技术领域Technical Field
本发明属于计算机技术领域，具体涉及一种冰雪运动场馆的典型负荷场景生成方法及系统。The present invention belongs to the field of computer technology, and in particular relates to a method and system for generating typical load scenarios of ice and snow sports venues.
背景技术Background Art
随着社会经济的发展，负荷种类的多样性急剧增加，伴随而来的不同负荷特性对于电网运行生产带来很大的困难。所以需要构建典型负荷场景，以辅助调度部门制定合理的运行方案，有利于系统安全稳定运行。一般区域性的典型负荷场景生成都是对大量原始负荷场景进行分类处理获得的，这些区域性的负荷存在一定的规律特征，例如商业负荷在营业时间内会形成较高的平段负荷，休息时间又会形成较小的低谷负荷，峰谷差及峰谷差率较大，负荷率相对较低。居民住宅负荷在工作日也会存在早晚双峰类型的负荷特性。由此可见，常见的商业负荷与居民住宅负荷都受到工作生活中的固定行为而产生了一定的规律。此时采用分类方法可以较好的提取出典型负荷场景。With the development of social economy, the diversity of load types has increased dramatically, and the different load characteristics that come with it have brought great difficulties to the operation and production of power grids. Therefore, it is necessary to construct typical load scenarios to assist the dispatching department in formulating reasonable operation plans, which is conducive to the safe and stable operation of the system. Generally, the generation of typical regional load scenarios is obtained by classifying and processing a large number of original load scenarios. These regional loads have certain regular characteristics. For example, commercial loads will form a higher flat load during business hours, and a smaller valley load during rest time. The peak-to-valley difference and peak-to-valley difference rate are large, and the load rate is relatively low. Residential loads will also have morning and evening double peak load characteristics on weekdays. It can be seen that common commercial loads and residential loads are subject to fixed behaviors in work and life and have certain regularities. At this time, the classification method can better extract typical load scenarios.
但是冰雪运动场馆的负荷受到赛事安排的影响较大，而赛事安排不存在较强的规律性，所以导致冰雪运动场馆的典型负荷场景生成不能简单的通过分类方法筛选出来。冰雪运动场馆存在不同的负荷类型，包括场馆照明用电、空调用电、索道缆车用电、新风机组用电、安保用电、电梯用电、通信设备用电、电动汽车充电负荷、造雪造冰设备负荷。其中电动汽车充电负荷、造雪造冰设备负荷受到赛事安排的影响较大，赛事期间场馆会有大量的运动选手、观众和场馆运行保障人员流动，所以电动汽车充电负荷会较高。赛事期间场馆造雪造冰设备也会根据赛事安排为竞赛场地提前进行造雪造冰的准备工作。并且电动汽车充电负荷与造雪造冰设备负荷占冰雪运动场馆的总负荷比例较大，其不确定性与重要性决定了冰雪运动场馆的典型场景生成不适用于常规的分类方法。However, the load of ice and snow sports venues is greatly affected by the arrangement of events, and there is no strong regularity in the arrangement of events, so the generation of typical load scenarios of ice and snow sports venues cannot be simply screened out through classification methods. There are different types of loads in ice and snow sports venues, including venue lighting electricity, air conditioning electricity, cable car electricity, fresh air unit electricity, security electricity, elevator electricity, communication equipment electricity, electric vehicle charging load, and snow and ice making equipment load. Among them, the electric vehicle charging load and snow and ice making equipment load are greatly affected by the arrangement of events. During the event, there will be a large number of athletes, spectators and venue operation and support personnel flowing in the venue, so the electric vehicle charging load will be higher. During the event, the venue's snow and ice making equipment will also prepare for snow and ice making in advance for the competition venue according to the event arrangement. In addition, the electric vehicle charging load and snow and ice making equipment load account for a large proportion of the total load of the ice and snow sports venues. Their uncertainty and importance determine that the generation of typical scenarios for ice and snow sports venues is not suitable for conventional classification methods.
常规的典型场景生成方法分为两类，包括场景缩减与场景生成两大类。场景缩减方法主要是对区域内负荷场景采用各种聚类算法、场景树法、后向缩减法等生成典型负荷场景，适用于具有一定规律性的负荷场景。场景生成方法通常采用ARMA、概率建模、马尔科夫链、神经网络等算法，通过学习负荷场景存在的共性，生成大量负荷场景，再对生成的场景进行削减。可以适用于具有不确定性特征的负荷场景的生成，但是该方法较为繁琐，通过计算机实现上述场景生成方法时，速度慢、计算时间较长。Conventional typical scenario generation methods are divided into two categories, including scenario reduction and scenario generation. The scenario reduction method mainly uses various clustering algorithms, scenario tree methods, backward reduction methods, etc. to generate typical load scenarios for load scenarios in the region, which is suitable for load scenarios with certain regularity. The scenario generation method usually uses algorithms such as ARMA, probability modeling, Markov chain, neural network, etc. to generate a large number of load scenarios by learning the commonalities of load scenarios, and then reduce the generated scenarios. It can be applied to the generation of load scenarios with uncertain characteristics, but this method is relatively cumbersome. When the above scenario generation method is implemented by computer, the speed is slow and the calculation time is long.
发明内容Summary of the invention
本发明的目的在于提供一种冰雪运动场馆的典型负荷场景生成方法及系统，以解决现有技术中，通过计算机实现背景技术中的场景生成方法时，速度慢、计算时间较长的问题。The purpose of the present invention is to provide a typical load scenario generation method and system for ice and snow sports venues, so as to solve the problems of slow speed and long calculation time when the scenario generation method in the background technology is implemented by computer in the prior art.
