The thesis focuses on the investigation of application of tangent modulus factor and strength utilization ratio function (SURF)to solution of structural fracture strength problems.

Strength utilization ratio function is a common solution about strength and stability probbms of steel structure under pressure.

A reward function for real traffic condition was proposed to control traffic dynamically and on real-time.

The design of reward function is one of difficulties in building reinforcement learning system.

To improve the performance of the reinforcement learning method on multi-agent systems,thinking about the characteristic of Keepaway that always ended with failure,based on the reference of the reward function design pattern in the pole-balance system,a new punitive reward function is redesigned.

Here we discuss its properties basing on the definition of the strong pseudoconvex function,and give its relationship with strong convex function.

A new simplified ductile spall model is presented using the redefined damage and the strength function given by Cochran-Banner.

The strength function given by Cochran-Banner was maintained using the redefined damage,and the correction concerning the volume of the mesh cells was realized considering it unnecessary to expect that it is much easier to open microcracks once they are formed than to strain the solid further.