Evaluation of robust control by universal learning networks

The characteristics of control system design using a universal learning network (ULN) are such that both the controlled systems and their controller are represented in a unified framework, and that the learning stage of the ULN can be executed by using not only first-order derivatives (gradient) but also the higher order derivatives of the criterion function with respect to parameters. ULNs have the same generalization ability as neural networks. So the ULN controller is able to control the system in a favorable way under an environment which is little different from the environment of the control system at the learning stage. However, stability cannot be sufficiently realized. In this paper, we propose a robust control method using a ULN and second-order derivatives of that ULN. Robust control, as considered here, is defined as follows. Even though the initial values of the node outputs are very different from those at the learning stage, the control system is able to reduce its influence to other node outputs and can control the system as in the case of no variation. In order to realize such robust control, a new term concerning the variation is added to the usual criterion function, and the parameters are adjusted so as to minimize the above-mentioned criterion function using second-order derivatives of the criterion function with respect to the parameters. Finally, it is shown that the ULN controller constructed by the proposed method works effectively in a simulation study of a non-linear crane system. This work was presented, in part, at the International Symposium on Artificial Life and Robotics, Oita, Japan, February 18–20, 1996     

Effect of metal cation type on the structure and properties of ethylene ionomers

Thermal properties by DSC, stiffness, melt viscosity, tensile properties, and dynamic mechanical properties were measured for the Na⁺, K⁺, Mg²⁺, Zn²⁺, Cu²⁺, Mn²⁺, and Co²⁺ salts of poly(ethylene-co-methacrylic acid) (EMAA). The changes in the structure and properties with increasing neutralization are larger in the alkaline and alkaline earth metal salts than in the transition metal salts. The stiffness shows a maximum at 33% neutralization in both the alkaline and alkaline earth metal salts, while no maxima are found up to 60% neutralization in the transition metal salts. The microphase separation of salt group aggregates is observed in both the alkaline and alkaline earth metal salts, but is not seen in the transition metal salts. These differences were attributed to both the stronger ionic interactions and the larger number of carboxyl groups associated with the alkaline and alkaline earth metal salts in the ordered structure of ionic salt groups (ionic crystallites). The mechanical properties measured at low strain, such as stiffness and yield stress, strongly depend on the degree of the crystalline order of the ionic crystallites. The high-strain properties, such as tensile strength and elongation at break, depend on the strength of the ionic interactions and the valence of the cation.     

Relationship between pulse sequences and signal intensity of joint fluid in the gradient-echo MR imaging

Platelet-activating factor (PAF) is a proinflammatory phospholipid mediator implicated in necrotizing enterocolitis. Regulation of PAF-acetylhydrolase (AH), the enzyme degrading PAF, is poorly understood. In this study we found that administration of a dose of PAF (1.5 microg/kg, i.v.), which does not cause gross intestinal injury, increased plasma and intestinal PAF-AH in the rat. Cycloheximide (CHX, 5 mg/kg, i.v.) reduced the activity of plasma (but not intestinal tissue) AH in control, as well as in PAF-injected rats, and aggravated systemic inflammation and tissue injury in the latter. The intestinal necrosis induced by PAF and CHX was ameliorated by posttreatment with WEB2170 (a PAF antagonist), indicating a role of endogenous PAF in mediating injury. Both WEB2170 and anti-TNF antibody reduced PAF-induced AH activity in intestinal tissue, but not in the plasma. Allopurinol largely prevented the injury induced by PAF and CHX, but had no effect on the up-regulation of AH. We conclude: 1) de novo protein synthesis is required to maintain physiologic AH level in the plasma; 2) PAF up-regulates plasma and intestinal AH activity; 3) CHX enhances the injurious effect of PAF; 4) endogenous PAF and TNF also play a role in the up-regulation of intestinal AH; the former probably mediating the intestinal injury by PAF; and 5) reactive oxygen species may mediate the injurious effect of PAF plus CHX, but do not contribute to the regulation of AH by PAF.     

Learning of symbiotic phenomena between agents by using neural networks

Recently, numerous attempts have been made by researchers to understand the essence of complex phenomena (complex systems). In this paper, we consider biological systems in nature as being among the most complex systems. The purpose of this paper is to propose a method of realizing symbiotic phenomena such as mutual benefit, competition, and exploitation more generally than the Lotka–Volterra equation by using neural networks. © 2002 Wiley Periodicals, Inc. Electr Eng Jpn, 140(1): 77–88, 2002; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.1172 Copyright © 2002 Scripta Technica     

在线仿真系统在催化裂化装置上的应用

介绍新开发的催化裂化装置在线仿真系统的硬件结构、仿真模型技术及功能。该系统由于采用了催化裂化反应—再生系统半(?)理模型以及硬件与装置DOS上位联网,直接读取装置实时数据作为仿真运算初值的配置方案、具有实时准确的优点,既可用于操作培训,又可用于生产预测。     

