This paper studies the seepage flow mathematical model of three-area composite reservoir under three kinds of outer boundary conditions (infinite boundary, constant pressure boundary and closed boundary), in which influences of well-bore storage and skin factor are not taken into consideration. On the basic of theory of similar structure of solution of boundary value problem of differential equation, this paper obtain the solution of the seepage flow model of three-area composite reservoir. The study is not only conducive to further analyze the inherent law of the solution and solve corresponding application problems, but also easy to compile corresponding analysis software.

This paper presents a percolation model for the composite reservoir, in which quadratic-gradient effect, well-bore storage, effective radius and three types of outer boundary conditions: constant pressure boundary, closed boundary and infinity boundary are considered. With Laplace transformation, the percolation model was linearized by the substitution of variables and obtained a boundary value problem of the composite modified zero-order Bessel equation. Using the Similar Constructive Method this method, we can gain the distributions of dimensionless reservoir pressure for the composite reservoirs in Laplace space. The similar structures of the solutions are convenient for analyzing the influence of reservoir parameters on pressure and providing significant convenience to the programming of well-test analysis software.

Based on the similar structure theory, we conduct rigorous mathematical derivation and proof for the boundary value problem of a class of differential equation . We put forward a new algorithm------Similar Structure Algorithm to solve such boundary value problems. The similar structure algorithm only includes arithmetic and logical operation, hence it can be oriented towards computer. According to the similar structure algorithm, an example is given to conduct a simulation experiment by the application of computer software at the end of the paper. And we also observe and analysis the variational laws of the laboratory result by changing the coefficient of the boundary condition.

The similar structure of solution for the boundary value problem of second order linear homogeneous differential equation has been studied. Based on the analysis of the relationship between similar structure of solution, its kernel function, the equation and boundary conditions, similar constructive method (shortened as SCM) of solution is obtained. According to the SCM, the similar structure of solution and its kernel function are constructed for the mathematical model of homogeneous reservoir which considers the influence of bottom-hole storage and skin effect under the infinite outer boundary condition. The SCM is a new and innovative way to solve boundary value problem of differential equation and seepage flow theory, which is especially used in Petroleum Engineering.

For the problem of inaccuracy of temperature control caused by large inertia, time variability; pure lag in the heating furnace, this article combined the learning mechanism of neural network control with the human thinking and reasoning of fuzzy control, establishing the 3 d fuzzy neural network system. Neural network is used to implement membership function, and drive the fuzzy reasoning. Using neural network, fuzzy modeling, achieve the goal of refinement fuzzy rules. Then I established the temperature control model of heating furnace which based on the fuzzy neural network. After using the model, the control accuracy and the uniformity of the slab temperature has been improved; the temperature difference between head and tail of slab has been reduced; which have a positive impact on the reducing of fuel consumption in the heating furnace, the improvement of yield in production line and the achievement of energy conservation and environmental protection.