A gradual neural-network algorithm for jointly time-slot/code assignment problems in packet radio networks

A gradual neural network (GNN) algorithm is presented for the jointly time-slot/code assignment problem (JTCAP) in a packet radio network in this paper. The goal of this newly defined problem is to find a simultaneous assignment of a time-slot and a code to each communication link, whereas time-slots and codes have been independently assigned in existing algorithms. A time/code division multiple access protocol is adopted for conflict-free communications, where packets are transmitted in repetition of fixed-length time-slots with specific codes. GNN seeks the time-slot/code assignment with the minimum number of time-slots subject to two constraints: (1) the number of codes must not exceed its upper limit and (2) any couple of links within conflict distance must not be assigned to the same time-slot/code pair. The restricted problem for only one code is known to be NP-complete. The performance of GNN is verified through solving 3000 instances with 100-500 nodes and 100-1000 links. The comparison with the lower bound and a greedy algorithm shows the superiority of GNN in terms of the solution quality with the comparable computation time.     

Control strategy of leveling load power fluctuations based on fuzzy neural networks by tuning coefficients of learning rate with genetic algorithm

The effective usage of the power facilities can be realized by leveling the fluctuating active power and compensating the reactive power. The fuzzy and fuzzy neural network control strategy of superconducting magnetic energy storages (SMES) was proposed for this purpose. The control results depend on the values of coefficients of learning rate in fuzzy neural networks. Therefore, it is desirable to obtain better control results by tuning the coefficients of learning rate to their optimum values. In this paper, the control strategy based on an autotuning of scaling factors with neural network and tuning of coefficients of learning rate of neural network with genetic algorithm is proposed for leveling load fluctuations. Encoding and decoding of coefficients of learning rate and selection, crossover, and mutation within genetic operations are shown, and crossover rate and mutation rate are discussed. Through these methods, we can achieve a better leveling of load power fluctuation by using fuzzy neural network with genetic algorithm. © 1998 Scripta Technica, Electr Eng Jpn, 125(1): 65–72, 1998     

A power smoothing control method for a photovoltaic generation system using a water electrolyzer and its filtering characteristics

This paper proposes a novel power smoothing control method for a photovoltaic generation (PV) system using a water electrolyzer (ELY). This method realizes that the long‐period components of the PV power fluctuation are supplied to the power grid and the short‐period components of the PV power fluctuation are supplied to the ELY. The power supplied to the ELY is converted to hydrogen for fuel cell vehicles. The results of the power smoothing control method are evaluated in terms of the power in the load frequency control (LFC) band and the ability of generating hydrogen. Furthermore, filtering characteristics of the proposed method are clarified. As the result, the proposed method has reduced the power in LFC band by 83.7% compared to the PV power without the power smoothing control. The filter characteristics of the proposed method have shown that fluctuating components with periods less than 1000 seconds, in which the step voltage regulator does not perform, can be sufficiently eliminated from the PV power fluctuation components. When applying the proposed method for a PV system rated at 800 kW, the ability of generating hydrogen was about 2.30% of the hydrogen required for one hydrogen station.     

A maximum neural network algorithm for route selection problems in multihop radio networks

In a multihop radio network, packets are transmitted from course nodes to destination nodes by activating several links between nodes. Each node can either send a packet to, or receive a packet from, at most one of its adjacent nodes simultaneously. To minimize the transmission time for given requests, the problems must be solved by selecting a transmission route for each request (the routing problem) and by finding a link activation schedule (the link activation problem). The routing problem is decomposed into two subproblems: the candidate extraction problem and the route selection problem. In this paper, we propose a neural network algorithm using the maximum neuron model for the route selection problem. We verify through simulations that our algorithm finds better solutions in a shorter time than the existing algorithms. We also probe the NP‐hardness of this problem. © 1999 Scripta Technica, Electr Eng Jpn, 129(2): 78–86, 1999     

Selective adsorption of biladien-ab-one and zinc biladien-ab-one to mesoporous silica

