Site selection and capacity determination of fast and slow charging piles for electric vehicles based on Improved Northern Goshawk Optimization

Driven by "carbon peak and carbon neutrality" goals, the rapid proliferation of electric vehicles (EVs) has triggered supply-demand conflicts in charging infrastructure and distribution network security challenges. This paper proposes an EV fast-slow charging pile site selection and capacity model integrating road network load characteristics with distribution network constraints, solved by an Improved Northern Goshawk Optimization (INGO) algorithm. Based on trip chain theory, an EV spatio-temporal load model is built. Charging hotspots and candidate sites are identified via Voronoi diagrams and road network accessibility analysis. INGO achieves multi-objective balance. Case studies using a modified distribution network and road network model validate the approach. Results show INGO converges faster than PSO and NGO, optimally matching travel trajectories with grid capacity, avoiding overload, and balancing construction costs, waiting time, and network losses.