Steam turbine blade sequencing method based on TILS algorithm
-
摘要:
通过优化汽轮机叶片的安装顺序,来减少安装后的残余不平衡量。对此提出一种阈值式迭代局部搜索(threshold iterative local search,TILS)算法,该算法在迭代局部搜索(iterative local search,ILS)算法基础上,采用阈值限定扰动与随机扰动相结合的方法来跳出局部最优解,减少了平均到达局部最优解所需的迭代步数。实验证明,该方法可以在短时间内找到一个近似最优叶片排序组合,相对于ILS算法,搜索效率提高了20%以上。计算得到的合成质径积的近似最优解,相对于现有分组排序法、遗传算法、云自适应遗传算法(CAGA)等方法,分别减小到其最优解的0.33%~31%,且计算时间也大幅度减小。
-
关键词:
- 汽轮机 /
- 叶片排序 /
- 组合优化 /
- 迭代局部搜索(ILS) /
- 阈值式迭代局部搜索(TILS) /
- 局部搜索
Abstract:To reduce the residual unevenness after installation by optimizing the installation sequence of turbine blades, a threshold iterative local search algorithm (TILS) was proposed. Based on the iterative local search algorithm (ILS), this algorithm adopted the combination of threshold limited disturbance and random disturbance to escape from local optimum, which reduced the number of iteration steps required to reach the local optimal solution on average. Experiments demonstrated that this method can find the approximately optimal combination of blade sequences in a relatively short time, which improved the search efficiency by more than 20% compared with the ILS algorithm. Compared with existing group sorting methods, genetic algorithms, and cloud adaptive genetic algorithm (CAGA) algorithms, the approximately optimal solution of synthetic mass product calculated by this algorithm was reduced to 0.33%—31%, and the computation time was significantly shorter.
-
表 1 叶片质径积数据
Table 1. Blade mass-diameter product data
序号 质径积/(g·mm) 序号 质径积/(g·mm) 1 7241085.5 45 7090879.5 2 7238714.5 46 7087313.5 3 7236525 47 7080382 4 7233798.5 48 7080174 5 7231219.5 49 7079576.5 6 7230890.5 50 7074955.5 7 7230769.5 51 7071000 8 7229661.5 52 7065479 9 7227740.5 53 7064908 10 7217787.5 54 7061610.5 11 7212837.5 55 7061134 12 7167020.5 56 7060321 13 7162070.5 57 7058434 14 7151372.5 58 7058373.5 15 7146059.5 59 7058347 16 7145757 60 7057681.5 17 7141828 61 7056486.5 18 7141741 62 7056097 19 7140304 63 7055257.5 20 7139578 64 7053484 21 7136798.5 65 7047789 22 7133675 66 7045331 23 7132022.5 67 7041765 24 7130404 68 7035915 25 7129659 69 7032946.5 26 7128725 70 7030099 27 7128040.5 71 7028628 28 7126146 72 7028299 29 7125514.5 73 7028030.5 30 7120995.5 74 7023201.5 31 7117509 75 7021522.5 32 7116609 76 7017983 33 7113372 77 7017472.5 34 7112022 78 7016814.5 35 7109469.5 79 7012556.5 36 7108872 80 7007848.5 37 7108811.5 81 7006256.5 38 7106353.5 82 6992824.5 39 7099993 83 6991535 40 7097119 84 6988116.5 41 7095587.5 85 6985987.5 42 7094211 86 6982119 43 7093137 87 6971232 44 7091900.5 88 6961729 表 2 ILS与TILS结果对比
Table 2. Comparing the results of ILS and TILS
求解时间/s 算法 平均局部最优个数 平均最小局部最优/(g·mm) 平均迭代步数 84±0.5 ILS 5183.1 0.47013 19.29 TILS 6335.65 0.37078 6.9729 1470±1 ILS 51812.75 0.118915 19.3 TILS 67401.55 0.114015 6.98301 表 3 各方法求解结果
Table 3. Solution results of each method
实验编号 使用方法 叶片数量 合成质径积的最小值/(g·mm) 相对比例/% 求解时间 CPU 1 分组排序法[2] 33 0.2500 7.20 ≈6 min i7-7500U TILS 0.0180 20 s i7-12700k 2 分组排序法[2] 43 0.0300 31.00 ≈10 min i7-7500U TILS 0.0093 2 min i7-12700k 3 分组排序法[2] 63 2.9400 3.54 ≈18 min i7-7500U TILS 0.1041 2 min i7-12700k 4 遗传算法[17] 43 0.4709 1.93 TILS 0.0091 30 s i7-12700k 5 CAGA[18] 32 99.6300 0.33 >20 s TILS 0.3242 10 s i7-12700k -
[1] WILF H S. Algorithms and complexity[M]. 2nd ed. Natick,US: AK Peters/CRC Press,2002. [2] 朱本刚,王曙光,张恩慧,等. 一种透平动叶片排序方法[J]. 燃气轮机技术,2018,31(3): 39-42. ZHU Bengang,WANG Shuguang,ZHANG Enhui,et al. A method of turbine buckets arrangement[J]. Gas Turbine Technology,2018,31(3): 39-42. (in ChineseZHU Bengang, WANG Shuguang, ZHANG Enhui, et al. A method of turbine buckets arrangement[J]. Gas Turbine Technology, 2018, 31(3): 39-42. (in Chinese) [3] 吴根红. 基于遗传算法的汽轮机叶片重力距平衡排序方法[J]. 内燃机与配件,2020(11): 95-96. WU Genhong. Ranking method of steam turbine blade gravity distance balance based on genetic algorithm[J]. Internal Combustion Engine & Parts,2020(11): 95-96. (in ChineseWU Genhong. Ranking method of steam turbine blade gravity distance balance based on genetic algorithm[J]. Internal Combustion Engine & Parts, 2020(11): 95-96. (in Chinese) [4] 李丹丹,陈勇,于达仁. 