Offshore wind power plants (WPPs) using AC submarine cables connected to the power grid may experience harmonic resonance and harmonic amplification that are not common in onshore wind farms under the action of the distributed capacitance of submarine cables, which seriously threatens the safe and reliable operation of offshore WPPs. To investigate the influencing factors of harmonic amplification in offshore WPPs, a scalable state space-based modeling method for large-scale offshore WPPs were established. Moreover, the root locus method was used to study how the factors including that grid short-circuit capacities, cable parameters and the number of wind turbine generators (WTGs) affect the resonance modes. Finally, an offshore WPP in Jiangsu was built on the MATLAB/Simulink simulation platform, and the analysis results verify the correctness of the theoretical analysis.
随着电动汽车和快速充电站的快速发展,电力网络与交通网络之间的耦合日渐增强.作为激励信号,充电价格是引导车辆用户的路径与充电选择的有效手段.为挖掘电动汽车充电负荷的空间灵活性,提出一种电网-充电运营商协同优化架构.首先,基于离散选择模型,对电力-交通融合网络进行建模,完成计及用户决策的异质性与不完全理性的充电负荷计算.之后,通过电网调度与充电运营商的合作,实现充电价格与充电容量的协同优化,对车辆用户的行为模式进行引导,以发掘电动汽车充电负荷的空间可平移潜力.最后,计及充电运营商在合作中的收益损失,基于纳什谈判机制对合作中产生的净收益进行合理分配.算例表明,所提协同运行架构有效地降低了电网的供电成本,同时从个体理性角度保证了合作联盟的稳定性.
光伏发电功率的预测对电网稳定以及安全地运行有重要意义,提出一种基于长短期记忆网络(long short term memory,LSTM)数字孪生体的预测模型,通过数字孪生体模型实现光伏发电功率的精准预测.数字孪生体分为物理空间与数据空间,首先根据物理空间得到的气象孪生数据由LSTM算法获取初步的预测功率,同时更新历史气象数据库.然后在气象数据库中找到相似日,对比相似日的预测功率和实际功率,对初步的预测功率进行误差修正,得到最终光伏功率预测值.文中所提的数字孪生体实现了物理实体与数据驱动的连接,同时物理实体可进行自我学习和更新,因此相较于传统的光伏预测结果更为精确,通过仿真算例进一步证实数字孪生体预测的准确性.
Improving the prediction accuracy is a key problem for wind power probability prediction research. The multisource numerical weather prediction were integrated to reduce the prediction error, the temporal pattern attention was used to select the input information adaptively, the temporal convolutional network was used to extract the multi-time scale probability features, and the mixed Beta distribution was used to construct the prediction probability information. The simulation results show that the convergence of model training can be improved effectively by integrating multi-source numerical weather prediction with temporal pattern attention, and the prediction results have higher accuracy.
针对电动汽车充电站选址定容问题,提出了一种多场景下计及配电网、充电站和用户多主体经济利益模型.首先通过对比配电网在正常运行环境和极端天气条件下的运行经济性与负荷损失成本,对电动汽车充电站进行预选址;其次以充电站预选方案以及节假日、工作日交通流量分布差异为基础,综合考虑充电站与用户端经济性对充电站站址容量进行优化;采用粒子群算法以及Voronoi图联合增加局部寻优效果,进一步优化电动汽车充电站选址定容结果.最后利用某地区的实际算例进行仿真分析,结果验证了所提电动汽车充电站规划方案的可行性和有效性.
针对能量传输效率对充电距离的敏感性导致节点充电效率低的问题,构建一种基于动态引力场的高能效无线充电调度策略.首先,考虑传感节点自身的能量状态,预测充电设备到达时的节点剩余能量,为评估出现死亡的节点建立价值节点集合;其次通过引入动态引力场理论,提出了一种基于动态引力场融合的多节点充电算法,考虑价值节点的剩余能量、消耗功率以及距基站的距离来定义节点的引力范围,根据引力场融合原则来选择更优的充电位置.最后对文中充电算法进行了仿真,并与多节点充电算法(multi-node rechargeable algorithm,MRA)和网格节点聚类(grid node clustering,GNC)算法进行了对比,验证了该充电算法能有效提高传感网络的能量效用,在保证合理充电时延的同时避免节点能量的过度损耗,显著增强网络的可持续性.
