state estimation in electric power systems based on adaptive neuro-fuzzy system considering load uncertainty and false data

Control center of modern power system utilizes state estimation as an important function. in such structures, voltage phasor of buses is known as state variables that should be determined during operation. to specify the optimal operation of all components, an accurate estimation is required. hence, various mathematical and heuristic methods can be applied for the mentioned goal. in this paper, an advanced power system state estimator is presented based on the adaptive neuro-fuzzy interface system. indeed, this estimator uses advantages of both artificial neural networks and fuzzy method simultaneously. to analyze the operation of estimator, various scenarios are proposed including impact of load uncertainty and probability of false data injection as the important issues in the electrical energy networks. in this regard, the capability of false data detection and correction are also evaluated. moreover, the operation of presented estimator is compared with artificial neural networks and weighted least square estimators. the results show that the adaptive neurofuzzy estimator overcomes the main drawbacks of the conventional methods such as accuracy and complexity as well as it is able to detect and correct the false data more precisely. simulations are carried out on ieee 14-bus and 30-bus test systems to demonstrate the effectiveness of the approach.