为实现上述目的，采用如下技术方案：In order to achieve the above purpose, the following technical solutions are adopted:
一种冰雪运动场馆的典型负荷场景生成方法，包括如下步骤：A typical load scenario generation method for an ice and snow sports venue comprises the following steps:
将冰雪运动场馆的负荷类型分为不确定性负荷和基础负荷；The load types of ice and snow sports venues are divided into uncertain loads and basic loads;
对不确定性负荷进行建模，并通过蒙特卡罗法模拟生成不确定性负荷日场景集；Model the uncertain loads and generate a set of uncertain load daily scenarios through Monte Carlo simulation;
将基础负荷的功率数据时序叠加，构建基础负荷日场景集，采用近邻传播聚类算法对基础负荷日场景集进行聚类分析，获得典型基础负荷日场景集；The power data time series of the base load are superimposed to construct a base load daily scenario set, and the base load daily scenario set is clustered and analyzed using the nearest neighbor propagation clustering algorithm to obtain a typical base load daily scenario set;
对不确定性负荷日场景集和典型基础负荷日场景集进行组合与数据叠加，构建出涵盖不确定性负荷和基础负荷的组合叠加日场景集；The uncertain load daily scenario set and the typical basic load daily scenario set are combined and superimposed to construct a combined superimposed daily scenario set covering uncertain load and basic load;
采用K-means聚类算法对组合叠加日场景集进行聚类分析，生成典型负荷日场景集，依据该典型负荷日场景集构建冰雪运动场馆的典型负荷场景。The K-means clustering algorithm is used to perform cluster analysis on the combined superimposed daily scene set to generate a typical load day scene set. Based on this typical load day scene set, the typical load scenario of ice and snow sports venues is constructed.
进一步的，所述不确定性负荷包括电动汽车负荷和造雪造冰设备；所述基础负荷包含场馆照明用电、空调用电、索道缆车用电、新风机组用电、安保用电、电梯用电、通信设备用电。Furthermore, the uncertain load includes electric vehicle load and snow and ice making equipment; the basic load includes electricity for venue lighting, air conditioning, cable car, fresh air unit, security, elevator, and communication equipment.
进一步的，电动汽车负荷包括小型家用电动汽车充电负荷和大型电动巴士的充电负荷；Furthermore, the electric vehicle load includes the charging load of small household electric vehicles and the charging load of large electric buses;
1)小型家用电动汽车充电负荷模型表示为PEV,car(t)；1) The charging load model of a small household electric vehicle is expressed as PEV,car (t);
PEV,car(t)＝Ncar,tpcarQcar(St＝1)PEV,car (t)=Ncar,t pcar Qcar (St =1)
式中：Ncar,t代表t时刻内小型家用电动汽车数量；pcar为单辆小型家用电动汽车的充电功率；Qcar(St＝1)为小型家用电动汽车处于充电状态的概率；Where: Ncar,t represents the number of small household electric vehicles at time t; pcar is the charging power of a single small household electric vehicle; Qcar (St = 1) is the probability that a small household electric vehicle is in a charging state;
2)大型电动巴士的充电负荷模型表示为PEV,bus(t)：2) The charging load model of a large electric bus is expressed as PEV,bus (t):
PEV,bus(t)＝Nbus,tpbusQbus(St＝1)PEV,bus (t)＝Nbus,t pbus Qbus (St ＝1)
式中：Nbus,t代表t时刻内电动公交汽车数量；pbus为单辆电动公交汽车的充电功率；Qbus(St＝1)为电动公交汽车处于充电状态的概率。Where: Nbus,t represents the number of electric buses at time t; pbus is the charging power of a single electric bus; Qbus (St = 1) is the probability that the electric bus is in the charging state.
进一步的，采用蒙特卡罗法模拟采样生成小型家用电动汽车充电负荷日场景sccar，1＝(PEV，car(1),PEV，car(22),…PEV,car(96)和大型电动巴士的充电负荷日场景scbus，1＝(PEV，bus(1),PEV，bus(2),…PEV,bus(96))并重复模拟Nsc,car次构建小型家用电动汽车充电负荷日场景集重复模拟Nsc,bus次构建大型电动巴士的充电负荷日场景集Nsc,car为根据小型家用电动汽车充电负荷模型随机抽样生成的小型家用电动汽车充电负荷场景的数量；Nsc,bus为根据大型电动巴士的充电负荷模型随机抽样生成的大型电动巴士充电负荷场景的数量。Furthermore, the Monte Carlo method is used to simulate and sample the small household electric vehicle charging load daily scenario sccar, 1 = (PEV, car (1), PEV, car (22), ... PEV, car (96) and the large electric bus charging load daily scenario scbus, 1 = (PEV, bus (1), PEV, bus (2), ... PEV, bus (96)) and repeat the simulation Nsc, car times to construct the small household electric vehicle charging load daily scenario set Repeat the simulation Nsc,bus times to build a daily scenario set for charging load of large electric buses Nsc,car is the number of small household electric vehicle charging load scenarios randomly sampled from the small household electric vehicle charging load model; Nsc,bus is the number of large electric bus charging load scenarios randomly sampled from the large electric bus charging load model.
进一步的，所述造雪造冰设备负荷模型Pice(t)表示为：Furthermore, the snow and ice making equipment load model Pice (t) is expressed as:
式中Pice(t)为t时段所有造雪设备的总功率，t＝1，2，...96；Pice,k(t)为第k组造雪设备在t时段的功率，k＝1，2，…nice；nice为造雪设备的总组数，将在同一区域内的造雪造冰的负荷定义为一组造雪设备。In the formula, Pice (t) is the total power of all snow-making equipment in period t, t = 1, 2, ... 96; Pice,k (t) is the power of the kth group of snow-making equipment in period t, k = 1, 2, ...nice ;nice is the total number of groups of snow-making equipment, and the snow-making and ice-making load in the same area is defined as a group of snow-making equipment.
