Resource allocation for smart grid communication based on a multi-swarm artificial bee colony algorithm with cooperative learning☆
Introduction
Demand response is an effective method to provide frequency regulation to the power grid, which requires two-way communication between the utility company and the consumers in short time intervals (Kilkki et al., 2014). In smart grid, the two-way communication is implemented by a hierarchical network, including home area networks (HAN), neighborhood area networks (NAN), and wide area networks (WAN) (Bouhafs et al., 2012). In the hierarchical network, a data aggregate unit (DAU) is deployed to gather the metering data from the consumers and forward the control commands of the utility company. It is shown that the packets loss at the DAU causes the response errors, which further increase the cost of the utility company (Niyato and Wang, 2012, Niyato et al., 2011, Niyato et al., 2013, Zheng et al., 2011, Zheng et al., 2015).
In recent years, cooperative diversity has been demonstrated to implement spatial diversity and reduce the bit error rate for demand-side communication (Omar et al., 2016). The basic idea is that the relays help the DAU transmits data to the gateway. In Ahmed et al. (2013), a relay selection mechanism was proposed to improve the end-to-end packet delivery latency, throughput, and energy efficiency of the HAN and NAN with smart relays. In Ma et al. (2017b), the resource allocation problem of the demand-side cooperative relaying network was studied based on the bargaining models and solutions. Since relaying represents the resource sharing between the relays and the DAUs, the relay selection and the resource allocation should be dealt with simultaneously. As far as we know, this problem has not been studied. In this paper, we jointly consider the relay selection and the power allocation problem, i.e., which DAU the relay should help and how the relay should allocate its power among the DAUs. When there are multiple DAUs and multiple relays, the resource allocation problem will be more complicated due to the interactions among the DAUs.
This study formulates the resource allocation of the smart grid communication as a combinatorial optimization problem. The artificial bee colony (ABC) algorithm is effective for solving the complex optimization problem (Xiang and Zhou, 2015, Kishor et al., 2016). Recently, several approaches have been proposed to improve the searching ability of the ABC algorithm by modifying the neighborhood update strategy (Li et al., 2012, Zhong et al., 2017, Cui and Gu, 2015). However, the cooperation among different bees have not been studied in the research of ABC algorithm. Inspired by the multi-swarm particle swarm optimization (PSO) (Xu et al., 2015), a dynamic multi-swarm artificial bee colony (MS-ABC) algorithm with cooperative learning was developed in this work. Specifically, the penalty function method transforms the nonlinear programming problems into an unconstrained optimization problem, and the MS-ABC algorithm combines the dynamic multi-swarm optimizer with a new cooperative learning strategy. It is observed that the MS-ABC algorithm based on penalty function shows good performances under different parameters.
The rest of the paper is organized as follows. We give the problem formulation in Section 2 and establish the optimization model in Section 3. The MS-ABC algorithm is presented in Section 4, and simulation results are shown in Section 5. Finally, we draw conclusions in Section 6.
Section snippets
Demand-side cooperative communication network
We consider a cooperative communication network model as shown in Fig. 1, which consists of DAUs and Gateways assisted by relays. The system is frequency division multiplexing and each DAU is allocated an orthogonal channel, over which the DAU-to-Gateway and the DAU-to-relay communications take place. The relay relationship are defined as a matrix , where the element means that node preforms as a relay for DAU , and 0 otherwise. Based on the conclusion in Kadloor and
System model and solutions
In this section, we first establish the cost optimization model and then derive the optimal relay assignment and power allocation.
The problem is equivalent to selecting the optimal relay and power allocation such that the cost to utility company is minimized, and and should be subject to the individual power constraints in (5) and the total relay power constraints in (6). This is cast into the following optimization problem:
We
MS-ABC Algorithm
In this section, we use a hybrid ABC (Artificial Bee Colony) algorithm based on the penalty function to solve the combinatorial nonlinear programming problem. The ABC algorithm is a swarm intelligent optimization algorithm inspired by honey bee foraging. Artificial Bee Colony algorithm, originally proposed by Karaboga (Karaboga, 2005), is one of the latest swarm-based-heuristic approaches in the literature. The ABC algorithm can approach the optimal solution fast and is effective for optimizing
Simulation results
In the simulation, the total system bandwidth is set to be MHz, the arriving rates of the DAU are from 0 Mbit/s to 2 Mbit/s, the probability of correct transmission from the gateway to the consumers is , and the base price of the AGC service is $/MW. The path loss exponents is , the coefficient , the initial penalty factor , the maximum estimation error , and the estimated value . The transmission power of the DAU is from 3 W to 3.5 W, the
Conclusions
This paper considered the relay assignment and power allocation problem in a communication network with multiple DAUs and multiple relays. The relay assignment and power allocation was formulated as the nonlinear programming problem. We transformed the nonlinear programming problem to the unconstrained optimization problem by using the penalty function method, then utilized the improved ABC algorithm to obtain the sub-optimal solution. Simulation results illustrates that the cooperative
Acknowledgments
This research was supported in part by National Natural Science Foundation of China under Grants 61573303 and 61503324, in part by Natural Science Foundation of Hebei Province, China under Grants F2016203438 and E2017203284, and in part by Project Funded by China Postdoctoral Science Foundation under Grant 2016M601282.
References (25)
- et al.
An improved discrete artificial bee colony algorithm to minimize the makespan on hybrid flow shop problems
Neurocomputing
(2015) - et al.
