Multi-Time Scale Optimal Scheduling of a Photovoltaic Energy Storage Building System Based on Model Predictive Control

Building emission reduction is an important way to achieve China’s carbon peaking and carbon neutrality goals.Aiming at the problem of low carbon economic operation of a photovoltaic energy storage building system,a multi-time scale optimal scheduling strategy based on model predictive control(MPC)is proposed under the consideration of load optimization.First,load optimization is achieved by controlling the charging time of electric vehicles as well as adjusting the air conditioning operation temperature,and the photovoltaic energy storage building system model is constructed to propose a day-ahead scheduling strategy with the lowest daily operation cost.Second,considering inter-day to intra-day source-load prediction error,an intraday rolling optimal scheduling strategy based on MPC is proposed that dynamically corrects the day-ahead dispatch results to stabilize system power fluctuations and promote photovoltaic consumption.Finally,taking an office building on a summer work day as an example,the effectiveness of the proposed scheduling strategy is verified.The results of the example show that the strategy reduces the total operating cost of the photovoltaic energy storage building system by 17.11%,improves the carbon emission reduction by 7.99%,and the photovoltaic consumption rate reaches 98.57%,improving the system’s low-carbon and economic performance.

Deep Learning for Multivariate Prediction of Building Energy Performance of Residential Buildings

In the quest to minimize energy waste,the energy performance of buildings(EPB)has been a focus because building appliances,such as heating,ventilation,and air conditioning,consume the highest energy.Therefore,effective design and planning for estimating heating load(HL)and cooling load(CL)for energy saving have become paramount.In this vein,efforts have been made to predict the HL and CL using a univariate approach.However,this approach necessitates two models for learning HL and CL,requiring more computational time.Moreover,the one-dimensional(1D)convolutional neural network(CNN)has gained popularity due to its nominal computa-tional complexity,high performance,and low-cost hardware requirement.In this paper,we formulate the prediction as a multivariate regression problem in which the HL and CL are simultaneously predicted using the 1D CNN.Considering the building shape characteristics,one kernel size is adopted to create the receptive fields of the 1D CNN to extract the feature maps,a dense layer to interpret the maps,and an output layer with two neurons to predict the two real-valued responses,HL and CL.As the 1D data are not affected by excessive parameters,the pooling layer is not applied in this implementation.Besides,the use of pooling has been questioned by recent studies.The performance of the proposed model displays a comparative advantage over existing models in terms of the mean squared error(MSE).Thus,the proposed model is effective for EPB prediction because it reduces computational time and significantly lowers the MSE.

Novel Hybrid Physics‑Informed Deep Neural Network for Dynamic Load Prediction of Electric Cable Shovel

Electric cable shovel(ECS)is a complex production equipment,which is widely utilized in open-pit mines.Rational valuations of load is the foundation for the development of intelligent or unmanned ECS,since it directly influences the planning of digging trajectories and energy consumption.Load prediction of ECS mainly consists of two types of methods:physics-based modeling and data-driven methods.The former approach is based on known physical laws,usually,it is necessarily approximations of reality due to incomplete knowledge of certain processes,which introduces bias.The latter captures features/patterns from data in an end-to-end manner without dwelling on domain expertise but requires a large amount of accurately labeled data to achieve generalization,which introduces variance.In addition,some parts of load are non-observable and latent,which cannot be measured from actual system sensing,so they can’t be predicted by data-driven methods.Herein,an innovative hybrid physics-informed deep neural network(HPINN)architecture,which combines physics-based models and data-driven methods to predict dynamic load of ECS,is presented.In the proposed framework,some parts of the theoretical model are incorporated,while capturing the difficult-to-model part by training a highly expressive approximator with data.Prior physics knowledge,such as Lagrangian mechanics and the conservation of energy,is considered extra constraints,and embedded in the overall loss function to enforce model training in a feasible solution space.The satisfactory performance of the proposed framework is verified through both synthetic and actual measurement dataset.

