微電網(wǎng)潛在調(diào)節(jié)能力評估及微網(wǎng)群能量調(diào)度方法研究
發(fā)布時間：2018-12-15 12:27
【摘要】：目前,小型風／光分布式發(fā)電以微電網(wǎng)的形式作為一個雙向可調(diào)度單元從低壓配電網(wǎng)接入,已成為電力新能源技術(shù)的一個重要的發(fā)展趨勢。對配電網(wǎng)(或微網(wǎng)群)而言微電網(wǎng)是作為一個整體雙向可控單元來參與電能配送,是配電系統(tǒng)的新成員。微電網(wǎng)作為一個整體可控單元,由于其內(nèi)部發(fā)電單元大都是隨機性很強的可再生能源發(fā)電電源,所以其整體對外所呈現(xiàn)的源/載特征既具有隨機性又與微電網(wǎng)的控制相關(guān),使得外部電網(wǎng)將整個微電網(wǎng)作為一個可控單元進行調(diào)度成為難點。也就是說,外電網(wǎng)對微電網(wǎng)內(nèi)部的發(fā)用電設備通常無法直接進行調(diào)度和控制,因此,如何量化評估微電網(wǎng)的這種隨機源/載特性,如何對微電網(wǎng)入口交換能量進行調(diào)度是一個業(yè)內(nèi)亟待解決的問題。本文圍繞微電網(wǎng)中隨機電源和負載預測、微電網(wǎng)潛在調(diào)節(jié)能力評估和微網(wǎng)群實時能量優(yōu)化調(diào)度幾個方面的理論和方法進行了深入研究,主要成果如下：(1)針對傳統(tǒng)預測方法趨勢性和波動性不能兼顧的問題,提出了一種基于均生函數(shù)-最優(yōu)子集回歸的超短期風電功率預測方法,該方法通過對原始樣本序列進行兩次差分處理,生成均生函數(shù)序列和累加序列,兼顧了風速的趨勢性和波動性特點,因而具有較高的預測精度和速度。該方法同樣適用解決負荷超短期預測的趨勢性和波動性問題,通過仿真計算驗證了該方法對超短期風電功率和負荷預測的有效性。(2)基于統(tǒng)計學理論提出了描述微電網(wǎng)可調(diào)節(jié)能力的4個狀態(tài)變量,風險備用電量、風險缺電量、已存電量和存電空間,為上級電網(wǎng)對微電網(wǎng)的能量調(diào)度奠定了基礎。在此基礎上提出了微電網(wǎng)潛在調(diào)節(jié)能力的評估指標和基于狀態(tài)空間的計算方法。并通過蒙特卡洛仿真實驗模擬微電網(wǎng)在各狀態(tài)空間運行時儲能單元對源/載隨機性的消納作用,驗證了上述理論的正確性。(3)以微電網(wǎng)可調(diào)節(jié)能力作為決策變量,首次建立了以微電網(wǎng)入口交換電量為控制變量微網(wǎng)群實時能量優(yōu)化調(diào)度數(shù)學模型,提出了將該模型轉(zhuǎn)換為最小費用最大流網(wǎng)流模型的方法。對微網(wǎng)群優(yōu)化調(diào)度指令執(zhí)行運行效果的仿真模擬實驗結(jié)果表明,通過微網(wǎng)群優(yōu)化調(diào)度可以實現(xiàn)可再生能源最大化利用,在微網(wǎng)群離網(wǎng)模式下能夠使群體總棄能和減載電量最小,并網(wǎng)模式下能夠使微網(wǎng)群與上級電網(wǎng)交互電量最小。(4)將提出的基于風險電量的微電網(wǎng)潛在調(diào)節(jié)能力評估方法和微網(wǎng)群實時能量調(diào)度方法應用于課題示范工程廣西貴港市三里一中微網(wǎng)群系統(tǒng),完成了示范微電網(wǎng)潛在調(diào)節(jié)能力評估和微網(wǎng)群能量調(diào)度策略的制定,由此對所提出的上述理論和方法進行了實用驗證。
[Abstract]:At present, small wind / light distributed generation is connected from low voltage distribution network as a bidirectional dispatching unit in the form of micro-grid, which has become an important development trend of new power energy technology. For distribution network (or micro-grid group), microgrid is a new member of distribution system, which is a whole bi-directional controllable unit to participate in power distribution. As a whole controllable unit, the internal generating units of microgrid are mostly renewable energy generation power sources with strong randomness, so the source / load characteristics of microgrid are both random and related to the control of microgrid. It is difficult for the external grid to dispatch the whole microgrid as a controllable unit. That is to say, the power generating equipment inside the microgrid can not be directly scheduled and controlled by the foreign power network. Therefore, how to quantitatively evaluate the random source / load characteristics of the microgrid, How to dispatch the energy exchange at the entrance of microgrid is an urgent problem to be solved. In this paper, the theories and methods of stochastic power supply and load prediction, potential regulation capability evaluation and real-time energy scheduling of microgrid group are studied in detail. The main results are as follows: (1) aiming at the problem that the trend and volatility of traditional forecasting methods can not be taken into account, an ultra-short-term wind power prediction method based on average generation function and optimal subset regression is proposed. In this method, the original sample sequence is processed by two differential processes, and the mean generating function sequence and the cumulative sequence are generated, which take into account the trend and volatility of wind speed, so it has high prediction accuracy and speed. This method is also suitable for solving the trend and volatility problems of ultra-short term load forecasting. The simulation results show that the proposed method is effective for ultra-short-term wind power and load forecasting. (2) based on the statistical theory, four state variables, risk reserve power consumption, risk power shortage, are proposed to describe the adjustable capacity of microgrid. The stored energy and storage space lay the foundation for the energy dispatching of the microgrid in higher power grid. On this basis, the evaluation index of potential regulation capability of microgrid and the calculation method based on state space are proposed. The simulation results of Monte Carlo simulation are used to simulate the absorption of the energy storage unit to the source / load randomness when the microgrid is running in each state space, which verifies the correctness of the above theory. (3) the adjustable ability of the microgrid is taken as the decision variable. For the first time, a mathematical model of real-time energy optimal dispatching for microgrid group with the input exchange quantity as control variable is established, and a method for converting this model to the maximum drift-flow model with minimum cost is proposed. The simulation results show that the optimal scheduling of microgrid group can maximize the utilization of renewable energy and minimize the total energy loss and load reduction in the off-grid mode. In the grid-connected mode, the interaction between microgrid group and higher grid can be minimized. (4) the methods of potential regulation ability evaluation and real-time energy dispatching of micro-grid group proposed based on risk electricity quantity are applied to the demonstration project. The system of three li and one medium microgrid in Guigang city, Guangxi, China, The evaluation of potential regulation ability of microgrid and the formulation of energy dispatching strategy for microgrid group are completed. The theory and method mentioned above are verified practically.
