基于RAGA的PPC模型在城市公园绿地景观生态美学评价中的应用

Application of RAGA-based PPC model in eco-aesthetics assessment of urban park green space landscape

基于公众对生态环境的日益重视,以生态美学思想为指导,提出城市公园绿地景观生态美学评价的指标体系;同时利用基于RAGA的PPC模型,将高维数据转换到低维子空间,通过寻求最优投影方向及投影函数值来实现对城市公园绿地景观的生态美学评价。选取大庆市10个具有代表性的公园绿地为研究对象,运用该评价指标体系及模型进行景观生态美学评价。结果表明:总体景观空间多样性、植被色彩多样性、建筑与小品分布合理性、树种多样性、水体可亲近性、植被长势自然性、地形自然度、建筑与小品体量适宜度指标对大庆市公园绿地景观生态美学效果的影响较大,是今后大庆市公园绿地景观生态建设、改造与管理上应作为重点考虑的因素;在10个公园绿地中,油田乐园、城市森林公园最具生态美,而乘风广场、世纪大道(高新技术开发区段)附属带状绿地的生态美较差。应用该模型可以避免传统方法存在的一些不足,也为处理景观评价中具有模糊性、不确定性的高维数据的综合评判、排序、寻优等方面研究提供了新的思路与方法。

Based on the increasing public awareness of environmental issues and with the guidance of eco-aesthetics ideology,the authors proposed an eco-aesthetics assessment index system of urban park green space landscape,and applied the RAGA-based PPC model to transform the multi-dimension data into low-dimension space,aimed to search for the optimum projection direction and the projection function value to realize the eco-aesthetics assessment of urban park green space landscape.The assessment index system and the model were then used to make landscape eco-aesthetics assessment of ten representative urban park green spaces in Daqing City.The results showed that the general landscape space diversity,vegetation color diversity,building and furniture arrangement rationality,tree species diversity,water body＇s approachability,vegetation growth naturalness,landform naturalness,and building and furniture mass fitness had greater effects on the park green space landscape in Daqing City,which should be paid much attention in the landscape ecological construction,renovation,and management of the park green spaces in the future.Among the ten urban park green spaces,Oil Field Paradise and Urban Forest Park had the best eco-aesthetics effect,whereas Chengfeng Square and Century Avenue （high-tech development zone）-attached banding green space had the worse one.The model could avoid some deficiencies in traditional methods,and also,provide new idea and method for the study of comprehensive assessment,sequence analysis,and optimization of landscape assessment with fuzzy and uncertain high-dimension data.