Maximizing performance of a grid-connected photovoltaic (PV)-fuel cell hybrid system by use of a two-loop controller is discussed. One loop is a neural network controller for maximum power point tracking, which extracts maximum available solar power from PV arrays under varying conditions of insolation, temperature, and system load. A real/reactive power controller (RRPC) is the other loop. The RRPC achieves the system's requirements for real and reactive powers by controlling incoming fuel to fuel cell stacks as well as switching control signals to a power conditioning subsystem. Results of time-domain simulations prove not only the effectiveness of the proposed computer models of the two-loop controller but also its applicability for use in stability analysis of the hybrid power plant.

Two-loop controller for maximizing performance of a grid-connected photovoltaic-fuel cell hybrid power plant

Artificial neural networks and their applications to power systems—a bibliographical survey

Summary form only given as follows. This paper presents a new approach for real-time tuning the parameters of a conventional power system stabilizer (PSS) using a radial basis function (RBF) network. The RBF network is trained using an orthogonal least squares (OLS) learning algorithm. Investigations reveal that the required number of RBF centers depends on spread factor, /spl beta/ and the number of training patterns. Studies show that a parsimonious RBF network can be obtained by presenting a relatively smaller number of training patterns, generated randomly and spreadover the entire operating domain. Investigations reveal that the dynamic performance of the system with an RBF network adaptive PSS (RBFAPSS) is virtually identical to that of an artificial neural network based adaptive PSS (ANNAPSS). The dynamic performance of the system with RBFAPSS is quite robust over a wide range of loading conditions and equivalent reactance X/sub c/.

Radial basis function (RBF) network adaptive power system stabilizer

Reactive power/voltage control in a distribution substation is investigated in this work. The purpose is to determine proper capacitor on/off status and suitable load tap changer (LTC) positions for the 24 hours in the next day. To reach this goal, an artificial neural network (ANN) is designed to reach a preliminary dispatch schedule for the capacitor and LTC. The inputs to the ANN are main transformer real power and reactive power and primary and secondary bus voltages and the outputs are the desired capacitor on/off status and LTC tap positions. The preliminary dispatch schedule is further refined by fuzzy dynamic programming in order to reach the final schedule. To demonstrate the effectiveness of the proposed method, reactive power/voltage control is performed on a distribution substation in Taipei, Taiwan. Results from the example show that a proper dispatch schedule for capacitor and LTC can be reached by the proposed method in a very short period.

A combined artificial neural network-fuzzy dynamic programming approach to reactive power/voltage control in a distribution substation

Application of recurrent, neural networks in the design of an adaptive power system stabilizer (PSS) is presented in this paper. The architecture of the proposed adaptive PSS has two recurrent neural networks. One functions as a tracker to learn the dynamic characteristics of the power plant and the second one functions as a controller to damp the oscillations caused by the disturbances. In the proposed approach, the weights of the neural networks are updated on-line. Therefore, any new information available during actual control of the plant is considered. Simulation studies show that the artificial neural network (ANN) based PSS can provide very good damping over a wide range of operating conditions.

An adaptive power system stabilizer based on recurrent neural networks

A new approach using an artificial neural network is proposed to adapt power system stabilizer (PSS) parameters in real time. A pair of online measurements i.e., generator real-power output and power factor which are representative of the generator's operating condition, are chosen as the input signals to the neural net. The outputs of the neural net are the desired PSS parameters. The neural net, once trained by a set of input-output patterns in the training set, can yield proper PSS parameters under any generator loading condition. Digital simulations of a synchronous machine subject to a major disturbance of a three-phase fault under different operating conditions are performed to demonstrate the effectiveness of the proposed neural network. >

Yuan-Yin Hsu

Papers

Tuning of power system stabilizers using an artificial neural network