进一步的，利用历史数据构建出每组造雪设备在不同时段负荷的通用概率分布模型fice(kt)，在此基础上采用蒙特卡洛法抽样模拟生成第k组造雪设备在t时段的功率Pice,k(t)；Furthermore, the historical data are used to construct a general probability distribution model fice (kt ) of the load of each group of snowmaking equipment in different time periods. On this basis, the Monte Carlo method is used to sample and simulate the power Pice,k (t) of the kth group of snowmaking equipment in time period t.
式中kt代表t时刻内造雪设备k，αice,k、βice,k和γice,k是概率分布模型fice(kt)的形状参数，取值范围分别为αice,k>0，βice,k>0，-∞<γice,k<∞；Where kt represents the snowmaking equipment k at time t, αice,k , βice,k and γice,k are the shape parameters of the probability distribution model fice (kt ), and their value ranges are αice,k >0, βice,k >0, -∞<γice,k <∞ respectively;
采用蒙特卡罗法模拟采样生成造雪造冰设备充电负荷日场景scice，1＝(Pice(1)，Pice(2)，…Pice(96))；并重复模拟Nsc,ice次构建造雪造冰设备负荷日场景集Nsc,ice为根据造雪造冰设备负荷模型随机抽样生成的造雪造冰设备负荷场景数量。The Monte Carlo method is used to simulate and sample the daily scenario of snow-making and ice-making equipment charging load scice,1 = (Pice (1), Pice (2), ... Pice (96)); and the simulation is repeated Nsc,ice times to construct the daily scenario set of snow-making and ice-making equipment load. Nsc,ice is the number of snow-making and ice-making equipment load scenarios generated by random sampling based on the snow-making and ice-making equipment load model.
进一步的，将基础负荷包含的所有类型负荷日场景按照时序对应进行叠加，生成基础负荷日场景scbase，1＝(pbase(1),pbase(2),…pbase(96))Furthermore, all types of load daily scenarios included in the base load are superimposed in time sequence to generate the base load daily scenario scbase, 1 = (pbase (1), pbase (2), ... pbase (96))
将基础负荷日场景构建成负荷日场景集Nsc,base代表基础负荷日场景的数量；Construct the base load day scenario into a load day scenario set Nsc,base represents the number of base load day scenarios;
采用近邻传播聚类算法对基础负荷日场景集SCbase进行聚类分析，获得典型基础负荷日场景kbase是由近邻传播聚类算法自动生成的最优数量；获得典型基础负荷日场景集The nearest neighbor propagation clustering algorithm is used to perform cluster analysis on the base load daily scenario set SCbase to obtain the typical base load daily scenario kbase is the optimal number automatically generated by the neighbor propagation clustering algorithm; obtain the typical base load daily scenario set
进一步的，将生成的电动汽车负荷日场景集以及造雪造冰设备负荷日场景集典型基础负荷日场景集进行随机组合，并按时序将负荷数据叠加，构建出涵盖冰雪运动场馆所有用电类型的电动汽车、造雪造冰设备、基础负荷的组合叠加日场景集SC＝(sc1，sc2，…，sci，…scn)，i∈n，n代表组合叠加日场景的数量n＝Nsc,ice×Nsc,car×Nsc,bus×kbase。Furthermore, the generated electric vehicle load daily scenario set And snow and ice making equipment load day scenario set Typical base load day scenario set Random combinations are performed and the load data are superimposed in time series to construct a combined superposition daily scenario set SC = (sc1 , sc2 , … , sci , … scn ) covering all electricity consumption types of ice and snow sports venues, including electric vehicles, snow and ice making equipment, and basic loads, where i∈n, n represents the number of combined superposition daily scenarios n = Nsc,ice ×Nsc,car ×Nsc,bus ×kbase .
进一步的，采用K-means聚类算法对电动汽车、造雪造冰设备、基础负荷的组合叠加日场景集SC＝(sc1，sc2，…，sci，…scn)进行聚类分析，共分成K类，提取出具有代表性的冰雪运动场馆典型负荷日场景集：Furthermore, the K-means clustering algorithm is used to perform cluster analysis on the combined superposition daily scenario set SC = (sc1 , sc2 , …, sci , …scn ) of electric vehicles, snow and ice making equipment, and basic loads, and a total of K categories are divided to extract a representative daily scenario set of typical loads of ice and snow sports venues:
SCtypical＝(sc1,sc2,…,scK)，1<K<nSCtypical =(sc1 ,sc2 ,…,scK ), 1<K<n
K为预设值。K is the preset value.
本发明提供的另一个技术方案是：Another technical solution provided by the present invention is:
一种用于所述典型负荷场景生成方法的系统，包括：A system for the typical load scenario generation method, comprising:
负荷分类模块，用于将冰雪运动场馆的负荷类型分为不确定性负荷和基础负荷；Load classification module, used to classify the load types of ice and snow sports venues into uncertain loads and basic loads;
负荷日场景集生成模块，用于分别生成不确定性负荷和基础负荷的日场景集；对不确定性负荷进行建模，并通过蒙特卡罗法模拟生成不确定性负荷日场景集；The load daily scenario set generation module is used to generate daily scenario sets of uncertain load and basic load respectively; model the uncertain load and generate the uncertain load daily scenario set through Monte Carlo simulation;
基础负荷日场景集生成模块，用于将基础负荷的功率数据时序叠加，构建基础负荷日场景集，采用近邻传播聚类算法对基础负荷日场景集进行聚类分析，获得典型基础负荷日场景集；The basic load daily scenario set generation module is used to superimpose the power data time series of the basic load to construct the basic load daily scenario set, and use the nearest neighbor propagation clustering algorithm to perform cluster analysis on the basic load daily scenario set to obtain a typical basic load daily scenario set;
日场景集叠加模块，用于对不确定性负荷日场景集和典型基础负荷日场景集进行组合与数据叠加，构建出涵盖不确定性负荷和基础负荷的组合叠加日场景集；The daily scenario set superposition module is used to combine and superimpose the uncertainty load daily scenario set and the typical basic load daily scenario set to construct a combined superposition daily scenario set covering the uncertainty load and the basic load;
典型负荷日场景集模块，用于采用K-means聚类算法对组合叠加日场景集进行聚类分析，生成典型负荷日场景集，依据该典型负荷日场景集构建冰雪运动场馆的典型负荷场景。The typical load day scenario set module is used to perform cluster analysis on the combined superposition day scenario set using the K-means clustering algorithm to generate a typical load day scenario set, and construct a typical load scenario for ice and snow sports venues based on the typical load day scenario set.