【學位授予單位】：中國農(nóng)業(yè)大學
【學位級別】：博士
【學位授予年份】：2014
【分類號】：TM73
[Abstract]:At present, small wind / light distributed generation is connected from low voltage distribution network as a bidirectional dispatching unit in the form of micro-grid, which has become an important development trend of new power energy technology. For distribution network (or micro-grid group), microgrid is a new member of distribution system, which is a whole bi-directional controllable unit to participate in power distribution. As a whole controllable unit, the internal generating units of microgrid are mostly renewable energy generation power sources with strong randomness, so the source / load characteristics of microgrid are both random and related to the control of microgrid. It is difficult for the external grid to dispatch the whole microgrid as a controllable unit. That is to say, the power generating equipment inside the microgrid can not be directly scheduled and controlled by the foreign power network. Therefore, how to quantitatively evaluate the random source / load characteristics of the microgrid, How to dispatch the energy exchange at the entrance of microgrid is an urgent problem to be solved. In this paper, the theories and methods of stochastic power supply and load prediction, potential regulation capability evaluation and real-time energy scheduling of microgrid group are studied in detail. The main results are as follows: (1) aiming at the problem that the trend and volatility of traditional forecasting methods can not be taken into account, an ultra-short-term wind power prediction method based on average generation function and optimal subset regression is proposed. In this method, the original sample sequence is processed by two differential processes, and the mean generating function sequence and the cumulative sequence are generated, which take into account the trend and volatility of wind speed, so it has high prediction accuracy and speed. This method is also suitable for solving the trend and volatility problems of ultra-short term load forecasting. The simulation results show that the proposed method is effective for ultra-short-term wind power and load forecasting. (2) based on the statistical theory, four state variables, risk reserve power consumption, risk power shortage, are proposed to describe the adjustable capacity of microgrid. The stored energy and storage space lay the foundation for the energy dispatching of the microgrid in higher power grid. On this basis, the evaluation index of potential regulation capability of microgrid and the calculation method based on state space are proposed. The simulation results of Monte Carlo simulation are used to simulate the absorption of the energy storage unit to the source / load randomness when the microgrid is running in each state space, which verifies the correctness of the above theory. (3) the adjustable ability of the microgrid is taken as the decision variable. For the first time, a mathematical model of real-time energy optimal dispatching for microgrid group with the input exchange quantity as control variable is established, and a method for converting this model to the maximum drift-flow model with minimum cost is proposed. The simulation results show that the optimal scheduling of microgrid group can maximize the utilization of renewable energy and minimize the total energy loss and load reduction in the off-grid mode. In the grid-connected mode, the interaction between microgrid group and higher grid can be minimized. (4) the methods of potential regulation ability evaluation and real-time energy dispatching of micro-grid group proposed based on risk electricity quantity are applied to the demonstration project. The system of three li and one medium microgrid in Guigang city, Guangxi, China, The evaluation of potential regulation ability of microgrid and the formulation of energy dispatching strategy for microgrid group are completed. The theory and method mentioned above are verified practically.
【學位授予單位】：中國農(nóng)業(yè)大學
【學位級別】：博士
【學位授予年份】：2014
【分類號】：TM73
【相似文獻】
相關(guān)期刊論文 前10條
1 鄭漳華;艾芊;;微電網(wǎng)的研究現(xiàn)狀及在我國的應用前景[J];電網(wǎng)技術(shù);2008年16期
2 麻敏;韓繼明;麻秀慧;陳衍妍;;淺談微電網(wǎng)[J];科技資訊;2009年15期
3 樓書氫;李青鋒;許化強;劉魯?shù)?;國外微電網(wǎng)的研究概況及其在我國的應用前景[J];華中電力;2009年03期
4 楊為;丁明;畢銳;高研;丁銀;;微電網(wǎng)實驗平臺的設計[J];合肥工業(yè)大學學報(自然科學版);2010年01期
5 袁清芳;周作春;陳艷霞;李香龍;陳國鋒;;微電網(wǎng)發(fā)展應對策略[J];農(nóng)村電氣化;2010年10期
6 安智敏;羅時光;;微電網(wǎng)的概念及發(fā)展[J];內(nèi)蒙古石油化工;2011年21期
7 王t，
本文編號：2380629
本文鏈接：http://sikaile.net/kejilunwen/dianlilw/2380629.html
教材專著
