📄 中文摘要
污水处理厂消耗全球1-3%的电力，因此准确的能耗预测对优化运行和可持续发展至关重要。虽然机器学习模型能提供点预测，但它们缺乏可解释的不确定性量化，这对于安全关键基础设施中风险感知的决策至关重要。本研究开发了一种区间二型自适应神经模糊推理系统（IT2-ANFIS），该系统不仅生成能耗预测，还能提供伴随这些预测的信任区间，从而量化预测的不确定性。IT2-ANFIS通过模糊逻辑规则和隶属函数，能够对不确定性进行建模和推理，使其在处理复杂非线性系统和不精确数据时表现出色。该方法的核心优势在于其端到端的可解释性，模糊规则可以直接映射到专家知识或物理过程，使得预测结果及其不确定性更容易被操作人员理解和信任。通过在实际污水处理厂数据集上进行评估，IT2-ANFIS在预测精度和不确定性量化方面均优于传统机器学习模型。其生成的信任区间能够有效反映预测的可靠性，为决策者提供了风险评估的关键信息。这种可解释的不确定性量化能力，使污水处理厂的能源管理决策能够更好地平衡效率、成本和风险，例如在电力价格波动或设备故障风险较高时，可以根据不确定性区间调整运行策略，从而提高系统的韧性和可持续性。
📄 English Summary
Explainable Uncertainty Quantification for Wastewater Treatment Energy Prediction via Interval Type-2 Neuro-Fuzzy System
Wastewater treatment plants consume 1-3% of global electricity, making accurate energy forecasting critical for operational optimization and sustainability. While machine learning models provide point predictions, they lack explainable uncertainty quantification essential for risk-aware decision-making in safety-critical infrastructure. An Interval Type-2 Adaptive Neuro-Fuzzy Inference System (IT2-ANFIS) is developed to generate energy consumption predictions along with confidence intervals, thereby quantifying the uncertainty associated with these predictions. IT2-ANFIS leverages fuzzy logic rules and membership functions to model and infer uncertainty, exhibiting superior performance in handling complex nonlinear systems and imprecise data. The core advantage of this method lies in its end-to-end explainability; fuzzy rules can be directly mapped to expert knowledge or physical processes, making the prediction results and their uncertainties more comprehensible and trustworthy for operators. Evaluated on real-world wastewater treatment plant datasets, IT2-ANFIS outperforms traditional machine learning models in both prediction accuracy and uncertainty quantification. The generated confidence intervals effectively reflect the reliability of the predictions, providing crucial information for risk assessment to decision-makers. This explainable uncertainty quantification capability enables wastewater treatment plant energy management decisions to better balance efficiency, cost, and risk. For instance, during periods of electricity price volatility or high equipment failure risk, operating strategies can be adjusted based on the uncertainty intervals, thereby enhancing system resilience and sustainability.