本发明技术方案带来的有益效果如下：The beneficial effects brought by the technical solution of the present invention are as follows:
本发明方法将场景缩减与场景生成两种方法相结合。针对不确定性负荷采用场景生成方法，构造大量不确定性负荷场景。针对基础负荷采用聚类算法，筛选出具有代表性的规律性典型负荷场景。将两类场景进行组合叠加，构造出能涵盖所有可能性的冰雪运动场馆场景集合，最后通过聚类算法提取出具有代表性的冰雪运动场馆典型负荷场景，提升了计算机计算效率，节省了时间，并且为调度和运维人员制定应急预案提供了参考依据，从而保障重要赛事期间用电设备的安全稳定运行。并且目前为止该研究领域没有类似的多类型负荷场景组合叠加聚类的典型场景生成方法。The method of the present invention combines the two methods of scenario reduction and scenario generation. The scenario generation method is used for uncertain loads to construct a large number of uncertain load scenarios. The clustering algorithm is used for basic loads to screen out representative regular typical load scenarios. The two types of scenarios are combined and superimposed to construct a set of ice and snow sports venue scenarios that can cover all possibilities. Finally, representative ice and snow sports venue typical load scenarios are extracted through the clustering algorithm, which improves computer computing efficiency, saves time, and provides a reference for dispatching and operation and maintenance personnel to formulate emergency plans, thereby ensuring the safe and stable operation of electrical equipment during important events. And so far, there is no similar typical scenario generation method for combining and superimposing clusters of multiple types of load scenarios in this research field.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
构成本申请的一部分的说明书附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
图1为本发明实施例中典型负荷场景生成方法的流程示意图。FIG1 is a flow chart of a typical load scenario generation method according to an embodiment of the present invention.
具体实施方式DETAILED DESCRIPTION
下面将参考附图并结合实施例来详细说明本发明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.
以下详细说明均是示例性的说明，旨在对本发明提供进一步的详细说明。除非另有指明，本发明所采用的所有技术术语与本申请所属领域的一般技术人员的通常理解的含义相同。本发明所使用的术语仅是为了描述具体实施方式，而并非意图限制根据本发明的示例性实施方式。The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.
本发明实施例提供了一种冰雪运动场馆的典型负荷场景生成方法及系统，如图1所示，冰雪运动场馆的典型负荷场景生成方法，主要分为四部分。第一部分是对的不同回路所采集的负荷类型进行分类，并将其分为不确定性负荷和基础负荷两类。第二部分是分别生成两类负荷的日场景集，对不确定性负荷包含的电动汽车负荷和造雪造冰设备负荷进行建模，并通过蒙特卡罗法模拟生成电动汽车负荷日场景集和造雪造冰设备负荷日场景集。对基础负荷所包含的常规负荷，例如场馆照明用电、空调用电、索道缆车用电、新风机组用电、安保用电、电梯用电、通信设备用电等具有一定使用规律的公共服务用电称为基础负荷，其功率数据可以通过相应的回路信息进行采集。将所有采集到的基础负荷的功率数据时序叠加，构建基础负荷日场景集。在此基础上采用近邻传播聚类算法对基础负荷日场景集进行聚类分析，获得典型基础负荷日场景集。第三部分是对电动汽车负荷日场景集、造雪造冰设备负荷日场景集和基础负荷日场景集进行组合与数据叠加。构建出涵盖冰雪运动场馆所有用电类型的电动汽车+造雪造冰设备+基础负荷的组合叠加日场景集。第四部分是采用K-means聚类算法对大量的电动汽车+造雪造冰设备+基础负荷的组合叠加日场景进行聚类分析，生成冰雪运动场馆典型负荷日场景集。The embodiment of the present invention provides a typical load scenario generation method and system for ice and snow sports venues. As shown in FIG1 , the typical load scenario generation method for ice and snow sports venues is mainly divided into four parts. The first part is to classify the load types collected by different loops and divide them into two categories: uncertain loads and basic loads. The second part is to generate daily scenario sets for the two types of loads respectively, model the electric vehicle load and snow and ice making equipment load contained in the uncertain load, and generate the electric vehicle load daily scenario set and the snow and ice making equipment load daily scenario set by Monte Carlo simulation. The conventional loads contained in the basic load, such as the electricity for venue lighting, air conditioning, cable car, fresh air unit, security, elevator, communication equipment, etc., which have certain usage rules, are called basic loads. Their power data can be collected through the corresponding loop information. The power data of all collected basic loads are superimposed in time series to construct a basic load daily scenario set. On this basis, the neighbor propagation clustering algorithm is used to perform cluster analysis on the basic load daily scenario set to obtain a typical basic load daily scenario set. The third part is to combine and superimpose the daily scenario set of electric vehicle load, the daily scenario set of snow and ice making equipment load and the daily scenario set of basic load. A combined superimposed daily scenario set of electric vehicles + snow and ice making equipment + basic load covering all types of electricity consumption in ice and snow sports venues is constructed. The fourth part is to use the K-means clustering algorithm to perform cluster analysis on a large number of combined superimposed daily scenarios of electric vehicles + snow and ice making equipment + basic load to generate a typical load daily scenario set for ice and snow sports venues.
