Improved grey-based approach for power demand forecasting

Grey theory is a multidisciplinary and generic theory to cope with systems of poor or deficient information. We proposed in this paper an improved grey method (GM) to overcome the disadvantages of the general GM(1,1). In the improved GM(1,1), a new background value formula is deduced and Markov-chain sign estimation is imbedded into the residual modification model. We tested the efficiency and accuracy of our model by applying it to the power demand forecasting in Taiwan. Experimental results demonstrate the new method has obviously a higher prediction accuracy than the general model.

Design Optimization and Operating Performance of S-CO_(2) Brayton Cycle under Fluctuating Ambient Temperature and Diverse Power Demand Scenarios

The supercritical CO_(2)(S-CO_(2)) Brayton cycle is expected to replace steam cycle in the application of solar power tower system due to the attractive potential to improve efficiency and reduce costs.Since the concentrated solar power plant with thermal energy storage is usually located in drought area and used to provide a dispatchable power output,the S-CO_(2) Brayton cycle has to operate under fluctuating ambient temperature and diverse power demand scenarios.In addition,the cycle design condition will directly affect the off-design performance.In this work,the combined effects of design condition,and distributions of ambient temperature and power demand on the cycle operating performance are analyzed,and the off-design performance maps are proposed for the first time.A cycle design method with feedback mechanism of operating performance under varied ambient temperature and power demand is introduced innovatively.Results show that the low design value of compressor inlet temperature is not conductive to efficient operation under low loads and sufficient output under high ambient temperatures.The average yearly efficiency is most affected by the average power demand,while the load cover factor is significantly influenced by the average ambient temperature.With multi-objective optimization,the optimal solution of designed compressor inlet temperature is close to the minimum value of35℃ in Delingha with low ambient temperature,while reaches 44.15℃ in Daggett under the scenario of high ambient temperature,low average power demand,long duration and large value of peak load during the peak temperature period.If the cycle designed with compressor inlet temperature of 35℃ instead of 44.15℃ in Daggett under light industry power demand,the reduction of load cover factor will reach 0.027,but the average yearly efficiency can barely be improved.

Daily power demand prediction for buildings at a large scale using a hybrid of physics-based model and generative adversarial network

Power demand prediction for buildings at a large scale is required for power grid operation.The bottom-up prediction method using physics-based models is popular,but has some limitations such as a heavy workload on model creation and long computing time.Top-down methods based on data driven models are fast,but less accurate.Considering the similarity of power demand patterns of single buildings and the superiority of generative adversarial network(GAN),this paper proposes a new method(E-GAN),which combines a physics-based model(EnergyPlus)and a data-driven model(GAN),to predict the daily power demand for buildings at a large scale.The new E-GAN method selects a small number of typical buildings and utilizes EnergyPlus models to predict their power demands.Utilizing the prediction for those typical buildings,the GAN then is adopted to forecast the power demands of a large number of buildings.To verify the proposed method,the E-GAN is used to predict 24-hour power demands for a set of residential buildings.The results show that(1)4.3%of physics-based models in each building category are required to ensure the prediction accuracy;(2)compared with the physics-based model,the E-GAN can predict power demand accurately with only 5%error(measured by mean absolute percentage error,MAPE)while using only approximately 9%of the computing time;and(3)compared with data-driven models(e.g.,support vector regression,extreme learning machine,and polynomial regression model),E-GAN demonstrates at least 60%reduction in prediction error measured by MAPE.

RS-SVM forecasting model and power supply-demand forecast

A support vector machine (SVM) forecasting model based on rough set (RS) data preprocess was proposed by combining the rough set attribute reduction and the support vector machine regression algorithm, because there are strong complementarities between two models. Firstly, the rough set was used to reduce the condition attributes, then to eliminate the attributes that were redundant for the forecast, Secondly, it adopted the minimum condition attributes obtained by reduction and the corresponding original data to re-form a new training sample, which only kept the important attributes affecting the forecast accuracy. Finally, it studied and trained the SVM with the training samples after reduction, inputted the test samples re-formed by the minimum condition attributes and the corresponding original data, and then got the mapping relationship model between condition attributes and forecast variables after testing it. This model was used to forecast the power supply and demand. The results show that the average absolute error rate of power consumption of the whole society and yearly maximum load are 14.21% and 13.23%, respectively, which indicates that the RS-SVM forecast model has a higher degree of accuracy.

