Elsevier

Journal of Cleaner Production

Volume 87, 15 January 2015, Pages 216-226
Journal of Cleaner Production

Environmental/economic scheduling of a micro-grid with renewable energy resources

https://doi.org/10.1016/j.jclepro.2014.09.088Get rights and content

Highlights

  • Proposing an environmental/economic model for optimal energy management.

  • Considering renewable energy resources.

  • Presenting a multi-objective mathematical programming optimization.

  • Reporting superior solutions in the case of cost, emission and execution time.

Abstract

Nowadays, modern power grids are moving towards using alternative energy sources, as the fossil fuel reserves are exhausting and international societies worry about the environmental effects by conventional electric power generation systems. In this regard, this paper implements the scheduling of energy sources in a micro-grid including different power sources like photo voltaic, fuel cell, battery units, wind turbine and micro turbine. Besides, this paper proposes a multi-objective framework for the optimal scheduling of a micro-grid in order to concurrently minimize the total operation cost and minimize the emission caused by generating units. For this end, lexicographic optimization and hybrid augmented-weighted epsilon-constraint method have been employed to solve the multi-objective optimization problem and generate the Pareto optimal solutions. The decision making process has been performed using a fuzzy technique. Afterwards, the effectiveness and the efficiency of the proposed method are verified using two case studies. Besides, the results of the proposed method are compared to those obtained from other methods recently employed. Finally, it is indicated that incorporating the lexicographic optimization and hybrid augmented epsilon-constraint technique lead to superior solutions in the case of operation cost, emission and execution time.

Introduction

Distributed Generation (DG) is defined as generating units in small-scale, installed and operate next to the consumers. From different technologies of DG units, the renewable ones are widely used, as they are considered green, sustainable and free energy sources. On the other hand, excessive employment of DGs may cause severe challenges to the electric grid in the case of secure and efficient operation of the power system (Görbe et al., 2012, Lidula and Rajapakse, 2011, Jiayi et al., 2008). In this regard, Micro-Grid (MG) is a concept having the capability to provide the required infrastructures in order to integrate small-scale DGs into the large electric power system, while it satisfies the future needs of power grids (Cardenas et al., 2014, Rozali et al., 2014, Battaglini et al., 2009).

In recent years, a lot of researches have been done focusing on the operation of MGs. So far, many optimization techniques have been used to solve the problem of optimal operation scheduling considering different loading conditions and different objectives. One of the common problems encountered by some previous optimization algorithms is to determine the most economical units to dispatch. Many optimization techniques have been reported in literature to consider cost or profit to find the overall power dispatch of MGs. The optimal energy and spinning reserve bids have been modeled using bi-level bidding by Moghimi et al. (2013), while Dali et al. (2010) and Hawkes and Leach (2007) have proposed an economic framework to meet the electricity demand and operational constraints. Zhang et al. (2013) have taken into consideration the distributed economic dispatch for a MG with high penetration of Renewable Energy Sources (RESs) and demand-side management which operates in grid-connected state, while the objective function is minimization of the net cost of MG, comprising DG and distributed storage costs, utility of dispatchable loads and the worst-case transaction cost caused by the uncertainty of RESs. The optimal operation scheduling of renewable-powered MGs has been solved by Zhao et al., 2014a, Zhao et al., 2014b to determine the Unit Commitment (UC) and the associated dispatch with the least cost while satisfying demand and requirements of system operation. It should be noted that the aforementioned papers have used single-objective framework disregarding the issues of greenhouse gas emission. Despite the performed research works in the area, more comprehensive MG operation models are still needed. It is noted that traditional pure economic scheduling no longer meets the requirements of optimal operation of MGs after passage of the Clean Air Act Amendments in 1990 and it is required to consider the issues relating to greenhouse gas emission (Norouzi et al., 2014a, Ahmadi et al., 2012, Le et al., 1995).