下面结合具体的实施例子进行进一步的阐述：The following is further explained in conjunction with specific implementation examples:
S1、对冰雪运动场馆的不同回路所采集的负荷类型进行分类，并将其分为不确定性负荷和基础负荷两类。其中，不确定性负荷由电动汽车负荷和造雪造冰设备负荷组成，因为不确定性负荷受个人使用情况和赛事安排的影响较大，由于冰雪运动场馆赛时主要由运动员、观众和保障人员组成，所以电动汽车负荷主要考虑小型家用电动汽车和大型电动巴士。造雪造冰设备负荷主要考虑水泵和造雪造冰设备的负荷。基础负荷包含场馆照明用电、空调用电、索道缆车用电、新风机组用电、安保用电、电梯用电、通信设备用电等具有一定使用规律的公共服务用电。S1. Classify the load types collected from different circuits of ice and snow sports venues, and divide them into two categories: uncertain load and basic load. Among them, the uncertain load is composed of electric vehicle load and snow and ice making equipment load. Because the uncertain load is greatly affected by personal usage and event arrangements, and because ice and snow sports venues are mainly composed of athletes, spectators and support personnel during competitions, the electric vehicle load mainly considers small household electric vehicles and large electric buses. The snow and ice making equipment load mainly considers the load of water pumps and snow and ice making equipment. The basic load includes public service electricity with certain usage patterns, such as venue lighting electricity, air conditioning electricity, cable car electricity, fresh air unit electricity, security electricity, elevator electricity, and communication equipment electricity.
S2、对不确定性负荷包含的小型家用电动汽车充电负荷和大型电动巴士的充电负荷分别进行建模。S2. Model the charging load of small household electric vehicles and large electric buses included in the uncertain load separately.
1)小型家用电动汽车充电负荷建模：1) Modeling of charging load for small household electric vehicles:
充电起始时间是小型家用电动汽车充电负荷建模的重要影响因素。每日最后一次行程结束时刻为小型家用电动汽车最早充电起始时刻，The charging start time is an important factor affecting the charging load modeling of small household electric vehicles. The end time of the last trip of the day is the earliest charging start time for small household electric vehicles.
小型家用电动汽车充电负荷建模中，何时开始充电与何时结束充电对模型建立至关重要。通常情况下我们将每天最后一次使用电动汽车的结束时刻认定为充电的起始时刻。通过统计全年数据的概率分布函数拟合为fend,car(tcar)In the modeling of charging load for small household electric vehicles, when to start charging and when to end charging are crucial to model building. Usually, we consider the end time of the last use of the electric vehicle every day as the start time of charging. The probability distribution function of the statistical data throughout the year is fitted as fend,car (tcar )
式中，tcar表示电动汽车的结束时间。μend,car与σend,car可以通过拟合大量小型家用电动汽车的结束时刻数据获得。Where tcar represents the end time of the electric vehicle.μ end,car and σend,car can be obtained by fitting the end time data of a large number of small household electric vehicles.
充电时长同样为小型家用电动汽车充电负荷建模的重要因素，通常情况下我们把小型家用电动汽车每天行驶的里程总和作为该日行驶的里程，通过对大量经验数据的研究，搭建了行驶里程的指数型概率分布模型fcar(scar)Charging time is also an important factor in modeling the charging load of small household electric vehicles. Usually, we take the total mileage of small household electric vehicles per day as the mileage of that day. Through the study of a large amount of empirical data, we build an exponential probability distribution model of mileage fcar (scar )
式中scar代表小型家用电动汽车该日的行驶里程，μcar,mile是统计数据形成的模型参数，其数值由历史数据拟合生成。Where scar represents the mileage of a small household electric car on that day, μcar,mile is a model parameter formed by statistical data, and its value is generated by fitting historical data.
充电功率还受到小型家用电动汽车数量的影响，但是由于充电数量具有随机性，我们可以利用历史数据构建出每个时间段内汽车数量的概率分布模型fcar(ncar,t)The charging power is also affected by the number of small household electric vehicles, but since the number of charging vehicles is random, we can use historical data to construct a probability distribution model for the number of vehicles in each time period: fcar (ncar,t )
式中ncar,t代表t时刻内小型家用电动汽车数量分布，αcar、βcar和γcar是概率分布模型fcar(ncar,t)的形状参数，取值范围分别为αcar>0，βcar>0，-∞<γcar<∞。具体取值由历史数据拟合生成。Where ncar,t represents the number distribution of small household electric vehicles at time t, αcar , βcar and γcar are the shape parameters of the probability distribution model fcar (ncar,t ), and their value ranges are αcar >0, βcar >0, -∞<γcar <∞. The specific values are generated by fitting historical data.
在行驶里程和充电起始时间相互独立的情况下建立小型家用电动汽车充电负荷模型PEV,car(t)The charging load model PEV,car (t) of a small household electric vehicle is established when the driving mileage and charging start time are independent of each other.
PEV,car(t)＝Ncar,tpcarQcar(St＝1)PEV,car (t)=Ncar,t pcar Qcar (St =1)
式中Ncar,t代表t时刻内小型家用电动汽车数量。pcar为单辆小型家用电动汽车的充电功率(kW)；Qcar(St＝1)为小型家用电动汽车处于充电状态的概率。Where Ncar,t represents the number of small household electric vehicles at time t. pcar is the charging power (kW) of a single small household electric vehicle; Qcar (St = 1) is the probability that a small household electric vehicle is in a charging state.
