Digitally controlled power systems: how much intelligence is needed and where it should be

The generation of power for telecommunications systems does not inherently require computer control. However, the use of microprocessors can dramatically enhance the capabilities of both power converters and peripheral system functions. Proper application of intelligent control to a power system adds desirable features, improves system reliability, makes the system easier to use, lowers both equipment and operating costs, and pushes performance to levels that would be impossible otherwise. Misapplication of intelligence can have the opposite effects. Intelligent control can be effectively applied to power converters, low voltage disconnect controllers, overall system control, user interfaces, and the monitoring of voltage, current, and temperature. Remote monitoring and full remote control are made possible through the use of intelligent control. Manufacturing, purchasing, installation, inventory, and maintenance information can also be electronically embedded within the system. In practice, the proper application of intelligent control to a power system will seek an optimal balance of the often conflicting design goals of lower cost, higher reliability, greater simplicity, improved performance, and added features. Distributed control, distributed communications, and the use of small, dedicated controllers are effective in achieving these goals. This paper offers some practical examples of where added intelligence has enhanced existing products. It discusses the advantages and disadvantages of intelligent control, the location and extent of the intelligence, and some caveats on how far to push the envelope.