唐求 吳娟 邱偉 沈潔 滕召勝
摘? ?要:針對雙線性變換法在設(shè)計聲級計頻率計權(quán)數(shù)字濾波器時存在固有頻率失真問題,提出一種基于改進(jìn)帝國競爭算法的數(shù)字IIR濾波器設(shè)計方法. 為避免帝國競爭算法出現(xiàn)早熟收斂而陷入局部最優(yōu)的問題,在帝國競爭算法同化階段引入混沌函數(shù)來增大搜索范圍,與此同時,在帝國競爭階段引入克隆進(jìn)化算子,引導(dǎo)算法向IIR濾波器參數(shù)最優(yōu)解方向搜索,得到改進(jìn)帝國競爭算法. 在研究聲級計A、C計權(quán)的IIR濾波器誤差來源的基礎(chǔ)上,利用改進(jìn)帝國競爭算法對聲級計頻率計權(quán)數(shù)字IIR濾波器系數(shù)進(jìn)行尋優(yōu)求解,構(gòu)建基于改進(jìn)帝國競爭算法的頻率計權(quán)數(shù)字IIR濾波器優(yōu)化模型. 仿真與實驗結(jié)果表明,本文提出的數(shù)字濾波器設(shè)計方法精度較高,且濾波器的誤差能控制在10-3 dB數(shù)量級范圍內(nèi). 在噪聲環(huán)境下不同聲信號級進(jìn)行的頻率計權(quán)測試結(jié)果表明,改進(jìn)帝國競爭算法測試的聲信號級的計權(quán)誤差能維持在10-2 dB數(shù)量級范圍內(nèi),完全滿足國家標(biāo)準(zhǔn)GB/T 3241—2010對1級聲級計的設(shè)計要求.
關(guān)鍵詞:聲級計;頻率計權(quán);數(shù)字IIR濾波器設(shè)計;帝國競爭算法;混沌函數(shù);克隆進(jìn)化
中圖分類號:TB52? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 文獻(xiàn)標(biāo)志碼:A
Abstract:Aiming at the frequency distortion problem in the design of frequency weighting digital filter based on bilinear transformation,an evolutionary method based on Modified Imperialist Competitive Algorithm(MICA)has been proposed to design digital IIR filter. In order to help the algorithm to escape from local minima,this paper introduced a chaotic function to make the search range wider in the assimilation operation of Imperialist Competitive Algorithm (ICA). Meanwhile,a clone evolution operator was introduced in the competition operation,guiding the search for global optimization efficiently. Then,the optimization model of modified ICA of the filter in sound-level meter was designed based on the research of the source of error in IIR filter. The coefficients of frequency weighted were searched based on the MICA. The results of both simulation and application show the performance of the design method to find better solution,indicating that the proposed method can significantly improve the precision and the error can be controlled within the order of 10-3 dB. Finally,the test of frequency weighting under different acoustic signal level with noise verifies that the error of the MICA test can be maintained in the order of 10-2 dB,which fully meets the design requirements of sound level meter (Class 1) in the national standard of GB/T 3241-2010.
Key words:sound level meter;frequency weighting;digital IIR filter design;Imperialist Competitive Algorithm(ICA);chaotic function;clone evolution
人耳對響度相同、頻率成分不同的聲音產(chǎn)生不同的聽覺感受,為了模擬人耳的聽覺特性,需在聲級計中設(shè)計一種頻率計權(quán)網(wǎng)絡(luò)修正聲音信號,使其對不同頻率信號具有與人耳相同的靈敏度[1]. 因此,頻率計權(quán)是聲級計實現(xiàn)噪聲測量的一項重要計量指標(biāo)[2]. IEC 61672規(guī)定1級聲級計必須實現(xiàn)A、C頻率計權(quán)功能[3].
近年來,全數(shù)字式聲級計得到廣泛應(yīng)用[4],但針對聲級計頻率計權(quán)數(shù)字濾波器的設(shè)計研究較少. 頻率計權(quán)數(shù)字濾波器的實現(xiàn)可以選擇無限沖激響應(yīng)(IIR)數(shù)字濾波器和有限沖激響應(yīng)(FIR)數(shù)字濾波器[5]. 對于相同的濾波精度,與FIR濾波器相比,IIR濾波器所用的階數(shù)少,存儲單元也較少[6].