Significant researches have been conducted on operation of MGs, while they have used Meta heuristic optimization methods to solve multi-objective environmental/economic operation of MGs. Due to the population-based search capability, simplicity, and convergence speed, Meta heuristic optimization methods are widely used for solving multi-objective optimization problems like environmental/economic operation of MGs (Moghaddam et al. (2011). Basak et al. (2012) have scrutinized a techno-economic factor for Distributed Energy Resources (DERs) on the basis of the impact of their generation on the network losses, while making an attempt to decrease the operation cost and emission generation. Moghaddam et al. (2012) have used a Fuzzy Self-Adaptive Particle Swarm Optimization (FSAPSO) algorithm for economic/emission dispatch of the generating units in a typical MG. The optimal operation scheduling of a typical MG with RESs along with a back-up MT/FC/battery hybrid power source has been done using an Adaptive Modified Particle Swarm Optimization (AMPSO) algorithm by Moghaddam et al. (2011), taking into account the cost and emission as two objective functions. Aghaei and Alizadeh (2013) have presented an optimal operation of a MG which is based on Combined Heat and Power (CHP), while the two objectives considered are cost and emission. This MG includes the energy storage system, three types of thermal power generating units and demand response programs. A multi-objective, intelligent energy management framework has been presented by Chaouachi et al. (2013) to minimize the operation cost and the emission of a MG considering its pre-operational variables as future availability of renewable energies and load demand. Zhao et al., 2014a, Zhao et al., 2014b have used an approach based on Genetic Algorithm (GA) to deal with the sizing optimization problem for MGs operating in stand-alone mode with multiple objectives consisting of life-cycle cost minimization, maximization of RESs penetration and minimization of emission. A linear programming technique has been used by Quiggin et al. (2012) to model a MG system comprising RESs, energy storage and demand response programs while it is found that MGs with contemporary technologies are able to remarkably reduce the CO2 emissions. An energy management model has been presented by Chen et al. (2013) to determine the optimal operating strategies associated with the maximum profit for a MG system in Taiwan. It is found that using an efficient power generation MG system would decrease the greenhouse gas emissions. The problem of economic/environmental dispatch of DG-based MGs has been solved using a chaotic quantum GA by Liao (2012). The problem of optimal size, design and operation of a hybrid, renewable energy based MG has been determined by Hafez and Bhattacharya (2012.), in order to minimize the life cycle cost taking into consideration the environmental concerns. The performance of a MG has been assessed by Zhang et al. (2014), using a performance metric with respect to the electricity price, emission and service quality, while a weighting factor is assigned to each one. With the weighting factors set by Zhang et al. (2014), performance metric was further applied to MGs operating in stand-alone, grid-tied and networked modes. Alagoz et al. (2012) have investigated the tree-like user-mode network architecture, which provides flexible, observable, and controllable network architecture for reliable and efficient energy delivery under uncertain conditions. A comprehensive model has been presented by Hemmati et al. (2014) for MG operating in stand-alone mode, while a multi-cross learning-based chaotic differential evolution algorithm has been used to solve the economic/environmental optimization problem. The optimal operation of WTs and other DERs operating in an interconnected MG has been presented by Motevasel and Seifi (2014) through an expert energy management system. The main aim beyond using the presented approach is to determine the optimal set points of DERs and storage devices in order to concurrently minimize the total operation cost and the net greenhouse gas emission. Motevasel et al. (2013) have utilized an intelligent energy management system for optimal operation of a MG that was based on CHP generation over a 24-hour horizon to simultaneously minimize the total operation cost and the net greenhouse gas emission. A stochastic environmental/economic multi-objective framework has been proposed by Niknam et al. (2012) using teaching-learning based optimization to obtain the Pareto optimal solutions. Mohamed and Koivo (2011) have proposed a generalized formulation for a MG to determine the optimal operating strategy and cost minimization scheme, while minimizing the greenhouse gas emissions using a modified game theory. It should be noted that the aforementioned papers have used the weighted sum method to convert multi-objective problems into single objective problem. Weighted sum method is the method commonly used for economic/environmental management problems compared with other previously established optimization approaches (Moghaddam et al., 2012, Moghaddam et al., 2011). However, a well-organized method to deal with multi-objective problems is epsilon-constraint technique having a main objective function selected among all objective functions. This technique has several significant merits over the conventional weighted sum method that merges the objective functions of the multi-objective problem into one objective function using weighted sum. The main advantages of the epsilon-constraint technique can be summarized as follows (Norouzi et al., 2014a, Mavrotas, 2009):

  • i.

    Using the weighted sum method, only efficient extreme solutions can be generated, while the epsilon-constraint technique has the capability to generate non-extreme efficient solutions.

  • ii.

    In the case of multi-objective problems, unlike the weighted sum method, the epsilon-constraint technique is able to generate unsupported efficient solutions.

  • iii.

    The weighted sum method requires the scaling of the objective functions, while it is not needed in the epsilon-constraint method.

Therefore, the epsilon-constraint method seems to be a good choice to solve multi-objective optimization problems.

