The requirement of accommodating more distributed generations (DGs) at distribution voltage level has encouraged distribution network operators to utilize their feeder capacity more effectively. This requires the availability of the various network measurements such as voltages, currents, loads, voltage control settings and DG outputs. However these quantities are not directly available in the distribution network control centre. Therefore, to control voltage and power flow in the network, the estimations of these quantities are required. This consists of the monitoring and control of the network operation by application of modern distribution management system (DMS) at the primary substation. The state estimator takes all the available network measurement information, together with a parameterized network model and estimates state of the system in operational time scale. The estimator outputs are then fed into the main control functions and other asset management tools.
Although the application of State Estimation (SE) is very common task at transmission system, the practical application are not common at distribution level. This is due to the fact that the operation, topology and design at distribution level differ from those at transmission level. The untransposed three-phase circuits, unbalanced loads, shorter lines with higher ratio of R/X, and the existence of discrete control options such as transformer tap positions illustrates some of those differences.
These specificities have motivated this piece of research to consider some of the key issues in distribution system state estimation and to develop algorithms to tackle them. This thesis investigated in detail the criterion for identifying suitable solvers for the distribution system state estimation (DSSE) while considering the specific characteristic of the distribution network such as discrete tap position by exploring new optimization methods which are likely to be useful for practical implementation. Some of the research findings have already been disseminated through invited conference panel and IEEE journal.