电力系统负荷预测综述

负荷预测对电力系统非常重要,是电力系统进行规划、调度的基础,也是电力系统安全、稳定和经济运行的保障。精确的负荷预测是电力系统一直追求的目标,各种现代的新兴算法、方法运用到负荷预测之中,不同的预测方法由于自身的特点对于负荷预测的适用性也不尽相同,对于不同方法在负荷预测中的综述就显得很有必要。讨论了国内外的负荷预测的研究现状,分析了进行电力系统负荷预测的多种传统方法和现代智能方法,并总结出各种预测方法的优缺点和适用性,对于电力系统在选择负荷预测方法时具有一定参考价值。最后,对智能电网下的几种特定负荷预测场景进行了介绍,以这些角度去看待负荷预测问题,得出适用于这些场景的负荷预测方法,对于未来的负荷预测的发展也进行了展望。     

Empirical Mode Decomposition based ensemble deep learning for load demand time series forecasting

Load demand forecasting is a critical process in the planning of electric utilities. An ensemble method composed of Empirical Mode Decomposition (EMD) algorithm and deep learning approach is presented in this work. For this purpose, the load demand series were first decomposed into several intrinsic mode functions (IMFs). Then a Deep Belief Network (DBN) including two restricted Boltzmann machines (RBMs) was used to model each of the extracted IMFs, so that the tendencies of these IMFs can be accurately predicted. Finally, the prediction results of all IMFs can be combined by either unbiased or weighted summation to obtain an aggregated output for load demand. The electricity load demand data sets from Australian Energy Market Operator (AEMO) are used to test the effectiveness of the proposed EMD-based DBN approach. Simulation results demonstrated attractiveness of the proposed method compared with nine forecasting methods.     

A novel combined model based on hybrid optimization algorithm for electrical load forecasting

Accurate electrical load forecasting always plays a vital role in power system administration and energy dispatch, which are the foundation of the smooth operation of the national economy and people’s daily life. Thinking from this vision, many scholars have made great efforts to seek suitable optimization algorithms to improve the performance of existing forecasting algorithm. However, most of the studies ignore the inherent disadvantages of single optimization algorithm, which leads to sub-optimal forecasting performance. Therefore, a novel electric load forecasting system was successfully proposed in this paper by the combination of data preprocessing, hybrid optimization algorithms, and several single classical forecasting methods, which successfully overcomes the defects of single traditional forecasting models and achieves higher forecasting accuracy than that of single model optimization. Besides, the 30 min interval data of Queensland, Australia from March to April is used as illustrative examples to evaluate the performance of the developed model. The results of tests demonstrate that the proposed hybrid model can better approximate the actual value, and it can also be employed as a useful tool for smart grids dispatching planning.     

Short-term electric power load forecasting using feedforward neural networks

Abstract: This paper presents the results of a study on short‐term electric power load forecasting based on feedforward neural networks. The study investigates the design components that are critical in power load forecasting, which include the selection of the inputs and outputs from the data, the formation of the training and the testing sets, and the performance of the neural network models trained to forecast power load for the next hour and the next day. The experiments are used to identify the combination of the most significant parameters that can be used to form the inputs of the neural networks in order to reduce the prediction error. The prediction error is also reduced by predicting the difference between the power load of the next hour (day) and that of the present hour (day). This is a promising alternative to the commonly used approach of predicting the actual power load. The potential of the proposed method is revealed by its comparison with two existing approaches that utilize neural networks for electric power load forecasting.     

基于改进自适应卡尔曼滤波的电力大负荷预测与计费研究

电力大负荷预测是电力公司进行高效电力系统规划和运行的重要基础;为了提高电力负荷预测精度进而更加有效地估计电力计量与计费,创新地提出了一种基于改进的自适应卡尔曼滤波(AKF)的电力大负荷计量计费预估方法;分析了电力负荷预测研究现状,针对传统卡尔曼滤波算法不足,引入自适应遗忘因子对卡尔曼滤波算法进行改进,建立数学模型、整定因子调整模型关键参数,得到电力大负荷数据的预测值,最终通过计量计费转换公式得用电量以及电费计量预估值;仿真结果表明：基于AEKF的电力大负荷计量预测方法的负荷预测结果与实际结果误差小于1.35%,电力计费预测结果与实际结果相对误差小于1.263%;应用实例证明：基于AEKF的电力大负荷计量计费预估方法,能够提高电力公司的调度效率12%,增加电费营收5.3%～12.2%。     

