Long range peak demand forecasting under conditions of high growth

An extended logistic model with a varying asymptotic upper bound for long-range peak demand forecasting is described. The model has been applied to a typical fast growing system, the Saudi Consolidated Electric Company. The forecasts are compared with actual demands and with those obtained from classical forecasting methods. The model gave relatively accurate peak demand forecasts compared with other classical methods. The model with a single load observation is capable of producing several peak demand forecasts corresponding to different levels of maximum temperature and various levels of social activity. The forecasts produced by the model were also stable irrespective of the length of the ex-post simulation period     

Short-term electric load forecasting based on Kalman filtering algorithm with moving window weather and load model

This paper presents a novel time-varying weather and load model for solving the short-term electric load-forecasting problem. The model utilizes moving window of current values of weather data as well as recent past history of load and weather data. The load forecasting is based on state space and Kalman filter approach. Time-varying state space model is used to model the load demand on hourly basis. Kalman filter is used recursively to estimate the optimal load forecast parameters for each hour of the day. The results indicate that the new forecasting model produces robust and accurate load forecasts compared to other approaches. Better results are obtained compared to other techniques published earlier in the literature.     

山洪灾害预警研究进展与展望

山洪灾害预警是当前世界各国防洪减灾研究的热点问题。基于国内外山洪灾害预警研究的总结分析,文章系统梳理了山洪灾害预警的技术内涵与关键环节,包括预警数据获取、预警模型计算、预警指标确定、预警不确定分析等四个方面。目前山洪灾害预警研究存在以下不足：多源预警数据融合分析有待加强;复杂下垫面条件下降雨径流关系研究不够深入;山洪灾害预警预报精度差;山洪灾害动态预警机制仍需完善。未来,攻克山洪灾害形成机理,构建山洪灾害立体感知体系预警系统,给出山洪灾害预警多指标的阈值区间,可作为山洪灾害预警发展的总体方向。     

基于RBF神经网络与模糊控制的短期负荷预测

针对电力系统短期负荷预测,综合考虑温度、日期类型和天气等因素对短期电力负荷的影响,建立了径向基函数（Radial?Basis?Function,RBF）神经网络和模糊控制相结合的短期负荷预测模型。该模型利用RBF神经网络的非线性逼近能力对预测日负荷进行了预测,并采用在线自调整因子的模糊控制对预测误差进行在线智能修正。实际算例表明RBF神经网络与模糊控制相结合提高了预测精度。     

动态等维新息模型在中长期负荷预测中的应用

负荷预测是选取电力系统规划变电站及发电厂主变压器容量的主要依据,以实际某市供电负荷为例,建立起动态等维新息模型,把原始数据和预测的数据进行组合,利用回归模型,得到模型的预测方程,并从模型预测结果的相对误差和模型精度等级进行对比,说明结合动态等维新息模型,能使中长期电力负荷预测结果更靠近实际负荷值。     

基于累积式自回归动平均法和反向传播神经网络的短期负荷预测模型

针对电力系统短期负荷的特点建立了将累积式自回归动平均法(autoregressive integrated moving average,ARIMA)和采用反向传播算法(back propagation,BP)的神经网络法相结合的短期负荷预测模型。该模型利用ARIMA方法对线性时间序列逼近能力强的特点首先对预测日负荷进行预测,然后应用BP神经网络方法对预测结果进行修正,因此克服了单一算法存在的不足。应用该模型对某地区电网进行负荷预测,结果表明该方法的预测效果较好     

Overview and comparison of long-term forecasting techniques for a fast developing utility: part I

The intrinsic uncertainties associated with demand forecasting become more acute when it is required to provide an invaluable dimension to the decision-making process in a period characterized by fast and dynamic changes. In this paper, estimates of the peak demand, pertaining to a typical fast growing system with inherit dynamic load characteristics, and also a normal developing system, is derived from the classical long-term forecasting methods. These demand estimates are compared with corresponding actual values. Then, a proposed model based on demand characteristics of fast developing utility is obtained to yield best fit. Afterwards, improved modeling of the system load characteristics using a knowledge-based expert system, described in a companion paper (Part II), will demonstrate better forecasts compared with forecasts obtained by direct applications of classical techniques.     

