The paper derives techniques for online and offline computer analysis of the power system in the presence of large amounts of stochastic-source (wind, solar, tidal etc.) generation and develops an extension of the probabilistic simulation (Beleriaux-Booth) technique for use in generation expansion planning. The online techniques are employed for the assessment of spinning reserve and unloadable generation requirements and the prediction of probable system frequency deviations. The offline methods are used for energy production estimation taking account of correlation effects and the required minimum levels of conventional generation that must be on the system for regulating (load following) the continuously varying stochastic-source power generation.

The paper deals with some aspects of automatic generation control of the two-area hydrothermal system provided with classical controllers. A comprehensive procedure for continuous-and discrete-mode optimisation of integral controllers using an integral squared error criterion is suggested. Investigations reveal that the optimal integral gains achieved through continuous-mode analysis are totally unacceptable in the discrete mode for the sampling periods used in practice. Moreover, for all practical purposes, the optimum integral gains in the discrete mode can be achieved by neglecting the generation rate constraints from the mathematical model, in contrast to the case in the continuous mode. Analysis also highlights significant differences in the dynamic performance of the hydrothermal system due to step-load perturbation in either of the areas. An attempt is also made to recommend an optimum sampling period.

A method is presented for the accurate simulation of the hydraulic and mechanical aspects of a hydroelectric power-generating system. The representation is suitable for incorporation into an overall power-system model. In the development of the detailed model, the hydraulic equations are solved by the method of characteristics, and the hydromachine is represented by its performance curves. Linear turbine transfer functions are subsequently derived from the detailed model using the method of fast Fourier transforms. Comparisons are made between the formulations, and these show that the range of applicability of the approximate linear models is limited. Representative results from two practical systems are collected together to illustrate the use of the models in power-system studies.

Stable control of the frequency in an electrical power system is essential to ensure continuity of supply. It is determined by the combined effect of the speed governors fitted to the generating sets throughout the system. The power output response of a steam-driven set to sudden loss in generating capacity is initially fast, but, after a few seconds, loss of boiler pressure reduces the output of the set. On the other hand, the response of a hydroelectric generator, although initially less rapid (due to the need to accelerate the water column), can be maintained at a high level for as long as storage water is available. Hydrosets are therefore particularly useful for system-frequency control provided they are fitted with governors that ensure a fast, but stable, response to a sudden change of load. In practice, governor settings are usually made during commissioning on the basis that the set will be stable when fully loaded and electrically isolated from the rest of the system. Attention, in the paper, is therefore given to the transient speed response of a single, isolated, governed hydrogenerator operating at, or near, full load.

The multilevel control of large systems intended to distribute between several control levels, the calculation of optimum set points and the use and calculation of optimum control laws are based upon the concept of decomposition-co-ordination of the overall optimisation problem. For the nonseparable problem examined, it is necessary to use a fomulation of the decomposition-co-ordination principle different from that used for separable problems. Decomposition methods (recognition of coupling variables, introduction of pseudovariables) and coordination algorithms are described briefly. The problem of optimum power scheduling in an hydroelectric system, representing an example of a typical nonseparable optimisation problem, is discussed. Various decomposition co-ordination methods are proposed, and a concrete example is examined which shows the effectiveness of multilevel control techniques applied to very highly nonseparable problems.

A coupled electromechanical and hydrodynamic time-domain simulation of a direct-drive generator connected to a heaving buoy for wave energy conversion is presented. The system is based on a novel direct-drive power take-off unit referred to as snapper. The simulation is based primarily in MATLAB using its built-in ordinary differential equation solvers. These solvers act on the data derived from electromagnetic finite element analysis and from the WAMIT wave interaction simulation software. Test results of a generator prototype for comparison with the electromechanical simulation are presented. Results from wave tank tests of a full system incorporating the power take-off are also provided for comparison with the hydrodynamic model.

In Brazil, SHP (Small Hydroelectric Plant) has been used as alternative to generate cleaner and renewable energy. These small power plants are between 1 and 30 MW and are located in rural areas. The SHP should be supervised from an OCC (Operational Control Centre) located, usually, at one substation. For that, several data communication technologies can be used as: Leased Line, Private Line, GPRS (General Packet Radio Service), among others. This paper presents the study how the PLC can be used as alternative service for monitoring and control of SHP, describing the functional and nonfunctional requirements to be met to the PLC specification. In addition, the paper focuses attention on the new strategies required for controlling and monitoring at Smart Grids and shows how PLC could have a key role among the technologies that can be applied. (4 pages)

This chapter presents an overview of the main issues associated with a wave energy generation system from a power system's standpoint. Issues specifically related to the time profile of power exported from a wave energy power plant are considered, and the impact of this fluctuating power on the power system performance is addressed. Some of these issues are covered in greater depth in future chapters. The need for reactive power compensation equipment, particularly in far offshore farms, is considered and addressed. The off-grid type of operation is also described. Most of the principles are illustrated with simplified models of wave energy generators (WEGs) with sinusoidal type outputs.

This study proposes a unified power flow controller-based oscillation damping controller for use with a permanent magnet synchronous generator-based offshore power farm. A novel intelligent damping controller (NIDC) was used to increase the stability of power control and improves the performance, where the proposed NIDC consists of the adaptive critic network, the functional link-based Elman neural network (FLENN), the genetic ant colony optimisation algorithm (GACO) and proportional–integral-derivative (PID) linear damping controller. The PID damping controller analyses complex eigenvalues based on the theory of modern control. The node connecting weights of the FLENN and critic network are trained online. A GACO approach is developed to adjust the learning rates and thus improve the learning ability. The proposed NIDC can achieve better damping characteristics, and the internal power fluctuations can also be effectively alleviated.