Multi-objective Optimization Procedure Research Articles

Aerodynamic optimization of high-rise building with façade ribs through CFD-based multi-objective optimization algorithm

Façade appurtenances is an effective aerodynamic optimization measure for alleviating the wind loads on high-rise buildings in modern cities, while the optimized layout of façade ribs necessary for practical engineering remains unclear. The conventional method through numerical simulations and wind tunnel tests can only obtain limited optimal schemes of ribs, the optimization process is time-consuming and laborious, an optimization algorithm can be used to effectively evaluate the optimal arrangement of façade ribs. In present study, a multi-objective optimization procedure coupling large eddy simulation (LES), back propagation neural network (BPNN), and non-dominated sorting genetic algorithm (NSGA-II) is applied to evaluate the optimal arranged parameters of façade vertical ribs. The optimization procedure can quickly identify the optimal location parameter b and extensional depth d of façade ribs that producing minimum wind loads acting on building and ribs. The findings indicate that there are complex non-linear relationships between the ribs arrangement parameters and the objective wind loads. The mean drag C‾d and fluctuating lift C'l of models of windward ribs and upstream sidewall ribs have opposite trend. The relationships between wind forces and arranged parameters of the downstream sidewall and leeward ribs are relatively complicated due to the vortex shedding. Genetic algorithm NSGA-II can effectively evaluate the optimal trade-off solution among multi-objective wind loads. The ranges of optimal layout parameters b/D and d/D are 0.14–0.17, 0.073–0.077, the ratio of b/d is about 2, and the optimal arrangement of façade ribs well balances the total wind forces on the model and wind forces on the ribs. The 3D numerical simulations demonstrate that the optimal layout of the ribs can considerably improve the flow structure and wind load, the maximum reduction ratio of the C‾d and C'l are achieved 25 % and 56 %, respectively. The obtained optimized layout parameters can provide reference for engineers when actually using the facade ribs.

A stepwise scheme for the optimization implementation of lazy wave dynamic cables

The dynamic cables are usually designed to present a complex layout to improve service life and safety in commercial scale floating applications. It is urgent to establish consistent design philosophy and analysis methodologies to provide cost-efficient design requirement. A stepwise scheme for the optimization implementation of lazy wave dynamic cables is proposed and performed with a mass-damping equivalence model. A complicated multi-objective optimal procedure is implemented progressively under hydrostatic state, regular wave state and stochastic sea state. It is validated through a lazy wave dynamic cable (LWDC) from a cylindrical FPSO system under extreme conditions. Stepwise upgraded calibrations based on the coupling of multi-objective optimization modular NSGA-II and lumped mass FEM are applied to assess every sub-objective in defined steps. It is shown that the result is far beyond expectation than initial design after the application of the progressive solution system.

Soil-dependent stochastic optimization of base isolation system with negative stiffness inerter damper subjected to non-stationary seismic excitation

The negative stiffness inerter damper (NSID) has demonstrated excellent performance in mitigating vibrations in frame structures. However, analysis and optimization of BIS-NSID considering both soil conditions and non-stationary seismic excitation are rarely reported. This study proposes a multi-objective performance optimization procedure (MPOP) to investigate the optimal parameters of BIS-NSID under non-stationary seismic excitation considering firm, medium and soft soil conditions. The non-stationary seismic excitation is expressed as Clough-Penzien model with time modulation function. The analytical solution for the root mean square responses of BIS-NSID are derived considering non-stationary seismic excitation. The pareto optimal fronts is utilized to minimize the BIF displacement and superstructure acceleration. The optimal design is further validated under artificial seismic records. The results illustrate the validity of the MPOP and highlight the advantage of the BIS-NSID for all soil conditions. The demand for inertance and damping values are greater in soft soil conditions, compared to firm and medium soil condition.