在碳中和、碳达峰的目标推动下,需求侧管理已逐渐成为降低碳排放的一种重要途径.为了更好地对电网进行需求侧管理,提出了一种考虑碳流追踪和用户碳排放程度评级的两阶段电力消费绿色责任证书分配模型,激励用户更好地承担碳减排责任.模型的一阶段以系统发电成本最小选为目标函数进行电力系统优化调度,确定机组的最优组合出力和系统潮流,进而构建碳流追踪模型以得到用户碳排放相关指标;二阶段综合考虑用户碳排放量和碳势建立用户碳排放程度评级体系,并以全网碳排放程度差异最小为目标进行绿色责任证书分配,满足碳减排责任主体共同而有差别地进行责任分摊.用CPLEX求解器对改进的IEEE14节点算例求解,仿真结果表明该方法能有效评估用户的碳排放程度情况,并实现对绿色责任证书的有效分配,降低了高碳用户的等效碳排放状况.
为促进新能源消纳,实现虚拟电厂的价值最大化,以风电商、光伏商和储能商为收益主体构建虚拟电厂参与电力交易.首先,构建区块链下的包含蓄电池和超级电容的风、光、储虚拟电厂结构;其次,基于区块链中共享的发电数据、用能数据和运行数据等参数,提出混合储能充放电策略和以混合储能成本最低为目标的容量优化配置方法,并采用粒子群算法求解;再次,以区块链应用程度、风险偏好、合作意愿及参与程度为因素改进Shapley值法,建立收益修正模型,并分析区块链下基于智能合约的收益分配过程;最后,通过算例仿真及结果分析,验证所建模型及方法的可行性与适用性.
为解决风电机组在风电功率平抑和故障穿越 2方面的不足,针对基于混合储能的直驱风力发电系统,提出一种同时兼顾风电功率平抑和故障穿越的复合功率控制策略.一方面,提出具有功率误差反馈环的改进型二阶滤波功率分配方法,实时修正超级电容和蓄电池储能的功率响应指令,提高目标功率分配精度的同时改善跟踪控制效果,实现风电功率平抑的同时延长储能介质使用寿命;另一方面,提出网侧变流器(grid side convertor,GSC)和混合储能共同作用的复合功率控制策略,实时修正各控制量的功率响应指令并快速清除直流母线上的冗余功率,提高风电机组的故障穿越能力,使风电系统基本不受电网故障的影响.
传统的电力系统暂态电压稳定评估模型存在 2方面问题:故障过程中的关键信息难以捕捉、暂态稳定样本与失稳样本不平衡导致模型对多数类样本存在倾向性.为此,提出了基于改进深度残差网络的电压稳定预警模型.首先,为了捕捉故障过程中的关键信息,在残差网络中嵌入卷积注意力模块,通过对时间通道与空间通道的双重注意力来挖掘电力系统动态轨迹中潜在的时空关系;其次,针对训练过程中模型倾向于多数类样本的问题,引入基于梯度平衡机制的损失函数来减小不平衡样本对评估结果的影响;第三,为了强化模型对数据特征的提取能力,将传统卷积核替换为非对称卷积模块.最后,通过在IEEE39节点系统上接入 2种不同风电占比进行测试,进一步验证所提方法在暂态电压稳定评估中的优异性能.
以风光为代表的新能源具有出力不确定性、波动性,聚合新能源发电资源的虚拟电厂(virtual power plant,VPP)在优化调度时难以避免VPP实际出力与调度计划间的偏差,进而导致VPP运营风险增加,收益降低.提出了考虑新能源不确定性风险的VPP双层调度策略.首先,建立了考虑VPP内部资源聚合效益的上层模型;其次,针对新能源出力不确定性,采用条件风险价值量化虚拟电厂调度风险,建立了考虑VPP收益和风险的下层多目标调度模型;最后,通过算例验证了调度模型的有效性,能够为VPP提供兼顾收益与风险的调度辅助决策.
