International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 12 | Dec 2024
p-ISSN: 2395-0072
www.irjet.net
Supervisory Controller Design in Smart Grid Using Compressed Sensing Sepideh Alaei Varnousfaderani 1Student, Dept of Computer Engineering, Azad University, Isfahan, Iran
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Smart grids have emerged as a promising
applied to electrical signals [3]-[5], including simulated data collected from Phasor Measurement Units (PMUs) [3], though not specifically within a smart grid model. Lossless compression methods, such as those defined by IEEE 1159 [6], achieve exact signal reconstruction but offer limited compression efficiency. The Discrete Cosine Transform (DCT) has also been utilized [7], but its effectiveness is limited to signals with sinusoidal properties.
solution to meet increasing electricity demand and reduce environmental pollution. However, the high volume of data transmission and storage poses significant challenges, including delays, interference, increased power consumption, and storage limitations. This paper investigates data compression as a solution, focusing on Compressed Sensing due to its efficiency and low computational complexity. A smart grid model comprising wind turbines, solar panels, and battery banks connected to a DC bus is analyzed, with local controllers for generators and a supervisory controller for power allocation based on load demand, environmental conditions, and generator constraints. Data acquisition and compression units are incorporated to optimize data handling. Simulation results demonstrate that Compressed Sensing significantly reduces data volume while preserving signal accuracy.
Recently, Compressed Sensing (CS) has emerged as a promising technique for simultaneous signal sampling and compression [5]-[6]. CS exploits the sparsity of signals in one domain to achieve high-precision reconstruction from a reduced number of measurements in another domain [7]. It has been successfully implemented in applications such as large-scale wireless sensor networks [8]-[9], biological signals like ECG and EEG [10]-[11], and video rate control [12], demonstrating reduced communication costs and increased network capacity. Despite its success in these areas, CS has not been widely applied to electrical signals in smart grids. Given the sparsity of electrical signals in the DCT domain, CS presents a compelling method for addressing data volume challenges in smart grids.
Key Words: Smart Grid, Data Compression, Compressed Sensing, Renewable Energy, Power Allocation.
1.INTRODUCTION Smart grids have garnered significant global attention due to their numerous advantages, such as providing efficient and reliable power delivery systems [1], integrating renewable and alternative energy sources through automated control [2], and enabling real-time monitoring with offline analysis capabilities [3]. As a result, substantial research has been conducted to expand smart grid capabilities and address its challenges.
This paper introduces a smart grid model comprising wind turbines, solar panels, and battery banks connected to a DC bus, managed through a two-level control architecture. The first level employs local sliding mode controllers for individual generators and a control algorithm for battery banks. The second level features a supervisory controller to allocate power references based on demand, environmental conditions, electricity prices, and generator constraints, along with a monitoring unit for system oversight.
One major challenge in smart grids is the high volume of transmitted and stored data due to the extensive data exchange required among various control units and computational systems [4]. High data volume can lead to transmission issues such as delays, interference, excessive power consumption, increased network traffic, and demands for additional bandwidth. In terms of storage, limited space presents significant constraints. Data compression is a practical and efficient solution to address these issues. It offers three key benefits [4]: significant reduction of data volume, preservation of essential information, and high-precision reconstruction of data at the receiver.
High data volumes transmitted to the supervisory and monitoring units necessitate the introduction of two additional components: the Data Acquisition and Compression Unit (DACU) and the Data Analysis Unit (DAU). These units enable the application of compression algorithms to transmitted and stored data. Simulation results demonstrate that Compressed Sensing effectively reduces data volume while maintaining high-precision signal reconstruction, ensuring optimal performance of the supervisory controller and overall system efficiency.
Various data compression techniques have been explored for smart grids. Wavelet transform has been
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 727