• Journal of Electrical Engineering and Technology
• /
• 제13권4호
• /
• pp.1724-1731
• /
• 2018

The conventional polynomial neural network (PNN) is a classical flexible neural structure and self-organizing network, however it is not free from the limitation of overfitting problem. In this study, we propose a space search-optimized polynomial neural network (ssPNN) structure to alleviate this problem. Ranking selection is realized by means of ranking selection-based performance index (RS_PI) which is combined with conventional performance index (PI) and coefficients based performance index (CPI) (viz. the sum of squared coefficient). Unlike the conventional PNN, L2-norm regularization method for estimating the polynomial coefficients is also used when designing the ssPNN. Furthermore, space search optimization (SSO) is exploited here to optimize the parameters of ssPNN (viz. the number of input variables, which variables will be selected as input variables, and the type of polynomial). Experimental results show that the proposed ranking selection-based polynomial neural network gives rise to better performance in comparison with the neuron fuzzy models reported in the literatures.

• 지능정보연구
• /
• 제15권4호
• /
• pp.53-64
• /
• 2009

본 논문에서는 실루엣 동영상으로부터 보행 분석에 사용되는 형태 기술자의 성능을 평가하였다. 형태 기술자는 모멘트기술자(MD), 푸리에기술자(FD), 저나이크기술자(ZD)를 사용하였다. 먼저 성능 평가 지수인 비대칭 지수 AI와 주기지수 PI를 정의하였다. 이 지수는 보행은 주기적으로 주어지며, 이를 표현한 형태 기술자도 같은 보행 주기를 가져야 하는 것을 기반으로 한다. 또한 지수는 형태의 특성을 정확하게 나타내는 인식 성능보다는 형태 기술자들에 의해 얼마나 잘 표현되었는가를 평가하는 것이다, 성능평가결과, ZD가AI = 1.09, PI = 2.21로 대체로 무난한 성능으로 나타났으며, FD가 MD보다 AI = 0.93로 하나의 주기내에서의 성능이 우수하였다. 반면 MD는 FD보다 PI = 2.37로 반복적인 특징이 높게 나타났다. 또한 형태 기술자는 5~10개 정도의 파라미터 개수를 사용하는 것이 좋은 성능을 가짐을 보였다.

• 대한전자공학회논문지
• /
• 제17권6호
• /
• pp.38-43
• /
• 1980

시간에 따라 변하지 않는 선형 다변수 시스템의 출력 궤환 제어를 이산시간계에서 최적화 하기 위한 필요 조건을 유도했다. Quadratic performance index를 사용했다. 결과는 출력의 비례적분 궤환의 최적이득을 구하는 데에도 쉽게 적용할 수 있다. 예제를 들어 최적 궤환이 득을 구했다.

• 한국운동역학회지
• /
• 제25권2호
• /
• pp.157-165
• /
• 2015

Object : The goals of this research were to make Performance Enhanced Model(PE) taken the largest performance index (PI) through artificial variation of principle components calculated by principle component analysis for trial data, and to verify the effect through comparing kinematic factors between trial data (Raw) and PE. Method : Ten subjects (5 men, 5 women) were recruited and 80% of their maximal record was considered. The PI is a regression equation. In order to develop PE, we extracted Principle components from trial position data (by Principle Components Analysis (PCA)). Before PCA, we made 17 position data to 3 row matrix according to components. We calculated 3 eigen value (principle components) through PCA. And except Y (medial-lateral direction) component (because motion of Y component is small), principle components of X (anterior-posterior direction) and Z (vertical direction) components were changed as following. Changed principle components = principle components + principle components ${\times}$ k. After changing the each principle component, we reconstructed position data using the changed principle components and calculated performance index (PI). A Paired t-test was used to compare Raw data and Performance Enhanced Model data. The level of statistical significance was set at $p{\leq}0.05$. Result : The PI was significantly increased about 12.9kg at PE ($101.92{\pm}6.25$) when compared to the Raw data ($91.29{\pm}7.10$). It means that performance can be increased by optimizing 3D positions. The difference of kinematic factors as follows : the movement distance of the bar from start to lock out was significantly larger (about 1cm) for PE, the width of anterior-posterior bar position in full phase was significantly wider (about 1.3cm) for PE and the horizontal displacement toward the weightlifter after beginning of descent from maximal height was significantly greater (about 0.4cm) for PE. Additionally, the minimum knee angle in the 2-pull phase was significantly smaller (approximately 2.7cm) for the PE compared to that of the Raw. PE was decided at proximal position from the Raw (origin point (0,0)) of PC variation). Conclusion : PI was decided at proximal position from the Raw (origin point (0,0)) of PC variation). This means that Performance Enhanced Model was decided by similar motion to the Raw without a great change. Therefore, weightlifters could be accept Performance Enhanced Model easily, comfortably and without large stress. The Performance Enhance Model can provide training direction for athletes to improve their weightlifting records.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1991년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 22-24 Oct. 1991
• /
• pp.238-242
• /
• 1991

This paper presents an approach to self-tuning PI control of dynamic plants, based on fuzzy logic application. A fuzzy logic composed of linguistic conditional statements is employed by defining the relations of input-output variables of the controller. In the synthesis of a fuzzy logic controller, one of the most difficult problem is the selection of linguistic control rules and parameters. To overcome this difficulty, self-tuning fuzzy PI controller (STFPIC) with a hierarchical structure in which the fuzzy PI controller is assigned as the lower level and the rule modification and parameter adjustment as the higher level. The rules and parameters are generated by the adjustment of membership function through performance index(PE). In this paper, the algorithm for of the controller performance is estimated by means of computer simulation.

