Computer-aided protective system hazard analysis

Based on a fault tree for a plant and on two fault trees for its protective system, the computer code, PROTECT, produces time profiles of expected numbers of normal trips, spurious trips, and destructive hazards. Input parameters to PROTECT are (1) minimal cut sets of the fault trees; (2) mean time to failures of basic components; (3) failure characteristics of operators; (4) scheduled maintenance intervals for plant and protective system; and (5) repair times from normal and spurious trips. Using the code, we can design the protective systems and maintenance policies, keeping both destructive hazards and spurious trips below acceptable levels.     

Top-down Algorithm for Obtaining Prime Implicant Sets of Non-Coherent Fault Trees

An algorithm for finding all the prime implicant sets is given for non-coherent fault trees involving gates other than simple AND and OR, e.g., EOR and NOT. The sets are a generalization of min cut sets and can be used in quantitative and/or qualitative system reliability analysis. The algorithm is a top-down analysis and avoids sum of product expressions of top event, which usually involve a large number of terms. Each step of the algorithm is clearly defined and it is proven that all prime implicant sets can be obtained. The algorithm is efficient, and rather complicated trees can be handled manually.     

Calcium and magnesium deficiency-induced atrophy of muscle and calcium accumulation in the spinal cord

Four consecutive patients with mycosis fungoides received cyclic chemotherapy using cyclophosphamide, vincristine, and prednisone (COP). In one case, there was complete remission of disease for 15 months. In another, there was 50% regression of tumors with healing of the ulcerated surfaces. The third patient showed complete clearing of scaling and redness of the skin. In the fourth patient, the lesions progressed.     

Optimal sulfur dioxide gas discharge control for Osaka city,Japan

This paper, using actual data, establishes a dynamical model of sulfur dioxide pollution and gives an optimal discharge control policy for Osaka city in Japan.From a mass balance of the gas, a linear scalar state equation is obtained. In spite of the simplicity of the model, it can represent the effect of wind speed, inversion layer, diffusion, boundary concentration and discharge intensity of the pollutant sources. Six unknown parameters in the model and system noise variance are identified by maximizing likelihood function by Flexible Polyhedron Search. This makes the model reasonably representative.A quadratic loss function for the control problem is the weighted sum of two terms; the loss caused by the discharge control and the loss due to the air pollution. The wind speed varies at random according to a Markov transition matrix, considering actual emergency control situations. The optimal control policy is obtained by Dynamic Programming, resulting in linear feedback control dependent on the wind speed, involving a Riccati equation. A practical method of the determination of the weighting coefficient in the loss function is proposed.Since the model is a scalar state system, the whole calculation can be performed with a relatively small computer. This ease of computation and the Markov property of the wind speed suggest that the proposed method is applicable to the actual emergency discharge control.     

Conceptual designing of reduced-moderation water reactor with heavy water coolant

The conceptual designing of reduced-moderation water reactors, i.e. advanced water-cooled reactors using plutonium mixed-oxide fuel with high conversion ratios more than 1.0 and negative void reactivity coefficients, has been carried out. The core is designed on the concept of a pressurized water reactor with a heavy water coolant and a triangular tight lattice fuel pin arrangement. The seed fuel assembly has an internal blanket region inside the seed fuel region as well as upper and lower blanket regions (i.e. an axial heterogeneous core). The radial blanket fuel assemblies are introduced in a checkerboard pattern among the seed fuel assemblies (i.e. a radial heterogeneous core). The radial blanket region is shorter than the seed fuel region. This study shows that the heavy water moderated core can achieve negative void reactivity coefficients and conversion ratios of 1.06–1.11.     

Medical image diagnosis of kidney regions by deep feedback GMDH-type neural network using principal component-regression analysis

The deep feedback Group Method of Data Handling (GMDH)-type neural network is applied to the medical image recognition of kidney regions. In this algorithm, the principal component-regression analysis is used for the learning calculation of the neural network, and the accurate and stable predicted values are obtained. The neural network architecture is automatically organized so as to fit the complexity of the medical images using the prediction error criterion defined as Akaike’s Information Criterion (AIC) or Prediction Sum of Squares (PSS). The recognition results show that the deep feedback GMDH-type neural network algorithm is useful for the medical image recognition of kidney regions, because the optimum neural network architecture is automatically organized.     

Effects of reductive annealing on insulating polycrystalline thin films of Nb-doped anatase TiO2: recovery of high conductivity

We studied the effects of reductive annealing on insulating polycrystalline thin films of anatase Nb-doped TiO₂ (TNO). The insulating TNO films were intentionally fabricated by annealing conductive TNO films in oxygen ambient at 400 ℃. Reduced free carrier absorption in the insulating TNO films indicated carrier compensation due to excess oxygen. With H₂-annealing, both carrier density and Hall mobility recovered to the level of conducting TNO, demonstrating that the excess oxygen can be efficiently removed by the annealing process without introducing additional scattering centers.