Moving Regression Filtering With Battery State of Charge Feedback Control for Solar PV Firming and Ramp Rate Curtailment

Abstract

The surge in energy demand needs to be met with environmentally pleasant resources to reduce the production of greenhouse gases. Solar Photovoltaic (PV) power is a widespread choice as it is accessible in plenty and is comparatively inexpensive. However, the large-scale penetration of intermittent PV power causes multiple variabilities in the grid such as frequency issues and voltage deviations. To counteract these instabilities, Battery Energy Storage System (BESS) is integrated into the grid as it reduces the PV fluctuations and promotes optimal operation. Nevertheless, storage systems are expensive, and thus smoothing filters are coupled with the BESS for cost reduction and power smoothing. Formerly, traditional filters such as Low Pass Filters (LPF), Moving Average (MA), and Moving Median (MM) filters have been proposed for power smoothing. However, these filters have inadequate power tracking capabilities particularly with longer window sizes (W.S) and time constants, which subsequently depreciates the storage system performance. To compensate for the delayed power tracking, larger energy storage systems are required which in turn adds to the overall operational costs. This paper proposes the Moving Regression (MR) filter combined with state of charge (SoC) feedback control for solar PV variability reduction, reduced time delay, decreased battery charging/discharging power, and ramp rate. Simulation results attest that the MR filter achieves better solar power smoothing without increasing the BESS capacity. Also, the performance of the MR filter is less affected with the increase in window sizes. Additionally, the execution of the introduced smoothing filter was found to be superior when assessed against the LPF, MA, MM, Savitsky-Golay (SG), and the Gaussian filter. In comparison to the SG (W.S = 53), the MR (W.S = 45) filter reduces the battery charging/discharging power by approximately 30.48% and as opposed to the LPF (T.C = 48), the peak SoC is reduced by 19.1%.INDEX TERMS Battery energy storage system, low pass filter, machine learning, moving average filter, moving regression filter, renewable energy, solar power smoothing.

Solar Power Generation System With Power Smoothing Function

“…The output power from a solar power generation system (SPGS) changes significantly due to environmental factors [1][2][3][4][5][6][7][8][9][10][11][12]. These environmental factors change with the weather and seasons and cannot be controlled.…”

“…To suppress upward and downward fluctuations of the SPGS, the rapid power regulation technology is required to temporarily store and release power to stabilize the power output from the SPGS. Since battery set has the advantages of small size, quick absorption and release of electrical energy and flexible operation, it has considerable potential as a power regulation device for the SPGS [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. In general, the control concept for smoothing the output power of SPGS is that the battery energy storage system supplies the difference between the average value and the instantaneous value for the output power of SPGS.…”

“…In general, the control concept for smoothing the output power of SPGS is that the battery energy storage system supplies the difference between the average value and the instantaneous value for the output power of SPGS. The average value for the output power of SPGS can be calculated by low-pass filters [3][4][5][6], moving average filters [3], [4], [7], Saviztky-Golay filtering [8] and moving regression filter [9]. Since the instantaneous value for the output power of SPGS is rare equal to its average value, the charging/discharging time of the battery set is long.…”

“…The output from a solar cell array is DC power and the battery set stores power in DC form, so a power conversion interface is needed for integrating solar cell array or battery set into the power grid for DC-AC power conversion . The configuration of SPGS and battery energy storage system can be divided into AC coupling [9][10][11][12] and DC coupling [4], [8], [13][14][15][16]. For the AC coupling configuration, the SPGS and the battery energy storage system (BESS) are respectively connected to the grid.…”

Comprehensive Survey of Various Energy Storage Technology Used in Hybrid Energy

“…In recent years, renewable energy sources have increased to a high level, thus challenging the generation-demand balance [6]. In addition, renewable energy generation has been increasing.…”