In practice, monthly temperature records play a major role calculating the heating and cooling degree-days of the air conditioner management in buildings. Classically, heating and cooling degree calculations are based on the comparison of outdoor temperature measurements with a standard level called the base temperature. Truncating the temperature series at a constant base temperature causes heating (deficits) and cooling (surpluses) degrees. Since outdoor temperature records are of random character, future predictions with probabilistic and statistical methods are convenient means for calculations. All previous methodologies in the literature are based on crisp logic with simple mathematical calculations. This paper proposes a new methodological approach by converting the actual temperature record probability distribution function (PDF) into a probabilistic standardized degree indicator (SDI) in normal (Gaussian) PDF form with zero mean and unit standard deviation. Furthermore, as a new evaluation methodology fuzzy logic inference calculations are provided. SDI is divided into seven classes, three for cooling, heating, and a single class for uncooled-unheated states. Heating and cooling degree changes according to SDI values are presented with crisp and fuzzy sets. The application of the methodology is made for Kadıköy urban area on the Asian side of Istanbul, Turkey. In the case of fuzzy set SDI air conditioned operation, it has been determined that there is cooling energy saving of approximately 25% compared to the crisp set. The same is true for the heating degree-day side.

在实践中，每月温度记录在计算建筑物空调管理的加热和冷却度数天数时起着重要作用。传统上，加热和冷却程度的计算是基于室外温度测量值与称为基础温度的标准水平的比较。在恒定的基础温度下截断温度序列会导致加热（不足）和冷却（盈余）度数。由于室外温度记录具有随机性，因此用概率和统计方法预测未来是方便的计算手段。文献中所有以前的方法都是基于简单的数学计算的清晰逻辑。本文提出了一种新的方法，通过将实际温度记录概率分布函数 (PDF) 转换为具有零均值和单位标准偏差的正态（高斯）PDF 形式的概率标准化度数指标 (SDI)。此外，作为一种新的评估方法，提供了模糊逻辑推理计算。SDI 分为七类，三类用于冷却、加热，单类用于非冷却-非加热状态。根据 SDI 值的加热和冷却程度变化以清晰和模糊的集合呈现。该方法应用于土耳其伊斯坦布尔亚洲一侧的 Kadıköy 市区。在模糊设置 SDI 空调运行的情况下，已确定与清晰设置相比，制冷能源节省约 25%。