Energy consumption prediction model of typical buildings in hot summer and cold winter zone of China

To overcome the shortcomings of the energyconsumption prediction models in the application during thedesign stage, a quick prediction model for energy consumptionis proposed based on the decoupling method. Taking typicalresidential and office buildings in hot summer and cold winterzones as research objects, the influence factors on buildingenergy consumption are classified into intrinsic factors andoperational factors on the basis of the heat transfer principle.Then, using the intrinsic factors as the fundamental variablesand operational factors as the modified variables, the quickprediction model for the buildings in typical cold and hot zonesis proposed based on the decoupling method and the accuracyof the proposed model is verified. The results show thatcompared to the simulation results of EnergyPlus, the relativeerror of the prediction model is less than 1.5% ; comparedwith the real operating data of the building, the relative erroris 13.14% in 2011 and 8.56% in 2012 due to the fact that thecoincidence factor becomes larger than the design value about16% in 2011 and 13% in 2012. The finding reveals that theproposed model has the advantages of rapid calculationcompared with EnergyPlus and Design Builder when predictingbuilding energy consumption in building designs. The energyconsumption prediction model is of great practical value inoptimal operation and building designs.

Research on Electricity Consumption Model of Library Building Based on Data Mining

With the exponential development of Chinese population,the massive energy consumption of buildings has recently become an interest subject.Although much research has been conducted on residential buildings,heating ventilation and air conditioning(HVAC),little research has been conducted on the relationship between student’s behavior,campus buildings,and their subsystems.Using classical seasonal decomposition,hierarchical clustering,and apriori algorithm,this paper aims to provide an empirical model for consumption data in campus library.Smart meter data from a library in Beijing,China,is adopted in this paper.Building electricity consumption patterns are investigated on an hourly/daily/monthly basis.According to the monthly analysis,electricity consumption peaks each year around June and December due to teaching programs,social exams,and outdoor temperatures.Hourly data analysis revealed a relatively stable consumption pattern.It shows three different types of daily load profiles.Daily data analysis demonstrated a high relationship between HVAC consumption and building total consumption,with a lift value of 5.9.Furthermore,links between temperature and subsystems were also discovered.Through a case study of library,this study provides a unique insight into campus electricity use.The results could help to develop operational strategies for campus facilities.

Distributed Model Predictive Load Frequency Control of Multi-area Power System with DFIGs

Reliable load frequency control LFC is crucial to the operation and design of modern electric power systems. Considering the LFC problem of a four-Area interconnected power system with wind turbines, this paper presents a distributed model predictive control DMPC based on coordination scheme. The proposed algorithm solves a series of local optimization problems to minimize a performance objective for each control area. The generation rate constraints GRCs, load disturbance changes, and the wind speed constraints are considered. Furthermore, the DMPC algorithm may reduce the impact of the randomness and intermittence of wind turbine effectively. A performance comparison between the proposed controller with and without the participation of the wind turbines is carried out. Analysis and simulation results show possible improvements on closed-loop performance, and computational burden with the physical constraints. © 2014 Chinese Association of Automation.

Characteristics of electricity consumption of different industry types considering atmospheric condition

Reducing greenhouse gases (RHG) is going on actively in the international movement. In the field of architecture, RHG is an inevitable work. To establish a plan for RHG, firstly we need to reduce energy consumption. Greenhouse gas generated by energy consumption is the main cause of global warming. For this we should know that how much electricity consumption we use. The research targets of this study are commercial buildings with various businesses. Their electricity consumption was analyzed by business units rather than buildings. Each business was divided into 13 sectors according to industrial classification and electricity consumption was analyzed for each industry. For commercial buildings, the electricity consumption is done by the private sector and construction management is an autonomy system in private instead of an integrated management system. In this study, we classified and analyzed the electricity consumption characteristics according to collected data, analyzed the relationship between the electricity consumption with atmospheric temperature through SPSS, and developed an electricity prediction model.

