Distribution system planners endeavor to supply economical and reliable electricity to customers. It is important to design, operate and maintain reliable power systems with lowest cost and highest benefit. Reliability improvement and loss reduction are two important goals for electrical distribution companies. These companies follow, consider and test a lot of technologies, optimization programs, etc. to bring above economic benefits and provide electricity with high quality and reliability and prevent interruptions in system because cost of interruptions and power outages can result severe economic impact on utility and customers.
With recent advances in technology, use of distributed generation (DG) in the power distribution system can provide the most economical solution and keep network in proper situation. A lot of Papers and studies have been carried out in recent years to present methodologies in DG placement and sizing.
One of the criteria to search the optimal DG allocation is minimizing power loss or reliability improvement.
Several papers have been published that address the use of artificial intelligence algorithms, analytical approaches or load flow approaches to optimize DG placement [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11] and [12] based on minimizing power loss. Authors in [1] and [2] solve the problem by analytical approach [3], employs non-linear programming [4], uses combination of genetic algorithm and simulated annealing [5] and [6], present genetic algorithm [7], submits tabu search method and [8] uses fuzzy approach for optimization of its algorithm [9] and [10], apply load flow approaches [11], uses sequential optimization and [12] uses heuristic approach.
All papers presented in [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11] and [12] deal important problems and weaknesses that are listed on below mentioned clauses:
•
All the simulations performed in [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11] and [12] address a static load condition. Objective function optimization based on a single load point, such as the peak load, may not provide reliable results.
•
Reliability aspects in above mentioned papers are not considered while applying DGs to a distribution system can contribute to improving system reliability.
•
DG placement in network has not been considered with evaluating reliability and loss at the same time.
Also some papers have appreciated approaches in their methodologies like [13], but considering static load condition in their concepts may not lead to satisfactory results.
This paper tries to overcome above mentioned weakness and proposes a novel algorithm to optimize objective function. To follow this proper purpose, first time-varying loads are taken into account then multi-objective function are considered based on a cost/benefit form that enhance benefits of DG allocation in system to compensate system loss, system reliability and cost of purchased power from transmission line along the planning period. Finally; to solve this multi-objective problem a novel approach based on dynamic programming is used. In addition DGs are considered as constant power source such as photo cells, fuel cells or gas generators. Also in this paper, purchased active power price from transmission grid varies in different time of day and also cost of energy not supplied for different customers (residential, commercial and industrial) varies in different time of the day.
In the following sections, load modeling is presented in Section 2, mathematical formulation is explained in Section 3, and objective function is submitted in Section 4, dynamic programming method is illustrated in Section 5 and a case study is reported in Section 6. Finally, the conclusions of the paper are summarized in Section 7.
In this paper, introducing dynamic programming as an optimization tool, a novel method has been presented to find best location for distributed generation installation in the network with variable load model which results maximum profit. Load of the network has been modeled in different level and the problem has been optimized considering existed constrains on permanent operation of the distribution system. From studied results it has been derived that reliability and loss of the network are drastically depends on location of the consumer, demanded power of the network, type and capacity of the distributed generations and their location in the network. In addition network condition have great effects on DG allocation in system.
DG allocation has another advantage in addition to mentioned benefits which cannot be neglected. These beneficial effects include improvement in voltage profile of the load point and locating network buses in allowable limit. Another advantage is reducing of power flow in feeders because of compensating loss and part of required power of load points of the network. It decrease stress of the feeders especially feeders that they are next to high voltage distribution substation. This increases duration of life time of the equipment.
Therefore DGs can get technical and financial benefits as indicated in simulation results if allocated in proper locations with appropriate sizes.
In future works there are extra considerations that must be careful attention about network upgrading problem and reduction of purchased reactive power in presence of DG in network. This mentioned factors may affect the results of DG allocation.
