Protection of Different Configured Transmission Lines and Frequency Control of Microgrid Using Artificial Intelligent Techniques

Abstract

Protection and control are crucial to maintain the stable operation of the power systems. Growth in global electricity consumption due to urbanization and industrialization, demands enhanced generation and transmission capacities with higher order network configurations. Power transmission network plays a vital role in transmitting power to distant areas or load centres. Often, renewable energy integration into power systems is encouraged due to low carbon emissions. There is a steady growth in the amount of renewable energy generation every year. The geographically dispersed nature and intermittent generation of renewables require increased transmission capabilities to move excess energy to distant load centres. Rising power demand and renewable integration are a challenge to the power system's protection and control. In order to have improved system stability, reduced service disruptions, and enhanced power delivery efficiency, the protection of transmission lines and frequency control of the system are vital. This work focuses on artificial intelligent protection schemes for various transmission line configurations (double circuit three-phase, single circuit six-phase, and single circuit three-phase), to ensure reliable and secure power transmission and control strategy for frequency control of microgrid. The main aim of the transmission line protection scheme is to identify and isolate the fault as quickly as possible to maintain the stability of the system. The quick detection and classification of faults help the repairmen or maintenance crew to improve the service restoration time. An intelligent protection scheme is proposed based on a single fuzzy inference system and discrete Fourier transform towards the faulty phase detection and classification on the mutually coupled double circuit lines. This proposed protection technique uses the magnitude of the pre-processed current information measured at the sending end bus only. This is implemented in the MATLAB/Simulink environment on a 400 kV, 50 Hz, and 300 km double circuit transmission line model. The efficacy of the proposed scheme has been tested by performing a wide range of simulation studies concerning different types of faults viz. common short circuit faults, cross-country and evolving faults, and high impedance faults. Typical fault scenarios viz. current transformer saturation, noisy environment, and faults occurring during power swing scenarios with variations in different fault parameters and operating conditions were also studied. The results presented confirm that the proposed method detects/classifies all types of faults within one cycle time and is reliable with a detection/classification accuracy of 99.75%. It is found immune to the variations in fault vi parameters and for varying operating conditions. Also, it is not affected by the zero-sequence mutual impedance of the line and does not require any training and communication link. Furthermore, the performance is also appraised with other training-based protection schemes. The enhanced power transfer capability is possible with the six-phase transmission system but it did not gain popularity due to the lack of a proper protection scheme to secure the line for 120 types of different possible short circuit faults. This work presents a comprehensive protection scheme utilizing discrete wavelet transform (db4 mother wavelet) and artificial neural networks (ANNs). Levenberg-Marquardt algorithm is used for training the ANNs. This protection scheme requires only pre-processed current information of the sending end bus. For fault detection and classification of all 120 types of faults, a single ANN module is implemented with six inputs and six outputs. For the estimation of fault location in each phase, 11 ANN modules with six outputs are used viz. one for each of the 11 types of combination of faults. The proposed protection scheme is implemented on a six-phase Allegheny power transmission system using MATLAB/Simulink platform. The simulation results prove its efficiency and effectiveness in detecting and classifying all types of faults with varying parameters. All fault types are detected/classified within one cycle time and the detection/classification accuracy is found to be 99.76%. It is found that the performance of the fault location estimation modules is better with the training data and moderate with the testing data. The integration of renewable energy sources (RES), such as solar and wind power, into power systems presents unique challenges for transmission line protection. Traditional distance protection schemes may not be adequately sensitive or adaptable to the dynamic characteristics of RES-connected lines. To address these challenges, this work proposes an intelligent novel protection scheme that combines the fuzzy logic system for fault detection/classification with regression-based bagged ensemble learning for fault location estimation. The proposed scheme utilizes voltage signals of the bus connected to renewable energy sources processed with discrete Fourier transform (DFT) to extract relevant features for fault diagnosis. A Mamdani based fuzzy inference system is implemented to analyze the DFT-extracted features and make decisions regarding fault occurrence and type. A bagged ensemble learning approach, incorporating multiple regression trees, is employed to accurately estimate the fault location along the transmission line. The performance and efficacy of the proposed protection scheme are verified through extensive vii MATLAB/Simulink simulations on the transmission line model integrated with renewable energy sources (solar and wind). The simulations were carried out considering the variations in fault parameters with different solar irradiations and wind speeds. The results demonstrate that the scheme effectively detects and classifies various fault types in one cycle time, even under dynamic RES generation conditions. The proposed scheme achieved 99.56% accuracy in fault detection/classification confirming its reliable operation. Further, the proposed fault location estimation approach approximates the fault location within ±5% error band and the Chi-square test is performed to assess its reliability. However, apart from the protection of transmission lines, there is another equally concerned issue as much as protection i.e., frequency control of microgrid. Microgrid (MG) is a combination of diesel engine generators, renewable energy sources, loads and various energy storage systems. The low inertia of the microgrid system, stochastic loads and intermittent/discontinuous generation of renewables create complications in the frequency control of microgrid. Massive frequency deviations will cause stability and reliability problems and sometimes may lead to microgrid blackouts. A more rugged and efficient control action is needed to ameliorate the frequency stability of the microgrid. Therefore, a multi-stage PID controller whose parameters are optimized by the moth flame optimization (MFO) algorithm is proposed to control the frequency of an islanded Bella Coola microgrid. This microgrid has renewable energy sources and coordinated control of plug-in hybrid electric vehicles with diesel engine generators. Some popular meta-heuristic based PID control techniques viz. PSO-PID, TLBO-PID, and GOA-PID are also applied to assess the superior performance of the MFO algorithm. The effectiveness of the proposed control method is evaluated on the Bella Coola microgrid to obtain its dynamic response considering the simultaneous changes in renewable energy sources, load, and parametric uncertainties. The dynamic response of the microgrid is enhanced significantly which is confirmed through MATLAB/Simulink simulation results. Moreover, the proposed multi-stage PID controller is robust towards parametric uncertainties of microgrid and plug-in hybrid electric vehicles as compared to other PID controllers. The stability and comparison analysis prove that the proposed method works efficiently.