城市光伏密集接入地区楼宇空调虚拟机组组合及联动策略研究

Recently, there are more and more distributed photovoltaic applications relying on the construction of buildings in cities. Meanwhile, the central air-conditioning or decentralized air-conditioning load has an increasing proportion of the total load of local building. On the other hand, the issue on the overlapping peaks of low-voltage distribution transformer has been brought by the wide-range fluctuation of distributed photovoltaic system, as a result, the average utilization ratio of distribution network equipment is commonly low. This project is based on the application background above. Differing from only the issue on photovoltaic consumption is focused in the previous research, this project will take the building for study and will follow the implementation route of characteristic analysis, virtual unit combination and typical scene application. Firstly, the controllable parameters and feasible control strategy of air-conditioning in diverse types and of diverse industries will be discussed. Then, the practically schedulable capacity of air-conditioning in each building will be evaluated considering the operating constraints. Finally, the equivalent models of virtual peak load units consist of air-conditioning units will be established and the implementing scheme of fine combination optimization regulation will be proposed in two scenes, which are day-ahead peak load shifting and within-day peak load shaving respectively. Traditionally, the way of adding the capacity of the units will be utilized solely to meet the short-term peak demand, however, this way will lead to the problem of the continuous increase of power generation and supply cost. This project will not only solve this problem but also explore a scheduling method of “micro-load” in distribution network. From the traditionally extensive load scheduling to the fine load scheduling, this scheduling method will lead to excavate the flexible regulation potential of demand-side resources. Therefore, this project has importantly scientific significance and application value for reasonably and orderly alleviating the pressure of power network.

目前城市中依托楼宇建设的分布式光伏应用越来越多，而楼宇的集中式中央空调以及分布式空调负荷占比也越来越重。分布式光伏出力的大幅波动给低压配变带来峰上叠峰问题，普遍导致配网设备平均利用率低下。本项目申请即是基于此实际应用背景，但区别于传统只关注光伏的消纳问题，提出以楼宇为对象，按照从特性分析、虚拟组合再到场景应用的实施路线，重点面向不同行业不同类型的空调，研究其可控参量及可行控制策略，基于运行约束条件评估实际可调度容量，继而实现空调虚拟机组的等效作用模型及精细化组合优化调控的实施方案，参与电网的“日前填谷”与“日内削峰”。通过本课题研究，一方面可以避免传统单纯依靠增加机组容量来满足短暂高峰用电导致发供电成本不断上升的问题，一方面可以探索配电网的“微负荷”调度方法，引领性挖掘电网需求侧资源的柔性调控潜力，由传统粗放式负荷调度走向精细化负荷调度，对于合理、有序缓解电网压力有显著科学意义和应用价值。

随着我国空调应用的日益广泛，城市夏季电网高峰负荷在近十几年来急速增长。城市中，公共楼宇空调负荷占据了总负荷的很大一部分比例且易于集中调控，因此，充分挖掘用电高峰时段内大规模公共楼宇空调负荷的调峰潜力，并采用合理的调控方式削减其高峰用电，对改善我国夏季电网负荷特性和实现电网的安全、稳定、经济运行具有重要意义。本项目围绕楼宇空调虚拟组合及应用的相关实际问题展开研究，并取得了以下成果：.1）分析了商场、写字楼和宾馆三种典型行业类型公共楼宇的空调系统夏季日负荷曲线特性，分别归纳了三种类型公共楼宇的空调负荷集群参与电网调峰的典型场景；分析了影响公共楼宇空调系统用电负荷的气象因素，进而提出了一种计及综合气象因素累积效应的公共楼宇空调短时基线负荷预测方法。.2）综合考虑空调系统的运行效率和用户的人体舒适度需求，分别对不同类型的公共楼宇空调负荷的降负荷控制策略及其调控效果进行了详细分析，重点提出了一种基于末端设备轮停的公共楼宇中央空调系统降负荷控制策略。.3）在考虑中央空调系统制冷主机的运行效率的基础上，研究了单幢公共楼宇的中央空调系统可调度容量的计算方法；为了便于规模化公共楼宇空调负荷的统一调控，分别提出了非蓄冷式和蓄冷式中央空调系统的分档调控方案。.4）研究了大规模公共楼宇空调负荷参与电网调峰的组合调度模型及约束条件；基于对公共楼宇空调负荷采用的直接负荷控制补偿机制，提出了计及大规模空调负荷和常规调峰机组共同参与的最佳调峰组合方式。.5）结合传统发电机组模型，以系统运行总成本最小化为目标建立了公共楼宇中央空调负荷广泛参与下的电网日前调度模型，并以南京市某区域电网为例对模型的调峰效果和经济性进行了仿真验证。阐述了可控负荷分层思想和对于微电网紧急控制的作用，提出了公共楼宇内基于可控负荷分层的微电网紧急控制优化模型及方法。.6）应用递阶层次结构，从技术、经济和环保等三个方面，建立了空调负荷参与虚拟调峰响应指标体系，并应用层次分析法和模糊综合评价法求解。