Geometry Optimization of Industry Buildings for Energy-Driven Design Development in Bangladesh

Even though the manufacturing industry consumes roughly 54% of total available energy globally, little consideration has been devoted to optimizing energy in the early stages of industry design, particularly in densely populated cities. With the increased demand for green buildings, energy performance has a greater influence on design results. As a result, this paper provides an envelope optimization technique that can assist architects and computational designers in analyzing the environmental performance of various alternatives and developing optimal design solutions, especially for Bangladesh and the building of these regions. First, an existing industrial site in Dhaka, Bangladesh's capital was chosen as a case study, and a hypothetical industry building, as well as its surroundings, were parametrically developed. Then, for the optimization method, the design factors linked to the building envelopes were chosen. Finally, a Multi-objective Optimization (MOO) procedure was utilized for defining performance measures including daylighting, energy and comfort measures, UDI, EUI, and PPD. According to the MOO results, the UDI may be enhanced by 25.286% as compared to the least favorable scenario. Consequently, the EUI may decreased by 38.718 kWh/m2 while the PPD can be increased by 41.78%. The geometric configuration of East, West, North, and South played a significant role when designing the industrial building. According to the analysis, the geometric configuration of a South WWR of 50%, a West WWR of 30%, an East WWR of 70%, and a sill height of 0.75m is the most feasible option. A statistical analysis of design factors and performance measures demonstrated that the window-to-wall ratio, particularly on south walls, has the greatest impact on industrial building design in densely populated areas. The proposed approach is expected to be used by architects and municipal planners to develop design metrics based on simulation results.

Intelligent optimization of eco-friendly H2/freshwater production and CO2 reduction layout integrating GT/rankine cycle/absorption chiller/TEG unit/PEM electrolyzer/RO section

The present study introduces an innovative approach tailored for gas turbine power plants to provide residential freshwater and energy, while also taking into account environmental concerns. The innovative design integrates sophisticated combustion technologies and effective heat recovery systems to enhance energy conversion processes and mitigate emissions, thereby facilitating more sustainable and environmentally-conscious power generation. The proposed system demonstrates capability in generating electrical power, producing desalinated water, and providing cooling capacity, thereby presenting a feasible substitute to conventional gas turbine systems. This research investigates the fundamental case of a gas turbine power plant and compares it to a modified system that includes heat recovery and hydrogen blending. The proposed system has been rigorously designed, analyzed, and optimized, with performance outcomes evaluated in comparison to the base case. Numerous effective parameters are taken into consideration in the exploration of the system's performance and the implementation of a multi-objective optimization procedure. The aims of this study encompass the optimization of overall exergy efficiency, the minimization of total cost rate, and the reduction of normalized CO2 emissions. The findings indicate a noteworthy enhancement in the suggested system when compared to the conventional base system. The overall exergy efficiency saw an increase from 31. 34–4273 %, accompanied by a decrease in the total cost rate from 1847 to 1514 $/h, and a reduction in the CO2 emission rate from 2. 22 kg/s to 1. 62 Moreover, through a comparative analysis, it is evident that augmenting the production capacity of the modified system results in a significant decrease in the power plant's payback period by 0. 8 years. The outcomes of this study offer significant insights for the creation of power generation systems that are both more effective and environmentally sustainable.

Surface quality improvement of AZ31 Mg alloy by a combination of modified Taguchi method and simple multi objective optimization procedure

Surface quality improvement of AZ31 Mg alloy by a combination of modified Taguchi method and simple multi objective optimization procedure

Design and Performance Assessment of Base Isolated Structures Supplemented with Vibration Control Systems

