Load recovery plays a key role in the safety and performance optimization of mechanical structures, and its accuracy is affected by a variety of factors such as sensor position and direction. Most current sensor optimization methods are based on discrete simulation data, resulting in the optimization quality dependent on the degree of data discretization. This article aims at recovering mechanical structure load using strain response, construct surrogate model to achieve a continuous mapping relationship between the data, and obtain the globally optimal strain gage placement through an optimization method. Firstly, based on the limited simulation data, the comprehensive strain information of the structural surface is obtained by using the spatial strain conversion principle and the local coordinate systems construction method. Secondly, based on the linear superposition relationship between strain response and load, the strain-load coefficient matrix is obtained, and the surrogate model between the strain gage placement and the strain-load coefficient is constructed. Then, the minimization of the variance of the load recovery is taken as the objective function for optimization, and the optimal result is obtained. Finally, both simulation and engineering case studies are carried out. The results verify that the proposed method is capable of accurately recovering the load with an error within 10–3%. Moreover, the load recovered at the optimal placement based on the strain response is less susceptible to perturbations. Its corresponding load recovery error is at least 70% lower than that of other placement methods, which indicates a high level of robustness. The proposed method also provides reference and technical support for optimizing strain gage placement on complex surfaces and major equipment.
LiSWangSChengY, et al. Strain/displacement field reconstruction and load identification of bracket arm structure of high-speed maglev train based on superposition conversion algorithm. Measurement2024; 224: 113945.
2.
TangQXinJJiangY, et al. Novel identification technique of moving loads using the random response power spectral density and deep transfer learning. Measurement2022; 195: 111120.
3.
HuaGWuYTangH, et al. Damage decoupling analysis of pump truck structure on multi-load. Noise Vib Worldw2011; 42(11): 20–24.
4.
ChengYLiZWangG, et al. Strain field reconstruction of crossbeam structure based on load-strain linear superposition method. Smart Mater Struct2021; 30(7): 075020.
5.
WangLZhangHWangY, et al. A data-driven force-thermal coupling load identification method considering multi-source uncertainties of structural characteristics and measuring noises. Comput Methods Appl Mech Eng2024; 425: 116937.
6.
HanekomMVenterG.Inverse method for static load reconstruction with automatic filtering for optimal sensor placement. Struct Multidiscipl Optim2022; 65(11): 1–15.
7.
ZhongJLiuHYuL.Sparse regularization for traffic load monitoring using bridge response measurements. Measurement2019; 131: 173–182.
8.
DhingraAKHunterTGGuptaDK.Load recovery in components based on dynamic strain measurements. J Vib Acoust2013; 135(5): 051020. DOI: 10.1115/1.4024384.
9.
MorrisBKDavisRB.Optimal design of strain sensor placement for distributed static load determination. Inverse Probl2023; 39(12): 125017.
10.
XuSDengXTiwariV, et al. An inverse approach for pressure load identification. Int J Impact Eng2010; 37(7): 865–877.
11.
CoatesCWThamburajP.Inverse method using finite strain measurements to determine flight load distribution functions. J Aircr2008; 45(2): 366–370.
12.
KobiskeADhingraA. Inverse approaches for recovery of distributed loads. Thesis University of Wisconsin, 2015.
13.
GuptaDKDhingraAK.Input load identification from optimally placed strain gages using D-optimal design and model reduction. Mech Syst Signal Process2013; 40(2): 556–570.
14.
NieminenVTuohinevaAAutioM.Wheel load reconstruction using strain gauge measurements on the bogie frame for strain prediction and fatigue assessment. Int J Fatigue2023; 170: 107533.
15.
GuptaDKDhingraAK.Load reconstruction technique using D-optimal design and markov parameters. Shock Vib2015; 2015: 1–9.
16.
AsterRCBorchersBThurberCH.Parameter estimation and inverse problems. Amsterdam: Elsevier, 2018.
17.
OkuboNTanabeSTatsunoT. Identification of forces generated by a machine under operating condition. In: Proceedings of the 3rd international modal analysis conference (IMAC), 1985, pp.920–927.
18.
MasroorSAZacharyLW.Designing an all-purpose force transducer. Exp Mech1991; 31(1): 33–35.
19.
ZhangSYangJLiY, et al. Identification of bearing load by three section strain gauge method: theoretical and experimental research. Measurement2013; 46(10): 3968–3975.
20.
AugustinePDhingraAKGuptaDK.Dynamic moving load identification using optimal sensor placement. In Proceedings of the ASME 2015 International Mechanical Engineering Congress and Exposition, Houston, TX, 13–19 November 2015. DOI: 10.1115/imece2015-50701.