式中fend与fT分别为行程结束时刻和小型家用电动汽车充电时长的概率密度函数；Tmax为充电时长T的积分上限，取Tmax＝10h，scar为日行驶里程(km)；ccar为小型家用电动汽车行驶每千米的耗电量(kWh/km)Where fend and fT are the probability density functions of the end time of the trip and the charging time of a small household electric vehicle, respectively; Tmax is the integral upper limit of the charging time T, Tmax = 10h, scar is the daily mileage (km); ccar is the power consumption of a small household electric vehicle per kilometer (kWh/km)
在时刻t内充电的小型家用电动汽车数量是随机的，所以采用时刻t内小型家用电动汽车的期望值作为数量Ncar,tThe number of small household electric vehicles charged at time t is random, so the expected value of small household electric vehicles at time t is used as the number Ncar,t
2)大型电动巴士的充电负荷模型：2) Charging load model for large electric buses:
大型电动巴士与小型家用电动汽车的建模流程基本一致，但是由于大型电动巴士属于商业运行车辆，所以在充电起始时刻概率分布模型的参数、行驶里程的概率分布模型的参数上存在区别。The modeling process of large electric buses and small household electric vehicles is basically the same. However, since large electric buses are commercial vehicles, there are differences in the parameters of the probability distribution model of the charging start time and the parameters of the probability distribution model of the mileage.
大型电动巴士的充电起始时刻概率分布函数拟合为fend,bus(tbus)The probability distribution function of the charging start time of a large electric bus is fitted as fend,bus (tbus )
式中，tbus表示电动巴士的结束时间。μend,bus与σend,bus可以通过拟合大量电动巴士的结束时刻数据获得。Where tbus represents the end time of the electric bus.μ end,bus and σend,bus can be obtained by fitting the end time data of a large number of electric buses.
电动巴士行驶里程的概率分布模型fbus(sbus)服从正态分布：The probability distribution model of electric bus mileage fbus (sbus ) follows a normal distribution:
式中sbus代表电动巴士该日的行驶里程，σs与μs是统计数据形成的模型参数，其数值由历史数据拟合生成。Where sbus represents the mileage of the electric bus on that day, σs and μs are model parameters formed by statistical data, and their values are generated by fitting historical data.
充电功率还受到电动巴士数量的影响，但是由于充电数量具有随机性，我们可以利用历史数据构建出每个时间段内汽车数量的概率分布模型fbus(nbus,t)The charging power is also affected by the number of electric buses, but since the number of electric buses is random, we can use historical data to construct a probability distribution model for the number of cars in each time period: fbus (nbus,t )
式中nbus,t代表t时刻内电动巴士的数量分布，αbus、βbus和γbus是概率分布模型fbus(nbus,t)的形状参数，取值范围分别为αbus>0，βbus>0，-∞<γbus<∞。具体取值由历史数据拟合生成。Where nbus,t represents the number distribution of electric buses at time t, αbus , βbus and γbus are the shape parameters of the probability distribution model fbus (nbus,t ), and their value ranges are αbus >0, βbus >0, -∞<γbus <∞. The specific values are generated by fitting historical data.
在行驶里程和充电起始时间相互独立的情况下建立电动公交汽车充电负荷模型PEV,bus(t)Establish the charging load model of electric bus PEV,bus (t) when the driving mileage and charging start time are independent of each other
PEV，bus(t)＝Nbus，tpbusQbus(St＝1)PEV, bus (t) = Nbus, t pbus Qbus (St = 1)
式中Nbus，t代表t时刻内电动公交汽车数量。pbus为单辆电动公交汽车的充电功率(kW)；Qbus(St＝1)为电动公交汽车处于充电状态的概率。Where Nbus,t represents the number of electric buses at time t. pbus is the charging power of a single electric bus (kW); Qbus (St = 1) is the probability that the electric bus is in the charging state.
式中fend与fT分别为行程结束时刻和电动公交汽车充电时长的概率密度函数；Tmax为充电时长T的积分上限，取Tmax＝16h，sbus为日行驶里程(km)；cbus为电动公交汽车行驶每千米的耗电量(kWh/km)Where fend and fT are the probability density functions of the end time of the trip and the charging time of the electric bus, respectively; Tmax is the integral upper limit of the charging time T, Tmax = 16h, sbus is the daily mileage (km); cbus is the power consumption of the electric bus per kilometer (kWh/km)
在时刻t内充电的电动公交汽车数量是随机的，所以采用时刻t内电动公交汽车的期望值作为数量Nbus，tThe number of electric buses charged at time t is random, so the expected value of electric buses at time t is used as the number Nbus, t
通常情况下，历史数据的采集频率为15min一次，一天包含96个数据。所以，采用蒙特卡罗法模拟采样生成小型家用电动汽车充电负荷日场景sccar，1＝(PEV，car(1)，PEV，car(2)，…PEV，car(96))和大型电动巴士的充电负荷日场景scbus，1＝(PEV，bus(1)，PEV，bus(2)，…PEV，bus(96))并重复模拟Nsc，car次构建小型家用电动汽车充电负荷日场景集和模拟Nsc，bus次大型电动巴士的充电负荷日场景集Nsc，car为根据小型家用电动汽车充电负荷模型随机抽样生成的小型家用电动汽车充电负荷场景的数量。Nsc，bus为根据大型电动巴士的充电负荷模型随机抽样生成的大型电动巴士充电负荷场景的数量。Normally, the frequency of historical data collection is once every 15 minutes, and one day contains 96 data. Therefore, the Monte Carlo method is used to simulate and sample to generate the daily charging load scenario of small household electric vehicles sccar, 1 = (PEV, car (1), PEV, car (2), ... PEV, car (96)) and the daily charging load scenario of large electric buses scbus, 1 = (PEV, bus (1), PEV, bus (2), ... PEV, bus (96)) and repeat the simulation Nsc, car times to construct the daily charging load scenario set of small household electric vehicles. and simulated Nsc, bus charging load daily scenario set for large electric buses Nsc,car is the number of small household electric vehicle charging load scenarios randomly sampled based on the small household electric vehicle charging load model. Nsc,bus is the number of large electric bus charging load scenarios randomly sampled based on the large electric bus charging load model.