A least squares support vector machine model optimized by moth-flame optimization algorithm for annual power load forecasting

Annual power load forecasting is essential for the planning, operation and maintenance of an electric power system, which can also mirror the economic development of a country to some extent. Accurate annual power load forecasting can provide valuable references for electric power system operators and economic managers. With the development of Energy Internet and further reformation of electric power market, power load forecasting has become a more difficult and challenging task. In this paper, a new hybrid annual power load forecasting model based on LSSVM (least squares support vector machine) and MFO (Moth-Flame Optimization algorithm) is proposed, which the parameters of LSSVM model are optimally determined by the latest nature-inspired metaheuristic algorithm MFO. Meanwhile, the rolling mechanism is also employed. The forecasting results of China’s annual electricity consumption indicate the proposed MFO-LSSVM model shows much better forecasting performance than single LSSVM, FOA-LSSVM (LSSVM optimized by fruit fly optimization), and PSO-LSSVM (LSSVM optimized by particle swarm optimization). MFO, as a new intelligence optimization algorithm, is attractive and promising. The LSSVM model optimized by MFO can significantly improve annual power load forecasting accuracy.     

Anomalous load profile detection in power systems using wavelet transform and robust regression

In modern power systems, a clean database of historical electric loads is crucial for system analysis, planning studies, load forecasting, and decision making. To this end, the clean database should be free of anomalies while preserving the power consumption behavior of special events like holidays. In recent years, model-based anomaly detection methods have been successfully used for anomaly detection and correction in electrical load. In these methods, the collected load data are first compared with the predicted load obtained from a forecasting model; if the difference is greater than a specific limit, the collected load data are replaced with the predicted value. One of the disadvantages of these methods is the consideration of load profiles of the days with higher levels of difficulty in load forecasting, such as holidays, as anomalous load profiles and replacing them with predicted values, which results in fading some important characteristics of the load series. Therefore in this paper, a two-stage anomaly detection method is proposed to alleviate this problem. In the first stage, the wavelet transform is used to identify suspected load profiles. In the second stage, a robust regression method is used to further inspect the suspected load profiles and decide whether they are anomalies or not. By performing experiments on three datasets (a utility of Iran, New York City, and a North American utility) and comparing the results with four other methods, it’s shown that the proposed method decreases false detection of anomalies while accurately correcting anomalous load profiles. Results of load forecasting with the Iranian dataset corrected by the proposed method showed an average of 2.28% and 1.58% error on the holidays and total days, respectively, which is less than the other four methods.     

基于PCA与改进BP神经网络相结合的电网中长期负荷预测

本文在标准反向传播神经网络的基础上,提出一种结合主成分分析法和改进的误差反向传播神经网络的方法来对电网中长期的电力负荷进行预测。首先利用主成分分析法对电力负荷的影响因素进行特征提取,有效地降低数据样本的维度,消除数据的冗余和线性信息,保留主要成分作为模型的输入数据。然后在标准的神经网络的反向传播环节中引入动量项和陡度因子。两种方法的结合有效地解决了网络收敛速度慢和容易陷入局部最小值的问题。将此方法应用于济源市的中长期电力负荷预测,实验结果表明,基于主成分分析法与改进的反向传播神经网络相结合的方法比常用的标准的反向传播神经网络、基于多变量的时间序列网络及时间序列网络具有更高的计算效率和预测精度,证明提出的预测模型在电力负荷预测中是有效的。     

灰色变异粒子群算法在电力负荷预测中的应用

由于电力负荷量是电力系统发展的基础,因此提高电力负荷量预测的准确性有利于电力系统的快速发展. 本文利用粒子群算法优化参数的良好性能和灰色预测法适合预测不确定因素影响系统的优势,提出了灰色变异粒子群组合预测模型来预测电力负荷量,提高了电力负荷预测的精度,并通过实例对组合预测模型的预测精度和有效性进行了分析. 结果表明,此组合预测模型的精度优于单一的灰色预测模型,且优于其他几种预测算法,该组合模型能很好地预测电力负荷量,为电力系统的决策和发展提供了可靠的科学数据.     

