Dynamic model of distribution network cell using artificial neural network approach

The goal of this research is to design a dynamic model of distribution network cell (DNC) using artificial neural network (ANN) approach. The development of dynamic model using artificial intelligent (AI) has been increased over the year which some of AI method is more complicated like fuzzy system that needs fuzzy rule to develop the model. So, there is need for other method that is more simple and accurate. In this project, the DNC model using ANN was developed using MATLAB programming. This research consists of two models which are for active and reactive power based on frequency and input voltage. The active and reactive power responses from the ANN model were compared to the response from the full DNC model at various type of disturbance. The response of full DNC model was obtained from the UK 11 kV distribution network model. The model was developed using DigSILENT PowerFactory software. The performance of the ANN model was validated by calculating the value of root means square error (RMSE) and the best fit value. Later, the performance of the ANN model was also compared to the fuzzy model and system identification model. The results obtained show that the ANN model was more simpler as no parameters involved in developing the equivalent model. The efficiency was also good based on the high best fit value compared to fuzzy system model and system identification model. In conclusion, the equivalent dynamic model of DNC based on ANN system approach was successfully developed.