面向能源互联网的零碳园区智能感知设备优化规划方法

Energy Internet-oriented Optimization Planning Method for Intelligent Sensing Equipment of Zero-Carbon Park

面向能源互联网的零碳园区以新能源为主体,汇集了高比例风/光/生物质等可再生能源、氢发电、煤电等能源形式。然而,目前零碳园区设备数据状态感知研究较少。为合理规划零碳园区中用于数据收集与分析的智能感知设备,保证零碳园区能源系统可靠、安全、优质、低碳和经济运行,提出一种面向能源互联网的零碳园区智能感知设备优化规划方法。首先,分析零碳园区状态感知设备的要求,制定了智能感知设备优化规划的原则,考虑投资成本、维护成本、故障成本等方面,提出了零碳园区智能感知设备优化规划的数学模型;其次,为实现所制定数学模型的准确求解,提出一种灰狼-教与学混合优化(grey wolf and teaching-learning hybrid optimization, GWO-TLBO)算法;最后,以一个零碳园区的实际案例作为仿真验证,验证了文章所提出的零碳园区智能感知设备优化规划方法可显著降低生命周期成本,与现有智能算法的对比实验表明所提出的GWO-TLBO算法具有较高的求解精度。

Energy internet-oriented zero-carbon park is dominated by new energy sources, bringing together a high percentage of renewable energy sources such as wind, biomass and solar energy, hydrogen generation, coal power and other forms of energy. However, there is little research on data state sensing of equipment in zero-carbon parks. In order to reasonably plan intelligent sensing devices for data collection and analysis in zero-carbon parks and ensure the reliable, safe, high-quality, low-carbon and economic operation of energy systems in zero-carbon parks, this paper proposes an energy Internet-oriented optimal planning method for intelligent sensing devices in zero-carbon parks. Firstly, the paper analyzes the requirements of state sensing devices in zero-carbon parks, formulates the principles of intelligent sensing device optimization planning, considers the investment cost, maintenance cost and failure cost, and proposes a mathematical model for intelligent sensing device optimization planning in zero-carbon parks. Secondly, in order to realize the accurate solution of the formulated mathematical model, the paper proposes a grey wolf and teaching-learning hybrid optimization(GWO-TLBO) algorithm. Finally, a practical case of a zero-carbon park is used as a simulation example to verify that the proposed intelligent sensing device optimization planning method for zero-carbon park can significantly reduce the life-cycle cost. The comparison experiments with existing intelligent algorithms show that the proposed GWO-TLBO has the highest solution accuracy.