A Decision-Support System Based on Particle Swarm Optimization for Multiperiod Hedging in Electricity Markets

This paper proposes a particle swarm optimization (PSO) approach to support electricity producers for multiperiod optimal contract allocation. The producer risk preference is stated by a utility function (U) expressing the tradeoff between the expectation and variance of the return. Variance estimation and expected return are based on a forecasted scenario interval determined by a price range forecasting model developed by the authors. A certain confidence level alpha is associated to each forecasted scenario interval. The proposed model makes use of contracts with physical (spot and forward) and financial (options) settlement. PSO performance was evaluated by comparing it with a genetic algorithm-based approach. This model can be used by producers in deregulated electricity markets but can easily be adapted to load serving entities and retailers. Moreover, it can easily be adapted to the use of other type of contracts.     

文化微粒群神经网络在用电量预测中的应用

为了提高传统神经网络在中长期用电量负荷预测中的速度和预测精度,将文化算法、微粒群算法融入神经网络中,设计了文化微粒神经网络模型;将该模型用于我国某地区中长期用电量预测建模,采用了滚动时间窗技术处理输入输出数据,进一步优化模型数据输入量.该方法综合了微粒群算法的全局寻优能力和文化算法的演化优势.通过与传统的灰色预测模型以...     

粒子群优化的灰色模型在中长期负荷预测中的应用

针对GM(1,1)模型的局限性及在负荷预测中存在的问题,提出了一种基于粒子群优化的灰色模型.粒子群算法是一种新的全局优化算法,有很强的全局寻优能力,用它来优化灰色模型的背景值及初始值修正值,能较好地提高电力系统中长期负荷预测的精度.在虚拟仪器LabVIEW平台上进行仿真,验证了所提方法的有效性.     

基于粒子群支持向量机的短期电力负荷预测

在分析支持向量机SVM(Support VectorM ach ine)回归估计方法参数性能的基础上,提出粒子群算法PSO(Partic le Swarm Optim ization)优化参数的SVM短期电力负荷预测模型。PSO算法是一种新型的基于群体智能的随机优化算法,简单易于实现且具有更强的全局优化能力。用所建立的负荷预测模型编制的M atlab仿真程序,对某实际电网进行了短期负荷预测,结果表明预测精度更高。     

基于VMD和PSO-SVR的短期电力负荷多阶段优化预测

为降低短期负荷序列的非线性以提升预测精度,提出一种基于多阶段优化的变分模态分解（variational mode decomposition, VMD）和粒子群算法优化支持向量回归（particle swarm optimization support vector regression, PSO-SVR）的短期电力负荷预测模型。第1阶段采用VMD优化和预处理原始负荷序列,分解获得多个较为平稳的模态分量。第2阶段利用相空间重构优化重组各序列分量,并针对各分量分别建立支持向量回归（support vector regression,SVR）预测模型。第3阶段将粒子群算法（particle swarm optimization,PSO）用于优化SVR模型内部参数,便于更好地进行训练和预测。最后累加所有序列的预测值,实现短期电力负荷预测。研究结果表明：所提方法可以取得更高的预测精度。     

基于VMD和PSO-SVR的短期电力负荷多阶段优化预测

为降低短期负荷序列的非线性以提升预测精度,提出一种基于多阶段优化的变分模态分解（variational mode decomposition, VMD）和粒子群算法优化支持向量回归（particle swarm optimization support vector regression, PSO-SVR）的短期电力负荷预测模型。第1阶段采用VMD优化和预处理原始负荷序列,分解获得多个较为平稳的模态分量。第2阶段利用相空间重构优化重组各序列分量,并针对各分量分别建立支持向量回归（support vector regression,SVR）预测模型。第3阶段将粒子群算法（particle swarm optimization,PSO）用于优化SVR模型内部参数,便于更好地进行训练和预测。最后累加所有序列的预测值,实现短期电力负荷预测。研究结果表明：所提方法可以取得更高的预测精度。     

基于Attention-GRU的短期电价预测

通过分析得出电价与负荷具有相关性,因此在电价预测模型中需要考虑实时负荷的影响。在此基础上针对前馈神经网络不能处理时序数据的缺陷与LSTM神经网络预测速度慢的问题,提出了一种基于Attention-GRU (Attention gated recurrent unit, Attention-GRU)的实时负荷条件下短期电价预测模型。该模型充分利用电价的时序特性,并采用Attention机制突出了对电价预测起关键性作用的输入特征。以美国PJM电力市场实时数据为例进行分析,通过与其他几种预测模型相比,验证了该方法具有更高的预测精度;与LSTM神经网络相比具有更快的预测速度。     