基于数值天气预报的风能预测系统

随着世界范围内风电事业的飞速发展,大量大容量风电机组直接接入高压输电网络,是对电网安全运营、电能质量保证的重大挑战,迫切需要使用风能预测系统来对风电机组的发电功率进行预测。提出一种基于数值天气预报以及人工神经网络的混合型风能预测系统。该系统以基于数值天气预报的风速和风向预测数据作为输入数据,以自组织神经网络作为预处理模型,以径向基函数网络模型作为预测模型,并依据特定风电机组或风场的发电量的历史数据对输出数据进行修正。用该预测系统对河北某风电场进行了实例计算,得到可接受的预测结果,表明系统可行。     

Short-term electricity prices forecasting in a competitive market: A neural network approach

This paper proposes a neural network approach for forecasting short-term electricity prices. Almost until the end of last century, electricity supply was considered a public service and any price forecasting which was undertaken tended to be over the longer term, concerning future fuel prices and technical improvements. Nowadays, short-term forecasts have become increasingly important since the rise of the competitive electricity markets. In this new competitive framework, short-term price forecasting is required by producers and consumers to derive their bidding strategies to the electricity market. Accurate forecasting tools are essential for producers to maximize their profits, avowing profit losses over the misjudgement of future price movements, and for consumers to maximize their utilities. A three-layered feedforward neural network, trained by the Levenberg-Marquardt algorithm, is used for forecasting next-week electricity prices. We evaluate the accuracy of the price forecasting attained with the proposed neural network approach, reporting the results from the electricity markets of mainland Spain and California.     

Short-term load forecasting with local ANN predictors

A new technique for artificial neural network (ANN) based short-term load forecasting (STLF) is presented in this paper. The technique implemented active selection of training data, employing the k-nearest neighbors concept. A novel concept of pilot simulation was used to determine the number of hidden units for the ANNs. The ensemble of local ANN predictors was used to produce the final forecast, whereby the iterative forecasting procedure used a simple average of ensemble ANNs. Results obtained using data from two US utilities showed forecasting accuracy comparable to those using similar techniques. Excellent forecasts for one-hour-ahead and five-days-ahead forecasting, robust behavior for sudden and large weather changes, low maximum errors and accurate peak-load predictions are some of the findings discussed in the paper     

Short term load forecasting using wavelet transform combined with Holt–Winters and weighted nearest neighbor models

Short term load forecasting (STLF) is an integral part of power system operations as it is essential for ensuring supply of electrical energy with minimum expenses. This paper proposes a hybrid method based on wavelet transform, Triple Exponential Smoothing (TES) model and weighted nearest neighbor (WNN) model for STLF. The original demand series is decomposed, thresholded and reconstructed into deterministic and fluctuation series using Haar wavelet filters. The deterministic series that reflects the slow dynamics of load data is modeled using TES model while the fluctuation series that reflects the faster dynamics is fitted by WNN model. The forecasts of two subseries are composed to obtain the 24 h ahead load forecast. The performance of the proposed model is evaluated by applying it to forecast the day ahead load in the electricity markets of California and Spain. The results obtained demonstrate the forecast accuracy of the proposed technique.     

基于数值降雨预报信息的中期径流预报方法研究

随着数值气象预报水平的不断发展,考虑数值降雨预报信息有利于提高流域径流预报的精度,能够为水库未来的兴利调度决策提供可靠的信息支撑。为此,本文以浑江桓仁水库流域为研究实例,分别采用新安江模型和多元线性回归模型建立流域汛期和非汛期的中期旬径流预报模型,其中模型参数分别采用遗传算法和最小二乘法进行优化率定。在以上模型的基础上,采用美国全球预报系统发布的未来10 d数值降雨预报信息作为降雨输入,预报桓仁水库的中期旬径流。研究结果表明中期径流预报受降雨预报信息的不确定性影响,预报精度随预见期延长而降低,但仍高于传统不考虑降雨预报信息的中期径流预报。     

Neural network load forecasting with weather ensemble predictions

In recent years, a large amount of literature has evolved on the use of artificial neural networks (ANNs) for electric load forecasting. ANNs are particularly appealing because of their ability to model an unspecified nonlinear relationship between load and weather variables. Weather forecasts are a key input when the ANN is used for forecasting. This paper investigates the use of weather ensemble predictions in the application of ANNs to load forecasting for lead times from one to ten days ahead. A weather ensemble prediction consists of multiple scenarios for a weather variable. We use these scenarios to produce multiple scenarios for load. The results show that the average of the load scenarios is a more accurate load forecast than that produced using traditional weather forecasts. We use the load scenarios to estimate the uncertainty in the ANN load forecast. This compares favorably with estimates based solely on historical load forecast errors.     

Four methods for short-term load forecasting using the benefits of artificial intelligence

Four methods are developed for short-term load forecasting and are tested with the actual data from the Turkish Electrical Authority. The method giving the most successful forecasts is a hybrid neural network model which combines off-line and on-line learning and performs real-time forecasts 24-hours in advance. Loads from all day types are predicted with 1.7273% average error for working days, 1.7506% for Saturdays and 2.0605% for Sundays.     

