• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2299-2301
• /
• 2000

In this paper, the design of PID controller using Neural networks for the control of non-linear system is presented. First, non-linear system is identified using BPN(Backpropagation Network) algorithm. This identified model is connected to the PID controller and the parameters of PID controller are updated to the direction of reducing the difference between the identified model output and model reference output in arbitrary input signal. Therefore, identified model output tracks the model reference output in an acceptable error range and the parameters of controller are updated adaptively. The output of the system has a good performance in case of both noisy and noiseless model reference and we can control the system stable in off-line when the dynamics of the system is changed.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.33A no.2
• /
• pp.140-147
• /
• 1996

The output of the mach-zehnder optical fiber interferometer varies with birefringences existing in the figer arms. New equations experessing for the visibility of the mach-zehnder interferometer were derived in terms of a number of parameters related to the birefringences, the input SOP and the optical losses. Based on the equations the visibility of the interferometer was simulated in different cases. Some maximum visibility conditons were also presented.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.15 no.1
• /
• pp.70-77
• /
• 2001

This paper present a fuzzy mxlel of describing reacton features for ozon contactor in advanced water treatn-ent systems. Input and output variables are chosen by considenng the object of ozon processing and several parameters related to management of water quahty. Dissolved organic carbon concentration, $UV_{254}$ absorptIon and $KM_NO_4$ consumption are proposed as common variables in input and outp.lt variables. Furthermore own concentration, raw water's temperature and contact time are suggested as input variables, Membership hmctions for input variables have triangular type share and the grades in each lrembership function are determined by investigating process data gathered at pilot planl The decision parts of fuzzy model have linear combination form of input variables and coefficients included in such linear equations are computedd with process clata in the sense of least square error Also fuzzy trodel suggested in this paper is partitioned by 3 independent fuzzy rnxlels using the characteristics of having no interactions armng output variables. As a result, such fuzzy mxlel has rrerits in computation and comprehension. According to simulatIon results, fuzzy moIel's outputs give almost similar data to process output under same input conditions.

• Asian Journal of Innovation and Policy
• /
• v.13 no.1
• /
• pp.1-28
• /
• 2024

Enhancing the efficiency of research and development (R&D) is crucial for organizations to remain competitive and generate innovative solutions. Data Envelopment Analysis (DEA) has emerged as a powerful tool for evaluating R&D efficiency. However, traditional DEA models heavily rely on the selection of input and output variables, which can limit their effectiveness. To overcome this dependency and improve the robustness of DEA, this study proposes a novel methodology that integrates machine learning techniques with DEA for determining the most suitable input and output variables. The proposed approach is particularly relevant for specialized R&D fields, such as Radiation Emergency Medicine (REM). REM is a critical domain that deals with the medical and public health consequences of nuclear emergencies. The selection of REM as the focus of this study is motivated by several factors, including the unique challenges posed by the field, the potential for significant societal impact, and the need for efficient resource allocation in emergency situations. By leveraging machine learning algorithms, such as Support Vector Machines (SVM), the proposed methodology aims to identify the most relevant input and output variables for DEA in the context of REM. The integration of machine learning enables the DEA model to capture complex relationships and non-linearities in the data, leading to more accurate and reliable efficiency assessments. The effectiveness of the proposed methodology is demonstrated through a comprehensive evaluation using real-world REM data. The results highlight the superior performance of the machine learning-integrated DEA approach compared to traditional DEA models. This study contributes to the advancement of R&D efficiency assessment in specialized fields and provides valuable insights for decision-makers in REM and other critical domains.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1999.10a
• /
• pp.247-253
• /
• 1999

In this paper predictive PID control system using neural network (NNPPID) is proposed to control temperature system. NNPPID is composed of neural network predictor forecasts the future output of plant based on the present input and output of plant. Neural self-tuner yields parameters of PID controller. Experiments prove that NNPPID temperature control system has better performance than conventional PID control.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.361-366
• /
• 2013