基于PSO算法优化GRU神经网络的短期负荷预测

为了实现高精度的电力系统短期负荷预测,提出了基于粒子群优化(particle swarm optimization,PSO)算法优化门控循环单元(gated recurrent unit,GRU)神经网络的电力系统短期负荷预测模型。首先建立GRU神经网络,GRU神经网络采用了门控循环单元,与采用传统循环单元的传统循环神经网络相比,克服了传统循环神经网络中可能出现的梯度爆炸和梯度消失问题;继而采用具有较强全局优化能力的改进粒子群算法对GRU神经网络参数进行优化,有效提高模型的预测精度。通过实际算例仿真分析,并与传统的GRU神经网络预测模型以及反向传播(back propagation,BP)神经网络预测模型进行对比,验证了所提电力系统短期负荷预测模型具有较好的精度和稳定性。     

粒子群优化算法在电力系统中的应用

粒子群优化方法是一种基于群体智能的新型演化计算技术.它在函数优化、神经网络设计、分类、模式识别、信号处理、机器人技术等许多领域已取得了成功应用,但在电力系统中应用的研究起步较晚,关于它实际应用的报道尚不多见.文章较为全面地详述了粒子群优化方法在配电网扩展规划、检修计划、机组组合、负荷经济分配、最优潮流计算与无功优化控制、谐波分析与电容器配置、配电网状态估计、参数辨识、优化设计等方面应用的主要研究成果.随着粒子群优化理论研究的深入,它还将在电力市场竞价交易、投标策略以及电力市场仿真等领域发挥巨大的应用潜力.     

基于粒子群优化的非线性灰色Bernoulli模型在中长期负荷预测中的应用

将非线性灰色Bernoulli模型用于到中长期电力负荷预测,提出了优选模型参数的粒子群优化算法。该模型是将GM(1,1)模型与Bernoulli微分方程相结合的一种灰色模型,适用于对不同发展趋势曲线的预测。通过粒子群优化算法,以模型预测平均绝对百分误差最小为目标,选择最优的模型参数。采用不同测试数据以及实际电网负荷数据进行了验证,结果表明上述模型有很好的适应性及较高的预测精度。     

基于改进LS-SVM的短期电力负荷预测方法研究

针对电力负荷随机性强、稳定性差、预测精度不理想等问题,提出了一种基于粒子群优化PSO和最小二乘支持向量机LS-SVM的短期负荷预测方法。模型的输入因子是负荷数据和气象信息等。粒子群优化算法用于实现支持向量机参数的自动优化,建立了基于粒子群优化的最小二乘支持向量机短期负荷预测模型。通过仿真验证了改进前后预测模型的准确性和有效性,结果表明,改进的预测方法具有收敛性好、预测精度高、训练速度快的优点。本研究为我国短期负荷预测方法的发展提供了参考和借鉴。     

自适应变系数粒子群和径向基神经网络在短期电价预测中的应用(英文)

分析了传统的粒子群优化(particle swarm optimization,PSO)算法和径向基(radial basis function,RBF)神经网络的优缺点,提出一种自适应变系数粒子群优化算法(adaptive variable coefficients particle swarm optimizer,AVCPSO)。该算法与RBF神经网络结合形成自适应变系数粒子群-径向基(AVCPSO-RBF)神经网络混合优化算法。基于此优化算法,建立了短期电价预测模型,并利用贵州电网历史数据进行短期电价预测。仿真计算结果表明,AVCPSO-RBF混合优化算法在短期电价预测中优于传统RBF神经网络法和PSO-RBF神经网络方法,克服了上述2种方法的缺点,改善了RBF神经网络的泛化能力,具有输出稳定性好、预测精度高、收敛速度快等特点,使用该方法得到的各日预测电价的平均百分比误差可控制在2%以内,平均绝对误差最大值为1.652RMB/MW·h。     

A proposed grey model for short-term electricity price forecasting in competitive power markets

In order to improve the performance of traditional grey models for short-term electricity price prediction in competitive power markets, a novel grey model, named PGM(1,2,a,b), is proposed in this paper. In the proposed model, the reference sequence is determined with correlation method. Furthermore, considering the limitation of the least square method (LSM), particle swarm optimization algorithm (PSO) is adopted to identify the parameters instead of LSM. To demonstrate the superiority of the proposed model, public available data coming from Nordpool, California, and Ontario power markets were used for training and testing. Simulation results show that the proposed model is capable of forecasting short-term electricity price efficiently and accurately.     

Mixed price and load forecasting of electricity markets by a new iterative prediction method

Load and price forecasting are the two key issues for the participants of current electricity markets. However, load and price of electricity markets have complex characteristics such as nonlinearity, non-stationarity and multiple seasonality, to name a few (usually, more volatility is seen in the behavior of electricity price signal). For these reasons, much research has been devoted to load and price forecast, especially in the recent years. However, previous research works in the area separately predict load and price signals. In this paper, a mixed model for load and price forecasting is presented, which can consider interactions of these two forecast processes. The mixed model is based on an iterative neural network based prediction technique. It is shown that the proposed model can present lower forecast errors for both load and price compared with the previous separate frameworks. Another advantage of the mixed model is that all required forecast features (from load or price) are predicted within the model without assuming known values for these features. So, the proposed model can better be adapted to real conditions of an electricity market. The forecast accuracy of the proposed mixed method is evaluated by means of real data from the New York and Spanish electricity markets. The method is also compared with some of the most recent load and price forecast techniques.     

