关于微分方程的一个混合边界问题

F. G. Tricomi (1923- ), S. Gellerstedt (1935- ), F. I . Frankl (1945- ), A. V. Bitsadze and M. A. Lavrentiev (1950- ), M. H. Protter (1953- ) and most of the recent workers in the field of mixed type boundary value problems have considered only one parabolic line of degeneracy. The problem with more than one parabolic line of degeneracy becomes more complicated. The above researchers and many others have restricted their attention to the Chaplygin equation:K(y)·u_xx+u_yy =f(x,y) and not considered the "generalized Chaplygin equation:"Lu=K(y)·u_xx+u_yy+r(x,y)·u=f(x,y) because of the difficulties that arise when r₁=non-trivial(≠0). Also it is unusual for anyone to study such problems in a doubly connected region. In this paper 1 consider a case of this type with two parabolic lines of degeneracy, r₂=non-trivial(≠0).in a doubly connected region, and such that boundary conditions are prescribed only on the-exterior boundary" of the mixed domain, and 1 obtain umqueness results for quasnegutar solutions of the characteristic and non-characteristic Problem by applying the b,c energy integral method in the mixed domain.     

一类拟线性椭圆方程组的Dirichlet问题

The homogeneous Dirichlet problem(1) for quasilinear elliptic system in a bounded domain Ω is investigated in this paper. The existence of generalized solutions in [H₀¹(Ω)]^N is obtained by using the contructive Galerkin method. For the case of a_ij^lm=0 when i≠j, it is estatablished that such generalized solutions have bounded [L_∞(Ω)]N norm and possess Holeler continuity. Even in the particular case that f_i are independent of Du, our results have improved those of A. V. Lair [Ann. Mat. Pura Appl., 116(1978)], allowing b_i¹(x,u) and f_i(x,u) to have a growth in u arbitrarily close to 1.     

关于M/G/l随机服务系统中L.Takacs引理的证明

L. Takacs' Lemma (ef [1] PP47-48) is a very important tool for the investigation of the transient behavior of the M/G/1 System. But the original proof contains a gap when γ(s, w) is expanded into Lagrange series by means of Lagrange theorem as given in the appendix of [1]. This is due to the fact that φ(s)=∫₀^∞e^-sxdH(x), the Laplace transform of the service time distribution , may not be analytic at s=0. An example of such a distrbution function H(x) is gives and a specific Lagrange theorem is proved By using this specific theorem, the gap in the original proof of Takacs′ Lemms is eliminated.     

Congruence Pairs of Decomposable MS-Algebras*

In this paper, the authors first introduce the concept of congruence pairs on the class of decomposable MS-algebras generalizing that for principal MS-algebras (see [13]). They show that every congruence relation θ on a decomposable MS-algebra L can be uniquely determined by a congruence pair (θ₁, θ₂), where θ₁ is a congruence on the de Morgan subalgebra L^?? of L and θ₂ is a lattice congruence on the sublattice D(L) of L. They obtain certain congruence pairs of a decomposable MS-algebra L via central elements of L. Moreover, they characterize the permutability of congruences and the strong extensions of decomposable MS-algebras in terms of congruence pairs.     

一类离散分布参数的经验Bayes估计的收敛速度

In this paper we consider a family of discrete distributions f_θ(x)dμ(x), and suppose that the Bayes estimate of φ(θ) with respect to the priori distribution H∈H has a form dH(x) =(?)a_k(x)f(x+k)/f(x). where f(x)=∫f_θdH(θ) . we construct asequence of empirical Bayes estimates and establish its rate of convergence, and prove that under suitable conditions this rate of convergence can arbitrarily close to 1. we also give a counter-example to the main Theorem 2.1 of [5], and then declare that the "Theorem" does not hold.     

