A neural network approach to ship track-keeping control

This paper presents an on-line trained neural net work controller for ship track-keeping problems. Following a brief review of the ship track-keeping control development since the 1980's, an analysis of various existing backpropagation-based neural controllers is provided. We then propose a single-input multioutput (SIMO) neural control strategy for situations where the exact mathematical dynamics of the ship are not available. The aim of this study is to build an autonomous neural controller which uses rudder to regulate both the tracking error and heading error. During the whole control process, the proposed SIMO neural controller adapts itself on-line from a direct evaluation of the control accuracy, and hence the need for a “teacher” or an off-line training process can be removed. With a relatively modest amount of quantitative knowledge of the ship behavior, the design philosophy enables real time control of a nonlinear ship model under random wind disturbances and measurement noise. Three different track-keeping tasks have been simulated to demonstrate the effectiveness of the training method and the robust performance of the proposed neural control strategy     

Solar-planetary cycles in Jupiter's great red spot darkness

The change in the darkness of the Great Red Spot (GRS) of Jupiter (1894–1974) has been analysed with Fourier (FFT), Maximum Entropy and Power spectrum (Blackman-Tukey window) (PSA) methods of spectrum analysis. Significance, non-randomness and stationarity tests assigned high variance to periodicities of 33 ± 4, 13–15, about 11, 9 and 3 yrs. The highest correlation between solar activity and GRS darkness was found for the 14th and 16th solar cycle. The periodicities obtained are interpreted as the combined eftects of solar activity, planetary resonances and internal jovian mechanisms.     

Brief description of bare essentials of surface transfer and results from simulations with the HAPEX-MOBILHY data

The Bare Essentials of Surface Transfer (BEST) land surface scheme is briefly described and the key physical parameterisations discussed. Results are then presented to illustrate how the model performs, with forcing data for HAPEX-MOBILHY, compared to a series of other schemes in the simulation of evaporation and sensible heat. The implications of the models performance, and some indications of the future development of the scheme are provided. The basic version of BEST was found to overestimate evaporation for the HAPEX-MOBILHY data, simulating 816 mm yr⁻¹ compared to a range of 550 to 816 mm yr⁻¹ for all models. A modification to the β parameterisation reduced the evaporation to 759 mm yr⁻¹ which, although an improvement, is still probably too high.     

Identification of physical processes inherent in artificial neural network rainfall runoff models

The emergence of artificial neural network (ANN) technology has provided many promising results in the field of hydrology and water resources simulation. However, one of the major criticisms of ANN hydrologic models is that they do not consider/explain the underlying physical processes in a watershed, resulting in them being labelled as black‐box models. This paper discusses a research study conducted in order to examine whether or not the physical processes in a watershed are inherent in a trained ANN rainfall‐runoff model. The investigation is based on analysing definite statistical measures of strength of relationship between the disintegrated hidden neuron responses of an ANN model and its input variables, as well as various deterministic components of a conceptual rainfall‐runoff model. The approach is illustrated by presenting a case study for the Kentucky River watershed. The results suggest that the distributed structure of the ANN is able to capture certain physical behaviour of the rainfall‐runoff process. The results demonstrate that the hidden neurons in the ANN rainfall‐runoff model approximate various components of the hydrologic system, such as infiltration, base flow, and delayed and quick surface flow, etc., and represent the rising limb and different portions of the falling limb of a flow hydrograph. Copyright © 2004 John Wiley & Sons, Ltd.     

The Dergaon (H5) chondrite: Fall,classification, petrological and chemical characteristics,cosmogenic effects,and noble gas records

Abstract— A multiple fall of a stony meteorite occurred near the town of Dergaon in Assam, India, on March 2, 2001. Several fragments weighing <2 kg and a single large fragment weighing ～10 kg were recovered from the strewn field, which extended over several tens of square kilometers. Chemical, petrographic, and oxygen isotopic studies indicate it to be, in most aspects, a typical H5 chondrite, except the unusually low K content of ～340 ppm. A cosmic ray exposure of 9.7 Ma is inferred from the cosmogenic noble gas records. Activities of eleven cosmogenic radionuclides were measured. ²⁶Al and ²²Na activities as well as the ²²Na/²⁶Al activity ratio are close to the values expected on the basis of solar modulation of galactic cosmic rays. The low ⁶⁰Co activity (<1 dpm/kg) is indicative of a small preatmospheric size of the meteorite. Cosmic ray heavy nuclei track densities in olivine grains range from ～106 cm^?2 in samples from the largest fragment to approximately (4–9) × 105 cm^?2 in one of the smaller fragments. The combined track, radionuclide, and noble gas data suggest a preatmospheric radius of ～20 cm for the Dergaon meteorite.     

Simulating colour variations in pulsating sdB stars

After briefly recalling basic facts about the metallicity gradients inspiral galaxies, we summarize two recent N-body gas-dynamical modelsthat implement a non-instantaneous gas recycling and point our their potential for an improved treatment of the chemical evolution ingalactic disks.