城市日用水量主影响因素粗集理论分析方法

Principal factor analysis method for daily urban water consumption based on rough set theory

为克服传统粗糙集变精度算法和动态约简算法在城市日用水量主影响因素分析中的"伪"约简属性存在问题,利用权值累积评价因子的概念,提出新的城市日用水量属性约简求解算法.利用改进的属性约简算法对国内某城市日用水量观测系统的主影响因素进行分析讨论.实例分析表明:新的属性约简算法较大程度上克服了传统约简算法结果中"伪"约简属性的存在,实例城市的日用水量主影响因素为最高温度,相对湿度,阴晴量,星期量.预测验证结果也表明了所提出的主影响因素分析方法具有其合理性.

It is difficult to find the true principal factors for daily urban water consumption based on the conventional rough set algorithm because there are usually different results of attribute reduction for time series with different lengths or the same length with different samples in the same information system. To solve this problem, an advanced algorithm based on weighting-coefficient cumulative estimation is put forward, and the weighting coefficient is the result of variable precision rough set algorithm. Principal factors for daily water consumption in Hangzhou were analyzed using the new algorithm. Results show that the proposed algorithm for attribute reduction is helpful for distinguishing the fake attribute from dyanamic reduction sets, and the principal factors affecting daily urban water consumption of Hangzhou city are the maximum air temperature, the relative humidity, the index of weather and the index of weekday.