A comprehensive review for wind,solar,and electrical load forecasting methods

Wind power,solar power,and electrical load forecasting are essential works to ensure the safe and stable operation of the electric power system.With the increasing permeability of new energy and the rising demand response load,the uncertainty on the production and load sides are both increased,bringing new challenges to the forecasting work and putting forward higher requirements to the forecasting accuracy.Most review/survey papers focus on one specific forecasting object(wind,solar,or load),a few involve the above two or three objects,but the forecasting objects are surveyed separately.Some papers predict at least two kinds of objects simultaneously to cope with the increasing uncertainty at both production and load sides.However,there is no corresponding review at present.Hence,our study provides a comprehensive review of wind,solar,and electrical load forecasting methods.Furthermore,the survey of Numerical Weather Prediction wind speed/irradiance correction methods is also included in this manuscript.Challenges and future research directions are discussed at last.

Deep learning for time series forecasting:The electric load case

Management and efficient operations in critical infrastructures such as smart grids take huge advantage of accurate power load forecasting,which,due to its non-linear nature,remains a challenging task.Recently,deep learning has emerged in the machine learning field achieving impressive performance in a vast range of tasks,from image classification to machine translation.Applications of deep learning models to the electric load forecasting problem are gaining interest among researchers as well as the industry,but a comprehensive and sound comparison among different-also traditional-architectures is not yet available in the literature.This work aims at filling the gap by reviewing and experimentally evaluating four real world datasets on the most recent trends in electric load forecasting,by contrasting deep learning architectures on short-term forecast(oneday-ahead prediction).Specifically,the focus is on feedforward and recurrent neural networks,sequence-to-sequence models and temporal convolutional neural networks along with architectural variants,which are known in the signal processing community but are novel to the load forecasting one.

Data-Driven Load Forecasting Using Machine Learning and Meteorological Data

Electrical load forecasting is very crucial for electrical power systems’planning and operation.Both electrical buildings’load demand and meteorological datasets may contain hidden patterns that are required to be investigated and studied to show their potential impact on load forecasting.The meteorological data are analyzed in this study through different data mining techniques aiming to predict the electrical load demand of a factory located in Riyadh,Saudi Arabia.The factory load and meteorological data used in this study are recorded hourly between 2016 and 2017.These data are provided by King Abdullah City for Atomic and Renewable Energy and Saudi Electricity Company at a site located in Riyadh.After applying the data pre-processing techniques to prepare the data,different machine learning algorithms,namely Artificial Neural Network and Support Vector Regression(SVR),are applied and compared to predict the factory load.In addition,for the sake of selecting the optimal set of features,13 different combinations of features are investigated in this study.The outcomes of this study emphasize selecting the optimal set of features as more features may add complexity to the learning process.Finally,the SVR algorithm with six features provides the most accurate prediction values to predict the factory load.

基于动态自适应图神经网络的电动汽车充电负荷预测

电动汽车充电站负荷波动的不确定性与长时间预测任务给提升充电负荷预测精度带来巨大的挑战。文中提出一种基于动态自适应图神经网络的电动汽车充电负荷预测算法。首先,构建了一个充电负荷信息时空关联特征提取层,将多头注意力机制与自适应相关图结合生成具有时空关联性的综合特征表达式,以捕获充电站负荷的波动性;然后,将提取的特征输入到时空卷积层,捕获时间和空间之间的耦合关系;最后,通过切比雪夫多项式图卷积与多尺度时间卷积提升模型耦合长时间序列之间的能力。以Palo Alto数据集为例,与现有方法相比,所提算法在4种波动情况下的平均预测误差大幅降低。

考虑充电桩限制的电动汽车充电负荷预测

针对大规模电动汽车接入配电网,提出一种考虑充电桩限制的电动汽车充电负荷预测方法。首先从单辆电动汽车入手,分析恒流-恒压充电方式下的锂电池充电特性,并进行简化,给出单辆电动汽车的充电模型;再从整体研究用户行为,运用蒙特卡洛模拟法,给出电动汽车起始充电时间和充电时长的随机分布;最后提出居民小区电动汽车充电负荷曲线预测的建模方法。该建模方法可为未来电动汽车充电负荷预测提供借鉴和参考,指导区域配电网建设与发展。

