Improving the Accuracy of Short-term Forecasting of Electric

发布时间：2021-07-05 16:53

　　目前,国内外研发了许多种负荷预测模型和软件系统。多数模型的预测精度能够满足电力系统调度与用户需求,但在某些情况下,短期负荷预测结果并不总是很理想。因此,结合当地气象条件和自然光照的预测模型迫切需要发展。主要研究内容如下:（1）基于支持向量机和粒子群的算法,一个用于地区调度的短期电力系统负荷预测模型被建立。该方法将当地自然光作为影响预测精度的一个重要因素,因此可以提高预测精确度。（2）粒子群算法被用于优化支持向量回归模型的参数,其中空气温度和自然光照考虑作为影响因素。（3）通过对神经模糊网络和支持矢量机的预测结果比较显示,支持向量回归模型有最好的逼近性质,考虑电力系统负荷、空气温度和自然光照因素。课题的理论意义在于开发一种应用粒子群优化算法优化支持向量机的参数的预测模型。该方法建立了电力消费、空气温度和自然光照之间非线性关系以提高模型的预测精度。实际意义在于开发的模型可用来预测区域调度办事处分支机构的能源消费,批发发电公司和领土产生的公司、区域网公司,能源销售公司,以及在分派办事处的个别公司成员的批发或零售电力市场和权力。短期预测的计算机程序也被在MatLab环境下开发应用开发。 

【文章来源】：兰州交通大学甘肃省

【文章页数】：67 页

【学位级别】：硕士

【文章目录】：
Abstract
摘要
1. Review and analysis of modern methods and mathematical models to predict electricity consumption
    1.1 Classification of short-term load forecasting methods
    1.2 Statistical methods of forecasting
        1.2.1 Methods for regression
        1.2.2 Time series methods
        1.2.3 Methods based on wavelet transform of time series
    1.3 Methods of artificial intelligence
        1.3.1 Methods based on neural network models
        1.3.2 Methods based on fuzzy logic
        1.3.3 Support vector method
    1.4 Evolutionary algorithms
    1.5 Requirements for short-term forecasting of electricity consumption
    1.6 Main problems of short-term forecasting of electricity consumption
        1.6.1 Accuracy of the input - output relationship hypothesis
        1.6.2 Prediction of abnormal days
        1.6.3 Inaccurate weather forecast data
    1.7 Review of current literature on the problem of short-term power consumption forecasting
        1.7.1 Models of neural networks
        1.7.2 Models of neuro-fuzzy networks
        1.7.3 Model of wavelet transform
        1.7.4 Regression models
    1.8 Conclusions
2. Time series analysis of electricity consumption and its determinants
    2.1 Characteristics of the electrical load diagrams of the power system
    2.2 Time series of power consumption and influencing factors
    2.3 Seasonal and meteorological factors affecting power consumption
    2.4 Temperature and light: the analysis of their impact on power consumption in the control room operating area
    2.5 Random disturbances
    2.6 Conclusions
3. Modelling short term future energy consumption based on neural networks and evolutionary algorithms
    3.1 Short-term load forecasting using artificial neural network
    3.2 Short-term load forecasting using artificial neural networks and particle swarm optimization algorithm
    3.3 Short-term load forecasting using artificial neural networks and particle swarm optimization algorithm
        3.3.1 Data analysis and pre-processing
        3.3.2 The number of layers, neurons and transfer functions
        3.3.3 Training of built neural networks
        3.3.4 Architecture of the ANN for the operating zone
        3.3.5 The choice of input variables
        3.3.6 Building the structure of neural network
        3.3.7 Selection of data for training, testing and validation
        3.3.8 Simulation results
    3.4 Training the ANN on the basis of self-organization
        3.4.1 Dataset for the study
        3.4.2 Training of self-organizing maps
        3.4.3 The results of clustering and prediction
        3.4.4 Performance criteria
        3.4.5 Simulation results
    3.5 Conclusions
4. Models of future energy consumption based on neural fuzzy network and support vector method
    4.1 Predicting power consumption using adaptive neural fuzzy network
        4.1.1 The architecture of neuro-fuzzy model
        4.1.2 Hybrid algorithm for training neural networks
        4.1.3 Simulation result
    4.2 Energy consumption forecasting using support vector
        4.2.1 Simulation results
    4.3 Forecasting of power consumption based on the support vector method and particle swarm algorithm
        4.3.1 Load forecasting steps and processes
        4.3.2 A set of analysis parameters
        4.3.3 Simulation results
    4.4 Conclusions
Summarize
Acknowledgement
References
Research achievement during working for the degree


【参考文献】：
期刊论文
[1]基于神经网络和模糊理论的短期负荷预测[J]. 赵宇红,唐耀庚,张韵辉.  高电压技术. 2006(05)



本文编号：3266457