[Objectives]　To address the challenge of economic evaluation for the configuration of large temperature difference units in long-distance heating networks

this study proposes the concept of economic distance. The aim is to quantitatively analyze the applicability of configuring large temperature difference units with different network distances through a techno-economic optimization model. [Methods]　Based on the life cycle cost analysis

an economic model is established

incorporating investment and installation costs and operational costs. By integrating hydraulic and thermal constraints of the network

a bi-level optimization framework comprising upper-level planning and design and lower-level operational strategies is developed. The framework couples a mechanistic model with a deep neural network surrogate model to achieve optimal solutions. A case study of a long-distance heating network is conducted to investigate the effects of return water temperature

pipe diameter

and heating load on the economic distance

and the economic distances for configuring large temperature difference units are obtained. [Results]　The results indicate that the economic performance of configuring large temperature difference units significantly depends on network length

with a distinct economic distance existing. For a DN1000 pipe at a return water temperature of 30 ℃

the economic distance is 58.34 km. Beyond this distance

the savings in operational costs outweigh the increase in investment costs. The economic distance exhibits a non-linear correlation with pipe diameter. Excessively large or small pipe diameters will reduce the economic distance. An increase in either return water temperature or heating load will increase the economic distance. [Conclusions]　This bi-level optimization framework enables quantitative evaluation of the economic distance for configuring large temperature difference units in long-distance heating networks

thereby providing decision-making support for techno-economic analysis in engineering planning.