基于改进LSTM神经网络的电动汽车充电负荷预测

当前对电动汽车(EV)充电负荷预测的研究缺少真实的数据支撑,并且模型考虑场景过于简单,影响因素考虑不到位,预测结果缺乏说服力。基于此,提出一种考虑多种电动汽车充电负荷影响因素的电动汽车充电负荷预测方法。首先,考虑天气、季节、温度、工作日、节假日等因素对电动汽车充电负荷的影响,采用三标度层次分析法分析各影响因素权重;其次,建立LSTM神经网络预测模型,通过真实数据训练得到用于预测的LSTM神经网络模型,结合影响因素权重分析结果对预测模型进行修正,得到最终的改进LSTM神经网络负荷预测模型;最后,采用常州某小区的真实数据对所提预测方法进行试验验证。结果表明,所提方法可以实现电动汽车充电负荷的精确预测,且负荷预测结果可为有序充电策略研究提供参考。

计及电动汽车保有量增长需求的充电负荷预测

当前对电动汽车充电负荷的研究大多集中在短期演变,对长时间尺度下的发展情况并未有较多研究。文中提出一种电动汽车保有量增长需求的充电负荷预测模型。首先采用萤火虫算法优化电动汽车保有量灰色预测模型的相关参数,对某地区2023—2033年电动汽车保有量进行预测;其次,综合考虑保有量预测结果、用户出行链、行驶里程及充电起始时间,结合在不同温度下的电动汽车电池容量和充电效率搭建充电负荷预测模型;最后,对江苏省某地区2023—2033年电动汽车充电负荷进行仿真预测。仿真结果有效地预测了电动汽车在未来10年中保有量发展趋势以及考虑保有量增长需求的充电负荷。

基于可进化模型预测控制的含电动汽车多微电网智能发电控制策略

多微电网中的环境状态、控制资源及偶然事件均具有强不确定性,而电动汽车在参与电网削峰填谷的同时也给发电控制带来了挑战。为此,该文提出一种基于可进化模型预测控制(LBMPC)的含电动汽车多微电网发电控制策略。首先,基于控制器交互的多微电网互联结构,考虑了发电机端电压调节和负荷频率控制(LFC)之间的耦合关系,建立含电动汽车多微电网的发电控制模型;然后,设计了一种基于多智能体的控制器参数自适应算法:频率控制器以实时频偏和EV站输出功率边界为状态集,以模型预测控制(MPC)控制器的可调参数矩阵Q_(x)作为动作集,以频率偏差为奖励函数指标,电压控制器同理,从而实现MPC与PI控制器权重参数的自适应调整;最后,仿真结果表明,自动调压(AVR)回路增加了有功功率干扰,对LFC控制器提出了更高的要求,与传统控制和MPC算法相比,应用于控制器互联结构的可进化模型预测控制器能够在子微电网之间进行信息交换,并且根据环境状态实时更新控制器参数,显著提高了多微电网频率控制过程的鲁棒性和快速性。同时,与纯深度确定性策略梯度(DDPG)控制器相比,该文提出的双层控制结构在机器学习智能体出现故障无法正常输出动作时,能更好地保证系统的安全稳定运行。

西藏高原村镇住宅冬季典型热电负荷特征研究

针对当前村镇住宅负荷特征表征中缺少基于用能行为差异的热电负荷关联性分析,以西藏高原村镇居住建筑为研究对象,通过走访与调研归纳出典型建筑结构特征、人员活动等信息,提炼出8种典型用能情景。利用EnergyPlus模拟计算不同用能行为下典型居住建筑全年逐时热电负荷,根据k均值聚类分析提取村镇住宅热电负荷特征,并提出基于逐时负荷系数的动态负荷典型化方法。最后基于特征参数化分析得到用能行为差异下西藏高原村镇建筑各类用户的用能特征。

