutions for the lower and upper bounds of the fuzzy profit by applying well-developed theories of geometric programming. Examples are given to illustrate the whole idea proposed in this paper. © 2012 Elsevier Ltd. All rights reserved.


Liu S.-T.,VaNung University
Computers and Industrial Engineering | Year: 2014

Data envelopment analysis (DEA) is a method for measuring the efficiency of peer decision making units (DMUs). Recently DEA has been extended to examine the efficiency of two-stage processes, where all the outputs from the first stage are intermediate measures that make up the inputs to the second stage. Two-stage DEA models do not require a priori specification of input and output weights (or multipliers) and by letting these weights run freely, estimates of system and process efficiencies are obtained. This paper proposes a methodology for a fuzzy two-stage DEA model, where the weights are restricted in ranges and input-output data are treated as fuzzy numbers. The assurance region approach is utilized to restrict weight flexibility in the model. Based on Zadeh's extension principle, a pair of two-level mathematical programs is formulated to calculate the upper bound and lower bound of the fuzzy efficiency score. We then transform this pair of two-level mathematical programs into a pair of conventional one-level mathematical programs to calculate the bounds of the fuzzy efficiency scores. The examples of non-life insurance companies in Taiwan and IT impact on firm performance are illustrated to calculate the system and process efficiencies and how to derive their relationship when the data is fuzzy. © 2014 Elsevier B.V. All rights reserved.


A grey-prediction self-organizing fuzzy controller (GPSOFC) has been proposed to control robotic systems. It solves the problems caused by the inappropriate selection of parameters in a self-organizing fuzzy controller (SOFC) and eliminates the dynamic coupling effects between degrees of freedom (DOFs) in robotic systems. However, its stability is difficult to demonstrate. To overcome the stability issue, this study developed an enhanced adaptive grey-prediction self-organizing fuzzy sliding-mode controller (EAGSFSC) for robotic systems. The EAGSFSC not only solves the problem of a GPSOFC implementation by determining the stability of the system but also applies an adaptive law to modify the fuzzy consequent parameter of a fuzzy logic controller for manipulating a robotic system to improve its control performance. The stability of the EAGSFSC was proven using the Lyapunov stability theorem. To confirm the suitability of the proposed method, this study applied the EAGSFSC to manipulate a 6-DOF robot to determine its control performance. Experimental results showed that the EAGSFSC achieved better control performance than the GPSOFC as well as the SOFC for robotic motion control. © 2013 Elsevier Inc. All rights reserved.


Lian R.-J.,VaNung University
IEEE Transactions on Industrial Electronics | Year: 2013

A self-organizing fuzzy controller (SOFC) is proposed to control engineering applications. However, its stability is difficult to demonstrate. Therefore, the parameters of the SOFC must be carefully chosen to ensure that the system controlled by the SOFC can operate stably. To overcome the problem, this study developed an enhanced adaptive self-organizing fuzzy sliding-mode controller (EASFSC) for active suspension systems. Rather than using the output error and the error change of the system, the EASFSC uses a sliding surface and its differential as the input variables of a fuzzy logic controller (FLC) in the SOFC to generate a control input through fuzzy operation. It also applies an adaptive law to modify the fuzzy consequent parameter of the FLC in the SOFC to improve the stability of the system. The stability of the EASFSC was proven using the Lyapunov stability theorem. The EASFSC eliminates the problem of an SOFC implementation in determining the stability of the system and overcomes the difficulty of finding appropriate membership functions and fuzzy rules for designing an FLC. To confirm the applicability of the EASFSC, this study used the EASFSC to manipulate anactive suspension system to improve its control performance. Experimental results showedthat the EASFSC achieved better control performance than the SOFC for active suspensioncontrol. © 2012 IEEE.

Loading VaNung University collaborators
Loading VaNung University collaborators