Emergency demand response and minimization of losses in distribution systems

With the development of technology, the world has become more industrialized. This situation has increased the demand for energy. population growth has increased significantly in recent years due to new technologies, environmental concerns and user preference. In addition to this, there is also serious fluctuation during the day in the demand of electricity. To meet demand, electricity companies either increase their output or buy electricity from a wholesale market with a dynamic pricing structure. Such an operation leads to a low load factor (ratio of average load to peak load). From this point; it has been a remarkable research and application area in recent years that the control axioms are shifted to the consumer side by production, rather than to the management of the production according to the student, in particular the demand power is managed to decrease the peak demand. In this thesis, the necessary methods to remove the disruptive effects that would prevent the operation of the system in case of malfunctions in distribution systems were studied. In this study incentive based demand response program, in which consumers are included, was used. Faults that may occur in the square are considered as a single-purpose optimization problem and this problem is solved through the proposed algorithm under certain constraints. In the applied optimization method, the main purpose is to ensure that the voltage level in the bar is appropriate under normal operating conditions. In addition, minimizing power losses in the power system is another purpose of this thesis. In this context, Simulated Annealing Algorithm which is one of the intuitive optimization methods is used. This algorithm was developed in the MATLAB environment. When the boundary voltage profiles are considered as inequality constraint, the decision variables are the transformer tap changer, the power factor of the distributed generation systems and the load curve characteristics of the buses to be loaded. A balanced distribution system with 23 bus was used as the test system. This test system was utilized in the modeling of the OpenDSS simulation program. In this program, the test system data used is arranged so that the daily load flow based on the fixed point iteration method is performed. Loads in bars are modeled as constant strength and reactive and active power constraints are defined here. In order to evaluate the performance of the proposed method, many studies have been carried out to make corrections in the system according to the possible emergency conditions. Three corrections have been applied to the beloved test system. The first correction is achieved by setting the transformer setting level. In the present test system, the transformer was initially deactivated and the load flow was made. Active power loss, reactive power loss, single phase voltage profile and cost functions of objective functions are determined considering that there are no faults in the system. One of the 4 DGs working in the Test System was disabled between 16:00 and 21:00 hours. In addition, a load increase was observed at Bus-23 between 12:00 and 18:00. Only the transformer tap changer is added to the system in order to regulate the voltage profile and reduce the losses in the faulty system. The transformer tap changer is optimized between -10 and +10 and the voltage level is regulated. The second method is reactive power planning. System modeling has been done by adding a fixed transformer tap changer to the existing system. One of the 4 DGs working in the Test System was disabled between 12:00 - 15:00 hours and the other between 21:00 - 24:00 hours. Also during the day Bus-7 and Bus 16 had a significant pay rise. In order to correct the voltage level in the system and reduce the losses according to this fault condition, the reactive power drawn from the system is reduced by keeping the active power of the two DGs constant and adjusting the power factor. Simulated annealing method is applied to the system to adjust the parameters of the optimum reactive power factor. It has been identified. The third method is the load curtailment method. System modeling has been done by adding a fixed transformer setting changer to the existing system. One of the 4 DGs working in the Test System was disabled between 12:00 - 15:00 hours and the other between 21:00 - 24:00 hours. Also during the day Bus-7 and Bus 16 had a significant pay rise. In order to correct the level of voltage in the system and to reduce the losses according to this fault condition, a certain load was curtailed from the determined buses. Based on the load shape parameter as the load curtailment parameter, the load curtailment level is set at maximum 30%. A specific sale price is given according to the 24-hour load consumption values of the buses. A lower selling price has been set aside for spending too much power. Relatively unimportant loads have been identified to alleviate the problem in the system. In order to be able to pay for these loads, a specific fee offer has been determined for the relevant bus loads and the price of the bid has been offered at 20% of the sales price of the relevant buses. The payout interval for the interruption is set between 0 and 25 $ / h, and the voltage levels are set according to normal operating limits. Bus-13, Bus-19, Bus-21 were determined for loading and 5 parameters were given to the system with the Simulated Annealing algorithm to determine how much load can be loaded, and the optimum working point was found. The following results have been achieved with this thesis study.Voltage surges and distribution system power losses can be effectively reduced when distributed production units are added to the system. But the effect on voltage regulation and the size and location of the DG must be studied carefully.When the transformer is added to the system together with the DG, the worst case reduction of voltage regulation and power losses is further increased. Thus, central optimal control during system planning as well as the effective operation of these system components together is necessary to operate under stronger conditions.Voltage surges and power losses associated with the DGs connected to the Feeder's lower buses are less than those connected to the buses at the top of the federation. Better voltage profile and reduced power losses are related to the spread rate of the DG to the system (frequency of use).Different system planning for different types of DGs (Photovoltaic-Wind Turbine etc.) is required. The addition of the transformer tap changer has a significant effect on the voltage regulation and accordingly, it is better to work with other devices in combination.The demand response methods reduce the voltage drop of the feeders in the distribution system and increase the voltage level at the end of the bus.The types of Demand Response methods can vary depending on the type of failure and the load on the system. If there is a fault in the system, if the current fault is lighter, and there is no or less load in the system, the transformer tap changer can be used to reduce losses and voltage regulation can be maintained under normal conditions. If the fault in the system is slightly larger and loads are increasing, in addition to the current fixed transformer setting stage, if the active power of the working distributed generators (DG) in the system is kept constant and the reactive power is reduced, both loss can be reduced and voltage can be regulated. If the fault size in the system is too large to be solved by the previous methods and the loads in the system are too high, a certain load disposal agreement can be made with those who use the most appropriate load to load in the distribution systems and the losses can be reduced and voltage can be kept under normal conditions. Various optimization methods (classical, intuitive) can be applied to the problem of minimizing the stress profile and the losses in the evaluation of certain constraints and the objective function.Used as a single-purpose optimization method, SA can easily and efficiently deliver optimum candidate results.