基于改进的蚁群算法的汽轮机叶片安装优化排序研究[C]//中国智能自动化学术会议论文集: 第一分册. 北京,2011: 194-198. [5] 李岩,袁惠群. 基于改进DPSO算法的航空发动机失谐叶片排序[J]. 振动、测试与诊断,2013,33(增刊1): 149-153,227. LI Yan,YUAN Huiqun. Optimization on mistuned blades sorting based on improved discrete particle swarm optimization algorithm in aero-engine[J]. Journal of Vibration,Measurement & Diagnosis,2013,33(Suppl.1): 149-153,227. (in ChineseLI Yan, YUAN Huiqun. Optimization on mistuned blades sorting based on improved discrete particle swarm optimization algorithm in aero-engine[J]. Journal of Vibration, Measurement & Diagnosis, 2013, 33(Suppl.1): 149-153, 227. (in Chinese) [6] 王港华. 基于遗传算法的小规模TSP问题研究分析[J]. 物流工程与管理,2022,44(3): 111-114,29. WANG Ganghua. Research analysis of small-scale TSP problem based on genetic algorithm[J]. Logistics Engineering and Management,2022,44(3): 111-114,29. (in ChineseWANG Ganghua. Research analysis of small-scale TSP problem based on genetic algorithm[J]. Logistics Engineering and Management, 2022, 44(3): 111-114, 29. (in Chinese) [7] 迟宗正. 大规模优化问题的高效元启发式算法研究[D]. 辽宁 大连: 大连理工大学,2021. CHI Zongzheng. Researsch on efficient metaheuristic algorithm for large-scale optimization problems[D]. Dalian Liaoning: Dalian University of Technology,2021. (in ChineseCHI Zongzheng. Researsch on efficient metaheuristic algorithm for large-scale optimization problems[D]. Dalian Liaoning: Dalian University of Technology, 2021. (in Chinese) [8] 张静乐. 大规模组合优化问题蚁群算法应用研究[D]. 郑州: 郑州大学,2006. ZHANG Jingyue. Research on application of ant colony algorithm for large-scale combinatorial optimization problem[D]. Zhengzhou: Zhengzhou University,2006. (in ChineseZHANG Jingyue. Research on application of ant colony algorithm for large-scale combinatorial optimization problem[D]. Zhengzhou: Zhengzhou University, 2006. (in Chinese) [9] 李睿智. 局部搜索算法及其在组合优化问题中的应用[M]. 北京: 科学出版社,2019. LI Ruizhi. Local search algorithm and its application in combinatorial optimization problems[M]. Beijing: Science Press,2019. (in ChineseLI Ruizhi. Local search algorithm and its application in combinatorial optimization problems[M]. Beijing: Science Press, 2019. (in Chinese) [10] CROES G A. A method for solving traveling-salesman problems[J]. Operations Research,1958,6(6): 791-812. doi: 10.1287/opre.6.6.791 [11] LIN Shen. Computer solutions of the traveling salesman problem[J]. The Bell System Technical Journal,1965,44(10): 2245-2269. doi: 10.1002/j.1538-7305.1965.tb04146.x [12] REITER S,SHERMAN G. Discrete optimizing[J]. Journal of the Society for Industrial and Applied Mathematics,1965,13(3): 864-889. doi: 10.1137/0113056 [13] RAGHUNATHAN A,JHA N K. An iterative improvement algorithm for low power data path synthesis[C]//Proceedings of the 1995 IEEE/ACM International Conference on Computer-aided Design. New York: ACM,1995: 597-602. [14] BRITO J,OCHI L,MONTENEGRO F,et al. An iterative local search approach applied to the optimal stratification problem[J]. International Transactions in Operational Research,2010,17(6): 753-764. doi: 10.1111/j.1475-3995.2010.00773.x [15] WANG Yiyuan,LI Ruizhi,ZHOU Yupeng,et al. A path cost-based GRASP for minimum independent dominating set problem[J]. Neural Computing and Applications,2017,28(1): 143-151. [16] PULLAN W,HOOS H H. Dynamic local search for the maximum clique problem[J]. Journal of Artificial Intelligence Research,2006,25: 159-185. doi: 10.1613/jair.1815 [17] 贾金鑫,李全通,高星伟,等. 叶片质量矩优化排序中遗传算法的应用[J]. 航空动力学报,2011,26(1): 204-209. JIA Jinxin,LI Quantong,GAO Xingwei,et al. Application of genetic algorithm in optimizing arrangement of engine blades based on initial unbalance[J]. Journal of Aerospace Power,2011,26(1): 204-209. (in ChineseJIA Jinxin, LI Quantong, GAO Xingwei, et al. Application of genetic algorithm in optimizing arrangement of engine blades based on initial unbalance[J]. Journal of Aerospace Power, 2011, 26(1): 204-209. (in Chinese) [18] SUN Chuanzhi,XIAO Pinghuan,WANG Xiaoming,et al. Blade sorting method for unbalance minimization of an aeroengine concentric rotor[J]. Symmetry,2021,13(5): 832. doi: 10.3390/sym13050832