Photovoltaic energy network

Patent metadata

Field	Value
Publication number	US-12107539-B2
Application number	US-202418606697-A
Country	US
Kind code	B2
Filing date	Mar 15, 2024
Priority date	Oct 29, 2021
Publication date	Oct 1, 2024
Grant date	Oct 1, 2024

Abstract

A system, method, and solar photovoltaic (PV) network for solar PV variability reduction with reduced time delays and battery storage optimization are described. The system includes a Moving Regression (MR) filter; a State of Charge (SoC) feedback control; and a Battery Energy Storage System (BESS). The MR filter, SoC feedback control and BESS are configured to provide smoothing of solar PV variabilities. The MR filter is a non-parametric smoother that utilizes a machine learning concept of linear regression to smooth out solar PV variations at every time step.

First claim

The invention claimed is:

1. A photovoltaic energy network, comprising:
a plurality of wind turbines;
a photovoltaic array comprising a plurality of wired photovoltaic modules connected in series;
a Moving Regression (MR) filter;
a State of Charge (SoC) feedback control;
a Battery Energy Storage System (BESS); and
an electrical grid,
wherein the photovoltaic array receives solar light signals and generates an unsmoothed solar photovoltaic power output, wherein the unsmoothed solar photovoltaic power output is electrically coupled to the MR filter and the SoC feedback control, and the photovoltaic array has a boost converter;
wherein each of the MR filter, the SoC feedback control and the BESS are electrically coupled to provide a combined smoothed solar photovoltaic power output, wherein the smoothed solar photovoltaic power output is electrically coupled to the electrical grid, and
wherein the MR filter is a non-parametric smoother that is configured to smooth an electrical input with machine learning linear regression over a plurality of time steps, and
the plurality of wind turbines, the photovoltaic array, the MR filter, the SoC feedback control, the BESS, and the electrical grid are electrically connected,
wherein the MR filter is a non-parametric smoother that utilizes a machine learning concept of linear regression to smooth out solar photovoltaic variations at every time step, and
wherein based on a first window size of the MR filter, k neighboring points of a target value are used as training values for a linear regression algorithm.

2. The network of claim 1 , wherein the MR filter, the SoC feedback control and the BESS are configured to provide smoothing of solar photovoltaic variabilities.

3. The network of claim 1 , wherein the MR filter and the SoC feedback control receive the unsmoothed solar photovoltaic power output from the photovoltaic array.

4. The network of claim 1 , wherein the MR filter and is configured to reduce a power lag and a ramp rate for the photovoltaic energy network.

5. The network of claim 1 , wherein the MR filter and SoC feedback control are each configured to reduce a power lag and a ramp rate for the photovoltaic energy network.

6. The network of claim 1 , wherein the MR filter and SoC feedback control are configured to control charging and discharging of the BESS.

7. The network of claim 1 , further wherein based on a second window size larger than the first window size additional neighboring points are used for training the linear regression algorithm in the MR filter so that the MR filter provides a higher accuracy of a predicted smoothed value of solar photovoltaic power output in comparison to a low pass filter, a moving average filter, a double moving average filter, a moving median filter, a Savitsky-Golay filter, or a gaussian filter.

8. The network of claim 1 , wherein the renewable energy network has a higher degree of photovoltaic power smoothing and power tracking and decreased battery charging and discharging in comparison to a low pass filter, a moving average filter, a double moving average filter, a moving median filter, a Savitsky-Golay filter, or a gaussian filter.

Assignees

• Univ King Fahd Pet & Minerals

Inventors

• Khalid Muhammad
• Syed Miswar Akhtar

Classifications

• H02J2101/24

  Photovoltaics · CPC title

• H02J2101/28

  Wind energy · CPC title

• H02J2103/35

  Grid-level management of power transmission or distribution systems, e.g. load flow analysis or active network management · CPC title

• H02J7/933

  the cycle being controlled or terminated in response to electric parameters · CPC title

• H02S99/00

  Subject matter not provided for in other groups of this subclass · CPC title