A review of data mining and solar power prediction

Yeşilbudak, Mehmet; Çolak, Medine; Bayındır, Ramazan

doi:10.1109/icrera.2016.7884507

Cited by 34 publications

(25 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…When data are collected using data mining techniques, the producible amount of energy can be predicted with high accuracy. 15 However, the response is sensitive to the surrounding environment, emphasizing the importance of data management and analysis to operate the energy harvesting system efficiently. Thus, this study considers the necessary environmental configuration for data acquisition and analysis for the system design process.…”

Section: Related Workmentioning

confidence: 99%

Adaptive operating mode management model for efficient energy harvesting systems

Choi

Koo

Park

2020

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

Energy harvesting technology is becoming popular concerning efficient use of Internet of Things devices, which collect energy present in nature and use it to power themselves. Although the technology is eco-friendly, it is dependent on the vagaries of the surrounding environment; the amount of energy produced is sensitive to the weather and terrain, and intermittent power threatens the system’s stability. Thus, it is essential to collect data that can determine the circumstances of the surrounding environment. Furthermore, these systems should be designed efficiently for continuous energy harvesting. This efficiency can vary depending on the system’s configuration. Core voltage levels and frequencies typically influence efficiency. To maximize system efficiency, power management with an appropriate combination of controllable factors is necessary. We design an energy harvesting system for real-time data acquisition. We propose a methodology to guide the optimal operating power stage considering various adjustable factors for efficient operation. Also, we propose an adaptive operating power mode management model, which involves selecting the optimal operating power step and the transition to a low-power mode (LPM) during idle time. The proposed model was applied to an actual energy harvesting system to demonstrate its effectiveness and facilitated the operation of the harvesting system at low power.

show abstract

Section: Related Workmentioning

confidence: 99%

Adaptive operating mode management model for efficient energy harvesting systems

Choi

Koo

Park

2020

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

show abstract

“…Thus, the Updating judgment is mainly applied to renew the BPNN to guarantee accurate predictions. Here, ∆ 2 is calculated by function Equations (9) and (17) in Sections 3 and 4. We apply this both for SDM and NDM.…”

Section: Framework Of the Aspf In Sdm And Ndmmentioning

confidence: 99%

“…Although the forecasting results by any one of the above methods are not satisfactory overall, the forecasting accuracy can be further improved by merging them together. Some existing works focus on the combination of artificial neural network and data mining for more accurate solar power [17]. K-means clustering with nonlinear auto-regressive neural networks are adopted to forecast solar irradiance in [18], and k-means with artificial neural networks are used to predict solar irradiance in [19].…”

Section: Introductionmentioning

confidence: 99%

Adaptive Solar Power Forecasting based on Machine Learning Methods

et al. 2018

View full text Add to dashboard Cite

Due to the existence of predicting errors in the power systems, such as solar power, wind power and load demand, the economic performance of power systems can be weakened accordingly. In this paper, we propose an adaptive solar power forecasting (ASPF) method for precise solar power forecasting, which captures the characteristics of forecasting errors and revises the predictions accordingly by combining data clustering, variable selection, and neural network. The proposed ASPF is thus quite general, and does not require any specific original forecasting method. We first propose the framework of ASPF, featuring the data identification and data updating. We then present the applied improved k-means clustering, the least angular regression algorithm, and BPNN, followed by the realization of ASPF, which is shown to improve as more data collected. Simulation results show the effectiveness of the proposed ASPF based on the trace-driven data.

show abstract

“…Data mining is used in load forecasting, clustering, and decision tree classification whereas information about load demand ahead of time helps the utilities and electricity suppliers in many ways . Also, data mining methods were used for renewable energy power prediction such as solar power generation . On the other hand, prediction techniques have been used to compute an initial guess using known power flow solutions near the target state using Lagrange polynomial approximations .…”

Section: Introductionmentioning

confidence: 99%

Fast quasi-static time-series analysis and reactive power control of unbalanced distribution systems

Selim

Abdel‐Akher

Aly

et al. 2018

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Summary The paper develops a method for solving time‐series power flow for unbalanced distribution power systems equipped with on‐load tap changer and photovoltaic (PV) power generation. The proposed PV control is developed using a fuzzy logic controller. When the reactive power of the PV inverter is not enough, tap‐up or tap‐down signals are sent to the on‐load tap changer to control the voltage. The Lagrange function is implemented to predict the initial values of voltage magnitudes and phase angles during the simulation process. These initial values are corrected using power flow routine that adopts the forward/backward analysis method. The IEEE‐123 node test feeder is used to assess the validity of the developed continuous power flow solution. The results are compared with the traditional quasi‐static time‐series solution without prediction scheme. The simulation results prove that the developed time‐series model has superior performance. The effects of PV penetration and voltage regulator on the method performance and tap setting are also investigated.

show abstract

A review of data mining and solar power prediction

Cited by 34 publications

References 38 publications

Adaptive operating mode management model for efficient energy harvesting systems

Adaptive operating mode management model for efficient energy harvesting systems

Adaptive Solar Power Forecasting based on Machine Learning Methods

Fast quasi-static time-series analysis and reactive power control of unbalanced distribution systems

Contact Info

Product

Resources

About