Optimal Substation Placement in Radial Distribution Network Considering Load Fluctuation
CHUN YANG,WEI GUO,JIAN-YONG WANG,ZHE-YAO LI,CAN-BING LI
DOI: https://doi.org/10.12783/dteees/peems2019/34014
2020-01-01
DEStech Transactions on Environment, Energy and Earth Sciences
Abstract:It is an important aim of substation placement to reduce the energy losses in feeder lines. The energy losses mainly depend on the sizes of load, lengths of feeder lines and load fluctuation. Load fluctuation is ignored in the existed publications. In this paper, an optimization method of substation placement considering the load fluctuation is proposed to achieve the minimum total costs of energy losses in feeder lines. In the proposed method, the concept of loss equivalent duration under maximum load is introduced to quantitatively assess the impact of load fluctuation on the energy losses. Numerical results demonstrate the effectiveness of the proposed method. Introduction Optimizing substation placement is very important for distribution network planning, which has impact on the economy and reliability of the entire distribution network construction. A new model for large distribution systems is proposed in the [1]. The model considers the fixed and the variable costs of substations and feeders as well as the cost of the power losses. The [2] optimizes distribution substation locations by using operations research methods. An application of recent advances in optimization techniques for distribution substation siting and radial feeder planning is presented in the [3]. The maximum benefit is easily obtained based on the method. The [4] introduce an algorithm for large radial distribution systems. Other method applied to distribution system planning in the [5-12]. The substations are generally placed close to the load center. The center of gravity method, which is usually referred as the location-allocation method [13], is widely used to find the load center. Generally, load-distance products (LDS) is assumed to have an approximatively linear relationship with the power losses in feeder lines. The center of gravity method usually aims to obtain the minimum sum of LDS, and the optimal solution of this method is given in [14]. An improved method, known as L2DS, where the square of active power is used instead of active power, is given in [15]. The solution proposed in [15] is closer to the heaviest load point, and the active power losses in feeder lines are smaller. In the existed publications, maximum load is usually used to optimize the location of substations without considering load fluctuation. In this study, considering load fluctuation, the concept of loss equivalent duration under maximum load is introduced. Based on the concept, an optimization method of substation placement is proposed. The Formulation of the Proposed Method The energy losses in feeder lines can usually be calculated by (1). Loss av W P T (1) where WLoss denotes the annual energy losses in feeder lines, ΔPav denotes the average power losses in a year, and T is treated as 8760 hours. Using the maximum load to calculate the energy losses, the concept of loss equivalent duration under maximum load τ, is defined in (2). If the transmitted power in feeder lines maintains the maximum load, the electrical energy losses in a period of τ equal to the real electrical energy losses in the whole year. max Loss W P (2) where ΔPmax denotes the power loss at the maximum load. In order to get the value of τ, (1) can be transformed into (3). max max max av Loss av P W P T P T P FT P (3) where F (ΔPav/ΔPmax) denotes the loss factor. According to [16], loss factor can be determined by the Buller and Woodrow formula [16]. Comparing (2) and (3), τ can be obtained by (4), (5). 2 8760 [ (1 ) ] FT cf c f (4) max av P f P (5) where c denotes an empirically determined constant usually between 0.15 and 0.3, f denotes the load factor, Pav denotes the average load, and Pmax denotes the maximum load. The introduction of τ changes the fluctuant load into a constant. Using τ to calculate energy losses, the costs of energy losses in feeder lines can be obtained by (6), (7).