S3、对不确定性负荷包含的造雪造冰设备负荷进行建模。造雪造冰设备负荷主要包括水泵、造雪机、氨制冷机。由于造雪造冰设备负荷受赛事安排影响较大，而赛事安排是人为制定的，不确定性较大，需要进行概率性建模。将每一组造雪设备的负荷曲线累加可得到总充电负荷曲：S3. Model the load of snow-making and ice-making equipment included in the uncertain load. The load of snow-making and ice-making equipment mainly includes water pumps, snow-making machines, and ammonia refrigerators. Since the load of snow-making and ice-making equipment is greatly affected by the event arrangement, and the event arrangement is artificially made and has a large uncertainty, probabilistic modeling is required. The total charging load curve can be obtained by summing up the load curve of each group of snow-making equipment:
式中Pice(t)为t时段所有造雪设备的总功率，t＝1，2，…96。Pice,k(t)为第k组造雪设备在t时段的功率,k＝1，2，…nice。nice为造雪设备的总组数，我们将在同一区域内的造雪造冰的负荷定义为一组造雪设备。我们可以利用历史数据构建出每组造雪设备在不同时段负荷的通用概率分布模型fice(kt)，在此基础上采用蒙特卡洛法抽样模拟生成Pice,k(t)。Where Pice (t) is the total power of all snowmaking equipment in period t, t = 1, 2, ... 96. Pice,k (t) is the power of the kth group of snowmaking equipment in period t, k = 1, 2, ... nice . nice is the total number of snowmaking equipment groups. We define the snowmaking and icemaking load in the same area as a group of snowmaking equipment. We can use historical data to construct a general probability distribution model fice (kt ) of the load of each group of snowmaking equipment in different periods, and on this basis, use the Monte Carlo method to sample and simulate to generate Pice,k (t).
式中kt代表t时刻内造雪设备k，αice,k、βice,k和γice,k是概率分布模型fice(kt)的形状参数，取值范围分别为αice,k>0，βice,k>0，-∞<γice,k<∞。具体取值由历史数据拟合生成。Where kt represents the snowmaking equipment k at time t, αice,k , βice,k and γice,k are the shape parameters of the probability distribution model fice (kt ), and their value ranges are αice,k >0, βice,k >0, -∞<γice,k <∞. The specific values are generated by fitting historical data.
采用蒙特卡罗法模拟采样生成造雪造冰设备充电负荷日场景scice，1＝(Pice(1),Pice(2),…Pice(96))并重复模拟Nsc,ice次构建造雪造冰设备负荷日场景集Nsc,ice为根据造雪造冰设备负荷模型随机抽样生成的造雪造冰设备负荷场景数量。The Monte Carlo method is used to simulate and sample the daily scenario of snow and ice making equipment charging load scice,1 = (Pice (1), Pice (2), ... Pice (96)) and the simulation is repeated Nsc,ice times to construct the daily scenario set of snow and ice making equipment load. Nsc,ice is the number of snow-making and ice-making equipment load scenarios generated by random sampling based on the snow-making and ice-making equipment load model.
S4、基础负荷所包含的场馆照明用电、空调用电、索道缆车用电、新风机组用电、安保用电、电梯用电、通信设备用电的功率数据可以通过相应的回路信息进行采集。将基础负荷包含的所有类型负荷日场景按照时序对应进行叠加，生成基础负荷日场景scbase,1＝(pbase(1),pbase(2),…pbase(96))将大量基础负荷日场景构建成负荷日场景集Nsc,base代表基础负荷日场景的数量。并采用近邻传播聚类算法对基础负荷日场景集SCbase进行聚类分析，获得典型基础负荷日场景kbase是由近邻传播聚类算法自动生成的最优数量。获得典型基础负荷日场景集S4. The power data of venue lighting, air conditioning, cable car, fresh air unit, security, elevator, and communication equipment included in the basic load can be collected through the corresponding loop information. All types of load daily scenarios included in the basic load are superimposed according to the time sequence to generate the basic load daily scenario scbase,1 = (pbase (1), pbase (2), ... pbase (96)) A large number of basic load daily scenarios are constructed into a load daily scenario set Nsc,base represents the number of base load daily scenarios. The nearest neighbor propagation clustering algorithm is used to perform cluster analysis on the base load daily scenario set SCbase to obtain the typical base load daily scenario kbase is the optimal number automatically generated by the proximity propagation clustering algorithm. Get the typical base load daily scenario set
S5、将生成的电动汽车负荷日场景集造雪造冰设备负荷日场景集和典型基础负荷日场景集进行随机组合，并按时序将负荷数据叠加。构建出涵盖冰雪运动场馆所有用电类型的电动汽车+造雪造冰设备+基础负荷的组合叠加日场景集SC＝(sc1，sc2，…，sci，…scn)i∈n，n代表组合叠加日场景的数量n＝Nsc,ice×Nsc,car×Nsc,bus×kbase。S5. Generate the generated electric vehicle load daily scenario set Snow and ice making equipment load day scenario set and typical base load day scenario set Random combinations are performed and the load data are superimposed in time sequence. A combined superposition daily scenario set SC = (sc1 , sc2 , …, sci , …scn )i∈n covering all types of electricity consumption in ice and snow sports venues is constructed, where n represents the number of combined superposition daily scenarios n = Nsc,ice ×Nsc,car ×Nsc,bus ×kbase .
