RHINOS: a consultation system for diagnosis of headache and facial pain

This paper is intended to describe the expert system RHINOS for diagnosing patients whose chief complaints are headache and facial pain. RHINOS assists physicians in diagnosing the above-mentioned disease. For this purpose the authors surveyed diagnosing processes used by physicians to determine what advice physicians want. As a result, the author reached the conclusion that knowledge for diagnosing the above-mentioned disease should be categorized according to four types of rules: exclusive rules (if the patient has disease D, he must have symptoms S1, S2,....,Sn), inclusive rules (if the patient has symptoms S1, S2,...,Sn, he has disease D with a probability X (0 less than X less than 1)), associate rules (if the patient has symptoms S1, S2,...,Sn the probability that he has the disease D increases) and disease image (if the patient has disease D, he may have symptoms S1, S2,...,Sn). Through harmonious use of these four types of rules, RHINOS gives the advise that physicians want. RHINOS is widely available because it is implemented by Prolog-KABA which is operative on personal computers such as NEC PC9801 and IBM-PC.     

TARGET YIELDS AND HYDROGEN CONSUMPTION IN BROWN COAL LIQUEFACTION

A simple mechanistic model for brown coal liquefaction has been developed and used in conjunction with data from PDU operation to examine the factors which contribute to hydrogen consumption. It is found that after allowing for hydrogen generation requirements, the maximum achievable refined product yield from Morwell brown coal ((ssuming a naphtha/middle distillate ratio of 5:6 together with refined product H/C ratios of 1.8 for naphtha and 1.5 for middle distillate) is 46 - 48%, corresponding to a hydrogen consumption of 6.6 - 7.6%. Alternatively, if consideration of hydrogen generation is excluded (as in pilot plant operation), the maximum yield would be around 54%, corresponding to 6.6% net hydrogen consumption. These results suggest that an aim of maximizing product yield by decreasing Cl-C4 gas formation may not be as desirable as at first thought.     

Hydrocracking Marlim vacuum residue with natural limonite. Part 2: experimental cracking in a slurry-type continuous reactor

Disposable Australian iron-slurry (AL) and NiO-MoO₃-Al₂O₃ (NiMo) catalysts were used in hydrocracking experiments to convert Marlim vacuum residue (ML-VR) in a slurry-bed continuous flow reactor at temperatures of 440-460 °C, under a hydrogen pressure of 14.7 MPa and an LHSV of 0.5. The degree of conversion ranged from 54 to 83%, depending on the reaction temperature and catalyst used, with AL giving more complete conversion than NiMo. AL also proved more active in the removal of nickel. Hydrogen consumption was linearly correlated with conversion regardless of the catalyst used.     

Frenkel pair accumulation induced crystallization of amorphous MgAl2O4

The epitaxial and homogeneous irradiation induced re-crystallization of amorphous MgAl₂O₄ was studied by means of continuous Frenkel pair accumulation in the molecular dynamics framework. Present results point out that the re-crystallization induced by Frenkel pair accumulation appears in both cases to be thermally enhanced but non diffusive. It is governed by a local rearrangement of each point defect in the homogeneous case, while spontaneous Frenkel pair recombination process in the crystalline part or at the interface drives the re-crystallization in the epitaxial case.     

Effect of catalyst combination on growth of single-walled carbon nanotubes

A new effective catalysts combination of iron — nickel for alcohol CVD technique was found. This catalyst catalyzed well as well as the typical catalyst of iron — cobalt catalysts, but gave a different diameter distribution. Calculating their electrical density of states under the assumption of their solid lattice structures, the result was fairly consistent with experimental results. The number of electrical states near Fermi level that is considered to be important for catalytic reaction is enough and the DOS of iron – nickel catalyst was quite similar to that of cobalt unlike manganese – copper catalyst. Consequently, a blend of catalysts that has a similar DOS to cobalt and has enough states near the Fermi level can be a good catalyst for alcohol CVD.