Effect of Generator Participation in Optimal Reactive Power Dispatch for Multi-Objective Optimization
Optimal Reactive Power Dispatch (ORPD) plays a critical role in ensuring secure, efficient, and stable power system operation, particularly under increasing demand, network complexity, and renewable energy integration. While traditional single-objective optimization methods focus individually on minimizing power loss, improving voltage profiles, or enhancing voltage stability, they fail to account for the trade-offs among these conflicting objectives. To address this, a new multi-objective optimisation technique, designated as the Multi-Objective Embedded Immunized Accelerated Evolutionary Programming (MOEIAEP), is proposed in this study. The technique integrates immunization and adaptive acceleration mechanisms into conventional evolutionary programming and employs a weighted-sum approach to balance three fitness functions: power loss minimization, voltage stability enhancement, and minimum voltage improvement. The IEEE 57-Bus Reliability Test System is used as a testbed, with different generator participation (GP) scenarios evaluated by varying the number and location of reactive power-supporting generators. Results show that MOEIAEP consistently outperforms two benchmark techniques, namely MOEP and MOAIS, across all GP configurations, demonstrating the superior convergence and solution quality of the proposed method. For example, in test case IE57/w-127/d19:30 at Qd19 = 30 MVAR, MOEIAEP achieved a total fitness value (Ftotal) of 0.5754 at NoGP = 5, compared to 0.6819 (MOAIS) and 0.7494 (MOEP). Similarly, in test case IE57/343/56:30 at Qd56 = 30 MVAR, MOEIAEP attained Ftotal = 0.7743 at NoGP = 6, outperforming MOAIS (0.7929) and MOEP (0.8309). These findings underscore the scalability, robustness, and effectiveness of the proposed MOEIAEP in managing complex ORPD problems and affirm its potential for deployment in larger or more intricate power system environments.