A novel approach for distributed generation expansion planning considering its added value compared with centralized generation expansion

Highlights

• A novel approach for modeling the DGEP is presented in the regulated power system.

• The generation capacity of a primary GEP is modified to obtain an optimal DGEP.

• The optimal capacity, location and technology of the DG units are determined.

• An integrated objective function considering the all HV and LV costs is proposed.

Abstract

Technological progress of distributed generation (DG) units, global approach to reduce the pollution emissions, and creating the opportunity for local investors to participate in the generation expansion investment, are some of the important reasons to involve the DG as an alternative option in the GEP. In this regard, expansion of the DG units compared to the expansion of centralized generation units should be evaluated economically and technically. Therefore, this paper proposes a novel approach for modeling the composite centralized and distributed generation expansion planning problem in the regulated power system. An integrated objective function is presented to involve the most important decision-making factors overall the power system. The purpose is to find the optimal combination of generation expansion created by power plants and DG units to minimize the total costs related to generation expansion, operation, maintenance, fuel, emission, distribution loss and expected energy not supplied, for the decision-making period. The proposed model reveals as a mixed integer nonlinear programming where, genetic algorithm is applied to find the optimal solution which contains the capacity, technology and operation strategy of non-stochastic DG units that will be optimally allocated into some distribution feeders. Results confirm the effectiveness and superiority of the proposed approach.

Introduction

Increasing energy demand due to population growth and industry development can cause inadequacy in the existing power system [1]. In this regard, serving the electricity demand with an acceptable level of security, reliability and quality for the coming years is a challenging issue. To alleviate this problem, power system expansion planning should be performed in three sectors of power system including generation, transmission and distribution while satisfying some economic and technical constraints. In recent decades, the expansion of large-scale centralized generations has been seriously challenged for its high costs, environmental impacts, and some technical problems such as network loss and security concerns. In generation expansion sector, the competition of the generation companies in the deregulated power generation markets, lead to serve the growing demand while in a regulated vertical power system, this issue is performed by a responsible entity. In fact, in most developing countries, especially Asian and African countries, the restructure and deregulation of the global power industry have not been completely implemented. Therefore, generation, transmission and distribution expansion are still sub-tasks of a PSEP process performed by a regulated power utility. Additionally, due to rapid development of DG technologies and high penetration of them into distribution networks, the conventional PSEP procedure can also be affected even in this traditional and regulated structure of power system. Hence, in this environment, accurate techno-economic modeling of the effects of DG units on PSEP and introducing a practical and optimal model for DGEP considering its added value in CGE is of paramount importance. The proposed method in this research is also from the viewpoint of a responsible entity in a regulated, vertical structure of power system.

Due to the high penetration of DG units into active distribution networks, they can be considered as a suitable alternative for large-scale concentrated generations [2]. This feature will reduce the transferred power from centralized power plants into the distribution network. Therefore, it is important to consider the effect of DG units in solving the problems of GEP, TEP and DSEP. In this regard, the existing methods for modeling the effects of DG units on the power system planning can be divided into three categories [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19]. The first category presents some models for obtaining the effective and economic integrations of DG units to be considered in the planning of distribution networks [3], [4], [5], [6], [7]. In [4], a genetic algorithm based integrated dynamic DSEP model is presented considering the uncertainties related to wind turbines, load, and price. However, other DG technologies are not considered in this study. In [5], a metaheuristic-based DSEP problem is presented concentrating the installation of renewable-based DGs on the related feeders based on the Weibull probability distribution function. However, the operation costs are not considered in this work. In [6], a multistage DSEP problem is proposed based on genetic algorithm considering the DG curtailment under stress condition, but the operation and maintenance cost of DG units are not included into the objective function. In [7], the integration of wind and solar based DGs is analyzed by introducing a multi-objective multistage DSEP formulation based on $\epsilon$-constraint method. However, this study just analysis the integration of intermittent DG units

The second category of the literatures models the integration of DG units on the procedure of TEP [8], [9], [10], [11], [12]. In [9], a multi-objective optimization approach for flexible TEP considering DG impacts is presented. The uncertainties related to generation expansion, network load, and market prices are considered in this work. However, optimal operation strategies of DG units are not determined in this study. In [10], co-planning of transmission networks and merchant distributed energy resources are proposed based on a robust optimization method. From the viewpoint of power system regulators and planners, a tri-level optimization problem is presented to construct the new transmission lines and determine the optimal location of merchant DERs without determining their operation strategy in different load steps. Impact of decision-making criteria such as investment, operation, and unserved energy costs in TEP considering large shares of DERs are analyzed in [11]. However, the customer’s interruption cost and the emission costs are not considered at objective function. In [12], a constructive heuristic method is proposed for use in the problem of composite GEP and TEP in the presence of DG units. Considering the fuel supply constraints, maintenance of generating units, reliability constraints of generation and transmission facilities, are some features of the introduced approach. But in the proposed method, optimal capacity, and suitable technology of DGs are not considered.

