Abstract: The existing site selection and capacity determination methods only consider the voltage and network loss indicators after photovoltaic (PV) is connected, but insufficient consideration is given to the consumption effect after PV is connected, making it difficult to meet the development needs of new distribution networks. This paper proposes a new distribution network PV locating and sizing method that takes into account multi-dimensional performance evaluation. First, a multi-dimensional evaluation system covering power quality, power supply capacity, and photovoltaic access effects is formed, and a combined subjective and objective assessment system is constructed. Comprehensive evaluation method of access solutions based on weight-fuzzy evaluation. Secondly, the comprehensive evaluation was organically combined with the optimized locating and sizing process to form an evaluation-selection integrated PV site selection and capacity optimization model, using the non-dominated sorting genetic algorithm-ii (NSGA-II) achieves the optimal access location and capacity acquisition under the condition of differentiated PV output throughout the year. Finally, the numerical example verification shows that the proposed method not only significantly reduces the node voltage deviation and line transmission pressure, but also has better PV access characteristics and multi-index comprehensive performance within the scope of a more comprehensive technical performance evaluation, improving the rationality of PV configuration. safety and resource efficiency, which can effectively support the development needs of new distribution networks.