Abstract:

Addressing the difficulty of translating massive energy consumption data into practical energy-saving value and identifying systemic energy waste in the current energy consumption management for manufacturing processes， an energy waste identification and quantification method was proposed based on the PEP model. In this method， the unit of analysis was shifted from production equipment to the product's value creation process， and the PEP model was constructed to describe the energy consumption composition of a unit product across different processes. Subsequently， through statistical learning of historical production big data， a data-driven energy consumption benchmark representing the enterprise's best practice was generated. Finally， by benchmarking actual production energy consumption against this benchmark， precise quantification and diagnosis of energy waste at the process level were achieved. A gypsum board production case study demonstrates that the method can effectively quantify systemic energy waste， identifying optimization potentials of 16.93% for electricity and 17.69% for natural gas within 9 days. Furthermore， energy profile volatility analysis reveals the key insight that high-energy-consumption processes do not equate to high-waste-potential processes.

Key words: energy benchmark, manufacturing process, energy waste, data-driven, product energy profile （PEP）

摘要：

针对当前制造过程能耗管理中海量能耗数据难以转化为节能价值以及系统性能源浪费难以识别的问题，提出了基于产品能谱（PEP）模型的能源浪费识别与量化方法。该方法首先将分析单元从生产设备转向产品价值创造过程，构建描述单位产品各工序能耗构成的PEP模型；然后通过对历史生产大数据的统计学习，生成代表企业最优实践的数据驱动型能耗基准；最后将实际生产能耗与该基准进行对标，实现工序级能源浪费的精准量化与诊断。石膏板生产案例表明，该方法能够有效量化系统性能源浪费，在9天内识别出电力和天然气分别有16.93%和17.69%的优化潜力；能谱波动分析揭示了高能耗工序不等同于高浪费潜力工序的关键洞见。

关键词: 能耗基准, 制造过程, 能源浪费, 数据驱动, 产品能谱