基于气象数据可视化降维和多模加权组合的短期负荷预测

针对短期负荷预测中数据预处理的必要性和单一预测模型的局限性,提出了一种基于气象数据可视化降维和多模加权组合的短期负荷预测方法。该方法将可视化降维、模态分解降噪、单一预测模型和权重确定理论相结合,构建了气象数据降维、历史负荷分解、模态分量降噪和多模加权组合的短期负荷预测模型。通过设置3种对比实验环境,对某地区供电公司所提供的电力负荷和气象数据进行分析。预测结果及误差分析表明,所提短期负荷预测方法在保留高维气象因素本质特征结构的同时,能有效结合数据预处理方法及单一预测模型的特点,有效提升该地区电网负荷的预测精度。     

基于CNN-LSTM及深度学习的风电场时空组合预测模型

为了更好地预测风电场的风电功率,提取风电场相邻站点之间时空信息和潜在联系,提出了一种基于卷积神经网络(CNN)、互信息(mutual information, MI)法、长短时记忆网络(LSTM)、注意力机制(AT)和粒子群优化(PSO)的短期风电场预测模型(MI-CNN-ALSTM-PSO)。CNN用于提取不同站点的空间特征,LSTM则用于获取多个站点的风电数据的时间依赖信息,据此设计CNN-LSTM时空预测模型,并结合深度学习算法,如MI特征选择、 AT注意力机制、 PSO参数优化,对模型进一步改进。通过两个海岛风电场的实验数据分析可知,所提模型具有最优的统计误差,CNN-LSTM模型可以高效提取风电场时空信息并进行时间序列预测,而结合深度学习算法(MI、 AT和PSO)后的组合模型能进一步提高风电功率预测精度和稳定性。     

短期电力负荷预测的灰色-小波网络组合模型

短期电力负荷数据具有离散、无规则波动的特点,先利用灰色预测弱化其波动性,然后将负荷原始检测数据与其相对应的灰色预测数据进行重构后作为小波网络的训练样本,在此基础上建立基于灰色-小波网络组合模型的短期电力负荷预测新方法。该方法有效整合了灰色理论、小波分析和人工神经网络的优点,与传统BP网络相比,收敛速度更快,预测精度更高。仿真试验表明了该方法用于短期电力负荷预测的可行性和有效性。     

基于云模型的电力系统负荷组合预测

电力系统负荷预测的精确度决定着电网安全稳定、高效的运行.最小二乘支持向量机(LSSVM)被广泛应用电力系统负荷预测上,然而该方法在处理不确定性问题上有很多不足之处.为了更精确的选择核函数的参数,处理不确定性因素,提高短期负荷预测的精度,提出了一种将云模型、粒子群优化(PSO)和LSSVM相结合的组合模型.首先通过对各影响因子的不确定性分析,按不确定性高低将各影响因子分别应用Cloud-LSSVM和PSO-LSSVM进行预测,然后通过组合模型的加权计算的得到最终预测值.最后,通过仿真对比证明该模型能更好的处理不确定性,从而提高电力系统短期负荷预测精度.     

引入多级扰动的混合型粒子群优化算法

为解决粒子群优化算法易陷入局部最优值的问题,提出一种引入多级扰动的混合型粒子群优化算法.该算法结合两种经典改进粒子群优化算法的优点,即带惯性参数的标准粒子群优化算法和带收缩因子的粒子群优化算法,在此基础上,引入多级扰动机制：在更新粒子位置时,引入一级扰动,使粒子对解空间的遍历能力得到加强;若优化过程陷入“局部最优”的情况,则引入二级扰动,使得优化过程继续,从而摆脱局部最优值.使用了6个测试函数——Sphere函数、Ackley函数、Rastrigin函数、Styblinski-Tang函数、Duadric函数及Rosenbrock函数来对所提出的混合型粒子群优化算法进行仿真运算和对比验证.模拟运算的结果表明：所提出的混合型粒子群优化算法在对测试函数进行仿真时,其收敛精度和收敛速度都优于另外两种经典的改进粒子群优化算法;另外,在处理多峰函数时,本算法不易被局部最优值所限制.     

