||There are more than five hundred thousand boilers in China, and they are consuming a large portion of the nation's natural energy. Most of these boilers are controlled in old fashions and their efficiency is low. One of the most effective means of boiler efficiency enhancement is an improvement of the steam generation control system. Since the dynamics of a boiler display nonlinearity, nonminimum phase behavior, instability and time delay, the control performance using the classical approaches is generally unsatisfactory. The main objective of this research is to develop fuzzy control schemes for drum-boiler systems to achieve better performance. In the first part of the thesis, we investigate some of the existing drum-boiler mathematical models. The step responses of the different boiler models are examined and compared. They show different properties of the boiler system and in particular the swell and shrink effect of drum boilers. One of the boiler models will be used as the main framework in the following boiler control investigations. In the second part of the thesis, fuzzy control systems for drum pressure and drum water level are developed and compared with conventional approaches. First, simple fuzzy controllers are designed. The simulation results show that the simple fuzzy controllers have better performance than conventional PI controllers for setpoint tracking and have comparable results for input disturbance rejection. To speed up the responses of the control systems, feed-forward controllers are added to the feedback controllers. In the second fuzzy control methodology, two-level fuzzy controllers that use the feed-forward idea are explored. The fuzzy rule bases of the secondary control systems try to include the functioning of feed-forward control. The primary controllers could be simple PI controllers or the PI-like fuzzy controllers which we have developed early. Finally, an adaptive fuzzy system tuned PI control scheme is developed. Here, a fuzzy system is constructed to tune the parameters of the PI controller automatically. The results reveal that the adaptive fuzzy control system gives improved performance over the conventional PI controllers in both setpoint tracking and disturbance rejection.