Optimized ESS Operation for Peak Shaving based on Probabilistic Load Prediction

Kodaira, Daisuke; Yu, Byeonggu; Jung, Won-Wook; Han, Sekyung

doi:10.1109/isgt-asia.2018.8467980

Cited by 6 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AlHakeem et al 66 also used the bootstrap confidence interval method for PV and wind-generated energy forecasting and developed a management scheme for energy storage. Kodaira et al 32 proposed an optimisation algorithm based on probabilistic load prediction for an ESS-PV system for peak shaving. Due to the shortcomings of empirical mathematical models that include their inability to capture non-linear patterns existing within the data, the fact that assumptions should be established before using those models, and the poor performance as the data size increased, most researchers are now leaned towards the use of ML for developing prediction models.…”

Section: Discussionmentioning

confidence: 99%

“…Such an approach is rather important because previous studies have primarily dealt with improving the performance of single systems separately (eg, ESS or PV) which is likely suboptimal. [26][27][28][29][30][31][32][33][34] The system working principle is made of two modules: a forecasting module and a scheduling module. The forecasting module uses real-time weather information and is composed of two deep neural networks (DNN), forecasting models for EHP energy consumption and PV energy generation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A novel deep learning‐based integrated photovoltaic, energy storage system and electric heat pump system: Optimising energy usage and costs

et al. 2021

View full text Add to dashboard Cite

The use of photovoltaic (PV) systems has drawn attention as a solution to reduce the dependence on fossil fuel for building energy needs. Moreover, incorporating energy storage systems (ESSs) in PV systems can optimise electric energy costs by increasing dependency on PV-generated energy during electric peak load times. However, current ESSs have limited capacities making it difficult to fully maximise PV-generated energy. We propose a novel integrated energy-efficient system for PV, ESS and electric heat pump (EHP) that maximises the usage of PV energy, optimises ESS usage and reduces EHP energy consumption costs. The components of the proposed integrated system are linked with a deep learning (DL)-based algorithm that forecasts PV energy generation and energy demand of the EHP. The proposed system schedules the charging/discharging time of ESSs depending on peak load times, the forecasted EHP electric demand, and PV-generated energy. The data used were collected for 10 months from a retail shop equipped with an EHP and ESS. We found that the developed DL-based forecasting models for PV and EHP are accurate and reliable (ie, R 2 above 0.95). Also, the results show that the proposed integrated energy-efficient PV-ESS-EHP system saves 12% of the total annual electric costs, which corresponds to 1 285 291 Won. The proposed system ensures an efficient method to maximise PV-generated energy resulting in reduced dependency on fossil fuels for building energy needs.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A novel deep learning‐based integrated photovoltaic, energy storage system and electric heat pump system: Optimising energy usage and costs

et al. 2021

View full text Add to dashboard Cite

show abstract

“…ESS is also beneficial from a power system point of view, since it contributes to stabilizing the power grid. Thus, there have been many studies to minimize total electricity cost using ESS [1][2][3][4][5][6][7][8].…”

Section: Introductionmentioning

confidence: 99%

“…Obtaining the probability, however, is challenging in real situations. The work in [8] generates hourly forecasted load by probabilistic load distribution to alleviate the uncertainty. However, the authors in [8] assume that the probabilistic load distribution follows the normal distribution, which may not necessarily hold in general.…”

Section: Introductionmentioning

confidence: 99%

“…The work in [8] generates hourly forecasted load by probabilistic load distribution to alleviate the uncertainty. However, the authors in [8] assume that the probabilistic load distribution follows the normal distribution, which may not necessarily hold in general. Scenario-based approach generates the scenarios of the uncertainty from historical measurements [12][13][14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Practical Operation Strategies for Energy Storage System under Uncertainty

Kim

Choi

et al. 2019

Energies

View full text Add to dashboard Cite

Recent advances in battery technologies have reduced the financial burden of using the energy storage system (ESS) for customers. Peak cut, one of the benefits of using ESS, can be achieved through proper charging/discharging scheduling of ESS. However, peak cut is sensitive to load-forecasting error, and even a small forecasting error may result in the failure of peak cut. In this paper, we propose a two-phase approach of day-ahead optimization and real-time control for minimizing the total cost that comes from time-of-use (TOU), peak load, and battery degradation. In day-ahead optimization, we propose to use an internalized pricing to manage peak load in addition to the cost from TOU. The proposed method can be implemented by using dynamic programming, which also has an advantage of accommodating the state-dependent battery degradation cost. Then in real-time control, we propose a concept of marginal power to alleviate the performance loss incurred from load-forecasting error and mimic the offline optimal battery scheduling by learning from load-forecasting error. By exploiting the marginal power, real-time ESS charging/discharging power gets close to the offline optimal battery scheduling. Case studies show that under load-forecasting uncertainty, the peak power using the proposed method is only 22.4% higher than the offline optimal peak power, while the day-ahead optimization has 76.8% higher peak power than the offline optimal power. In terms of profit, the proposed method achieves 77.0% of the offline optimal profit while the day-ahead method only earns 19.6% of the offline optimal profit, which shows the substantial improvement of the proposed method.

show abstract