计及评价指标冲突的电动汽车充电负荷区间预测

电动汽车充电负荷具有强随机性,且受电池容量与用户用车行为影响。为有效预测充电负荷时序分布,本文提出一种计及评价指标冲突的充电负荷区间预测方法。首先,该方法分析日间充电负荷间时序相关性,并用强相关历史日充电负荷数据构建充电负荷预测所需的特征集。接着,采用弯曲高斯过程（warped Gaussian process , WGP）方法,并结合多种协方差函数来构建多个充电负荷区间预测模型。为解决多指标评价存在冲突和仅选择最优的一个预测模型会出现极端误差问题,本文应用面积灰关联决策方法,对各模型开展计及评价指标冲突的综合评价,并依据获取的面积灰关联贴近度,构建电动汽车充电负荷组合区间预测模型。实验结果表明,本文提出的方法能够获得更精确、覆盖率更高的充电负荷预测区间。     

组合预测模型在电力系统负荷预测中的应用研究

本文基于灰色预测模型、滑动平均模型和指数平滑模型这三种单一预测模型,采用方差-协方差策略,建立组合预测模型。然后结合老挝电力系统的概况,对老挝的全国年用电量进行预测和分析。结果表明,组合预测模型的预测精度明显高于各单一预测模型,即组合预测模型的相对误差小于各单一预测模型的相对误差,说明组合预测模型具有相当的适用性和优越性。     

计及光伏及风电并网的电力系统短期负荷预测

为提升光伏、风电等分布式能源大量接入电网后短期电力负荷的预测精度,促进电网消纳能力提升,本文对光伏出力及短期用电负荷采用小波——径向基函数(RBF)神经网络预测方法;对风力发电首先利用总体平均经验模态分解(EEMD)方法对其功率数据分解,再采用BP神经网络、RBF神经网络、小波神经网络、ELMAN神经网络四种神经网络预测方法进行预测,并用粒子群算法(PSO)和灰色关联度(GRA)修正。最后,利用等效负荷的概念,分析光伏、风力发电并网对于短期电力负荷预测的影响,并将三种模型有效结合,得到了考虑光伏及风力发电并网的电力系统短期负荷预测的等效负荷预测模型。实例分析表明,本文所提方法相较于其他方法在该预测项目上具有相对更高的预测精度。     

基于PSO-BP混合算法的短期电力负荷预测

将粒子群优化算法和BP神经网络算法相结合,形成粒子群一神经网络（PSO—BP）混合算法,建立了涉及各种影响因素的短期负荷预测模型。运用所建立的PSO-BP混合算法和BP算法的负荷预测模型进行短期负荷预测,比较所得结果可知,PSO-BP混合算法预测精度较高,效果较好。     

短期负荷预测方法综述

电力系统为电力用户提供可靠、优质、经济、环保电能的前提条件是能对电力负荷进行精确的预测。电网中的调度部门要根据短期负荷的预测结果来安排发电和供电计划,从而优化资源配置,提高经济效益。因此,短期负荷预测具有重大意义。为了提高负荷预测的准确性,较为全面的综述了短期负荷预测方法的研究状况。首先简述了短期负荷预测的意义、特点以及影响因素,综合叙述了短期负荷预测方法的历史发展。然后分别从数据预处理和组合预测两个方面总结了各个方法的研究现状和存在的问题。最后指出了当前短期负荷预测研究的主要问题以及下一步可能的研究方向。 得出的结论是对历史数据进行预处理后结合时下流行的机器学习算法能提高电力系统短期负荷预测的精度。     

基于多因素加法模型的中期电力负荷预测

提前准确预测所需电力负荷,做好电力规划是电力部门保证电力供应稳定不可或缺的重要环节.基于欧洲智能网络(EUNITE)竞赛电力数据和北美电力数据,提出一种多因素加法模型,进行中期电力预测.考虑到温度、假期、星期等因素对电力负荷产生不同的影响,拟合出这些因素与电力负荷之间的映射关系,相加得到电力负荷预测的函数.还比较了业界常用的7种不同的算法模型,使用6种不同指标对这些模型和多因素加法模型进行评估,实验结果发现,在这8种不同算法模型中,多因素加法模型有着更加精确的预测性能,运算速度比其他模型快,并且模型更加容易理解和解释.