基于K-Means城市分类算法的夜光遥感电力消费估算
|
张莉, 谢亚楠, 屈辰阳, 汪鸣泉, 常征, 王茂华
|
Estimation of electric power consumption using nighttime light remote sensing data based on K-Means city classification algorithm
|
ZHANG Li, XIE Yanan, QU Chenyang, WANG Mingquan, CHANG Zheng, WANG Maohua
|
|
表8 K-Means城市分类方法与传统分类方法的估算结果对比
|
Tab.8 Comparison of EPC estimation results between K-Means city classification method and traditional classification method
|
|
分类方法 | /% | | 高精度/% | 中精度/% | 低精度/% | /% | 不分类 | 42.37 | 59.04 | 47.53 | 29.28 | 23.19 | - | 经济分区 | 38.37 | 63.23 | 50.19 | 28.13 | 21.68 | 50.57 | 地理分区 | 34.05 | 53.16 | 52.47 | 28.13 | 19.40 | 52.09 | 人口规模 | 31.06 | 53.76 | 52.09 | 28.51 | 19.40 | 53.99 | 波士顿矩阵-城镇人口比例 | 32.62 | 57.61 | 51.34 | 30.41 | 18.25 | 56.27 | 波士顿矩阵-市辖区人口比例 | 32.03 | 59.20 | 53.99 | 27.37 | 18.64 | 58.55 | 波士顿矩阵-市辖区GDP比例 | 31.66 | 59.78 | 49.42 | 32.31 | 18.27 | 61.21 | 波士顿矩阵-市辖区第三产业GDP比例 | 32.13 | 58.09 | 50.95 | 31.56 | 17.49 | 55.13 | K-Means算法 | 32.02 | 57.04 | 53.99 | 26.99 | 19.01 | 57.80 |
|
|
|