天然气负荷预测及调峰方案研究
发布时间：2019-05-26 20:16
【摘要】：为改善能源结构和治理环境污染,作为清洁能源的天然气需求量逐年大幅增加,天然气供应也出现了诸多新问题,如何预测和规划天然气用量,保障平稳供气,提高供气方案的可靠性和经济性等均是实际工程中经常遇到的难题。论文主要对天然气的负荷预测、调峰方案及调压站优化配置进行研究。首先利用相关分析确定天然气短期用气负荷的影响因子,借助SPSS和MATLAB软件,用回归分析法和人工神经网络法分别对天然气短期负荷进行预测,结果表明采用BP人工神经网络的预测值和实际天然气负荷变化趋势基本一致。其次,基于天然气中期负荷的特点,以某市冬季高峰供气阶段为例,利用BP人工神经网络建立预测模型,进行中期负荷的预测,预测误差在10%以内,对今后工程应用上有指导作用。论文对天然气的长期负荷预测进行了研究,考虑天然气用户的用气特点和和用户等级,结合用气结构的优化调整,创新性的提出新的用户分类方法。利用分类用户的特点及相应的预测方法,分别对某市的居民用户、商业用户、采暖用户、汽车用户和工业用户的长期负荷进行预测,与实际年负荷相比误差为9.3%,属于高精度预测,表明分类预测方法适用于长期负荷预测。最后对天然气供气方案中的调峰储气和调压站优化配置做了研究,提出采用单位调峰量年作用成本进行经济分析,分析对比多种调峰方式,并计算某市的调峰量,提出适合该市的近远期调峰方案。对高中压调压站的优化配置建立数学模型,确定了最佳配置数量和作用半径,并以某市新建片区为例,该数学模型可为今后的工程规划提供理论依据。
[Abstract]:In order to improve the energy structure and control environmental pollution, the demand for natural gas as clean energy has increased greatly year by year, and many new problems have emerged in the supply of natural gas. How to predict and plan the amount of natural gas to ensure the stable gas supply. Improving the reliability and economy of gas supply scheme is a common problem in practical engineering. In this paper, the load forecasting, peak shaving scheme and optimal configuration of voltage shunting station are studied. Firstly, the influencing factors of short-term gas load of natural gas are determined by correlation analysis, and the short-term load of natural gas is predicted by regression analysis and artificial neural network with the help of SPSS and MATLAB software. The results show that the predicted value of BP artificial neural network is basically consistent with the change trend of actual natural gas load. Secondly, based on the characteristics of medium-term load of natural gas, taking the peak gas supply stage in winter as an example, the prediction model is established by using BP artificial neural network to predict the medium-term load, and the prediction error is less than 10%. It has a guiding role in engineering application in the future. In this paper, the long-term load forecasting of natural gas is studied. Considering the characteristics and user grade of natural gas users, combined with the optimization and adjustment of gas structure, a new user classification method is proposed innovatively. By using the characteristics of classified users and the corresponding prediction methods, the long-term loads of resident users, commercial users, heating users, automobile users and industrial users in a city are predicted respectively, and the error is 9.3% compared with the actual annual load. It belongs to high precision prediction, which shows that the classification forecasting method is suitable for long-term load forecasting. Finally, the optimal configuration of peak-shaving gas storage and voltage regulating station in natural gas supply scheme is studied, and the economic analysis is carried out by using the annual action cost of unit peak-shaving quantity, and various peak-shaving methods are analyzed and compared, and the peak-shaving quantity of a city is calculated. A short-term and long-term peaking scheme suitable for the city is put forward. A mathematical model is established for the optimal allocation of high and high pressure regulating stations, and the optimal allocation quantity and action radius are determined. Taking a newly built area of a city as an example, the mathematical model can provide a theoretical basis for future engineering planning.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU996
本文编号：2485603
[Abstract]:In order to improve the energy structure and control environmental pollution, the demand for natural gas as clean energy has increased greatly year by year, and many new problems have emerged in the supply of natural gas. How to predict and plan the amount of natural gas to ensure the stable gas supply. Improving the reliability and economy of gas supply scheme is a common problem in practical engineering. In this paper, the load forecasting, peak shaving scheme and optimal configuration of voltage shunting station are studied. Firstly, the influencing factors of short-term gas load of natural gas are determined by correlation analysis, and the short-term load of natural gas is predicted by regression analysis and artificial neural network with the help of SPSS and MATLAB software. The results show that the predicted value of BP artificial neural network is basically consistent with the change trend of actual natural gas load. Secondly, based on the characteristics of medium-term load of natural gas, taking the peak gas supply stage in winter as an example, the prediction model is established by using BP artificial neural network to predict the medium-term load, and the prediction error is less than 10%. It has a guiding role in engineering application in the future. In this paper, the long-term load forecasting of natural gas is studied. Considering the characteristics and user grade of natural gas users, combined with the optimization and adjustment of gas structure, a new user classification method is proposed innovatively. By using the characteristics of classified users and the corresponding prediction methods, the long-term loads of resident users, commercial users, heating users, automobile users and industrial users in a city are predicted respectively, and the error is 9.3% compared with the actual annual load. It belongs to high precision prediction, which shows that the classification forecasting method is suitable for long-term load forecasting. Finally, the optimal configuration of peak-shaving gas storage and voltage regulating station in natural gas supply scheme is studied, and the economic analysis is carried out by using the annual action cost of unit peak-shaving quantity, and various peak-shaving methods are analyzed and compared, and the peak-shaving quantity of a city is calculated. A short-term and long-term peaking scheme suitable for the city is put forward. A mathematical model is established for the optimal allocation of high and high pressure regulating stations, and the optimal allocation quantity and action radius are determined. Taking a newly built area of a city as an example, the mathematical model can provide a theoretical basis for future engineering planning.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU996
【参考文献】
相关期刊论文 前3条
1 田一梅,赵元,赵新华;城市煤气负荷的预测[J];煤气与热力;1998年04期
2 王金凤;董静雅;王德意;;城镇燃气储气方式研究[J];天然气技术与经济;2011年03期
3 贾承造;张永峰;赵霞;;中国天然气工业发展前景与挑战[J];天然气工业;2014年02期
相关硕士学位论文 前1条
1 李庆生;武汉市城市天然气负荷预测和调峰方式优化[D];华中科技大学;2006年
，本文编号：2485603
本文链接：https://www.wllwen.com/jingjilunwen/jianzhujingjilunwen/2485603.html
