The multi-parameter combination mind-evolutionary-based machine learning and its application

Xie, Kun; Mou, Changhua; Xie, Gang

doi:10.1109/icsmc.2000.884986

Cited by 5 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm proposed is based on the approach to problems with the GA and the analysis of the development of human thinking and the simulation of human society in the convergence and alienation of the phenomenon. Therefore, to some extent, it solves the slow convergence problem and other issues of the traditional algorithms with respect to local optimal …”

Section: Research On the Mppt Methods Based On The Bp Neural Network mentioning

confidence: 99%

See 1 more Smart Citation

Research on the voltage prediction of unmanned aerial vehicle photovoltaic modules based on new combination optimization algorithm

Wang

Shen

2019

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Summary Solar energy is a new energy source that is not only renewable but also available everywhere in the world. Thus, this new energy source is used on solar unmanned aerial vehicles. The use of solar photovoltaic systems (PVS) involves the conversion of solar energy into electricity. A photovoltaic cell is the core part of solar unmanned aerial vehicle (UAV) for providing energy. The solar panels are laid on the wings of solar unmanned aerial vehicles. This paper proposes a new MPPT controller for predicting the voltage to obtain the maximum power from the solar panel. In this paper, on the basis of studying the characteristics of photovoltaic cells, combined with the advantages of the mind evolutionary algorithm (MEA) and back propagation network, this article proposes a new type of BP network modeling structure based on the MEA, which is used for the modeling of photovoltaic cells. First, the MEA is constructed based on topology of a back propagation network. Then, this algorithm is used to obtain the optimal solutions, which are regarded as the initial weights and threshold values of the BP neural network. Finally, the simulation experiment is performed MATLAB software; comparing the different prediction results of the MEA optimization BP Network and genetic algorithm (GA) optimization BP neural network with the simple use of the BP neural network. The simulation results indicate that the optimization of the BP neural network by the MEA decreased the mean absolute percent error (MAPE) and root mean square error (RMSE) evaluation indicators by approximately 64% and 0.0524, respectively, compared with the back propagation neural network (BPNN). Overall, the MEA optimization BP neural network has high precision, small error, and a short training time.

show abstract

Section: Research On the Mppt Methods Based On The Bp Neural Network mentioning

confidence: 99%

“…Therefore, to some extent, it solves the slow convergence problem and other issues of the traditional algorithms with respect to local optimal. 45,46 The GA itself has some defects. (1) GA has no effective measures to ensure that a global search is conducted.…”

Section: Bp Neural Network With Mea Optimizedmentioning

confidence: 99%