Learning control system of lifting machine motors

Abstract

Process automation control by diagnostic electric motors in operation conditions allows to reduce to a minimum the damage from these consequences due to early detection of defects. The theory of diagnosticof lifting machine motors has not been completely developed yet. In practice, the control of technical state of the motors is mainly performed during scheduled maintenance, which does not reveal to detect originating defects and to prevent significant damage of motors up to their complete failure. The difficulty of obtaining diagnostic information is that the main functional units of electric motors are dependent. This means that physical damage in any unit results in malfunctions of other units. The main way of increasing the efficiency of the automated control system of lifting machine motors is giving it the properties of adaptability on the basis of ideas and methods of machine learning and pattern recognition. To increase the operational reliability and service life of a mine electric lifting machines the article offers an information and machine learning algorithm for extreme functional control systems with electric hyprnspherical classifier. Normalized Shannon entropy measure was used as a criterion for functional efficiency of leaning systems of the functional control.

Keywords

• information-extreme intellectual technology
• functional control
• learning matrix
• learning algorithm
• functional efficiency criteria
• electric drive
• mine hoisting engine