由于聲級計的頻率計權(quán)算法采用嵌入式系統(tǒng)實現(xiàn),要求計算量小,占用存儲空間少,故本文選用數(shù)字IIR濾波器設(shè)計頻率計權(quán). 其中,常用雙線性變換(Bilinear Transformation,BT)設(shè)計數(shù)字IIR濾波器[7]. 但雙線性變換是一種近似變換,存在固有的頻率失真[8],導(dǎo)致誤差較大. 為此,文獻(xiàn)[9]采用粒子群優(yōu)化算法(Particle Swarm Optimization,PSO)對A計權(quán)的數(shù)字IIR濾波器系數(shù)進(jìn)行搜索優(yōu)化,取得了明顯成效. 但PSO算法在優(yōu)化過程中容易出現(xiàn)早熟收斂而陷入局部極值點,從而得不到全局最優(yōu)解[10-11],尤其在加噪環(huán)境下,誤差更為明顯.
帝國競爭算法(Imperialist Competitive Algorithm,ICA)在濾波器的優(yōu)化設(shè)計中,全局搜索能力和信息不依賴能力均高于其他智能優(yōu)化算法[12]. 但該算法在系數(shù)搜索過程中也同樣存在早熟收斂等不足,導(dǎo)致優(yōu)化結(jié)果存在誤差[13]. 據(jù)此,本文提出一種改進(jìn)帝國競爭算法(Modified Imperialist Competitive Algorithm,MICA)的聲級計頻率計權(quán)數(shù)字濾波器設(shè)計方案. MICA在標(biāo)準(zhǔn)ICA算法的同化過程中添加混沌函數(shù)來增強(qiáng)算法的搜索可能性,引入克隆進(jìn)化算子來有效引導(dǎo)算法向最優(yōu)解方向搜索,最終得到濾波器的最優(yōu)系數(shù). 相比標(biāo)準(zhǔn)ICA,MICA具有搜索范圍廣,尋優(yōu)精度高和優(yōu)化性能好等特點.
本文針對雙線性變換法實現(xiàn)聲級計頻率計權(quán)存在的誤差,通過在ICA算法中添加混沌函數(shù)和引入克隆算子,設(shè)計MICA算法,并將MICA應(yīng)用到頻率計權(quán)數(shù)字IIR濾波器設(shè)計中. 仿真與實驗數(shù)據(jù)表明,在加噪環(huán)境下,不同聲信號級進(jìn)行的頻率計權(quán)誤差均能維持在10-2 dB數(shù)量級范圍內(nèi),符合1級聲級計的設(shè)計要求,證明了該方法的有效性.
4? ?結(jié)? ?論
本文針對雙線性變換法設(shè)計聲級計頻率計權(quán)數(shù)字濾波器時出現(xiàn)誤差較大的問題,提出了一種基于改進(jìn)帝國競爭算法的聲級計頻率計權(quán)數(shù)字IIR濾波器設(shè)計方法. 為避免標(biāo)準(zhǔn)ICA早熟收斂而陷入局部最優(yōu),在同化階段加入混沌函數(shù)以及帝國競爭階段引入克隆進(jìn)化算子,進(jìn)一步提高算法的收斂精度. 對A計權(quán)的測試結(jié)果表明,本文提出的改進(jìn)ICA算法,有效改善了雙線性變換法的誤差,優(yōu)化效果明顯. 且在加噪環(huán)境下,不同聲信號級的計權(quán)誤差均能維持在10-2 dB數(shù)量級范圍內(nèi),符合1級聲級計設(shè)計要求. 本文提出的方法不僅適用于聲級計的頻率計權(quán)優(yōu)化設(shè)計,也適用于其他采用雙線性變換設(shè)計數(shù)字濾波器引起的頻率特性失真問題,具有較高的實際應(yīng)用價值.
參考文獻(xiàn)
[1]? ?ANSI Sl.42-2001? Design response of weighting networks for acoustical measurements [S]. New York:Acoustical Society of America,2001.
[2]? ? 鐘波,孫慶生,王雪晶,等. 聲級計頻率計權(quán)特性自動檢定系統(tǒng)研究與實現(xiàn)[J]. 電聲技術(shù),2010,34(5):37—40.
ZHONG B,SUN Q S,WANG X J,et al. Design and realization of automatic calibration system on frequency weighting of sound level meter[J]. Audio Engineering,2010,34(5):37—40. (In Chinese)
[3]? ? IEC 61672-1? Sound level meters-Part 1:Specifications [S]. Geneva:International Electrotechnical Commission,2003.