Analysis on the Situation of Power Supply and Demand in Shandong

In the first half of 2007, the power industry in Shandongprovince continued to maintain a rapid growth momentum.The gross electricity consumption amounted to 121.25 TWh,14.4% higher over that in the same period of last year. The totalinstalled capacity reached 53.29 GW. It was expected that bythe end of 2007, the gross electricity consumption in Shan-dong would reach 260 TWh, increasing by 14.4% on ayear-on-year basis; the maximum load would reach 40.

计及成本和价值的长周期负荷侧调节能力投入策略

随着新型电力系统的发展,负荷侧调节能力建设日益受到关注,对于负荷侧调节能力成本、价值及其投入策略的研究,具有重要意义。基于近10 a来中国电力工业发展的基础数据,结合新型电力系统发展态势,分析了负荷侧调节能力需求的背景,通过对发用电两侧调节能力特点的比较,从电力系统整体角度梳理了负荷侧调节能力的投入所带来的增量价值,分析了负荷侧调节能力的应用价值盈余,进一步结合具体需求提出了3种长周期下负荷侧调节能力投入的优化策略模型,并进行了算例分析。结果显示:负荷侧调节能力的投入有利于降低电力系统的总成本、增加系统调节能力,但是由于其成本曲线的变化以及“非持久特性”等因素,其投入策略需要综合长周期市场发展需求和发用电两侧成本因素进行权衡,所提出的3种优化模型各有其适用场景。

计及需求响应的光热电站参与深度调峰的分层优化调度策略

从源、荷两侧挖掘系统调峰潜力,建立计及需求响应的光热电站参与深度调峰的分层优化调度模型。上层从负荷侧出发,提出一种基于负荷分类的价格需求响应模型,可有效缓解系统调峰压力;中层从电源侧出发,利用光热电站灵活的调节特性在深度调峰时段协调火电机组参与辅助调峰,构建以运行总成本最小为目标函数的日前调度模型;下层提出一种基于模型预测控制的日内动态调整模型,在滚动优化的同时,通过状态反馈环节实时调整光热电站储热装置充放热修正日前调度计划。仿真结果表明,所提调度策略在降低系统调峰成本的同时能有效抑制风光以及负荷的短时功率波动,在保证系统安全稳定运行的前提下提升风光消纳率。

A strategy for strengthening chaotic mixing of dual shaft eccentric mixers by changing non-Newtonian fluids kinetic energy distribution

Efficiently modulating the velocity distribution and flow pattern of non-Newtonian fluids is a critical challenge in the context of dual shaft eccentric mixers for process intensification,posing a significant barrier for the existing technologies.Accordingly,this work reports a convenient strategy that changes the kinetic energy to controllably regulate the flow patterns from radial flow to axial flow.Results showed that the desired velocity distribution and flow patterns could be effectively obtained by varying the number and structure of baffles to change kinetic energy,and a more uniform velocity distribution,which could not be reached normally in standard baffle dual shaft mixers,was easily obtained.Furthermore,a comparative analysis of velocity and shear rate distributions is employed to elucidate the mechanism behind the generation of flow patterns in various dual-shaft eccentric mixers.Importantly,there is little difference in the power number of the laminar flow at the same Reynolds number,meaning that the baffle type has no effect on the power consumption,while the power number of both unbaffle and U-shaped baffle mixing systems decreases compared with the standard baffle mixing system in the transition flow.Finally,at the same rotational condition,the dimensionless mixing time of the U-shaped baffle mixing system is 15.3%and 7.9%shorter than that of the standard baffle and the unbaffle mixing system,respectively,which shows the advantage of the U-shaped baffle in stirring rate.

考虑风电出力不确定性的多源联合系统双层优化调度

针对含风-火-储的多源联合系统,风电出力具有不确定性的特点,风机在特定时间段内的预测功率与实际功率之间存在误差,当风机实际出力无法满足调度计划中安排的功率时会导致系统经济效益大幅下降。为此,文中提出了考虑风电预测误差和需求侧响应的双层优化策略,上层模型以风电、火电和可平移负荷总运行成本最少为目标,采用改进粒子群算法(Improved Particle Swarm Algorithm, IPSO)制定火电和可平移负荷的最优调度策略,然后通过Gibbs法对风机最大出力预测误差的概率密度函数进行抽样获取一定量的样本,得到各样本上层电源的功率缺额;下层模型以储能和可中断负荷总运行成本最少为目标,采用线性规划方法对冲上层电源功率缺额,进而制定下层模型电源调度策略。在大量抽样样本背景下,通过对比各样本总成本函数值的期望和方差验证了所提双层优化策略的经济性和有效性。