The problem of short-term economic/environmental scheduling of a MG has been solved in this paper. The previously cited papers used Meta heuristic methods to solve this problem. It is worth to mention that the present paper employs a Multi-objective Mathematical programming (MMP) approach for this end. Moreover, to the best of authors' knowledge, there is no published paper proposing such framework using lexicographic optimization and hybrid augmented-weighted epsilon-constraint technique. It is noted that the results obtained are superior by others showing the effectiveness of the proposed method.

The main contributions of this paper can be listed as below:

  • 1.

    Proposing a multi-objective framework for short-term scheduling of a MG taking into consideration cost and emission as two objective functions.

  • 2.

    Incorporating the lexicographic optimization and hybrid augmented-weighted epsilon-constraint technique to simultaneously minimize cost and emission and generate Pareto optimal solutions and to determine the best compromise solution employing a fuzzy satisfying method.

  • 3.

    Exceptional solutions obtained from the presented method compared to recently published papers in the case of cost, emission and solution time.

The remainder of this paper is organized as below:

The problem formulation is proposed in Section 2 and Section 3 includes the description on lexicographic optimization and hybrid augmented epsilon-constraint method. Section 4 presents the simulation results with detailed discussion. Finally, Section 5 gives some relevant conclusions.

Section snippets

Mixed integer nonlinear programming (MINLP) formulation for MG

The two objective functions of the presented multi-objective framework for MG can be stated as follows:Multiobjectivefunctions={F1CostMinimizationF2EmissionMinimizationwhere, F1 and F2 denote the objective functions presented in details in the following.

Multi-objective mathematical programming (MMP)

Unlike the single optimization, multi-objective optimization problems deal with more than one objective function that are competing and conflicting. However, solving such problems result in a set of non-dominated solutions instead of a single optimal solution (Ahmadi et al., 2012). Generally, the following form represents a multi-objective optimization problem:minF(x)=(F1(x),,Fm(x))Ts.t:xX|where m denotes the total number of objective functions, x indicates a n-dimensional vector of decision

Case study and simulation results

Two different cases are used in this paper to implement the proposed method in order to show the efficiency and the effectiveness of the lexicographic optimization and augmented-weighted epsilon-constraint method. The system employed to solve the presented problem is a laptop computer with 2.4 GHz Pentium IV CPU and 3 GB RAM while SBB solver under GAMS (Brooke et al., 2012), has been used. The next section presents the results obtained through solving the proposed problem with two case studies.

Conclusion

This paper proposed a MINLP model for optimal scheduling of a MG over a 24-hour horizon. The optimal scheduling of a MG has been modeled in the form of multi-objective framework. The presented multi-objective problem has been solved by an analytic method using lexicographic optimization and hybrid augmented-weighted epsilon-constraint technique. The proposed method requires less solution time while results in better solutions in comparison with other methods from both economic and environmental

References (41)

  • A.D. Hawkes et al.

    Cost-effective operating strategy for residential micro combined heat and power

    Energy

    (2007)
  • M. Hemmati et al.

    System modeling and optimization for islanded micro-grid using multi-cross learning-based chaotic differential evolution algorithm

    Int. J. Electr. Power Energy Syst.

    (2014)
  • H. Jiayi et al.

    A review on distributed energy resources and microgrid

    J. Renew. Sust. Energy Rev.

    (2008)
  • G.C. Liao

    Solve environmental economic dispatch of smart microgrid containing distributed generation system – using chaotic quantum genetic algorithm

    Int. J. Elect. Power Energy Syst.

    (2012)
  • N.W.A. Lidula et al.

    Microgrids research: a review of experimental microgrids and test systems

    J. Renew. Sust. Energy Rev.

    (2011)
  • H. Mavalizadeh et al.

    Hybrid expansion planning considering security and emission by augmented epsilon-constraint method

    Int. J. Elect. Power Energy Syst.

    (2014)
  • G. Mavrotas

    Effective implementation of the ε-constraint method in multiobjective mathematical programming problems modified augmented

    Appl. Math. Comp.

    (2009)
  • A.A. Moghaddam et al.

    Multi-operation management of a typical micro-grids using particle swarm optimization: a comparative study

    Renew. Sust. Energy Rev.

    (2012)
  • A.A. Moghaddam et al.

    Multi-objective operation management of a renewable MG (micro-grid) with back-up micro-turbine/fuel cell/battery hybrid power source

    Energy

    (2011)
  • H. Moghimi et al.

    Stochastic techno-economic operation of power systems in the presence of distributed energy resources

    Elect. Power Energy Syst.

    (2013)
  • Cited by (0)

    View full text