This paper investigates the implementation of supplemental vibration control systems (VCS) in base isolated (BI) structures, to improve their dynamic performance. More specifically, the aim of the VCS is to reduce the base displacement demand of BI structures, and at the same time mitigate the superstructure seismic responses. The purpose of the examined VCS is dual, and for this reason a multi-objective optimization methodology is formulated for the design of the VCS. The examined vibration absorbers include modifications of the KDamper concept. The KDamper is an extension of the traditional Tuned Mass Damper (TMD), and introduces a negative stiffness (NS) element to the additional oscillating mass of the TMD. The generated NS force is exactly in phase with the inertia force of the added mass, thus, artificially amplifying it. This way, lighter configurations are possible with an enhanced damping behavior. These VCS are designed based on engineering criteria and manufacturing constraints, while the excitation input used in the multi-objective optimization procedure is selected from a dataset of artificial accelerograms, designed to be spectrum-compatible with the EC8 design acceleration response spectrum. The effectiveness of the examined VCS is also assess with real near-fault earthquake records, and a comparison is performed with TMD-based VCS having 50 times larger additional masses. The numerical results demonstrate the superiority of the KDamper-based VCS in improving the dynamic behavior of BI structures over other mass-related systems (TMD).

Optimal Constrained Groove Pressing Process Parameters Applying Modified Taguchi Technique and Multi-Objective Optimization

Most engineering problems are complicated, and developing mathematical models for such problems requires understanding the phenomena through experiments. It is well known that as processing parameters with assigned levels increase, so does the number of experiments. By minimizing the number of experiments, Taguchi’s method of experimental design will help to furnish the idea of full factorial experimental design. Taguchi’s method is more appropriate for single-objective optimization problems and needs modifications while dealing with multi-objective optimization problems. Aluminum alloys are in great demand in today’s automotive and aerospace sectors due to their low density, good corrosion resistance, and excellent machinability. The material is subjected to a constrained groove pressing (CGP) process to obtain microstructural grain refinement with enhanced mechanical behavior. This paper considers AA6061 material having major alloys such as silicon and magnesium. For this work, 3 CGP process parameters (viz., displacement rate, plate thickness and number of passes) are assigned 3 levels to each parameter, acquired the test data, viz., grain size (gs), micro hardness (hs), and tensile strength (ult) based on L9 orthogonal array of Taguchi. Using a modified version of Taguchi’s methodology, it is possible to estimate the range of grain size (gs), micro hardness (hs), and tensile strength (σult) for effective combinations of the CGP processing parameters and validate the results with existing test data. A more dependable and simpler multi-objective optimization procedure is used to choose the optimal CGP processing parameters.

Proposing a procedure for multi-objective optimization of cascade thermoelectric coolers to achieve maximum cooling capacity and COP

Abstract A novel approach is proposed for optimal designing two-stage thermoelectric coolers using the biogeography-based optimization algorithm. For the comprehensiveness of the research, both types of two-stage coolers, including electrically serial and separated stages, have been investigated. Using the proposed procedure, first by selecting the most important performance parameters of the cooler, i.e. COP and the cooling capacity as objective functions, the efficiency and the validity of the approach are proved by comparing the results with results of the previous researchers. In all the examined cases, the proposed approach has led to the enhancing of the cooler performance. The results of single-objective optimization showed that using the proposed method, the cooling capacity and COP can be improved up to 17 % and 8.5 % compared to the previous designs, respectively. In the following, multi-objective optimization of thermoelectric coolers is conducted, taking into account both the objectives of the COP and the cooling capacity. Finally, Pareto optimal solutions and the exergetic efficiency of the optimized cooler were calculated for different values of the weight factor of the multi-objective function. The results of multi-objective optimization showed that by increasing the COP weight factor in the objective function, the exergetic efficiency of the refrigeration system enhances.

An explainable machine learning-based probabilistic framework for the design of scaffolds in bone tissue engineering.