• 한국전기전자재료학회논문지
• /
• 제21권12호
• /
• pp.1124-1129
• /
• 2008

The object of this paper is to acquire the data related to insulation evaluation of hybrid vehicle traction motors. We made a comparative analysis on Insulation Resistance (IR), Dielectric Absorption Ratio (DAR), and Polarization Index (PI) of the motor stators. The experiment was carried out according to IEEE Std. 43 and IEC 60085-1 for insulation resistance test standard of rotating machinery. Test voltage of 500 V was applied between a phase and the enclosure. The IR and the DAR of used motors were lower than those of new ones. The DAR and the PI were $0.92{\sim}1.02$ and $0.74{\sim}1.1$, respectively and the result did not meet the recommendation basis 2 for insulation level H. From the experimental results, we could prepare parameters and basis for insulation evaluation of the traction motor stator by the comparative analysis of short-time insulation resistance changes, DAR and PI.

• 한국지능시스템학회논문지
• /
• 제4권2호
• /
• pp.35-40
• /
• 1994

Fuzzy logic controllers are often found to behave better than PI controllers. One of the major reasons for this is that the fuzzy logic inferences used can produce nonlinear type controllers. For some applicatioins, howeveer, linear fuzzy logic controllers also perofrm better than PI controllers. In this paper, we examine linear fuzzy logic controllers to show that the truncation effects of the fuzzy logic controllers make them perform much better than the PI controllers. In terms of a performance index we used, the truncation effects reduced the index value by up to 80% for examples we studied.

• Journal of Korean Neurosurgical Society
• /
• 제39권1호
• /
• pp.20-25
• /
• 2006

Objective : The authors analyze prospectively the result of transcranial doppler[TCD] in normal pressure hydrocephalus and compared its cerebral blood flow parameters to radionuclide cerebrospinal fluid[CSF] flow study, postoperative brain computed tomography[CT] findings and clinical outcome, and studied the relationship between cerebral hemodynamics and clinical performance. Methods : Twenty five patients with hydrocephalus undertook pre- and post-operative TCD but only preoperative CSF flow study was performed. Mean flow velocity[Vm], pulsatility index[PI] and resistance index[RI] were assessed through TCD and changes in ventricle size and cortical gyral atrophy were checked through brain CT. Results : Postoperative hydrocephalus showed an increase in Vm[ACA P=0.037, MCA P=0.034], decrease in PI[ACA P=0.019, MCA P=0.017] and decrease in RI [ACA P=0.017, MCA P=0.021] compared to preoperative TCD parameters in the postoperative improvement group. In the postoperative improvement group, postoperative TCD parameters correlated with CSF flow study grade [Vm : $R^2=-0.75$, PI : $R^2=0.86$, RI : $R^2=0.78$] and ventriculocranial ratio change correlated with PI change [$R^2=0.73$]. The convexity gyral atrophy and initial TCD parameters showed close relationship to outcome. Conclusion : PI and RI can be used as an indicator of post operative prognosis, and with the addition of CSF flow study values, can also be used as a tool to predict pre-op and post-op patient status and successful shunt surgery.

• Annals of Clinical Neurophysiology
• /
• 제20권2호
• /
• pp.79-84
• /
• 2018

Background: Pulsatility of cerebral arteries and aortic stiffness have been associated with white matter hyperintensities (WMH). We explored which is better correlated with the severity of WMH in a population with acute lacunar infarct. Methods: We included patients with acute small subcortical infarcts who underwent transcranial Doppler (TCD) and brachial ankle pulse wave velocity (baPWV). Exclusion criteria were any stenosis or occlusion on major cerebral arteries on magnetic resonance angiography; poor temporal insonation windows; ankle brachial index < 0.9; and atrial fibrillation. We assessed the performance of the pulsatility index of bilateral middle cerebral arteries (PI-MCA) and baPWV for predicting moderate-to-severe WMH, defined as an Age Related White Matter Changes score > 5, and then sought to find independent predictors using binary logistic regression analysis. Results: Eighty-three patients (56 males, mean age $61.5{\pm}11.4$) participated in the study. Uni-variate analysis showed old age and high PI-MCA were significantly correlated with moderate-to-severe WMH. However, baPWV was not associated with the severity of WMH. Multivariate analysis revealed old age (odds ratio per 1-year increase, 1.068; p = 0.044) and upper tertile of PI-MCA (odds ratio, 5.138; p = 0.049) were independently associated with moderate-to-severe WMH. Receiver-operating characteristics showed PI-MCA differentiated those with and without moderate-to-severe WMH with an area under the curve of 0.719. Conclusions: PI-MCA derived from TCD was better correlated with the severity of WMH than baPWV in a population with lacunar infarction. Pulsatility of cerebral arteries may better predict cerebral small vessel disease than the aortic stiffness index.

• 품질경영학회지
• /
• 제50권3호
• /
• pp.473-490
• /
• 2022

Purpose: The purpose of this study is to propose a new useful suggestion to monitor the stability of process by developing a stability ratio or index related to investigating how well the process is controlled or operated to the specified target. Methods: The proposed method to monitor the stability of process is building up a new measure index which is making up for the weakness of the existing index in terms of short or long term period of production. This new index is a combined one considering both stability and capability of process to the specification limits. We suppose that both process mean and process variation(or deviation) are changing on time period. Results: The results of this study are as follows: regarding the stability of process as well as capability of process, it was shown that two indices, called SI(stability index) and PI(performance index), can be expressed in two-dimensional X-Y graph simultaneously. This graph is categorized as 4 separated partitions, which are characterized by its numerical value intervals of SI and PI which are evaluated by test statistics. Conclusion: The new revised index is more robust than the existing one in investigating the stability of process in terms of short and long period of production, even in case both process mean and variation are changing.