【摘要】：局部通風機是礦山井下掘進工作面中的主要通風機械設備,能夠稀釋和排放工作面中的有毒有害氣體,保證給井下作業(yè)人員一個安全、可靠、良好的工作條件。目前,我國大多數(shù)金屬礦山井下局部通風機調(diào)速能力較差,智能化程度較低,并且多數(shù)是不調(diào)速的,風機一通電便開始長期無休止地恒速運轉(zhuǎn),其轉(zhuǎn)速不會隨有毒有害氣體濃度的變化而變化,常發(fā)生“一風吹”通風現(xiàn)象,不能針對井下實際情況來調(diào)整工作區(qū)域所需風量,這不僅浪費了大量電能,而且空氣質(zhì)量經(jīng)常不合格,直接影響到礦井的安全生產(chǎn)。針對礦山通風存在的問題,本文提出了將PID控制,模糊控制和神經(jīng)網(wǎng)絡相結(jié)合的控制策略,設計了一種基于ANFIS-PID控制的局部通風機風量自動調(diào)節(jié)系統(tǒng)。該方法能夠?qū)崟r調(diào)節(jié)工作區(qū)域所需風量,及時排出有毒有害氣體,同時還能節(jié)約電能。本文所設計的控制器是以有毒有害氣體的濃度偏差E和偏差變化率EC為輸入變量,PID控制器的輸出變量來模擬控制現(xiàn)場情況。在ANFIS中采用的是混合學習算法,即在向前運算中,保持所有條件參數(shù)不變,采用最小二乘法來改變結(jié)論參數(shù);在改進后的參數(shù)不變的情況下,采用誤差逆向傳播法來改變條件參數(shù),這樣就可以達到改變隸屬度函數(shù)形狀的目的,最后使整個樣本集的均方差達到系統(tǒng)所規(guī)定的精度要求。在MATLAB中利用Simulink搭建系統(tǒng)模型,對本文提出的ANFIS-PID控制算法進行仿真試驗,并與PID控制算法、自適應模糊PID控制算法進行仿真比較。仿真結(jié)果表明:ANFIS-PID控制器比其它兩種控制器的動態(tài)響應曲線好、響應時間短、無超調(diào)量、穩(wěn)態(tài)精度高和魯棒性強。研究表明:本文提出的ANFIS-PID控制策略,不需要被控對象的精確數(shù)學模型,在具有時滯大、時變性、非線性的風量調(diào)節(jié)系統(tǒng)中工作穩(wěn)定、適應性好、抗干擾能力強、魯棒性強,而且系統(tǒng)設計方法簡單,能方便用于實際工業(yè)控制中。
[Abstract]:Local ventilator is the main ventilation mechanical equipment in the underground driving face, which can dilute and discharge the poisonous and harmful gas in the working face, and ensure a safe, reliable and good working condition for the underground workers. At present, local fans in most metal mines in our country have poor speed regulation ability and lower degree of intelligence, and most of them do not adjust speed. As soon as the fan is electrified, the fan starts to run endlessly at a constant speed for a long time. The rotational speed does not change with the change of the concentration of toxic and harmful gases. The phenomenon of "one wind blowing" often occurs. It is impossible to adjust the required air volume in the working area according to the actual situation in the underground, which not only wastes a lot of electric energy. And air quality is often not qualified, directly affect the safety of mine production. Aiming at the existing problems of mine ventilation, this paper puts forward a control strategy combining PID control, fuzzy control and neural network, and designs an automatic control system of local fan air flow based on ANFIS-PID control. This method can adjust the air volume needed in the working area in real time, discharge toxic and harmful gases in time, and save electric energy at the same time. The controller designed in this paper uses the concentration deviation E of toxic and harmful gases and the rate of variation of deviation EC as the output variables of the pid controller to simulate the control field. In ANFIS, a hybrid learning algorithm is used, that is, in forward operation, all conditional parameters remain unchanged, and the least square method is used to change the conclusion parameters. The error reverse propagation method is used to change the conditional parameters so that the shape of the membership function can be changed and the RMS of the whole sample set can meet the precision requirements of the system. The ANFIS-PID control algorithm proposed in this paper is simulated and compared with the PID control algorithm and the adaptive fuzzy PID control algorithm by using Simulink to build the system model in MATLAB. The simulation results show that the ratio ANFIS-PID controller has better dynamic response curve, shorter response time, no overshoot, higher steady-state accuracy and better robustness than the other two controllers. The results show that the proposed ANFIS-PID control strategy does not need the precise mathematical model of the controlled object, and it is stable, adaptable, robust and robust in the air volume control system with large time-delay, time-varying and nonlinear. Moreover, the system design method is simple and can be used in practical industrial control conveniently.
【學位授予單位】：南華大學
【學位級別】：碩士
【學位授予年份】：2015
【分類號】：TD441

【相似文獻】

相關(guān)碩士學位論文 前1條

1 韓艷杰;基于ANFIS-PID控制的局部通風機風量自動調(diào)節(jié)系統(tǒng)仿真研究[D];南華大學;2015年

，

本文編號：2180296