SmartMicro Grid Energy System Management Based on Optimum Running Cost for Rural Communities in Rwanda

The governmental electric utility and the private sector are joining hands to meet the target of electrifying all households by 2024.However,the aforementioned goal is challenged by households that are scattered in remote areas.So far,Solar Home Systems(SHS)have mostly been applied to increase electricity access in rural areas.SHSs have continuous constraints to meet electricity demands and cannot run income-generating activities.The current research presents the feasibility study of electrifying Remera village with the smart microgrid as a case study.The renewable energy resources available in Remera are the key sources of electricity in that village.The generation capacity is estimated based on the load profile.The microgrid configurations are simulated with HOMER,and the genetic algorithm is used to analyze the optimum cost.By analyzing the impact of operation and maintenance costs,the results show that the absence of subsidies increases the levelized cost of electricity(COE)five times greater than the electricity price from the public utility.The microgrid made up of PV,diesel generator,and batteries proved to be the most viable solution and ensured continuous power supply to customers.By considering the subsidies,COE reaches 0.186$/kWh,a competitive price with electricity from public utilities in Rwanda.

支撑新能源电力系统灵活性需求的用户侧资源应用与关键技术

伴随着传统电力系统向以新能源为主体的新型电力系统转型升级,高比例新能源的接入和传统机组占比下降导致电力系统灵活性产生缺额。电动汽车、数据中心、P2X(power-to-X)等需求侧资源可以在多时间尺度为电力系统提供灵活性,是新型电力系统灵活性的重要组成部分。该文首先描述高比例新能源背景下输电网和配电网在不同时间尺度的灵活性需求,然后对比总结了典型需求侧资源的调节特性与应用实践,研究了需求侧资源在支撑高比例新能源输电网和配电网中运行、规划2个层面多时间尺度灵活性需求的应用场景。最后针对未来需求侧资源的研究重点,从聚合技术、电碳协同优化调控策略、可信调节能力量化和通信与信息交换标准等角度,对支撑新型电力系统需求侧资源应用的关键技术进行了总结与展望。

基于ISSA-LSTM模型的可再生能源电力需求预测

为了更精准地预测未来能源结构调整方向及成效,选用ISSA-LSTM组合预测模型对中国2023-2030年可再生能源的电力需求进行预测。首先,利用Circle混沌映射改进麻雀搜索算法(SSA)以提高搜索能力以及种群多样性;然后引入长短期记忆神经网络(LSTM)以有效捕捉可再生能源电力需求随机波动性和时序性;最后,通过ISSA-LSTM模型预测长期可再生能源的电力需求,验证测试集数据,并与其他传统模型进行对比。结果表明:ISSA-LSTM模型预测结果能够满足对可再生能源电力需求预测的精度要求;在未来2023-2030年可再生能源电力需求稳定,波动幅度不大,可达到全国用电量的1/3;利用Circle混沌映射改进策略能有效提升SSA寻优能力。与PSO算法相比,SSA算法寻找LSTM超参数最优解的能力更优,ISSA-LSTM模型预测可再生能源电力需求精度更高。

考虑保供电需求的光储微电网优化配置及电能质量评估

针对用户日益迫切的保供电需求以及光伏出力与负荷用电的随机性、波动性引发的电能质量的问题,建立了考虑保供电需求的光储微电网双层优化模型,并对规划配置的微电网电能质量进行评估。该模型以等效净负荷标准差和弃光率最小为上层优化目标,以微电网有功功率损耗最小为下层优化目标,以储能满足保供电负荷的保供电需求作为附加约束条件,采用遗传算法求解。然后设计源、网、荷三类指标,采用组合赋权法确定各指标权重,实现对微电网电能质量的综合评估。最后以改进的IEEE-33系统验证了所提模型的有效性。结果表明考虑保供电需求后能够提高微电网电能质量,并在一定范围内保供电负荷越大时越能提高微电网电能质量和用户满意度。

考虑需求响应和风电不确定性的电网调度

风电不确定性对电力系统的影响日益增加,文中提出一种考虑需求响应的电网源荷网经济调度方法。根据需求价格弹性理论和消费者心理学原理构建价格型需求响应模型,基于灵活补偿激励型需求响应充分考虑两类需求响应互补特性,增加两类需求响应对系统的备用约束。引入通用概率分布描述风电出力特性,以期望成本量化风电不确定性惩罚成本,并将目标函数中的非线性变量线性化处理,采用混合整数线性规划问题对模型进行求解。仿真结果表明,所提模型能够有效降低负荷的峰谷差,提高风电的消纳能力。