A comparative study of artificial bee colony algorithm
Appl. Math. Comput.
(2009) - et al.
On the performance of artificial bee colony (ABC) algorithm
Appl. Soft Comput.
(2008) - et al.
Agent-based modeling and simulation of a smart grid: a case study of communication effects on frequency control
Eng. Appl. Artif. Intell.
(2014) - et al.
NSABC: non-dominated sorting based multi-objective artificial bee colony algorithm and its application in data clustering
Neurocomputing
(2016) - et al.
Development and investigation of efficient artificial bee colony algorithm for numerical function optimization
Appl. Soft Comput.
(2012) - et al.
A dynamic multi-colony artificial bee colony algorithm for multi-objective optimization
Appl. Soft Comput.
(2015) - et al.
Dynamic multi-swarm particle swarm optimizer with cooperative learning strategy
Appl. Soft Comput.
(2015) - et al.
An improved artificial bee colony algorithm with modified-neighborhood-based update operator and independent-inheriting-search strategy for global optimization
Eng. Appl. Artif. Intell.
(2017) - et al.
Smart grid cooperative communication with smart relay
J. Commun. Netw.
(2013)
Links to the future: communication requirements and challenges in the smart grid
IEEE Power Energy Mag.
Cited by (25)
Application of effective gravitational search algorithm with constraint priority and expert experience in optimal allocation problems of distribution network
2023, Engineering Applications of Artificial IntelligenceCitation Excerpt :One is to determine an appropriate constraint processing strategy to search qualified configuration scheme meeting all constraints of distribution network, and the other is to determine an effective technology that can achieve high quality DG/SC configuration schemes after finding qualified ones. The typical constraint processing strategy in OAPDN research is penalty function method (PFM), which determines qualified schemes by introducing multiple penalty coefficients (Sharma et al., 2020; Ma et al., 2019; Ji et al., 2017). However, any inappropriate penalty coefficient may result in OAPDN schemes failing to achieve zero constraint-violation and the defects of PFM are more obvious in larger scale distribution networks.
Resource scheduling methods for cloud computing environment: The role of meta-heuristics and artificial intelligence
2022, Engineering Applications of Artificial IntelligenceAn adaptively balanced grey wolf optimization algorithm for feature selection on high-dimensional classification
2022, Engineering Applications of Artificial IntelligenceCitation Excerpt :In the past decades, many SI-based metaheuristics have been proposed, such as the Particle Swarm Optimization (PSO) (Kennedy and Eberhart, 1995), the Ant Colony Optimization (ACO) (Dorigo et al., 2006), the Artificial Bee Colony (ABC) (Karaboga and Basturk, 2007), the Whale Optimization Algorithm (WOA) (Mirjalili and Lewis, 2016), the Dragonfly Algorithm (DA) (Hammouri et al., 2020), the Firefly algorithm (FA) (Kahya et al., 2019) and etc. By combining learning operators or adopting a special learning mechanism, SI-based algorithms are fast and reliable for searching for the optimal solution to single-objective optimization problems (Ma et al., 2019; Tam et al., 2018; Boghdady et al., 2022) and approximating the true Pareto optimal solutions to multi-objective problems (Abbaszadeh Shahri et al., 2022; Lin et al., 2019). These works have established that SI-based algorithms have the characteristics of self-organization, flexibility and robustness (Cui and Gao, 2012; Derrac et al., 2011).
A powerful meta-heuristic search algorithm for solving global optimization and real-world solar photovoltaic parameter estimation problems
2022, Engineering Applications of Artificial IntelligenceA literature survey on load frequency control considering renewable energy integration in power system: Recent trends and future prospects
2022, Journal of Energy StorageCitation Excerpt :Hence for tuning of other controllers, current progressive soft computing algorithms are developed. More research in this domain explained above has been accomplished by the use of different soft computing approaches like artificial BEE colony algorithm [63,64], Bacterial foraging Techniques [65,66], Bat-inspired algorithm [67,68], cuckoo search optimization techniques [69,70], Differential Evolution algorithm [71,72], Genetic algorithm-based LFC [73,74], PSO techniques [75,76], A Quasi-oppositional Harmony search algorithm(QOHS) [77,78], TLBO (Teaching-learning -based optimization algorithm) [79,80], The firefly techniques [81,82] to examine the problem or the complication of the non-linearity of the interconnected power system. The economic Load forecasting approach [83,84] is the most resist job in the multi-area interconnected power system, but its application can encourage the performance of the LFC noticeably.
Artificial intelligence in sustainable energy industry: Status Quo, challenges and opportunities
2021, Journal of Cleaner ProductionCitation Excerpt :The AI technology controls the power fluctuations of the smart grid. AI models used to optimize smart grids (Di Santo et al., 2018), various kinds of smart grid technologies (Supriya et al., 2015), smart load control (Raza and Khosravi, 2015a), demand response (Hui et al., 2020), smart grid-enabled IoT (Al-Turjman and Abujubbeh, 2019), management of district-level load, detection of electricity theft in the smart grid environment, monitoring and control of smart grid (Sanchez-Hidalgo and Cano, 2018), allocation of smart grid resource (Ma et al., 2019), cyberattacks prevention (Gavriluta et al., 2020), effective delivery of energy (Hui et al., 2020), and so on. Around 75 billion IoT devices are expected to be used worldwide by 2025 (Rhodes, 2020).
- ☆
No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to https://doi.org/10.1016/j.engappai.2018.12.002..