基于JADE的多智能體動態(tài)博弈自動發(fā)電控制仿真平臺研究
發(fā)布時間：2018-09-09 19:31
【摘要】：隨著電網(wǎng)規(guī)模的不斷發(fā)展，當前電網(wǎng)已經(jīng)成為一個復雜互聯(lián)系統(tǒng)包含多個控制區(qū)域。電網(wǎng)供應的電能其質量評定的一個重要指標是頻率。合格的頻率是電網(wǎng)運行所必須的，其能確保電網(wǎng)的穩(wěn)定與安全。進行頻率控制的主要手段是采用自動發(fā)電控制。因此，自動發(fā)電控制所能達到的效果對電網(wǎng)供應的電能質量至關重要。由于運行時間的不同，如每天的用電高峰和低谷時，電力系統(tǒng)運行點會隨之改變。這就對電力系統(tǒng)控制性能提出了更高的要求。當某一區(qū)域的控制器在某一時段無法零化其控制誤差時，系統(tǒng)頻率和聯(lián)絡線功率交換將無法穩(wěn)定運行。因此，在AGC控制器中引入智能算法是解決復雜互聯(lián)系統(tǒng)多區(qū)域協(xié)調控制的一種可行的方法。這是因為智能算法相比以固定增益控制器來控制的傳統(tǒng)控制算法擁有更好的適應性。然而，，國內外并沒有統(tǒng)一標準的AGC智能仿真平臺以實現(xiàn)互聯(lián)電網(wǎng)多控制區(qū)域的協(xié)作控制和有效評估智能算法的控制效果。 為此，本文提出了基于JADE平臺開發(fā)新型多智能體動態(tài)博弈的自動發(fā)電控制技術，并實時模擬由于新能源接入而帶來的電網(wǎng)強隨機環(huán)境下的自動發(fā)電控制，主要內容包括以下方面： （1）本文簡要介紹了多智能體系統(tǒng)，以及多智能體系統(tǒng)在電力系統(tǒng)中的應用，并對自動發(fā)電控制進行了概述。 （2）本文在第三章中介紹了多智能體仿真平臺JADE的基本概念，并對JADE中智能體的行為和通信進行了詳細的介紹。 （3）本文在第四章中給出了所設計的基于JADE技術的多智能體自動發(fā)電控制仿真平臺，該平臺由頻率響應模型、實時數(shù)據(jù)傳輸模塊和多智能體控制算法模塊三部分組成。 （4）本文第五章中利用所設計的平臺進行了仿真，通過對仿真結果的分析，可知平臺的控制效果良好。此平臺可為未來復雜互聯(lián)多區(qū)域的控制協(xié)作算法的研究提供更多公正而可靠的評測，為設計和提升算法性能提供研究基礎。同時，該平臺的建立也為未來復雜多區(qū)域強隨機互聯(lián)系統(tǒng)的AGC控制手段的升級提供了一條走向實際應用的思路。
[Abstract]:With the continuous development of power grid scale, power grid has become a complex interconnected system including multiple control areas. Frequency is an important index of quality evaluation of electric energy supplied by power grid. Qualified frequency is necessary for the operation of power grid, which can ensure the stability and security of power grid. The main means of frequency control is automatic generation control. Therefore, the effect of automatic generation control is very important to the power quality of power supply. Because of the different operating time, such as the peak and low point of daily electricity consumption, the power system operation point will change. This puts forward higher requirements for power system control performance. When the controller in a certain region can not zero its control error at a certain time, the system frequency and tie-line power exchange will not operate stably. Therefore, the introduction of intelligent algorithm into AGC controller is a feasible method to solve the multi-area coordinated control of complex interconnected systems. This is because the intelligent algorithm has better adaptability than the traditional control algorithm controlled by fixed gain controller. However, there is no unified AGC intelligent simulation platform to realize the cooperative control of multi-control area and evaluate the control effect of intelligent algorithm. In this paper, a new multi-agent dynamic game based automatic generation control technology based on JADE platform is proposed, and the automatic generation control under the strong random environment of power grid caused by the new energy access is simulated in real time. The main contents are as follows: (1) this paper briefly introduces the multi-agent system and the application of multi-agent system in power system. An overview of automatic generation control is given. (2) in Chapter 3, the basic concept of JADE, a multi-agent simulation platform, is introduced. The behavior and communication of agents in JADE are introduced in detail. (3) in chapter 4, the simulation platform of multi-agent automatic generation control based on JADE technology is presented, which is based on frequency response model. The real-time data transmission module and the multi-agent control algorithm module are composed of three parts. (4) in the fifth chapter of this paper, we use the designed platform to simulate, and through the analysis of the simulation results, we can see that the control effect of the platform is good. This platform can provide more fair and reliable evaluation for the future research of control collaboration algorithm in complex interconnected multi-region, and provide the research foundation for designing and improving the performance of the algorithm. At the same time, the establishment of the platform also provides a way of practical application for upgrading the AGC control methods of complex multi-region strong stochastic interconnected systems in the future.
