Abstract: A community is not only an important place for residents to live but also an important part of urban development.In China, with the development of cities and people's attention to the improvement of human settlements, the renewal of existing communities has received increasing attention.In the process of updating the public space of the existing community, due to the lack of a more scientific and objective assessment, there may be problems such as strong subjectivity in renewal and transformation, insufficient information, superficial and current renewal, and a lack of thinking about green sustainability.The green evaluation of the existing community public space can improve the above situation.The green evaluation of community public space has the characteristics of integrity, convenience and dynamic adaptability.On the basis of sorting out the research on community sustainable development and green evaluation and clarifying the contents of green evaluation, this paper studies the construction of a green evaluation system for existing community public space.The green evaluation of the public space of the existing community adopts the analytic hierarchy process and constructs the analytic hierarchy process model.Based on the comparative analysis of the literature and the field investigation of several existing communities and according to the characteristics of public space in existing communities and green renewal needs, three levels of indicators for green evaluation are selected and established, including five level-1 indicators, 13 level-2 indicators and 26 level-3 indicators.Refer to relevant research to determine the quantitative standard of the three-level index level.By means of expert inquiry and comprehensive scoring between experts and residents, the score of each index at each level is obtained, the relative importance value of the index is calculated, and the weight of each index at all levels is generated by Yaahp software to determine the evaluation operation model so that the final score of the evaluation can be calculated.Four typical existing communities in Guangzhou are selected as time cases of public space green evaluation.The evaluation data of community public space are obtained by means of data query, measurement and investigation. The green evaluation system is used to obtain and analyze the green evaluation results of the public space of the above four existing communities to verify the feasibility of the evaluation system.After verification, the green evaluation system can objectively evaluate the green degree of the existing community public space, clearly understand and master its relative weaknesses and deficiencies from the evaluation, and draw common problems in the green aspect of the community public space.The conclusion shows that the evaluation system is objective, feasible and universal in a certain range. It can be used as a part of the current situation research before the renewal of the existing community public space and provide direction and guidance for the renewal. It can also be used as a reference for the effect investigation of community renewal. It can also be used to publicize and educate the green construction of community public space to make the community have the awareness of multiple subjects and understand the importance of green construction of community public space.Due to the limitations of sample location and the number of practical evaluation cases, the evaluation system still has the limitations of universality and no clear conclusion of the evaluation reference standard. It needs to be continuously studied, calibrated and improved in subsequent use to make the evaluation system more objective and practical.

Key words: existing community, public space, green evaluation, evaluation system

摘要： 我国既有社区的公共空间在更新过程中，往往由于缺乏较为科学客观的评估，可能产生更新改造主观性较强，更新停留在表面及当前，缺乏对绿色可持续的思考等问题。针对这一状况，结合文献对比分析和实地调研，运用层次分析方法，确立三个层级的指标构成，通过专家及居民综合评分确定各指标权重，构建既有社区公共空间的绿色评价体系。并运用该体系对广州市四个典型的既有社区公共空间进行实例分析，以验证该评价体系的可行性。经验证，该评价体系具有一定的客观性与可行性，可为既有社区更新提供客观的参考。

关键词: 既有社区, 公共空间, 绿色评价, 评价体系