The scale of DC microgrid systems continues to increase, and a large number of converters are integrated into the power grid, resulting in severe stability problems, and the traditional impedance analysis method has certain limitations in establishing an equivalent model of complex topological networks. To this end, this paper proposes an improved impedance analysis method based on the generalized impedance ratio,first popularizes the concept of traditional impedance ratio, establishes a system equivalent model according to the port characterristics, and derives the concept of generalized impedance ratio to characterize system stability. The system is further modeled with small signals, and a stability criterion based on the generalized impedance ratio is obtained for stability analysis. Finally, the correctness and effectiveness of the stability analysis of the proposed method in the multivariate DC microgrid system are verified by example analysis and time domain simulation.
以电动汽车为代表的需求侧资源通过聚合运营,将为电力系统提供大量的可调节资源,缓解系统平衡运行的矛盾.但其商业运营模式尚未成熟,关键原因之一是电动汽车聚合价值评估难,难以支撑相关市场交易机制的构建.通过引入价值网络建模方法,建立电动汽车聚合运营价值创造模式,构建电动汽车聚合运营价值网络模型.进而基于电动汽车聚合商视角,分析了聚合体系内多主体价值交换关系,提出了聚合模式下多主体价值获取路径.最后,结合国内某区域市场环境,对有无聚合商时的市场各主体效益进行分析.仿真结果表明电动汽车聚合运营使得各主体都能产生增量收益,在现有模式下聚合商获益最大.市场补偿系数、新能源消纳比例增量是影响各方受益的关键因素.
为促进公路交通碳减排、尽早实现碳达峰,首先对公路交通碳排放进行测算,在公路交通所消耗的汽油和燃油产生碳排放的基础上,还考虑到电动汽车充电量中的煤电产生的碳排放.然后,采用STIRPAT模型对公路交通碳排放影响因素进行分析,选取公路交通GDP、汽车保有量、充电量的煤电占比、公路货运周转量以及公路客运周转量这 6个因素来研究公路交通碳排放变化,并对这 6个影响因子的公路交通碳排放贡献率进行测算分析.最后,通过设定基准情景和集中减排情景,预测我国公路交通未来的碳排放量,为实现公路交通碳达峰提供发展思路.结果表明要在大力推广电动汽车的前提下,使电动汽车占比变化率在 2019-2030年间为 37.91%,2030年以后为 1.70%,并提高清洁能源比例,使电动汽车充电量的煤电占比达到-23%的变化率变化时,对于电动汽车充电碳排放而言,更加具有清洁低碳意义.同时也要保持公路客运周转量每年变化-4.14%的下降趋势以及维持公路货运量每年变化4.84%的速度缓慢增长,我国公路交通将有望在 2030年前实现碳达峰.
In the UHVDC Hierarchical connection system,when the AC system on the inverter side fails, the commutation failure of the high and low end converters may occur at the same time due to the delay in starting the commutation failure prevention control of the non-faulty layer. Therefore, a coordinated control strategy of high and low side converters based on the commutation time and voltage integral area is proposed.Taking advantage of the fact that the commutation failure prevention control of the faulty layer can respond to faults more quickly, the proposed strategy introduces the output of the commutation failure prevention control of the faulty layer into the non-faulty layer, so that the start time of the commutation failure prevention control of the non-faulty layer is advanced, increase the turn-off angle of the non-faulty layer converter; at the same time, the voltage dependent current order limiter control is modified to reduce the DC current during the fault process. A simulation model was built in PSCAD/EMTDC to verify the proposed strategy under different working conditions.The results show that this strategy can significantly improve the sensitivity of non-faulty layer converters to faults and reduce the risk of simultaneous commutation failures.