Prediction of process parameters is very important in parameter design. If predictions are fairly accurate, the quality improvement process will be useful to save time and reduce cost. The concept of dual response approach based on response surface methodology has widely been investigated. Dual response approach may take advantages of optimization modeling for finding optimum setting of input factor by separately modeling mean and variance responses. This study proposes an alternative dual response approach based on machine learning techniques instead of statistical analysis tools. A hybrid neural network-genetic algorithm has been proposed for the purpose of parameter design. A neural network is first constructed to model the relationship between responses and input factors. Mean and variance responses correspond to output nodes while input factors are used for input nodes. Using empirical process data, process parameters can be predicted without performing real experimentations. A genetic algorithm is then applied to find the optimum settings of input factors, where the neural network is used to evaluate the mean and variance response. A drug formulation example from pharmaceutical industry has been studied to demonstrate the procedures and applicability of the proposed approach.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.470-470
• /
• 2000

This paper describes a new method of identifcation of backlash nonlinear systems by use of M-sequence correlation method. In this method, we can obtain not only Volterra kernels of up to 3rd order of the nonlinear system, but also the width of the backlash element from observing the crosscorrelation between the input and the output. Here strictly speaking, a multi-valued nonlinear system such as backlash element can not be expressed by Volterra kernel representation mathematically. But in practice, we encounter many cases where it is difficult to treat them mathematically but they can be controlled from experience. So we here dare to suppose that backlash nonlinear system can be approximated by Volterra kernel representation. Simulations are carried out on a nonlinear system consisting of linear part plus backlash element. And Volterra kernels are measured. The output calculated from the observed Volterra kernels is in good agreement wi th the actual output. And we show that we can obtain the width of backlash element, which is one of the most important parameters, by observing the maximum value of crosscorrelation function between the input M-sequence and the output.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.214-218
• /
• 1992

In this paper, we propose a method for obtaining state-space realization form of two-dimensional transfer function matrices (2DTFM). It contains free parameters. And, we perform various consideration about it. Moreover, we present the conditions so that the state-space realization form exists.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.5
• /
• pp.705-711
• /
• 2013

In this paper, we introduce an design of multi-output fuzzy neural networks based on Interval Type-2 fuzzy set. The proposed Interval Type-2 fuzzy set-based fuzzy neural networks with multi-output (IT2FS-based FNNm) comprise the network structure generated by dividing the input space individually. The premise part of the fuzzy rules of the network reflects the individuality of the division space for the entire input space and the consequent part of the fuzzy rules expresses three types of polynomial functions with interval sets such as constant, linear, and modified quadratic inference for pattern recognition. The learning of fuzzy neural networks is realized by adjusting connections of the neurons in the consequent part of the fuzzy rules, and it follows a back-propagation algorithm. In addition, in order to optimize the network, the parameters of the network such as apexes of membership functions, uncertainty factor, learning rate and momentum coefficient were automatically optimized by using real-coded genetic algorithm. The proposed model is evaluated with the use of numerical experimentation.

• Design & Manufacturing
• /
• v.16 no.3
• /
• pp.1-8
• /
• 2022

In this study, an artificial neural network(ANN) was constructed to establish the relationship between process condition prameters and the qualities of the injection-molded product in the injection molding process. Six process parmeters were set as input parameter for ANN: melt temperature, mold temperature, injection speed, packing pressure, packing time, and cooling time. As output parameters, the mass, nominal diameter, and height of the injection-molded product were set. Two learning structures were applied to the ANN. The single-task learning, in which all output parameters are learned in correlation with each other, and the multi-task learning structure in which each output parameters is individually learned according to the characteristics, were constructed. As a result of constructing an artificial neural network with two learning structures and evaluating the prediction performance, it was confirmed that the predicted value of the ANN to which the multi-task learning structure was applied had a low RMSE compared with the single-task learning structure. In addition, when comparing the quality specifications of injection molded products with the prediction values of the ANN, it was confirmed that the ANN of the multi-task learning structure satisfies the quality specifications for all of the mass, diameter, and height.