The generic fiber of the universal deformation space associated to a tame Galois representation

We will study the generic fiber over of the universal deformation ring R _Q , as defined by Mazur, for deformations unramified outside a finite set of primes Q of a given Galois representation , E a number field, k a finite field of characteristic l. The main result will be that, if ˉρ is tame and absolutely irreducible, and if one assumes the Leopoldt conjecture for the splitting field E ₀ of , then defines a smooth l-adic analytic variety, near the trivial lift ρ₀ of ˉρ, whose dimension is given by cohomological constraints and as predicted by Mazur. As a corollary it follows that, in the cases considered here, R _Q is a quotient of by an ideal I generated by exactly m equations, where and . Under the above assumptions for and ˉρ odd, using ideas of Coleman, Gouvêa and Mazur it should now be possible to show that modular points are Zariski-dense in the component of , that contains the trivial lift ρ₀, provided this lift satisfies the Artin conjecture and E ₀ satisfies the Leopoldt conjecture. Furthermore, in the Borel case, we show that the Krull dimension of R _Q can exceed any given number, provided Q is chosen appropriately. At the same time, we present some evidence that despite this fact, one might however expect that the dimension of the generic fiber is given by the same cohomological formula as in the tame case. Received: 12 December 1997 / Revised version: 5 February 1998     

Fuzzy度量化——嵌入理论的一个应用

A fuzzy Pseudo-metric space is called fuzzy metric space iff it is sub-T₀ space. Suppose that (X,F) is a fuzzy topological space. Consider a relation ～between ordinary points of X: x～y iff for each λ≠0,x_λ∈y_λ and y_λ∈x_λ. The relation～is an equivalent relation. The corresponding fuzzy quotien space is a sub-T0 space and is called the associated sub-T₀ space of (X,F). in this paper, we establish the following theorem via fzzzy imbedding theory:Theorem Suppose that (X,F) is a fuzzy topological space with countable topological base. Then (1)(X,F) is fuzzy metrizable iff it is a fuzzy completely regular and sub-T₀ space.(2)(X,F) is a fuzzy completely regular space,then its associated sub-T₀ space is fuzzy metrizable.     

Tur´an Problems for Berge-(k, p)-Fan Hypergraph*

Let F be a graph. A hypergraph H is Berge-F if there is a bijection f : E(F) → E(H) such that e ? f(e) for every e ∈ E(F). A hypergraph is Berge-F-free if it does not contain a subhypergraph isomorphic to a Berge-F hypergraph. The authors denote the maximum number of hyperedges in an n-vertex r-uniform Berge-F-free hypergraph by ex_r(n, Berge-F).A (k, p)-fan, denoted by F_k,p, is a graph on k(p ? 1) + 1 vertices consisting of k cliques with p vertices that intersect in exactly one common vertex. In this paper they determine the bounds of ex_r(n, Berge-F) when F is a (k, p)-fan for k ≥ 2, p ≥ 3 and r ≥ 3.     

二值回归模型中自变量误差的处理

In this paper the binary regression model, in which the response Y is binary, i.e. Y takes the value 0 or 1 only, is considered. The conditional probability P{Y=1|X=x} is assumed to be of the form G(a₀+a₁x), where x is the measurement of the factor, a= (a₀,a₁)′ is the vector of the unknown parameter being estimated and G(a) = (1+e^a)^-1. The measurement x is assu-med to be contaminated when the value of x is out of certain limit. M-estimation is employed to estimate the unknown a of the model. In this paper, the consistent M-estimation is proved to be asymptotically normal and the optimal M-equation is obtained. Finally, a computation method is introduced to solve the M-equation and to get the optimal solution.     

混合Bitsadze-Lavrentév-Tricomi边值问题

It is well known that F. G. Tricomi (1923) is the originator of the theoryof boundary value problems for mixed type equations by establishing the Thicomi equation: y·u_xx+u_yy=0 which is hyperbolic for y < 0, elliptic for y=0. and parabolic for y= 0 and then applied it in the theory of transonic flows.Then A.V.Bitsadze together with M. A . Lavrent′ev (1950) established the Bitsadze Lavre nt′ev equation: sgn( y ) ·u_xx+u_yy=0 where sgn(y) = 1 for y > 0, = -1 for y<0, 0 for y=0 with the discontinuous coefficient sgn( y ) of u_xx, while in the case of Tricomi equation the corre sponding coefficient y is continuous. In this paper we establish the mixed Bitsadze Lavrent′ev Tricomi equation. Lu=K(y)·u_xx+sgn(x) ·u_yy+r(x,y)·u=f(x,y), where the coefficient K=K(y) of u_xx is increasing continuous and coefficient M=sgn(x) of u_yy discontinuous, r=r(x,y) is once continuously differentiable, f=f(x,y) continuous. Finally we prove the uniqueness of quasi regular solutions and observe that these new results can bbe applied in fluid dynamics.