MODEL IDENTIFICATION OF A SERVO-TRACKING SYSTEM USING FUZZY CLUSTERING

Author(s):  
ERIC M. NGUYEN ◽  
NADIPURAM R. PRASAD

This paper investigates the use of Fuzzy Clustering as a means for model identification of a complex and highly non-linear servo-tracking system when only observational data is available. The use of Fuzzy Clustering facilities automatic generation of rules and its antecedent parameters. The consequent of the model is then formulated in the form of Takagi, Sugeno and Kang (TSK), and its parameters determined by the Least Squares Method (LSM).

Author(s):  
Vladimir Grinkevich ◽  

The evaluation of the mathematical model parameters of a non-linear object with a transport delay is considered in this paper. A temperature controlled stage based on a Peltier element is an identification object in the paper. Several input signal implementations are applied to the input of the identification object. The least squares method is applied for the calculation of the non-linear differential equitation parameters which describe the identification object. The least squares method is used due to its simplicity and the possibility of identification non-linear objects. The parameters values obtained in the process of identification are provided. The plots of temperature changes in the temperature control system with a controller designed based on the mathematical model of the control object obtained as a result of identification are shown. It is found that the mathematical model obtained in the process of identification may be applied to design controllers for non-linear systems, in particular for a temperature stage based on a Peltier element, and for self-tuning controllers. However, the least square method proposed in the paper cannot estimate the transport delay time. Therefore it is required to evaluate the time delay by temperature transient processes. Dynamic object identification is applied when it is required to obtain a mathematical model structure and evaluate the parameters by an input and output control object signal. Also, identification is applied for auto tuning of controllers. A mathematical model of a control object is required to design the controller which is used to provide the required accuracy and stability of control systems. Peltier elements are applied to design low-power and small- size temperature stage . Hot benches based on a Peltier element can provide the desired temperature above and below ambient temperature.


2021 ◽  
Vol 906 (1) ◽  
pp. 012056
Author(s):  
Maria Mrówczyńska ◽  
Jacek Sztubecki ◽  
Zofia Ziçba ◽  
Izabela Wilczyńska

Abstract The geodetic monitoring of engineering structures, their displacements, and deformations, carried out permanently or periodically, allows obtaining information on the technical condition of facilities. The achieved information enables determining the necessary changes in using objects and minimizing future errors in the similar object’s design. The measurement results are subject to geometric interpretation based on the determined displacement parameters of the object’s shape and the approximation of the vector displacement field. Due to the influence of random factors characterized by a change in time and varying intensity, the deformation measurements performed during the operation of the facilities are of great importance for the safety of structures and engineering structures. In actual tasks of determining the object’s deformation and building a geometric model of displacements, the dominant method is the differential method, the advantage of eliminating systematic errors in measurement results while maintaining the geometric structure of the measurement and control network. The displacement’s geometric model, built based on measurements and calculations, can build a dynamic model of a building object, additionally considering such causes of deformation as, for example, own and usable weight, wind pressure, changes in ambient temperature, or ground vibrations. The article proposes approaches using the free alignment of linear and angular observations made in a geodetic network to determine horizontal displacements of an engineering object. This method may be necessary to study displacements of various parts of the object, thus analyzing its deformation. Free alignment allows for an optimal fit of the equalized network into the approximate network by imposing additional conditions (compared to the classic least squares method) on the vector of estimates of increments to approximate coordinates and the value of the covariance matrix. As an example of applying the proposed approach, the actual data received from the geodetic monitoring of the building structure was used. The structure was a road viaduct located along Wojska Polskiego Street in Bydgoszcz. The object of measurements and analyses was represented by finite sets of fixed points, subject to periodic observations over two years. The authors tested the effectiveness of the proposed algorithm and compared the obtained results with the values of horizontal displacements, which were calculated based on the classic study of geodetic monitoring results using the least-squares method. The accuracy analysis of the obtained values of the geodetic network horizontal displacements using free alignment and the least-squares method was also performed. The results indicate the possibility of using the presented approach to identify the geometric model of horizontal displacements without losing the accuracy of their determination.


2019 ◽  
Vol 5 (1) ◽  
pp. 361-363
Author(s):  
Fars Samann ◽  
Andreas Rausch ◽  
Thomas Schanze

AbstractIn biomedical engineering, dipole source localization is commonly used to identify brain activities from scalp recorded potentials, which is known as inverse problem of electroencephalography (EEG) source localization. However, this problem is fundamental in biomedical engineering, medicine and neuroscience. The EEG inverse problem is non-linear, in addition, it is ill-posed and the solver can be unstable, i.e. the solution is non-unique and it is highly sensitive to small changes of the measured signal (noise). For solving the EEG inverse problem iterative methods, like Levenberg-Marquardt algorithm, are usually considered. However, these techniques require good initial values and many electrodes N, since a large redundancy supports the finding of the right solution. Therefore, in this paper, a hybrid method of linear and non-linear modelling and least squares approach are proposed to overcome of these problems: the solutions calculated by means of a linear approximation of EEG inverse problems serve as initial values for solving the original non-linear model. In addition, independent component analysis (ICA) is combined with the proposed hybrid least squares method to separate different dipole sources from multiple EEG signals. The performance of the hybrid least squares method with and without ICA is measured in term of root mean square error. The simulation results show that the proposed method can estimate the location of dipole source with acceptable accuracy under high noise condition and small N comparing with linear least squares method considering larger N. Finally, it should be mentioned that the proposed method promises advantages in finding solutions of the EEG inverse problem effectively.


2020 ◽  
Author(s):  
João R. C. de Araújo ◽  
Saulo O. D. Luiz ◽  
Antonio M. N. Lima

Coiled polymer actuators have the characteristic to generate power by contracting when heated under strain. This work presents a project of a rotational joint driven by such actuators. Furthermore, the thermmomechanical model of the CPAs is identified by means of the least squares method. In addition, a model of the joint considering a viscous friction is suggested. Finally, the joint model is identified by means of a non linear optimization algorithm.


Sign in / Sign up

Export Citation Format

Share Document