State Estimation Of Distribution Systems With Distributed Generations

Distributed generations (DGs) are normally small generation units installed in distribution system near customers. DGs are expected to play a significant role in the electric power industry's future. DGs utilize not only traditional fossil fuel, such as diesel, natural gas, fuel oil, etc., but also renewable resource energy such as solar or wind power. They not only increase overall system efficiency but also reduce investments in traditional generation, bulk transmission and distribution facilities. And they are also helpful to loss reduction and reliability enhancement. With gradually increasing proportion of distributed generation in distribution systems,traditional algorithms of state estimation face substantial challenges and new requests.This paper does some research on the state estimation for distribution systems with DGs.An improved branch-current-based state estimation method for unbalanced distribution systems considering DGs is developed and tested in this paper. This method chooses the magnitude and phase angle of the branch current as the state variables. The measurements include real and reactive branch power measurements, current magnitude measurements, power measurements and voltage magnitude measurements. In the algorithm measurements for DGs are included and utilized. The nodes where DG installed are classified into four types, PQ type, whose P and Q are constant; PV type, whose P and V are constant; PI type, whose P and I are constant; and P-Q(V) type, whose P is constant, V is variable and Q is limited by P and V. According to the characters of each node type, this paper proposes a method of dealing with these DGs. The essence of this method is to transform all kinds of DG nodes into the node type which can be directly treated by state estimation algorithms at each iterative step.Then the proposed algorithm is verified with 11-node test system and 34-node test system. Three cases are tested with different node type of DGs in several different conditions. The result proves the efficiency of the proposed method . Some conclusions are drawn from the test as follows:1) When the DGs connect to the system, the state estimation average absolute error has an obvious decrease, and the results become better; 2) When the DGs get farther to the root node, the results become better.3) When DGs have a low permeability rate in distribution system, and DGs'information doesn't exist in the system topology but existing in reality, the DGs have a little influence on the results.