建筑虚拟电厂参与需求响应市场的报量报价机制设计

高负荷密度的受端城市电网正面临日益严重的调节资源匮乏问题,城市建筑拥有大量中央空调、电动汽车等优质灵活资源,可通过聚合构建虚拟电厂参与城市电网的供需互动。随着需求响应市场的快速发展,灵活资源的市场化定价和交易成为趋势。为此,考虑到建筑和虚拟电厂两者的获益需求,设计了建筑虚拟电厂参与需求响应市场交易的报量报价机制。首先,根据建筑负荷的响应特性,将其分为无损可转移负荷、有损可转移负荷和有损可削减负荷,并分别提出其容量-成本的计算方法。然后,设计了同时保证建筑和虚拟电厂可靠收益的分配方法,以持续激励两者参与市场的积极性。在此基础上,建立了虚拟电厂参与市场交易的报价优化模型,实现不同场景下虚拟电厂的收益最大化。最后,通过算例仿真证明了所提机制在市场交易和收益分配方面的有效性。

Prediction of office building electricity demand using artificial neural network by splitting the time horizon for different occupancy rates

Due to the impact of occupants’activities in buildings,the relationship between electricity demand and ambient temperature will show different trends in the long-term and short-term,which show seasonal variation and hourly variation,respectively.This makes it difficult for conventional data fitting methods to accurately predict the long-term and short-term power demand of buildings at the same time.In order to solve this problem,this paper proposes two approaches for fitting and predicting the electricity demand of office buildings.The first proposed approach splits the electricity demand data into fixed time periods,containing working hours and non-working hours,to reduce the impact of occupants’activities.After finding the most sensitive weather variable to non-working hour electricity demand,the building baseload and occupant activities can be predicted separately.The second proposed approach uses the artificial neural network(ANN)and fuzzy logic techniques to fit the building baseload,peak load,and occupancy rate with multi-variables of weather variables.In this approach,the power demand data is split into a narrower time range as no-occupancy hours,full-occupancy hours,and fuzzy hours between them,in which the occupancy rate is varying depending on the time and weather variables.The proposed approaches are verified by the real data from the University of Glasgow as a case study.The simulation results show that,compared with the traditional ANN method,both proposed approaches have less root-mean-square-error(RMSE)in predicting electricity demand.In addition,the proposed working and non-working hour based regression approach reduces the average RMSE by 35%,while the ANN with fuzzy hours based approach reduces the average RMSE by 42%,comparing with the traditional power demand prediction method.In addition,the second proposed approach can provide more information for building energy management,including the predicted baseload,peak load,and occupancy rate,without requiring additional building parameters.

基于多维关联规则的用电负荷智能预测方法

用电负荷预测受到冗余数据影响,负荷预测值与实际值相差较大,因此提出基于多维关联规则的用电负荷智能预测方法。使用多维关联规则挖掘用电负荷频繁项集,获取全部用电负荷待预测数据,根据挖掘结果划分用电负荷种类。计算多维关联规则提升度,预处理冗余数据,生成待预测目标集。根据获取的用电序列,整合全部频繁项集,构建预测模型,并进行强关联学习。通过调整负荷数据训练收敛程度,获取用电负荷的最大、最小值。在用电设备节点中注入用电负荷预测多维关联规则修正数值,避免噪声数据影响预测结果。实验结果表明,该方法最大、最小负荷与实际数据,分别在9月30日和6月15日存在5 MW和0.3 MW的误差,说明该方法预测结果精准。

考虑环境因素的电动汽车充电站实时负荷预测模型

为了减少电动汽车大规模集成到电网造成的不利影响,提出了一种能够实现充电站充电负荷精准预测的方法。该方法利用LightGBM(light gradient boosting machine)与XGBoost(eXtreme gradient boosting)模型构建线下-线上组合模型。考虑充电负荷、时间、温度、天气等历史数据,利用LightGBM模型初步建立充电负荷线下预测模型;基于XGBoost模型,以线下预测模型输出负荷和实际负荷的误差为优化目标,实时变化的交通流量为协变量,建立线上预测模型,并对初步预测结果进行误差修正。某市实际充电站预测结果表明,相比于随机森林(RF)、LightGBM模型、XGBoost模型、多层感知机(MLP)以及LightGBM-RF组合模型,该组合模型具有更高的预测精度,同时可以准确预测不同充电站的实时充电负荷。