Abstract

In control theory, decomposition is a method according to which the studied system is divided into subsystems, the problem – into subproblems, etc., each of which is solved independently. The procedure of strict decomposition is based on a clearly formulated mathematical apparatus of decomposition of problems with block structure, as well as on various parametric methods. Heuristic decomposition methods allow an informal analysis of the specified tasks and make it possible to divide them on the basis of intuitive predictions about the interaction of individual tasks. Such a division can be based on various criteria, for example, the minimum of broken ties, technological or economic considerations, the division by stages of the management process, etc. The problem of optimization of rolling mills on a heavy-plate mill is multidimensional and characterized by several output indicators, each of which is a function of several controls. To solve this task it is necessary to replace a vector criterion with a scalar one. After that, there arises the problem of determining the weight coefficients of various indicators, which in the general case is solved at the level of expert assessments, and the efficiency of optimization of automatic control by this criterion is significantly determined by the experience of experts. In this situation, you can use the methods of heuristic decomposition, which allows you to analyze informally set tasks and make it possible to divide them on the basis of intuitive predictions about the interaction of individual tasks. The paper considers the formulation of the problem of optimization of rolling modes on heavy-plate mills by decomposing the general automation problem into a number of single-criteria problems using heuristic methods, which allows ensuring maximum efficiency of automation. The above scientific and technical solutions can be used in the development of automated control systems for rolling modes on the heavy-plate mill.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call