基于主成分分析-BP神经网络的风电备件需求预测

风电机组具有结构复杂,运维困难,且长期处于恶劣的工作环境的特点。风电备件的需求预测有助于为风电场配备最合适的备件数,以确保风电场的平稳、高效运行。构建主成分分析-反向传播(principal component analysis-back propagation,PCA-BP)模型,针对受多因素影响的复杂备件,先利用PCA将影响风电备件的要素进行筛选,再利用BP神经网络算法,得到最为精确的预测结果。比较自回归积分滑动平均(autoregressive integrated moving average,ARIMA)模型、BP神经网络预测和PCA-BP神经网络预测的结果。结果表明:PCA能显著降低神经网络预测误差,预测的精度为93.94%,高于BP神经网络预测的88.39%和ARIMA模型的85.31%,所以PCA-BP神经网络模型的预测精度准确且有可靠结果,能够适用于风机备件的需求预测。

不同参数高效超超临界机组经济性分析及建议

本文根据目前大量已投运二次再热机组运行经验,结合某660MW项目高效一次再热和二次再热不同的高效超超临界参数,结合对应的汽机热平衡图参数,通过对THA、80%THA、60%THA三个运行负荷进行了相关的机组经济性参数的计算和对比,通过分析提出了在选择一次再热或二次再热参数上的一些建议和结论。

基于负荷准线的5G基站虚拟电厂优化控制方法

在新型电力系统中,亟待深度挖掘需求侧资源以提升系统灵活性和新能源消纳能力。在“新基建”背景下,5G基站作为一种新型需求侧资源正迅速发展。研究如何在保证基站备用需求的前提下,由铁塔公司组建含大规模5G基站的虚拟电厂(virtual power plant,VPP)并常态化参与需求响应。首先,提出了考虑储能动态备用容量的5G基站运行可行域构建方法,建立了5G基站VPP的聚合模型。然后,建立了5G基站VPP响应负荷准线的日前优化模型,提出了适合对大规模5G基站进行协调控制的日内解聚合方法。最后,建立了含高比例新能源的区域电网仿真算例。仿真结果表明,聚合大规模基站参与准线型需求响应,可以显著降低5G基站的运行成本,同时提高电网的新能源消纳能力。

基于低碳需求响应的新型电力系统中源-网-荷-储协调的优化调度

【目的】在我国“双碳”目标的背景下,节能减排已成为当前电力行业的首要战略目标之一。电力系统除了要考虑系统安全可靠性和经济性之外,还需要将“碳减排约束”引入系统各环节中,将呈现全新的低碳运行和管理模式。【方法】针对新型电力系统碳计量与低碳调控运行问题,提出一种基于碳排放流的低碳需求响应调度策略。首先,分析新型电力系统源-网-荷-储各环节的碳排放影响因素,建立基于碳排放流理论精确的碳计量模型。其次,以动态碳排放因子为引导信号建立低碳需求响应机制,进行优化调度。【结果】通过算例验证:基于低碳需求响应的新型电力系统源-网-荷-储协调优化调度方法有效激发激发系统各环节减排潜力,降低碳排放量,提高新能源消纳率。

考虑超碳需求响应的综合能源系统低碳优化调度

西北地区新能源发展迅速,充分利用当地独有的光热资源优势,同时结合火电低碳化改造与新能源协同运行,有利于推动当地能源系统绿色低碳转型。为尽最大限度地提升系统的减碳能力,提出一种基于源荷协同降碳的超碳需求响应模型,将动态碳排放因子作为分时电价的惩罚因子,从而将源侧碳信号传递至荷侧,驱使用户侧进行低碳性状态转移。首先,在日前调度阶段,构建预调度-再调度两阶段调度运行机制,再调度根据预调度的系统状态信息进行超碳需求响应,来深度降低系统碳排量。其次,将光热电站引入综合能源系统,与风电场、碳捕集电厂协同运行,从而构建高比例新能源场景,来验证超碳需求响应在此场景下的减碳效益。最后,建立了基于超碳需求响应的预调度-再调度两阶段低碳调度模型。经算例仿真分析表明,所提源荷协同降碳的新思路能有效提高系统的减碳能力,深入挖掘系统的降碳空间,提升系统的经济效益。