Title of the Paper: Neural Network Control of the StatCom in Multimachine Power Systems
DOWNLOAD FULL PDF
Authors: Juan M. Ramirez Ruben Tapia O.
Abstract: This paper presents the application of neural networks for controlling the static synchronous compensator (StatCom) device. The primary duty of the StatCom is the regulation of the AC bus bar voltage where the device is connected. Additionally, a secondary task may be added to such device for obtaining a positive interaction with other controllers in order to mitigate low frequency oscillations. For this task, a neural network is proposed due to its simple structure, adaptability, robustness, considering the power grid nonlinearities. The applicability of the proposition is studied by digital simulation exhibiting satisfactory performance. Results of simulation for different disturbances and operating conditions demonstrate the effectiveness of the feedback variables selected in the control scheme.
Keywords: Controllers, damping, flexible AC transmission systems, neural networks.
Title of the Paper: Transmission Usage Allocation in Pool and Bilateral Trades Using Artificial Neural Network
DOWNLOAD FULL PDF
Authors: M.W. Mustafa, S.N. Khalid,
Abstract: This paper proposes methods to allocate transmission usage for pool and bilateral market models in deregulated power industry. This paper focuses on creating an appropriate artificial neural network (ANN) to allocate transmission usage for pool and bilateral trades separately in a simpler and faster manner. The modified IEEE 14-bus network is utilised as a test system to illustrate the effectiveness of the ANN output compared to that of conventional methods used as teachers. The basic idea is to use supervised learning paradigm to train the ANN. Downstream tracing procedure of Graph method is used as a teacher for calculating the contribution factors of individual generator to line flows under pool model. In bilateral model, circuit method is used as a teacher to decouple the line usages on the basis of transactions pairs. The descriptions of inputs and outputs of the training data for the ANN are easily obtained from the load flow results and methods used as teachers respectively. The structure of each ANN is designed to assess the extent of line usage by each generator while supplying to their respective customer. Most commonly used feedforward architecture has been chosen for the proposed ANN based transmission usage allocation technique. Almost all the system variables obtained from load flow solutions are utilised as an input to the neural network. Moreover, tan-sigmoid activation functions are incorporated in the hidden layer to realise the non linear nature of the transmission usage allocation. The proposed ANN based method provides promising results in terms of accuracy and computation time.
Artificial neural network, Bilateral trades, Graph method, Circuit method, Transmission usage allocation, Power pool.