供水管网DMA分区入口处的流量压力监测与评价数据集

该数据集主要包括供水管网流量压力数据，首先对提取监测序列数据进行数据清洗与预处理，采用范围校核、离群校核、突变校核、方差校核、信息冗余校核等多种算法自动识别无效数据，随后结合现场设备运行状态进一步对数据的有效性进行人工复核，减少机器的误判，最终将数据赋予有效、无效两类标签。对于无效数据，采用差值、LSTM预测模型进行合理化修复，生成高质量、可用于建模的连续流量压力数据集。该数据可支撑漏损分析、水力模型与优化调度、用水行为分析与负荷预测等。This dataset primarily comprises flow and pressure data collected from water supply networks. Initially, the extracted monitoring sequence data is subjected to data cleaning and preprocessing. A range of algorithms, including range check, outlier check, abrupt change check, variance check and information redundancy check, are utilized to automatically identify invalid data. Subsequently, manual review of data validity is further performed in conjunction with the operating conditions of on-site equipment to minimize machine misjudgments. Finally, all data samples are assigned two types of labels: valid and invalid. For invalid data samples, interpolation methods and LSTM prediction models are adopted to perform reasonable restoration, generating high-quality continuous flow and pressure datasets suitable for model training. This dataset can support multiple applications including leakage analysis, hydraulic modeling and optimal scheduling, water use behavior analysis and load forecasting.

搜集汇总

背景与挑战

背景概述

该数据集专注于供水管网DMA分区入口的流量和压力监测，包含超过31万条记录，每年更新，涵盖地点、时间、原始及修复后的流量压力数据。数据经过自动校核和人工复核，确保高质量，适用于漏损预警、水力模型训练和调度优化等水务行业应用，旨在提升城市供水系统的智能化管理水平和运行效率。