Subtype-specific differences in subcellular localization and chlorethylclonidine inactivation of alpha1-adrenoceptors

Chlorethylclonidine (CEC) inactivation has been used as one criterion to subclassify the alpha1-adrenoceptors (AR); however, the extent of CEC inactivation can vary depending on the CEC treatment. By constructing the FLAG-tagged (N-terminus) and green fluorescent protein (GFP)-fused (C-terminus) alpha1-ARs, we have determined the relationship between CEC sensitivity and the cellular localization of alpha1-AR subtypes using COS-7 cells. In GFP-expressing cells, flow cytometry analysis with anti-FLAG N-terminus antibody detected strong fluorescent signals in most of alpha1B-AR-expressing cells, but low signals in alpha1A-AR-expressing cells. Further examination with confocal microscopy showed that fluorescent signals densely localized intra-cellularly in alpha1A-AR-expressing cells, while most of alpha1B-AR localized on the cell surface. Furthermore, radioligand binding studies with [125I]HEAT showed that CEC (10 microM) treatment of intact cells inactivated approximately 30-40% of alpha1A-AR and >90% of alpha1B-AR, while the CEC treatment of membrane preparations resulted in >80% decrease in the alpha1A-AR density and >90% of alpha1B-AR density, respectively. The results showed that the hydrophilic alkylating agent CEC inactivated only alpha1-AR on the cell surface irrespective of its subtype, and that the subtype-specific sorting is a major determinant for CEC inactivation of alpha1-AR. Subtype-specific cellular localization suggests a new class of functional properties that may explain the signal and functional diversity of homologous alpha1-AR (as well as other G protein-coupled receptors) subtypes.     

Method for minimum‐loss reconfiguration of distribution system by tabu search

This paper proposes a loss minimum reconfiguration method by tabu search for an open‐loop radial distribution system with distributed generators. The problem is to find the optimal normal open sectionalizing switch positions which minimize the total distribution line losses subjected to the line/transformer capacity constraints and voltage constraint. Generally, the problem is mathematically formulated as a complex combinatorial optimization problem or mixed integer programming problem, and is solved by using mathematical programming method, heuristic algorithm, intelligent method, and so on. However, a satisfactory algorithm for power companies has not yet been attained in both computational burden and solution accuracy. Thus, in this paper, the authors propose a method to solve the above problem by using tabu search (TS). Reverse power flow caused by distributed generators can be included in the solution algorithm. TS is one of the meta‐heuristic algorithms, and sometimes has been evaluated to be better than genetic algorithm (GA) or simulated annealing (SA) from viewpoints of both computational speed and solution accuracy. In order to evaluate the validity and efficiency of the algorithm, several numerical examples are shown in this paper. © 2005 Wiley Periodicals, Inc. Electr Eng Jpn, 152(2): 18–25, 2005; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20086     

Hydrothermal soft chemical synthesis of BaTiO<Subscript>3</Subscript> and titanium oxide with cocoon-like particle morphology

The hydrothermal reactions of fibrous H₂Ti₄O₉ particles with Ba(OH)₂ solution in the presence of cationic surfactants of n-hexadecyltrimethylammonium hydroxide (HTMA-OH) and n-hexadecyltrimethylammonium bromide (HTMA-Br) were investigated in a temperature range of 150–250 °C. H₂Ti₄O₉ phase with layered structure was transformed to BaTiO₃ phase in the Ba(OH)₂–(HTMA-OH) and the Ba(OH)₂–(HTMA-Br) solutions, and partially transformed to anatase phase in the Ba²⁺-free HTMA-OH and HTMA-Br solutions by topotactic structural transformation reactions under the hydrothermal conditions. The cocoon-like BaTiO₃ and titanium oxide particles were obtained after the hydrothermal reactions in the Ba(OH)₂–(HTMA-OH) and HTMA-OH solution, respectively. These cocoon-like particles were formed by assembling the fibrous particles in the surfactant solutions.     

High-precision voltage-measuring method using fast sampled data for electric power systems

Control and protection equipment in power systems requires higher sensitivity and operational reliability to meet today's changing power system requirements. The voltage-measuring deviation requirement for advanced voltage and var control equipment is less than 0.1 percent under conditions of harmonic distortion in the voltage waveform and power-system frequency variation. Studies on digital signal processing suitable for electric power systems showed that these requirements are satisfied using fast sampling and very fast 32-bit floating point operations by a Digital Signal Processor (DSP). This paper describes the design philosophy of a high-precision power system voltage-measuring method using fast sampled data. In addition, total voltage-measuring deviation characteristics under a combination of the techniques are described along with digital filter characteristics, frequency-measuring deviation characteristics, frequency variation versus gain-compensation characteristics of a digital filter, and peak value operating principles.