Search

This research presents a novel approach for optimal active and reactive power control using permanent magnet synchronous generator with fully rated converter (PMSG‐FRC). The power output variations and wind energy intermittent performance create significant challenges for power system integration, operation, and control. A fast and reliable sensorless optimal wind speed (WS) control system‐based extreme learning machines and complete ensemble empirical mode decomposition with adaptive noise is developed. Also, an approximate entropy‐based complexity measure is used for online WS estimation. The fuzzy controller system is used for calculating the optimal wind turbine (WT) speed, power, and torque, and then adjusts WT by the optimal speed value. Moreover, the wind farm (WF) active and reactive power controller is utilised for obtaining the active and reactive power reference values for every region/feeder. The system is applied and integrated to power system grid by using IEEE standard models, and the speed of WT is adjusted by the required active power level and the estimated WS value to increase the maximum power capture from WF. The simulations results using different cases studies and power system voltage levels have confirmed the validity and accuracy of the proposed control algorithms for practical applications and demonstrated excellent performance for power system integration. [ABSTRACT FROM AUTHOR]

Artificial Neural Network (ANN) has been recognized as a powerful method for short‐term load forecasting (STLF) of power systems. However, traditional ANNs are mostly trained by gradient‐based learning algorithms which usually suffer from excessive training and tuning burden as well as unsatisfactory generalization performance. Based on the ensemble learning strategy, this paper develops an ensemble model of a promising novel learning technology called extreme learning machine (ELM) for high‐quality STLF of Australian National Electricity Market (NEM). The model consists of a series of single ELMs. During the training, the ensemble model generalizes the randomness of single ELMs by selecting not only random input parameters but also random hidden nodes within a pre‐defined range. The forecast result is taken as the median value the single ELM outputs. Owing to the very fast training/tuning speed of ELM, the model can be efficiently updated to on‐line track the variation trend of the electricity load and maintain the accuracy. The developed model is tested with the NEM historical load data and its performance is compared with some state‐of‐the‐art learning algorithms. The results show that the training efficiency and the forecasting accuracy of the developed model are superior over the competitive algorithms. [ABSTRACT FROM AUTHOR]

Intelligent system (IS) using synchronous phasor measurements for transient stability assessment (TSA) has received continuous interests recently. For post‐disturbance TSA, one pivotal concern is the response time, which was reported in the literature as a fixed value ranging from 4 cycles to 3 s after fault clearance. Since transient instability can develop very fast, there is a pressing need for faster response speed. This paper develops a novel IS to balance the response speed and accuracy requirements. A set of classifiers are sequentially organised, each is an ensemble of extreme learning machines (ELMs), whose inputs are post‐disturbance generator voltage trajectories and outputs are the classification on the stable/unstable status of the post‐disturbance system and an evaluation of the credibility of the classification. A self‐adaptive TSA decision‐making mechanism is designed to progressively adjust the response time, such that the IS can do the classification faster, thereby allowing more time for emergency controls. The ELM ensemble classifiers can also be updated by on‐line pre‐disturbance TSA results due to its very fast learning speed. Case studies on the New England system and IEEE 50‐machine system have validated the high efficiency and accuracy of the IS. [ABSTRACT FROM AUTHOR]

With a large number of inverter‐interfaced renewable power generation, fault‐induced delayed voltage recovery (FIDVR) events have become a serious threat to power system stability assessment. This study proposes a novel data‐driven method based on probabilistic prediction, ensemble learning, and multi‐objective optimisation programming (MOP) to rapidly predict the FIDVR severity index for real‐time FIDVR assessment. Distinguished from the existing single machine learning (ML) algorithm data‐driven method, the proposed method combines different randomised learning algorithms to acquire a more diversified ML outcome. The probabilistic prediction models the uncertainties existing in the prediction process, which quantifies the prediction confidence over a progressive observation window. Besides, the FIDVR can be evaluated through the time‐adaptive framework to achieve the best FIDVR speed and accuracy with the MOP framework. The simulation results on the New England 10‐machine 39‐bus system display its preponderance over the single ML, and also demonstrate its better speed and accuracy performance in FIDVR assessment. [ABSTRACT FROM AUTHOR]