Service Oriented Middleware for Smart Grid Resources Optimization

المشروع

تفاصيل المشروع

Description

We propose a new Service Oriented Architecture (SOA) for Smart Grid Resource Optimization (SGROM) based three different decision models: analytical, heuristic approach, and multi-criteria approach which allows Smart Grid Constituencies (SGC) such as Power Generators (PG), Power Transporters (PT), Power Distributors (PD), and Power Consumers (PC) to optimize their pay-offs using the smart grid (see figure 1). The proposed resource optimization management methods aim to support power consumers with sustainable energy from power distributors, transparently at the best price on a real time basis (variable pricing). It will also decide automatically for the power distributors the best power generators and connecting transporter lines based on the current power generation and transportation costs, the demand of the distributor, and the maximum power supply from the generator. Both SGROM smart engines analytical, heuristic and multi-criteria decision making aim at optimizing the profits and resource utilization over the whole smart grid. A comparative evaluation study includes simulation experiments of our new technique, as compared to the traditional grid system. Another comparative study will evaluate the three decision model approaches in terms of problem size, computational speed, and quality of the solution.

Layman's description

We propose a new Service Oriented Architecture (SOA) for Smart Grid Resource Optimization (SGROM) based three different decision models: analytical, heuristic approach, and multi-criteria approach which allows Smart Grid Constituencies (SGC) such as Power Generators (PG), Power Transporters (PT), Power Distributors (PD), and Power Consumers (PC) to optimize their pay-offs using the smart grid (see figure 1). The proposed resource optimization management methods aim to support power consumers with sustainable energy from power distributors, transparently at the best price on a real time basis (variable pricing). It will also decide automatically for the power distributors the best power generators and connecting transporter lines based on the current power generation and transportation costs, the demand of the distributor, and the maximum power supply from the generator. Both SGROM smart engines analytical, heuristic and multi-criteria decision making aim at optimizing the profits and resource utilization over the whole smart grid. A comparative evaluation study includes simulation experiments of our new technique, as compared to the traditional grid system. Another comparative study will evaluate the three decision model approaches in terms of problem size, computational speed, and quality of the solution.

Key findings

Several solutions have been proposed to manage the grid resources as part of the demand side management process. A comprehensive review of the existing demand side resource management solutions is presented in [13]. Most of the proposed systems restrict the resource optimization process to only two components of the grid system and do not include all the four SGCs described above. Some of the proposed solutions focus on developing scheduling algorithms to maximize the efficiency of the two components (generation, distribution) of the grid. Salinas et al. [6], proposed a dynamic energy management model with distributed energy resources. This model focuses on optimizing the resources between a PD and a PC with uncertain consumer load demand, distributed renewable energy resources and the energy storage devices owned by the consumer. The authors developed a real time pricing model to minimize the long-term average total cost to support consumer?s load demand. An energy scheduling model using a multi-objective function was presented in [14] with as objectives the minimization of the operation cost alongside the maximization of the minimum reserve in a day-ahead. The scheduling model focuses on different power generation resources including conventional power plants and renewable energy resources. Similarly, [15] presented a dynamic power management strategy for micro smart grid systems based on bidirectional power flows. The dynamic model focuses on residential applications where consumers are allowed to feedback the grid from their PV-Wind systems. Due to the heterogeneous nature of the hardware devices used in the electrical grid system, [7] proposed a software layer that abstracts this hardware heterogeneity and provides homogenous-looking appearance to the application layer of the smart grid. This middleware focuses on the data exchange between the smart devices and the smart applications to simplify the integration process and avoid dealing with data of different formats. Smart grid systems involve two way communications between the consumer and the grid through an advanced metering and monitoring system. The quality and reliability of the data collected is a key factor for optimizing the operations of the smart grid. Therefore, data mining and analytics tools are essential for the effective management and utilization of the available data [16]. An efficient utilization of the huge amount of data produced requires more data storage space with more sophisticated computational resources, which implies real need for Big Data analytics and high performance computing techniques. A comprehensive review of the state-of-the-art for the exploitation of big data tools for dynamic energy management in smart grid platforms is presented in [17].
عنوان قصيرNowadays, the general trend worldwide is towards upgrading existing power grids to Smart Grids. The key motivations behind modernizing the power grid are essentially the accommodation of future demand growth, the enhancement of efficiency, reliability, a
اختصارTTotP
الحالةلم يبدأ

بصمة

استكشف موضوعات البحث التي تناولها هذا المشروع. يتم إنشاء هذه الملصقات بناءً على الجوائز/المنح الأساسية. فهما يشكلان معًا بصمة فريدة.