S6、由于日场景数量较大，采用K-means聚类算法对电动汽车+造雪造冰设备+基础负荷的组合叠加日场景集SC＝(sc1，sc2，…，sci，…scn)进行聚类分析，共分成K类，提取出具有代表性的冰雪运动场馆典型负荷日场景集SCtypical＝(sc1，sc2，…，scK)，1<K<n，K可以根据任务需求人为设定。S6. Due to the large number of daily scenarios, the K-means clustering algorithm is used to perform cluster analysis on the combined superposition daily scenario set SC = (sc1 , sc2 , …, sci , …scn ) of electric vehicles + snow and ice making equipment + basic load, which is divided into K categories in total, and the representative daily scenario set of typical loads of ice and snow sports venues SCtypical = (sc1 , sc2 , …, scK ) is extracted, where 1<K<n, and K can be set manually according to task requirements.
由技术常识可知，本发明可以通过其它的不脱离其精神实质或必要特征的实施方案来实现。因此，上述公开的实施方案，就各方面而言，都只是举例说明，并不是仅有的。所有在本发明范围内或在等同于本发明的范围内的改变均被本发明包含。It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110461136.2ACN113128063B (en) | 2021-04-27 | 2021-04-27 | A typical load scenario generation method and system for ice and snow sports venues |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110461136.2ACN113128063B (en) | 2021-04-27 | 2021-04-27 | A typical load scenario generation method and system for ice and snow sports venues |
| Publication Number | Publication Date |
|---|---|
| CN113128063A CN113128063A (en) | 2021-07-16 |
| CN113128063Btrue CN113128063B (en) | 2024-08-16 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110461136.2AActiveCN113128063B (en) | 2021-04-27 | 2021-04-27 | A typical load scenario generation method and system for ice and snow sports venues |
| Country | Link |
|---|---|
| CN (1) | CN113128063B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114595929B (en)* | 2022-01-20 | 2025-01-24 | 南瑞集团有限公司 | Method, device and system for generating typical operation mode scenarios of power system |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580585A (en)* | 2019-09-11 | 2019-12-17 | 东南大学 | A load decomposition-based power user clustering power consumption behavior characteristic analysis method |
| CN110909911A (en)* | 2019-09-29 | 2020-03-24 | 中国农业大学 | Aggregation method for multidimensional time series data considering spatiotemporal correlation |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580585A (en)* | 2019-09-11 | 2019-12-17 | 东南大学 | A load decomposition-based power user clustering power consumption behavior characteristic analysis method |
| CN110909911A (en)* | 2019-09-29 | 2020-03-24 | 中国农业大学 | Aggregation method for multidimensional time series data considering spatiotemporal correlation |
| Publication number | Publication date |
|---|---|
| CN113128063A (en) | 2021-07-16 |
| Publication | Publication Date | Title |
|---|---|---|
| CN111311122B (en) | Charging station construction method for smart city | |
| CN109508826B (en) | Electric vehicle cluster schedulable capacity prediction method based on gradient lifting decision tree | |
| CN112418605A (en) | Optimal operation method for energy storage system of optical storage type charging station | |
| Liang et al. | A calculation model of charge and discharge capacity of electric vehicle cluster based on trip chain | |
| CN110674575A (en) | A Modeling Method for Charging Demand and Discharging Capacity of Electric Vehicle Clusters Based on Time Series Travel Sets | |
| CN103631642A (en) | Ecological simulation-based electric car battery charging and replacing service network simulation system and method | |
| CN110429596A (en) | The distribution network reliability evaluation method of meter and electric car spatial and temporal distributions | |
| Guner et al. | Distributed storage capacity modelling of EV parking lots | |
| CN114580251B (en) | Method and device for analyzing charging load of electric automobile in distribution transformer area | |
| Guner et al. | Impact of car arrival/departure patterns on EV parking lot energy storage capacity | |
| CN114398723A (en) | Minkowski-sum-based large-scale electric vehicle cluster characteristic analysis method and system | |
| CN113128063B (en) | A typical load scenario generation method and system for ice and snow sports venues | |
| CN109742755A (en) | Operation and scheduling method of active distribution network with electric vehicles based on node partitioning | |
| Fu et al. | Research on the stackelberg game method of building micro-grid with electric vehicles | |
| Chen et al. | Location planning of charging stations considering the total cost of charging stations and users | |
| CN115833207A (en) | Electrified railway vehicle-mounted energy storage configuration method, system, equipment and medium | |
| CN116502964A (en) | Method and device for monitoring and modeling carbon emission space-time distribution of electric automobile | |
| CN119398359A (en) | A non-cooperative game scheduling method and device for virtual power plant based on generalized energy storage | |
| CN119182157A (en) | Distributed multi-energy complementary system design and operation hybrid optimization method considering energy storage characteristics of electric automobile | |
| Dixon et al. | Characterization of electric vehicle fast charging forecourt demand | |
| CN118539420A (en) | EV charging load prediction method based on hierarchical quantum clustering and user portrayal | |
| CN106203719A (en) | A Load Forecasting Method for Connecting Electric Vehicles to Power Grid | |
| CN116702996A (en) | Electric bus energy prediction method based on typical working condition recognition | |
| CN110766933A (en) | An optimization method for dynamic planning of airport taxi pick-up area settings | |
| Li et al. | Behaviorally realistic model for analyzing the effect of V2G participation |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |