Solar Irradiance Forecasting in Remote Microgrids Using Markov Switching Model

Shakya, Ayush; Michael, Semhar; Saunders, Christopher T.; Armstrong, Douglas; Pandey, Prakash; Chalise, Santosh; Tonkoski, Reinaldo

doi:10.1109/pesgm.2018.8585869

Cited by 9 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the downside of potentially much slower convergence (though cut sharing algorithms such as those from Infanger and Morton (1996) help circumvent this issue), often a stochastic process cannot be properly modeled by a standard ARMA process. This is the case, for example, with solar power which may be best described by regime switching models (as in Shakya et al 2016). In our case we aim to model wind power with a hidden Markov model, and thus need an algorithm that will allow for general forms of interstage dependence.…”

Section: Sddp With Hidden Markov Models Quadratic Regularization and ...mentioning

confidence: 99%

Risk Directed Importance Sampling in Stochastic Dual Dynamic Programming with Hidden Markov Models for Grid Level Energy Storage

Durante,

Nascimento,

Powell

2020

Preprint

View full text Add to dashboard Cite

Power systems that need to integrate renewables at a large scale must account for the high levels of uncertainty introduced by these power sources. This can be accomplished with a system of many distributed grid-level storage devices. However, developing a cost-effective and robust control policy in this setting is a challenge due to the high dimensionality of the resource state and the highly volatile stochastic processes involved. We first model the problem using a carefully calibrated power grid model and a specialized hidden Markov stochastic model for wind power which replicates crossing times. We then base our control policy on a variant of stochastic dual dynamic programming, an algorithm well suited for certain high dimensional control problems, that is modified to accommodate hidden Markov uncertainty in the stochastics. However, the algorithm may be impractical to use as it exhibits relatively slow convergence. To accelerate the algorithm, we apply both quadratic regularization and a risk-directed importance sampling technique for sampling the outcome space at each time step in the backward pass of the algorithm. We show that the resulting policies are more robust than those developed using classical SDDP modeling assumptions and algorithms.

show abstract

Section: Sddp With Hidden Markov Models Quadratic Regularization and ...mentioning

confidence: 99%

Risk Directed Importance Sampling in Stochastic Dual Dynamic Programming with Hidden Markov Models for Grid Level Energy Storage

Durante,

Nascimento,

Powell

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…This hybrid model showed superior prediction performance compared to the Beta distribution model. In [11], a method to forecast one-dayahead solar irradiance based on the Markov switching model is presented for remote microgrids. Essentially, this approach utilizes available data for one-day-ahead solar irradiance forecasting to schedule energy resources in remote microgrids.…”

Section: Introductionmentioning

confidence: 99%

A deep attention-driven model to forecast solar irradiance

Dairi

Harrou

Sun

2021

2021 IEEE 19th International Conference on Industrial Informatics (INDIN)

View full text Add to dashboard Cite

Accurately forecasting solar irradiance is indispensable in optimally managing and designing photovoltaic systems. It enables the efficient integration of photovoltaic systems in the smart grid. This paper introduces an innovative deep attentiondriven model for solar irradiance forecasting. Notably, an extended version of the variational autoencoder (VAE) is introduced by amalgamating the desirable characteristics of the bidirectional LSTM (BiLSTM) and attention mechanism with the VAE model. Specifically, the introduced approach enables the conventional VAE's ability to model temporal dependencies by incorporating BiLSTM at the VAE's encoder side to better extract and learn temporal dependencies embed on the solar irradiance concentration measurements. In addition, the self-attention mechanism is embedded in the VAE's encoder side following the BiLSTM to highlight pertinent features. The performance of the proposed model is evaluated through comparisons with the recurrent neural network (RNN), gated recurrent unit (GRU), LSTM, and BiLSTM. Measurements of solar irradiance in the US and Turkey are used to evaluate the investigated models. Results confirm the superior performance of the proposed model for solar irradiance forecasting over the other models (i.e., RNN, GRU, LSTM, and BiLSTM).

show abstract

“…In the same class of statistical methods there exist other works that capture variability in solar PV power [16]- [18] and black box models like using artificial neural networks (ANN) [19]- [21] and support vector machines (SVM) [22] based pattern matching techniques to predict solar power when class labels are known. Additionally, there are also methods based on the Markovianity assumption of solar power such as [23]- [25] in order to forecast solar PV power.…”

Section: Introductionmentioning

confidence: 99%

A Model for Joint Probabilistic Forecast of Solar Photovoltaic Power and Outdoor Temperature

Ramakrishna

Scaglione

Vittal

et al. 2019

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

In this paper, a stochastic model with regime switching is developed for solar photo-voltaic (PV) power in order to provide short-term probabilistic forecasts. The proposed model for solar PV power is physics inspired and explicitly incorporates the stochasticity due to clouds using different parameters addressing the attenuation in power. Based on the statistical behavior of parameters, a simple regime-switching process between the three classes of sunny, overcast and partly cloudy is proposed. Then, probabilistic forecasts of solar PV power are obtained by identifying the present regime using PV power measurements and assuming persistence in this regime. To illustrate the technique developed, a set of solar PV power data from a single rooftop installation in California is analyzed and the effectiveness of the model in fitting the data and in providing short-term point and probabilistic forecasts is verified. The proposed forecast method outperforms a variety of reference models that produce point and probabilistic forecasts and therefore portrays the merits of employing the proposed approach.

show abstract

Solar Irradiance Forecasting in Remote Microgrids Using Markov Switching Model

Cited by 9 publications

References 0 publications

Risk Directed Importance Sampling in Stochastic Dual Dynamic Programming with Hidden Markov Models for Grid Level Energy Storage

Risk Directed Importance Sampling in Stochastic Dual Dynamic Programming with Hidden Markov Models for Grid Level Energy Storage

A deep attention-driven model to forecast solar irradiance

A Model for Joint Probabilistic Forecast of Solar Photovoltaic Power and Outdoor Temperature

Contact Info

Product

Resources

About