Accurate wind speed forecasting plays a critical role in farm planning for economic operation of power grid. New methods which lead to better performances are urgently demanded the energy management as there is no one model that qualified all cases. In this paper, novel hybrid developed by integrating Fast Ensemble Empirical Mode Decomposition (FEEMD) and Regularized Extreme Learning Machine (RELM) optimized Backtracking Search Algorithm (BSA) realize accurate short-term forecasting. This...

Abstract This paper focuses on the problem of adaptive output feedback fault tolerant control for a nonlinear hydro‐turbine governing system. A dynamic mathematical model system is established, which aims to investigate performance under servomotor delay and actuator faults. Then, estimation observer proposed achieve online real‐time diagnosis Based information, an observer‐based controller designed. Furthermore, reasonable assumptions, results demonstrate that closed‐loop can global...

Aimed at distinguishing different fault categories of severity rolling bearings, a novel method based on feature space reconstruction and multiscale permutation entropy is proposed in the study. Firstly, ensemble empirical mode decomposition algorithm (EEMD) was employed to adaptively decompose vibration signal into multiple intrinsic functions (IMFs), representative IMFs which contained rich information were selected reconstruct vector space. Secondly, (MPE) used calculate complexity...

The shafting systems of hydropower units work as the core component for conversion water energy to electric and have been running a long time in hostile hydraulic-mechanical-electrical-coupled environment-their vibration faults are frequent. How quickly accurately identify improve reliability unit is key issue. This study proposes novel fault identification framework, which divided into three coordinated stages: nonlinear modeling, signal denoising, holographic identification. A dynamical...

This paper proposes a novel fault diagnosis method for rolling element bearings based on the newly developed adaptive signal processing technique -- variational mode decomposition (VMD), combining with time-frequency feature extraction and support vector machine (SVM). For given samples of signals, VMD is firstly employed to decompose signals into collections intrinsic functions (IMFs). To extract more characteristics information, 20 features each IMF are calculated from time domain...

The noise-assisted method of ensemble empirical mode decomposition represents a significant improvement over the original for eliminating mixing problem. However, will generate some additional problems, including contamination residue noise in signal reconstruction and high computational cost. In this work, an improved method, combining complementary time-saving by over-sampling investigated signal, is proposed to solve these problems. By using added white can be eliminated completely adding...

Due to the reversible properties of pump turbine, its operation along "S" shape curve can lead unstable operation, like remarkable unit frequency oscillation, under unload condition. To solve problem, a robust non-fragile fractional order PID controller (FOPID) turbine governing system is proposed in this paper. Furthermore, combination bacterial foraging algorithm (BFA) and multi-scenario functions are employed select parameters for novel FOPID controller, which composed by time domain...

The fault diagnosis of generator units is critical to guarantee the high efficiency electric system. However, detailed samples are difficult obtain, and distribution usually shows characteristics unevenness unbalance, which may lead low precision. Nevertheless, it has been seldom considered in traditional classifier for until now. In this paper, a novel weighted support vector data description (SVDD) with fuzzy adaptive threshold decision proposed applied units. To tackle drawback that SVDD...

To solve the profit-based unit commitment (PBUC) problem occurred in deregulated power systems, we developed a new method by use of multi-agent system (MAS). The agents our PBUC-MAS utilized rule based and dynamic programming methods to optimization corporately, they could be located on different computers same computer network anywhere. Through presented approach, can help generation companies (GenCos) make decision scheduling generators, selling reserving market obtain their maximum...

HVDC lines may have use the same right of way as ac lines. This paper discusses general study methodology that can be used to investigate any coupling between dc and transmission share right-of-way. The capacitive inductive a line with long neutral conductor an system running parallel it were investigated. A digital simulation model phenomenon was developed induced fundamental frequency voltage on conductor. current in converter transformer secondary winding also modelled. (5 pages)

This paper proposes a time-frequency method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and Winger-Ville distribution (WVD) for hydro-turbine non-stationary vibration signals analysis.Unlike the traditional EMD EEMD, CEEMDAN can efficiently reduce mixing has better convergence.Base series of intrinsic functions (IMFs) obtained by using CEEMDAN, WVD effectively suppress interference cross-terms accurately depict characteristics original signals.The...

For reducing the economic losses of hydraulic generator units, fault detection and diagnosis is fundamental. A technique presented that uses pattern recognition cluster analysis. Faults units are classified from characteristic parameters using statistical analysis methods. parameter matrix standard faults has been developed for identifying vibration faults. The carried out in a real case. Results demonstrated proposed method good candidate to be used as an online tool units.

Fault identification under variable operating conditions is a task of great importance and challenge for equipment health management. However, when dealing with this kind issue, traditional fault diagnosis methods based on the assumption distribution coherence training testing set are no longer applicable. In paper, novel state method integrated by time-frequency decomposition, multi-information entropies, joint adaptation proposed rolling element bearings. At first, fast ensemble empirical...