(1. 江西理工大學(xué) 材料與化學(xué)工程學(xué)院，贛州 341000；
2. 中南大學(xué) 冶金科學(xué)與工程學(xué)院，長(zhǎng)沙 410083；
3. 江西理工大學(xué) 信息工程學(xué)院，贛州 341000)
摘 要: 基于已建立的銅閃速熔煉神經(jīng)網(wǎng)絡(luò)模型，以能耗費(fèi)用最低為目標(biāo)，在工藝指標(biāo)控制范圍內(nèi)，采用遺傳算法對(duì)銅閃速熔煉過程的工藝參數(shù)進(jìn)行了仿真優(yōu)化計(jì)算。結(jié)果表明，當(dāng)空氣、分配風(fēng)、工藝氧和中央氧的市場(chǎng)價(jià)格折合比值分別為0.05、0.1、0.4和0.45，精礦量為128 t，其成分(質(zhì)量分?jǐn)?shù))為Cu 20.61%、S 27.59%、Fe 24.72%、SiO2 11.64%和MgO 1.39%時(shí)，銅閃速熔煉工藝參數(shù)的遺傳優(yōu)化值為空氣15 011 m3、分配風(fēng)1 302 m3、工藝氧17 359 m3、中央氧1 000 m3、熔劑13.6 t；與實(shí)踐平均值相比，若采用優(yōu)化工藝參數(shù)控制，熔煉能耗費(fèi)用可降低4.6%。
關(guān)鍵字: 銅閃速熔煉；神經(jīng)網(wǎng)絡(luò)；遺傳算法；控制優(yōu)化
based on genetic algorithms
（1. Faculty of Material and Chemistry Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China;
2. School of Metallurgical Science and Engineering, Central South University,Changsha 410083, China;
3. Faculty of Information Engineering,Jiangxi University of Science and Technology, Ganzhou 341000, China）
Abstract:Based on the built neural network model, the technological parameters of copper flash smelting process were optimized to make energy consume the lowest by using genetic algorithms when the technological objects ranged in control scope. The simulation results show that the optimizing value of air is 15 011 m3, distribution wind is 1 302 m3, technological oxygen is 17 359 m3, central oxygen is 1 000 m3 and flux is 13.6 t, when the converted ratio of the marketable price of air is 0.05, distribution wind is 0.1, technological oxygen is 0.4, central oxygen is 0.45, and the concentrate mass is 128 t, the mass fractions of components of the concentrate are Cu 20.61%, S 27.59%, Fe 24.72%, SiO2 11.64% and MgO 1.39%, respectively. Compared with the practical average data, the energy consume can be reduced by 4.6% if the smelting process is controlled by adopting the optimizing technological parameters.
Key words: copper flash smelting; neural network; genetic algorithms; control optimization