The aim of the conventional GEP is to implement the centralized generation expansion scheme by a techno-economic model to meet the predicted demand increase in a long-term period. Different factors such as primary energy resources cost, fuel type, emission, and reliability constraints, have made the GEP to be a complicated optimization problem mathematically [13]. Powerful software WASP-IV is usually employed to determine the required optimal generation capacity to supply the forecasted load with minimum reliability indices such as LOLE, LOLP and etc. [14]. However, due to the assumption of single-node load center in this software, it is unable to determine the optimal locations of new power plant and allocating the generation capacity among them. The third category of approaches presented in the literatures examines the effect of DG units on the GEP procedure [15], [16], [17], [18], [19]. In [15], a computationally effective approach for optimizing the multi-period GEP problem with high integration of DERs is proposed. The model has been formulated as a large-scale MILP. Although, the capacity of required DERs is determined, but the installation location of DERs are not evaluated. In [16], the impact of high penetration of intermittent DER on the generation capacity investments considering the required operating reserve is studied. This work revealed that sitting and sizing strategies for operating reserves can bring down the total investment costs and benefits of the DERs integration. In [17], a multi-objective GEP problem for long-term planning focusing on the integration of DERs and storage devices such as electric vehicles, is presented aimed to minimize the centralized energy generation and maximize the DERs generation without considering their suitable technologies. In [18], an effective model for solving GEP problem by integrating large-scale DERs is proposed. The main purpose of this work is to present a meta-model assisted evolutionary algorithm approach to capture the operational challenges arising when a high penetration of intermittent DERs is considered. A complete review of different models presented for integrating DERs in the procedure of GEP problem is presented in [19]. Although many attempts to model the effects of DGs in solving the GEP problem can be found in the literature, to the best knowledge of the authors, no attempts have been reported on presenting a complete model for composite centralized and distributed expansion planning problem, simultaneously by considering the added value of different technologies of non-stochastic DG units on the predesigned GEP problem along with determining their optimal operation strategy in different Load Steps.

In this study, a novel approach for modeling the composite centralized and distributed generation expansion planning is proposed and presented in Fig. 1. The main aim of the proposed method is to evaluate that how the expansion of DG units within distribution networks, can modify the predesigned rated generation capacity obtained by a primary GEP to yield a more economic generation expansion strategy. From this point of view, a primary GEP, which has already specified the required centralized generation capacity and its construction location for a horizon (specific) year, is the input of the proposed method, while the optimal expansion plan of non-stochastic DG units (capacities, locations and technologies) along with modified capacities of the central power plants, are the output of the proposed method. The Fig. 1 illustrates that the primary GEP is the input of the DGEP and the optimal generation expansion, including the central and distributed generation, is the target to be achieved.

The proposed method specifies the optimal location, capacity, and technology type of the DGs which will be installed in the suitable feeders and be operated at different load levels to optimize the overall objective function. The proposed objective function comprises the most important traditional costs used in the expansion and operation problems. Investment costs, fuel costs, emission costs, operation and maintenance costs of both centralized and distributed generation, distribution and transmission networks loss costs and customer’s interruption costs are all considered in the proposed objective function. The objective function is calculated for a specified planning period, and the system is assumed to be unchanged during this period. The present value of the investment and operation costs are included in the objective function. This paper focuses on the competition between the centralized generation expansion and distributed generation expansion to minimize the total investment and operation costs. In summary, the main contributions of this paper are as follows:

• A novel approach for modeling the composite centralized and distributed generation expansion planning problem is proposed.

• An optimal combination of the required generation capacity which should be expanded along the planning horizon, is determined in a centralized and distributed manner, to satisfy the economic and environmental constraints.

• An integrated objective function considering the most important decision-making factors to find the optimal solution is formulated.

• The presented model helps investors to specify the optimal capacity and suitable technology of the non-stochastic DG units which should be allocated into the distribution feeders.

• Optimal operation strategy of DG units in different Load Steps of LDC is determined.

To confirm the superiority of the proposed approach, its performance is compared with recently published schemes which present the models for integrating the DG in the GEP. Table 1, compares the proposed composite CG&DG expansion planning scheme with other models presented in the literatures in nine aspects. These aspects are the specifying the most suitable DG technology integrated with modified GEP in different load level of LDC, determining the optimal DGC allocated to distribution feeders, optimum integration of non-stochastic DG in the GEP, optimum integration of renewable DG in the GEP, techno-economic evaluation of replacing the CG by DG units, determining the optimal combination of generation capacity provided by DG and CG, modeling the TNLC, DNLC & EENS in the optimum integration of DG in the GEP, and simple objective function or multiple objective function. As can be seen, the proposed method delivers better performance than the existing schemes in modeling all mentioned aspects. The majority of the existing schemes model the integration of renewable DGs in the GEP without determining the optimum capacity and location of DG units in distribution feeders. Additionally, the effect of DG integration in the GEP are not modeled on the TNLC and DNLC. Therefore, to the best knowledge of the authors, no attempts have been reported on presenting a complete model for composite CG & DG expansion planning problem, simultaneously by considering the added value of different technologies of non-stochastic DG units on GEP. Consequently, the proposed method has significant advantages over existing schemes.

The remainder of this paper is arranged as follows: problem definition and related assumptions are presented in Section 2. The proposed procedure for composite CG and DG expansion planning along with related flowchart are presented in Section 3. Section 4 provides the proposed objective function, load model and the related formulations. Section 5 provides the extensive numerical studies to evaluate the performance of the proposed approach. Eventually, Section 6 concludes the manuscript.