[4]? ? 楊昌棋,秦樹人,張躍俊. 虛擬式噪聲分析儀的數(shù)字計權(quán)與開發(fā)[J]. 重慶大學(xué)學(xué)報(自然科學(xué)版),2001,24 (5):59—61,66.
YANG C Q,QIN S R,ZHANG Y J. Digit weight and development of a virtual noise analyzer[J]. Journal of Chongqing University (Natural Science Edition),2001,24 (5):59—61,66. (In Chinese)
[5]? ? 姚佳旭,朱磊,潘楊,等. 基于STM32的級聯(lián)型IIR數(shù)字濾波器設(shè)計[J]. 電子測量技術(shù),2018,41(17):95—99.
YAO J X,ZHU L,PAN Y,et al. Design of cascaded IIR digital filters based on STM32[J]. Electronic Measurement Technology,2018,41(17):95—99. (In Chinese)
[6]? ? 劉強(qiáng),陳仁義,劉琳,等. 改進(jìn)的前饋FIR振動控制器[J].? 振動與沖擊,2009,28(2):107—110.
LIU Q,CHEN R Y,LIU L,et al. Improved feed forward FIR vibration controller [J]. Vibration and Shock,2009,28(2):107—110. (In Chinese)
[7]? ? 金暉,何潔. 頻率計權(quán)的全數(shù)字實現(xiàn)[J]. 儀器儀表學(xué)報,2006,27(S2):1495—1496.
JIN H,HE J. Digital design method of the frequency weighting [J]. Chinese Journal of Scientific Instrument,2006,27(S2):1495—1496.
[8]? ? KRISHNA B T. Design of fractional order differentiators using novel s to z transform[C]// Proceedings of the 2012 International Conference on Radar,Communication and Computing. Tiruvannamalai:IEEE,2012:268—271.
[9]? ? 唐求,賈楊威,滕召勝,等. 基于粒子群優(yōu)化的聲級計A計權(quán)設(shè)計[J]. 儀器儀表學(xué)報,2015,36(4):856—862.
TANG Q,JIA Y W,TENG Z S,et al. Design of A-weighting in sound-level meters based on PSO algorithm[J]. Chinese Journal of Science Instrument,2015,36(4):856—862. (In Chinese)
[10]? LUITEL B,VENAYAGAMOORTHY G K. Differential evolution particle swarm optimization for digital filter design [C]// 2008 IEEE Congress on Evolutionary Computation. Hong Kong:IEEE,2008:3954—3961.
[11]? 胡瑾秋,郭放,張來斌. 結(jié)合改進(jìn)PSO算法和LSSVM的化工異常工況超早期監(jiān)測預(yù)警研究[J]. 電子測量與儀器學(xué)報,2018,32(2):36—41.
HU J Q,GUO F,ZHANG L B. Study on ultra-early prediction of chemical abnormal situation based on improved PSO algorithm and LSSVM[J]. Journal of Electronic Measurement and Instrumentation,2018,32(2):36—41. (In Chinese)
[12]? ZHANG Y,WEI H,LIAO R,et al. A new support vector machine model based on improved imperialist competitive algorithm for fault diagnosis of oil-immersed transformers[J]. Journal of Electrical Engineering & Technology,2017,12(2):830—839.
[13]? 邵永亮,常軍. 運(yùn)用改進(jìn)帝國競爭算法識別結(jié)構(gòu)模態(tài)參數(shù)[J]. 噪聲與振動控制,2017,37(2):152—157.
SHAO Y L,CHANG J. Structural modal parameter identification based on improved imperialist competitive algorithm [J]. Noise and Vibration Control,2017,37(2):152—157. (In Chinese)
[14]? ANDRZEJ M,ANDRZEJ P. Digital-filter-based compensation of case effect in sound-level meters[J]. International Journal of electronics and telecommunications,2010,56(3):263—266.
[15]? ATASHPAZ G E,LUCAS C. Imperialist competitive algorithm:an algorithm for optimization inspired by imperialistic competition[C]// 2007 IEEE Congress on Evolutionary Computation. Singapore:IEEE,2007:4661—4667.
[16]? 何存富,王志,劉秀成,等. 基于GA-PSO混合算法的鋼桿磁特性參數(shù)識別方法[J]. 儀器儀表學(xué)報,2017,38(4):838—843.
HE C F,WANG Z,LIU X C,et al. Magnetic property parameter identification of steel pole based on GA-PSO hybrid algorithm [J]. Chinese Journal of Science Instrument,2017,38(4):838—843. (In Chinese)