Extreme learning machine (ELM) as a neural network algorithm has shown its good performance, such as fast speed, simple structure etc, but also, weak robustness is an unavoidable defect in original ELM for blended data. We present a new machine learning framework called “LARSEN-ELM” for overcoming this problem. In our paper, we would like to show two key steps in LARSEN-ELM. In the first step, ...

seismic ground motion is profoundly affected by geometrical and mechanical properties of soil deposits overlaying bedrock. local seismic ground response of saturated soil deposits was studied in literature by applying the effects of soil stress state and index properties on the strain-dependent normalized shear modulus reduction, g/g0, and damping ratio, d, curves in an equivalent linear analys...

Value prediction is an important subproblem of several reinforcement learning (RL) algorithms. In a previous work, it has been shown that the combination of least-squares temporal-difference learning with ELM (extreme learning machine) networks is a powerful method for value prediction in continuous-state problems. This work proposes the use of ensembles to improve the approximation capabilitie...

Measurements of local density and magnetic field fluctuations near the pedestal top, conditionally-averaged over the ELM cycle, have been made in MAST. A Doppler backscattering (DBS) system installed at MAST was used to measure intermediate-k (k⊥ρi ≈ 3 − 4) density fluctuations at the top of the pedestal. A novel diagnostic technique combining DBS with crosspolarization scattering (CP-DBS) enab...

In this paper, two simplified models of edge localized mode (ELM) power exhaust are developed, one based on the kinetic and the other on the fluid treatment of parallel losses. These models are found to capture many (though not all) of the salient features of kinetic simulations at substantial savings in both cost and complexity (CPU time in seconds versus days), making them ideal as real time ...

Density profiles in pedestal region (H-mode) are measured in HL-2A and the characteristics of the density pedestal are described. Cold particle deposition by Supersonic Molecular Beam Injection (SMBI) within the pedestal is verified. ELM mitigation by SMBI into the H-mode pedestal is demonstrated and the relevant physics is elucidated. The sensitivity of the effect to SMBI pressure and duration...

In recent years, forecasting has received increasing attention since it provides an important basis for the effective operation of power systems. this paper, a hybrid method, composed kernel principal component analysis (KPCA), tree seed algorithm based on Lévy flight (LTSA) and extreme learning machine (ELM), is proposed short-term load forecasting. Specifically, randomly generated weights bia...