In the modern welding technology of metal materials, the welding robot is a very important factor. And the welding wire feeding device is a very important part of the welding robot. The welding wire feed systems stabilization and reliability is essential for the welding quality. The resistance is caused by the draw of welding wire and other factors, which is highly nonlinear and stochastic. The wire feed rate and welding quality will be severely affected by the resistance. This paper investigates the method controlling the system according to the observations of the resistance, speed and other factors. There are the complex environmental factors in the welding process, such as the strong magnetic field and high current field, which are random. So the robust control algorithm based on the state space model is designed to keep the welding wire feed speed stable under disturbances. The robust control method can improve the speed of response and eliminate the effect of wire feed resistance to the wire feed speed. Even when disturbed, the system can quickly reach a stable wire feed speed.

Wireless sensor networks are widely deployed and thus are suffering increasing threats such as reactive bit jamming attacks. Such jamming can be alleviated by coded transmissions. This paper derives an optimal redundancy level required for Reed-Solomon code based transmissions and unveils the monotonic characteristic of coded transmission efficiency with respect to frame size, jamming strength, as well as communication protocol stack overhead.

The problem of computer control algorithm for the singular Von Neumann input-output model is researched. A kind of new mathematic method is applied to study the singular systems without converting them into general systems. A kind of stability condition under which the singular input-output model is admissible is proved with the form of linear matrix inequality. Based on this, a new state feedback stability criterion is established. Then the formula of a desired state feedback controller is derived.