The adsorption of linear tetrapyrroles and porphyrins to mesoporous silicas with pore diameters of 2.3, 2.7, 4.0, and 7.0 nm from a benzene solution was investigated. As linear tetrapyrroles, 1,15,21,22-tetrahydro-19-benzoyl-5,10,15-triphenyl-15-hydroxybilin-1-one (biladien-ab-one, 1a), its zinc complexes (2 and 3), 1,15,21,22-tetrahydro-19-(4-hexyloxybenzoyl)-5,10,15-tri(4-hexyloxy)phenyl-15-hydroxybilin-1-one (C6-biladienone, 1b), 1,15,21,22-tetrahydro-19-(4-nonyloxybenzoyl)-5,10,15-tri(4-nonyloxy)phenyl-15-hydroxybilin-1-one (C9-biladienone, 1c), 1,15, 21,22-tetrahydro-19-(4-dodecyloxybenzoyl)-5,10,15-tri(4-dodecyloxy)phenyl-15-hydroxybilin-1-one (C12-biladienone, 1d), and 1,15,21,22-tetrahydro-19-(3,5-didodecyloxybenzoyl)-5,10,15-tri(3, 5-di-dodecyloxy)phenyl-15-hydroxybilin-1-one (1e) were employed. As porphyrins, 5,10,15,20-tetraphenylporphyrin (4a) and 5,10,15,20-tetra(4-methoxyphenyl)porphyrin (4b) were used. The isothermal adsorption curves were determined, showing that the amount of adsorption decreases in the order, 1a > 1b > 1c > 1d > 2, while neither 1e, 4a, nor 4b were adsorbed to mesoporous silicas. Mesoporous silica with the diameter of 4.0 nm adsorbed the largest amounts of biladienones. The adsorption was not observed in THF, acetone or ethanol. Neither 1c nor 1d was adsorbed to mesoporous silica with the pore diameter of 2.3 nm. These results indicate that the absorption is driven by polar interactions such as hydrogen bonding, and size exclusion effects are observed even for a molecule with flexible alkyl chains. The amount of adsorbed biladienone increased as the temperature was raised from 25 to 35 °C. The adsorption of biladien-ab-ones at 80 °C caused dehydration of biladien-ab-one to bilatrien-abc-one due to the acidity of the mesoporous silica.     

A node configuration algorithm with bandwidth bottleneck resolution for WDM ring networks

The wavelength-division multiplexing (WDM) technology has been extensively studied and actually deployed as a key technology to solve global information explosions by multiplexing plural signal paths using different wavelengths into one optical fiber. Among WDM-based network architectures, this article focuses on a double-ring one for a next-generation metropolitan area network (MAN), where each node is normally equipped with two pairs of fast-tunable transmitters and wavelength-fixed receivers (TT-FR) as interface to two fibers. To enhance the network performance, additional transmitters/receivers at nodes and receiver wavelengths should be assigned optimally, so that the bandwidth bottleneck can be resolved both on nodes and fiber links. This article formulates this node configuration problem with bandwidth bottleneck resolution as a combinatorial optimization problem, and proves the NP-completeness of its decision version. Then, it presents the two-phase heuristic algorithm composed of a greedy method and a variable depth search (VDS) method. Our algorithm is evaluated through extensive network simulations, where the results support the effectiveness of this joint assignment approach.

The relationship between solid-state fluorescence intensity and molecular packing of coumarin dyes

The solid-state fluorescence intensity of coumarin dyes depends on the substituents present at the 4- and 7-positions. 7-(Diethylamino)coumarins showed higher solid-state fluorescence quantum yield (Φ_f(ss) = 0.29–0.40) than 7-aminocoumarins (0.01). In the case of julolydyl coumarins, a 4-methyltetramethyljulolydyl derivative also displayed high Φ_f(ss) (0.34), this being greater than that observed for both 4-(perfluoroalkyl)tetramethyljulolydyl (0.09 and 0.10) and 4-methyljulolydyl derivatives (0.01). X-ray crystallographic analysis suggested that coumarin dyes bearing network hydrogen bonds and/or π–π stacking show weak solid-state fluorescence whereas coumarin dyes having isolated monomer- and dimer-type stacking show intense fluorescence. 4-(Perfluoroalkyl)tetramethyljulolydyl derivatives displayed medium fluorescence intensity owing to isolated monomer-type packing with little intermolecular interactions operating between adjacent molecules.     

Generalized Analytical Model for a Thyristor-Controlled Variable Inductor

A generalized analytical model is presented for thyristor- controlled variable inductors. The proposed model can represent the performance of various types of thyristor phase-controlled variable inductors (VI's), e. g., conventional, UM-concept, magnetically coupled reactor, and asymmetrically firing control types, by introducing two parameters: m and ?By means of numerical computation using this model, the characteristics of variable inductors can be clarified, and their characteristics can be compared with ease. This analytical model is very useful for deciding circuit configuration and design.     

Neural-network steering control of an automated guided vehicle

With the progress in factory automation in industry has come the demand for higher speed drive and performance for an AGV (automated guided vehicle). A new steering control of the AGV by fuzzy control has been proposed to replace the PI control. However, much time is required to investigate the regulations and to adjust the scaling factors for excellent performance in the fuzzy control. In this paper, a new steering control for an AGV based on the neural network using the backpropagation method is proposed. The good steering control results by the fuzzy control are adopted for the teaching signal of the neural network. First, the effect of the number of learning and the learning errors on the steering control results are discussed by computer simulation using the AGV model. Further, the ability of generalization in the turning radius and the traveling speed also are investigated. It becomes clear that the AGV can travel along a designated route provided the neural network learns both the right and left turning at the maximum traveling speed and the minimum turning radius. Then it is proved by an experiment using the AGV constructed for the test that the proposed steering control method is very affective.