Automatic Generation Control (AGC) problem of a large interconnected power system has been studied by dividing the whole system into a number of control areas. A control area is defined as a power system, a part of a power system or a combination of systems to which a common generation control scheme is applied. The electrical interconnections within each control area are very strong as compared to the ties with the neighboring areas. All the generators in a control area swing in unison or coherently and it is characterized by a single frequency. In normal steady state operation each control area of a power system should strive to meet its own load demand and in addition each control area of a power system should participate in regulating the frequency of the system.

A lot of work concerning AGC of interconnected power system has been reported in past using several control strategies such as classical control, optimal control, suboptimal control, adaptive control and etc, to obtain better dynamic responses. Although many studies pertaining to controllers for thermal plants are available in the literature ,they are mostly concentrating on secondary control aspects of AGC .Very few works pertaining to primary control /governor control aspect .i.e proper selection of speed regulation parameter ‘R’ is found. However the works relate to selection of speed regulation parameter ‘R’ rather than optimization. Surprisingly no work has been reported on study of the importance tie line strength variations and it’s effect on system dynamic response. System response to unequal ties, which exist in practice in a multi area system has not been investigated. Sensitivity analysis of a multi area system and effects of tie line snapping are yet to be explored. Effect of two stage reheat in a multi area system is yet to be explored.

All the above intrinsic problems have been care fully investigated in the present work. The present work explores the problem of Automatic Generation Control of two area reheat thermal & multi area (three areas) reheat thermal system with Generation Rate Constraints (GRC).

Chapter -I introduce the AGC problem of interconnected power systems in general and present a critical survey of the past work concerning AGC. It clearly lays down the objectives and motivation of the research work presented in the thesis.

A maiden attempt has been made to introduce a new optimization technique called Bacterial Foraging Technique for integral controller gain optimization. The difficulty of optimizing the parameters of AGC controllers is mitigated by using Bacterial Foraging Technique which gives an opportunity optimize many parameters at a time. As optimization is simultaneous, the relative effects of variation of parameters are taken care of. Most important achievement under this technique is the optimization of ‘R’ parameter along with controller gains. Bacterial Foraging Technique has been applied to both two area and three area system for small perturbation in an area on individual basis and in all areas simultaneously. The responses obtained are compared with that of classical technique. Findings from the investigations are higher values of “R” can be safely taken with improving the system performance.

In a normal reheat system, single stage reheat has been considered in the past. With advent of modern technology two stage reheat system for improving efficiency are existing for bigger units Effect of two stage reheat has been considered for two area system in past. Effect of second stage reheat for multi area system has been discussed here ; responses comparison with that of single stage reheat is done. Analysis reveals that the two stage reheat turbine can be modeled as a single stage reheat turbine for all practical purposes.

From regulation and stability point of view tie lines are vital link between two connected areas. Tie line strength variations and its effect on the system response has not been discussed in the past. Responses for ±25% variation in tie strength and effects of unequal ties for a three area system are discussed. How system behaves when interconnecting tie line is snapped, and effect of auto reclosing are also analyzed. Analysis of the responses concludes that with high value of tie strength system performance detoriates. System performance is hardly affected if auto reclosing takes place within 5 cycles of snapping

Most of the earlier researchers have considered the problem of AGC for equal interconnected areas. However in a realistic situation, although a particular system can be represented by control areas, their capacities in general will be unequal. There are situation when anew control area of finite capacity may be interconnected to an already existing system of large capacity. In such cases, it would be of practical significance to investigate the effect of capacity ratios of two control areas on selection of their optimum controller settings. The present work (Chapter -II) highlights the condition for which the controller settings of individual areas can be optimized, treating the rest of the system having infinite inertia. Under such condition individual pool member can optimize their controller settings on individual basis. Effect of dead band on governor is also discussed, In the presence of GRC the governor dead band role is becoming quite insignificant. Selection of speed regulation parameters for two areas and three areas construed that higher value of ‘R’ can be safely taken without affecting the system performance.

Sensitivity analysis carried out for two area and three area system to establish that for a wide variation in system load (±25%) and system parameters (±25%) , the optimum integral controller gain changes. However there is hardly any change in system response is observed. It emphasizes the robustness of the controller to the changes in operating condition and system parameters.

Lastly chapter -VII, brings out the detailed conclusion of the entire work and the scope for further research.