【學位授予單位】：華南理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TM921.5
本文編號：2233357
[Abstract]:With the continuous development of power grid scale, power grid has become a complex interconnected system including multiple control areas. Frequency is an important index of quality evaluation of electric energy supplied by power grid. Qualified frequency is necessary for the operation of power grid, which can ensure the stability and security of power grid. The main means of frequency control is automatic generation control. Therefore, the effect of automatic generation control is very important to the power quality of power supply. Because of the different operating time, such as the peak and low point of daily electricity consumption, the power system operation point will change. This puts forward higher requirements for power system control performance. When the controller in a certain region can not zero its control error at a certain time, the system frequency and tie-line power exchange will not operate stably. Therefore, the introduction of intelligent algorithm into AGC controller is a feasible method to solve the multi-area coordinated control of complex interconnected systems. This is because the intelligent algorithm has better adaptability than the traditional control algorithm controlled by fixed gain controller. However, there is no unified AGC intelligent simulation platform to realize the cooperative control of multi-control area and evaluate the control effect of intelligent algorithm. In this paper, a new multi-agent dynamic game based automatic generation control technology based on JADE platform is proposed, and the automatic generation control under the strong random environment of power grid caused by the new energy access is simulated in real time. The main contents are as follows: (1) this paper briefly introduces the multi-agent system and the application of multi-agent system in power system. An overview of automatic generation control is given. (2) in Chapter 3, the basic concept of JADE, a multi-agent simulation platform, is introduced. The behavior and communication of agents in JADE are introduced in detail. (3) in chapter 4, the simulation platform of multi-agent automatic generation control based on JADE technology is presented, which is based on frequency response model. The real-time data transmission module and the multi-agent control algorithm module are composed of three parts. (4) in the fifth chapter of this paper, we use the designed platform to simulate, and through the analysis of the simulation results, we can see that the control effect of the platform is good. This platform can provide more fair and reliable evaluation for the future research of control collaboration algorithm in complex interconnected multi-region, and provide the research foundation for designing and improving the performance of the algorithm. At the same time, the establishment of the platform also provides a way of practical application for upgrading the AGC control methods of complex multi-region strong stochastic interconnected systems in the future.
【學位授予單位】：華南理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TM921.5
【參考文獻】
相關期刊論文 前9條
1 劉紅進,袁斌,戴宏偉,祁達才,焦連偉,倪以信,吳復立;多代理系統(tǒng)及其在電力系統(tǒng)中的應用[J];電力系統(tǒng)自動化;2001年19期
2 申群太;彭云建;張靜秋;;基于面向對象知識庫和產(chǎn)生式規(guī)則知識庫的電網(wǎng)調度操作票專家系統(tǒng)[J];電氣自動化;2002年02期
3 束洪春,唐嵐,董俊;多Agent技術在電力系統(tǒng)中的應用展望[J];電網(wǎng)技術;2005年06期
4 詹紅霞;顏泉;;Multi-agent系統(tǒng)在電力系統(tǒng)中的應用[J];重慶大學學報(自然科學版);2006年11期
5 胡舜耕;張莉;鐘義信;;多Agent系統(tǒng)的理論、技術及其應用[J];計算機科學;1999年09期
6 張汝波,顧國昌,劉照德,王醒策;強化學習理論、算法及應用[J];控制理論與應用;2000年05期
7 蔣麗娟;劉衛(wèi)國;;基于層次黑板模型的多Agent系統(tǒng)研究[J];計算機系統(tǒng)應用;2008年05期
8 楊煜普,李曉萌,許曉鳴;多智能體協(xié)作技術綜述[J];信息與控制;2001年04期
9 趙偉;白曉民;丁劍;方竹;李再華;;基于協(xié)同式專家系統(tǒng)及多智能體技術的電網(wǎng)故障診斷方法[J];中國電機工程學報;2006年20期
本文編號：2233357
本文鏈接：http://sikaile.net/kejilunwen/dianlilw/2233357.html