With increasing penetration of wind power, how to improve the operational stability of active distribution network while reducing operating costs has become an urgent problem to be solved. A robust coordinated active-reactive power optimization for the distribution network is established in this paper,aiming at reducing the operation costs and voltage deviation of the distribution network. The decision variables include the operation strategy of energy storage system, output of gas turbine and active management measures such as adjustment of transformer tap, capacitor banks. The MVEE（Minimum Volume Enclosing Ellipsoid） algorithm is used to transform the correlated wind power output scenarios into elliptic uncertain set constraints considering the correlation of wind turbine output,which improves the conservative property of the traditional interval uncertainty set of robust optimization. The conic duality theorem is used to transform the proposed robust model into bilevel mixed-integer second-order cone programming problem.The column constrained generation algorithm is used to solve the two-layer optimal scheduling problem. The simulation analysis is carried out on the improved IEEE 33-bus distribution network. The influence of wind power output correlation on the distribution network optimal scheduling is studied, and the economic and robustness of the proposed method are verified.
为了实现风-光-氢-储综合能源系统实时跟踪电网日前发电计划的优化调度,并考虑峰谷价电价规则,结合储能系统时空位移特性,使综合能源系统制氢、氢燃料电池发电经济效益最高,首先建立了风-光-氢-储各子系统数学模型及以电力经济最优为目标的日前优化调度模型,综合考虑系统功率平衡、运行状态、安全状态等指标约束,设定氢储系统优先出力对应的权重系数,采用智能群优化算法对系统目标函数进行迭代优化求解;最后根据某地风光负荷数据及分时电价信息,对所提优化调度方法在Matlab中进行分析验证,得到风-光-氢-储综合能源系统各子系统功率分配情况以实现最优经济性以及对电网日前调度指令的实时跟踪.
The investment in new energy power stations is vast, so the power generation efficiency is critical. The wake effect is an important factor affecting the power generation efficiency of wind farms. However, the existing wake optimization algorithms ignore the influence of wind speed change on the regulation range of tip speed ratio（TSR）, which may lead to the actual failure of optimization results. Considering that the feasible interval of the optimization variables of the turbine in the wake area will change with the optimization of the front turbines. This paper models that change and proposes an optimization algorithm considering the constraint change to improve the accuracy of the wake optimization of the wind farm. The effectiveness of the proposed algorithm is verified by comparing the simulation results of the two algorithms with or without constraint changes in the same environment.
As a complex energy system with multi inputs and outputs, the design, planning and operation management of integrated energy system are current research difficulties. In allusion to the integrated energy park system with coupled hydrogen energy storage, a comprehensive evaluation index considering energy efficiency, reliability, environmental protection and economy was proposed. Under specified operation strategy, an optimization method combining particle swarm optimization（abbr. PSO） with maximum rectangle method（abbr. MRM）was utilized to perform system capacity allocation. To verify the reliability of capacity allocation results, taking a certain comprehensive energy park system in Xi’an for example, the cooling, heating and electrical loads in this park area within a whole year were taken as load demand, on the basis of traditional operation strategy of determining electricity by heat and determining heat by electricity a comprehensive operation strategy was proposed. The system capacity allocation for three operation strategies was researched, and the characteristics of their comprehensive operation strategies were analyzed. Computing results show that comparing with determining electricity by heat and determining heat by electricity, the comprehensive evaluation index of the proposed comprehensive operation strategy were improved by 1.66% and 0.13% respectively.
针对冲击性负荷预测问题,提出了一种基于混沌多目标蚁狮优化算法(chaotic multi-objective antlion optimization algorithm,CMOALO)和核极限学习机(kernel extreme learning machine,KELM)的冲击性负荷预测模型.首先,为了降低预测难度,使用集合经验模式分解(ensemble empirical mode decomposition,EEMD)将原始冲击性负荷分解为一系列更为平稳的子序列.为了同时提升模型的预测精度和稳定性,提出了一种MOALO;其次,为进一步提高算法的解搜索能力,将MOALO与混沌运算融合,提出了CMOALO算法,将其用于优化KELM.最后通过某地区真实采集的冲击性负荷数据对所提出的EEMD-CMOALO-KELM模型进行验证.通过案例分析可知,所提出的冲击性负荷预测模型,无论是在预测精度还是预测稳定性方面,性能最好.
