The integration of photovoltaic (PV) systems into distribution networks plays a crucial role in reducing power losses and improving voltage stability. However, many existing optimization studies overlook time-varying loads, which are essential for accurately modelling the dynamic behaviour of industrial, residential, and commercial demand profiles. This study applies the Coyote Optimization Algorithm (COA) to identify the optimal placement and sizing of PV units in the IEEE 33 bus and IEEE 69 bus distribution systems under varying load conditions. Scenarios involving one, two, and three PV units are evaluated. The results show that the configuration with three PV units provides the greatest power loss reductions, reaching 43.39 percent in the 33-bus system and 45.81 percent in the 69-bus system, particularly in commercial load scenarios. In addition to minimizing losses, the optimized PV placement significantly improves voltage profiles, with voltage levels rising to 0.9977 per unit and 0.9987 per unit in the 33 bus and 69 bus systems, respectively. Industrial loads, which experience peak demand during evening hours, benefit less from PV integration, suggesting the potential value of energy storage systems. Overall, the findings confirm the effectiveness of COA in optimizing PV integration under realistic, time-dependent loading conditions and contribute to a more accurate and practical approach for planning future distribution networks. © 2025, IPMEDIA SDN BHD. All rights reserved.