21.
MitchellTJ.An algorithm for the construction of “D-optimal” experimental designs. Technometrics2000; 42(1): 48–54.
22.
GalilZKieferJ.Time- and space-saving computer methods, related to mitchell’s detmax, for finding d-optimum designs. Technometrics1980; 22(3): 301–313.
23.
AlamriASGeorgiouSStylianouS.Discrete choice experiments: an overview on constructing D-optimal and near-optimal choice sets. Heliyon2023; 9(7): e18256.
24.
LiuXYueRXChatterjeeK.Geometric characterization of D-optimal designs for random coefficient regression models. Stat Probab Lett2020; 159: 127–132.
25.
LiJYanJZhuJ, et al. K-BP neural network-based strain field inversion and load identification for CFRP. Measurement2022; 187: 110227.
26.
CooperSBDiMaioD.Static load estimation using artificial neural network: application on a wing rib. Adv Eng Softw2018; 125: 113–125.
27.
IriarteXAginagaJGainzaG, et al. Optimal strain-gauge placement for mechanical load estimation in circular cross-section shafts. Measurement2021; 174: 108938.
28.
LoutasTHBourikasA.Strain sensors optimal placement for vibration-based structural health monitoring. The effect of damage on the initially optimal configuration. J Sound Vib2017; 410: 217–230.
29.
LurieAI. Theory of elasticity. New York: Springer Science & Business Media, 2010.
30.
GreenAEZernaW.Theoretical elasticity. Courier Corporation, Oxford University Press: Oxford, UK, 1992.
31.
SongXLvLSunW, et al. A radial basis function-based multi-fidelity surrogate model: exploring correlation between high-fidelity and low-fidelity models. Struct Multidiscipl Optim2019; 60(3): 965–981.
32.
SongXLvLLiJ, et al. An advanced and robust ensemble surrogate model: extended adaptive hybrid functions. J Mech Des2018; 140(4): 1–9.
33.
BhattacharjeeKSSinghHKRayT, et al. Multiple surrogate-Assisted many-objective optimization for computationally expensive engineering design. J Mech Des2018; 140(5): 051403. DOI: 10.1115/1.4039450.
34.
LiuYWangSZhouQ, et al. Modified multifidelity surrogate model based on radial basis function with adaptive scale factor. Chin J Mech Eng2022; 35(1): 77. DOI: 10.1186/s10033-022-00742-z.
35.
KimYCarbonettoPStephensM, et al. A fast algorithm for maximum likelihood estimation of mixture proportions using sequential quadratic programming. J Comput Graph Stat2020; 29(2): 261–273.
36.
GillPEWongE.Sequential quadratic programming methods. New York, NY: Springer New York, 2011, pp. 147–224 DOI: 10.1002/9780470400531.eorms0984.
37.
MehmoodAZameerALingSH, et al. Integrated computational intelligent paradigm for nonlinear electric circuit models using neural networks, genetic algorithms and sequential quadratic programming. Neural Comput Appl2020; 32(14): 10337–10357.
38.
SunGSongXBaekS, et al. Robust optimization of foam-filled thin-walled structure based on sequential Kriging metamodel. Struct Multidiscipl Optim2014; 49(6): 897–913.
39.
SongXSunGLiG, et al. Crashworthiness optimization of foam-filled tapered thin-walled structure using multiple surrogate models. Struct Multidiscipl Optim2013; 47(2): 221–231.
40.
WangSLvLDuL, et al. An improved LHS approach for constrained design space based on successive local enumeration algorithm. In: 2019 IEEE 9th annual international conference on CYBER technology in automation, control, and intelligent systems (CYBER), 2019, pp.896–899. DOI: 10.1109/CYBER46603.2019.9066677.
41.
HammersleyJ.Monte Carlo methods. New York: Springer Science Business Media, 2013.
42.
ZhuZDuX.Reliability analysis with Monte Carlo simulation and dependent Kriging predictions. J Mech Des2016; 138(12): 1–11.
43.
KongXCaiBLiuY, et al. Optimal sensor placement methodology of hydraulic control system for fault diagnosis. Mech Syst Signal Process2022; 174: 109069.
44.
KaminskyALWangYPantK.An efficient batch K-fold cross-validation voronoi adaptive sampling technique for global surrogate modeling. J Mech Des2021; 143(1): 15.
45.
LeeSJChungISBaeSY.Structural design and analysis of CFRP boom for concrete pump truck. Mod Phys Lett B2019; 33(14n15): 1–5.