A hybrid intelligent approach for the prediction of electricity consumption

Power load forecasting is an essential tool for energy management systems. Accurate load forecasting supports power companies to make unit commitment decisions and schedule maintenance plans appropriately. In addition to minimizing the power generation costs, it is also important for the reliability of energy systems. This research study presents the implementation of a novel fuzzy wavelet neural network model on an hourly basis, and validates its performance on the prediction of electricity consumption of the power system of the Greek Island of Crete. In the proposed framework, a multiplication wavelet neural network has replaced the classic linear model, which usually appears in the consequent part of a neurofuzzy scheme, while subtractive clustering with the aid of the Expectation–Maximization algorithm is being utilized in the definition of fuzzy rules. The results related to the minimum and maximum load using metered data obtained from the power system of the Greek Island of Crete indicate that the proposed forecasting model provides significantly better forecasts, compared to conventional neural networks models applied on the same dataset.     

Mixed Models for Short-Run Forecasting of Electricity Prices: Application for the Spanish Market

Short-run forecasting of electricity prices has become necessary for power generation unit schedule, since it is the basis of every profit maximization strategy. In this article a new and very easy method to compute accurate forecasts for electricity prices using mixed models is proposed. The main idea is to develop an efficient tool for one-step-ahead forecasting in the future, combining several prediction methods for which forecasting performance has been checked and compared for a span of several years. Also as a novelty, the 24 hourly time series has been modelled separately, instead of the complete time series of the prices. This allows one to take advantage of the homogeneity of these 24 time series. The purpose of this paper is to select the model that leads to smaller prediction errors and to obtain the appropriate length of time to use for forecasting. These results have been obtained by means of a computational experiment. A mixed model which combines the advantages of the two new models discussed is proposed. Some numerical results for the Spanish market are shown, but this new methodology can be applied to other electricity markets as well     

一种基于数据挖掘的馈线可装容量模型分析方法

馈线作为配网运行最关键的设备之一,评估馈线供电能力是保障配网运行的重要手段。本文通过引入馈线组负荷同时系数和需要系数两个参数,构建计算模型,求解馈线可装容量以评估馈线供电能力。首先,通过聚类分析和神经网络预测等方法预测馈线组负荷同时系数。然后,将馈线各负荷根据其实际接入容量情况分为饱和负荷和未饱和负荷,采用灰色预测和神经网络相结合的组合预测方法计算未饱和负荷的需要系数。最后,将预测得到的两个系数代入馈线可装容量计算模型进行求解。实际算例分析表明:所提方法的计算结果具有一定的预测趋势,充分利用了馈线载流量,并兼顾了配电网运行的可靠性,对于指导电网营销部门业扩报装工作具有重要意义。     

Analysis of the value for unit commitment of improved loadforecasts

Load forecast errors can yield suboptimal unit commitment decisions. The economic cost of inaccurate forecasts is assessed by a combination of forecast simulation, unit commitment optimization, and economic dispatch modeling for several different generation/load systems. The forecast simulation preserves the error distributions and correlations actually experienced by users of a neural net-based forecasting system. Underforecasts result in purchases of expensive peaking or spot market power; overforecasts inflate start-up and fixed costs because too much capacity is committed. The value of improved accuracy is found to depend on load and generator characteristics; for the systems considered here, a reduction of 1% in mean absolute percentage error (MAPE) decreases variable generation costs by approximately 0.1%-0.3% when MAPE is in the range of 3%-5%. These values are broadly consistent with the results of a survey of 19 utilities, using estimates obtained by simpler methods. A conservative estimate is that a 1% reduction in forecasting error for a 10,000 MW utility can save up to $1.6 million annually     

基于分形的电力系统负荷预测

本文提出了一个基于分形的电力系统负荷预测新方法,它根据分形拼贴定理求取一个与负荷历史记录相近的吸引子的迭代函数系统（ＩＦＳ）以建立预测模型实现对未来电力负荷的预测,实例计算表明,该方法精度高,速度快,不存在收敛问题,且数据收集简便,因此有很好的实用价值。     

An adaptable automated procedure for short-term electricity loadforecasting

The Irish Electricity Supply Board requires forecasts of system demand or electrical load for: (a) one day ahead; and (b) 7-10 days ahead. Here, the authors concentrate on and give results only for one day ahead forecasts although the method is also applicable for 7-10 days ahead. A forecasting model has been developed which identifies a `normal' or weather-insensitive load component and a weather-sensitive load component. Linear regression analysis of past load and weather data is used to identify the normal load model. The weather-sensitive component of the load is estimated using the parameters of regression analysis. Certain design features of the short-term load forecasting system are important for its successful operation over time. These include adaptability to changing operational conditions, computational economy and robustness. An automated load forecasting system is presented here that includes these design features. A fully automated algorithm for updating the model is described in detail as are the techniques employed in both the identification and treatment of influential points in the data base and the selection of predictors for the weather-load model. Monthly error statistics of forecast load for only one day ahead are presented for recorded weather conditions