No abstract

Abstract-This paper presents results of our work in development of a genetic algorithm based path-planning algorithm for local obstacle avoidance (local feasible path) of a mobile robot in a given search space. The method tries to find not only a valid path but also an optimal one. The objectives are to minimize the length of the path and the number of turns. The proposed pathplanning method allows a free movement of the robot in any direction so that the path-planner can handle complicated search spaces.

Control of three-phase grid-connected voltage source converter under unbalanced grid faults greatly depends on the active and reactive powers processed by the converter. The instantaneous active power theory with sequence decomposition is employed to analyse the instantaneous power components, especially the second-order oscillation power. Study shows that the second-order oscillation power comprises two quadrature components, the cosine and sine terms, which are contributed by the average active power and the average reactive power, respectively. This finding sheds insight on the regulation of oscillation power under unbalanced grid conditions. Based on this observation, the authors propose a positive and negative sequence conductance and susceptance control scheme, which enables simple regulation of the active power or reactive power oscillation with the average active power and reactive power control. In addition, the authors investigate the relationship between the positive/negative sequence conductance and susceptance distribution factors with power oscillation and peak current. A maximum current limitation scheme is embedded into the current reference generation block for overcurrent protection. Numerical simulations and prototype measurements verify the accuracy of the analysis and the effectiveness of the control scheme.

No abstract

No abstract

Electricity price is an important indicator of the market operation. Accurate prediction of electricity price will facilitate the maximization of economic benefits and reduction of risks to the power market. At the same time, because of the excellent performance of deep learning models, using long-short term memory neural network (LSTM) and other deep learning models to predict time series has gradually become a research hotspot. In this paper, an optimized heterogeneous structure LSTM model is proposed to solve the problems of the single network structure and hyperparameter selection existing in the current research on LSTM. The heterogeneous structure LSTM is constructed based on the decomposed and reconstructed electricity price data, and sequence model-based optimization (SMBO) is used to optimize its hyperparameters. In order to verify the proposed model, online hourly forecasting and day-ahead hourly forecasting are tested on the electricity markets of Pennsylvania-New Jersey-Maryland (PJM). The experimental results show that the performance of the proposed model is much better than that of the general LSTM model and traditional models in accuracy and stability, providing a new idea for the use of LSTM for time series prediction. INDEX TERMS Long short-term memory neural network, neural network structure, hyperparameter optimization, time series analysis, electricity price forecasting.

No abstract

No abstract