基于改进粒子群-径向基神经网络模型的短期电力负荷预测

为了准确、快速、高效地预测电网短期负荷,提出了改进的粒子群–径向基神经网络算法。用改进的粒子群算法训练径向基神经网络,实现了径向基函数神经网络的参数优化。建立了短期电力负荷预测模型,综合考虑气象、天气、日期类型等影响负荷的因素进行短期负荷预测。算例结果表明,该算法优于径向基神经网络法和粒子群–径向基网络算法,克服了径向基网络和粒子群优化方法的缺点,改善了径向基神经网络的泛化能力,输出稳定,预测精度高,收敛速度快,平均百分比误差可控制在1.2%以内。     

PSO演化神经网络集成的边际电价预测新方法

为了克服神经网络模型结构和参数难以设置,学习算法收敛速度慢等缺点,提出了一种基于粒子群优化的演化神经网络集成新模型对日前交易电力市场的边际电价进行预测。该模型将边际电价预测问题转化为神经网络实际输出与预测输出误差最小化问题,首先采用粒子群优化算法把神经网络的结构和权重映射成问题空间中的粒子,通过粒子速度和位置更新方程进行粗学习,获得多个相对占优的神经网络结构和初始权重并构成神经网络集成预测模型,然后采用梯度学习算法和交叉验证对神经网络集成单元的权重进行细学习,并以误差最小的神经网络集成单元的输出作为神经网络集成预测模型的输出。运用此方法对加州日前交易电力市场的边际电价进行了日预测,结果表明其优于三层BP神经网络预测方法。     

考虑预警负荷的电动汽车充放电优化策略

预警负荷会严重影响电力系统的安全经济运行。面向参与车辆到电网(vehicle to grid,V2G)服务的电动汽车用户,综合考虑预警负荷、预警电价和充电激励措施对充放电过程的影响,提出基于改进粒子群算法(improved particle sw arm optimization,IPSO)的电动汽车充放电优化策略。通过计算预警负荷发生时的放电奖励,建立预警负荷电价模型、电池容量损耗模型,基于分时电价和放电激励制度建立用户充放电成本模型。此外,引入长短时记忆的概念,提出改进粒子群优化算法。在上述模型和算法的基础上,以最小化用户成本为优化目标,计及用户充电需求和充放电功率等约束,提出不同预警负荷情况下的充放电优化策略。在MATLAB中完成了仿真验证,结果表明,在已知预测预警负荷的前提下,采用文中的充放电优化策略能够提高电动汽车用户V2G参与度,有效降低用户成本,并缓解预警负荷发生时电网压力。     

A new ARMAX model based on evolutionary algorithm and particle swarm optimization for short-term load forecasting

In this paper, a new ARMAX model based on evolutionary algorithm and particle swarm optimization for short-term load forecasting is proposed. Auto-regressive (AR) and moving average (MA) with exogenous variables (ARMAX) has been widely applied in the load forecasting area. Because of the nonlinear characteristics of the power system loads, the forecasting function has many local optimal points. The traditional method based on gradient searching may be trapped in local optimal points and lead to high error. While, the hybrid method based on evolutionary algorithm and particle swarm optimization can solve this problem more efficiently than the traditional ways. It takes advantage of evolutionary strategy to speed up the convergence of particle swarm optimization (PSO), and applies the crossover operation of genetic algorithm to enhance the global search ability. The new ARMAX model for short-term load forecasting has been tested based on the load data of Eastern China location market, and the results indicate that the proposed approach has achieved good accuracy.     

基于TCSC技术和粒子群优化算法的电力系统阻塞疏导方法

提出了一种基于晶闸管控制的串连电容器(thyristor controlled series capacitor,TCSC)技术和粒子群优化算法的电力系统阻塞疏导方法。首先根据线路灵敏度分析确定安装TCSC的线路;然后提出了电力市场环境下电网中含有TCSC装置的阻塞疏导计算数学模型;最后运用粒子群优化算法对这一数学模型进行参数优化,达到疏导电网阻塞的目的。IEEE 14节点系统算例表明,基于TCSC技术进行电网阻塞疏导是有效、合理的。     

Energy procurement management for electricity retailer using new hybrid approach based on combined BICA–BPSO

These smaller consumers usually prefer to purchase at a constant price that is adjusted at most a few times per year. Electricity retailers are in business to bridge the gap between the wholesale market and these smaller consumers. This paper proposes a hybrid approach based on binary imperialist competitive algorithm (BICA) and binary particle swarm optimization (BPSO) to find the optimal energy procurement for electricity retailer with multiple procurement options such as self-generating facility, bilateral contracts and pool market purchase in restructured electricity markets. The solutions of these problems provide adequate information to obtain an energy procurement strategy for retailers. Also, the results of proposed hybrid approach are compared with conventional ICA and PSO methods. Test results show that the proposed hybrid approach is more effective and has higher capability in finding optimum solutions in comparison to ICA and PSO methods. A case study is used to illustrate the efficiency of the proposed hybrid approach.