Recently, 3D-printed biodegradable scaffolds have shown great potential for bone repair in critical-size fractures. The differentiation of the cells on a scaffold is impacted among other factors by the surface deformation of the scaffold due to mechanical loading and the wall shear stresses imposed by the interstitial fluid flow. These factors are in turn significantly affected by the material properties, the geometry of the scaffold, as well as the loading and flow conditions. In this work, a numerical framework is proposed to study the influence of these factors on the expected osteochondral cell differentiation. The considered scaffold is rectangular with a 0/90 lay-down pattern and a four-layered strut made of polylactic acid with a 5% steel particle content. The distribution of the different types of cells on the scaffold surface is estimated through a scalar stimulus, calculated by using a mechanobioregulatory model. To reduce the simulation time for the computation of the stimulus, a probabilistic machine learning (ML)-based reduced-order model (ROM) is proposed. Then, a sensitivity analysis is performed using the Shapley additive explanations to examine the contribution of the various parameters to the framework stimulus predictions. In a final step, a multiobjective optimization procedure is implemented using genetic algorithms and the ROM, aiming to identify the material parameters and loading conditions that maximize the percentage of surface area populated by bone cells while minimizing the area corresponding to the other types of cells and the resorption condition. The results of the performed analysis highlight the potential of using ROMs for the scaffold design, by dramatically reducing the simulation time while enabling the efficient implementation of sensitivity analysis and optimization procedures.

Multiobjective Robust PI Synthesis in Plants with Uncertain Poles

The use of multiobjective optimization procedures for control tuning has been constantly explored, presenting satisfactory results, and generating significant contributions to control engineering. In this way, the field of robust control can benefit significantly from the integration of multiobjective optimization design procedures for robust Proportional-Integral-Derivative (RPID) controller synthesis, enabling innovative solutions to complex control challenges. Due to this fact, this paper explores an approach to utilize multiobjective optimization algorithms in a simplified procedure, considering uncertain systems in the form of interval plants with uncertain poles, the Kharitonov theorem is used to obtain a region of stability as search space for the multiobjective algorithm, being the optimization process performed over a nominal system model. Two numerical cases and a flexible actuator model were used to demonstrate the procedure for RPI synthesis, resulting in robustly stable controllers with optimized performance over error and control signal effort measures.

Distributed event-triggered fuzzy control for nonlinear interconnected systems

This paper deals with the asynchronous distributed event-triggered control of continuous-time nonlinear interconnected systems. The nonlinear dynamics of the subsystems are represented by Takagi–Sugeno fuzzy models via the sector nonlinearity approach. Moreover, the nonlinear interconnections among the subsystems are assumed to be known and sector-bounded. In the event-triggered control setup, where the state information is asynchronously available to the local controllers only in specific time instants, it is necessary to deal with the asynchronous premise variables because they introduce extra difficulties to derive suitable co-design conditions. To deal with this issue, we propose a new triggering strategy such that the local event-triggering mechanisms (ETMs) counteract the effects of asynchronous premise variables. With this new distributed ETM, a co-design condition is proposed, and the existence of a strictly positive minimum inter-event time is proved to exclude Zeno behaviour. Moreover, a multi-objective optimization procedure is introduced to enlarge the estimate of the domain of attraction of the closed-loop equilibrium and minimize the number of transmissions provided by the ETMs. Finally, the approach is validated through the synchronization of interconnected oscillators.

Multi-objective optimization of additive manufactured functionally graded lattice structures under impact

A multi-objective optimization procedure to effectively design gradient lattice structures under dynamic loading is presented. It aims both at maximized energy absorption characteristics and a lightweight design. The reduction of the peak crushing force and a maximized specific energy absorption are simultaneously optimized. Design variables like the relative density and density gradient of the lattice structure are investigated, considering different topologies of the lattice unit cells. To model and simulate the lattice structures efficiently, a simplified beam-based simulation finite element model is introduced and validated against experimental tests. They cover the range from dynamic equilibrium up to compaction wave propagation as impact deformation phenomena. Based on large numbers of simulations, an artificial neural network is used to predict the energy absorbing characteristics for the given design parameters and to find the optimal configuration of the lattice structures. The neural network is trained using a multi response adaptive sampling algorithm allowing parallel simulation with automatically generated finite element models during each iteration step. Subsequently, a multi-objective genetic algorithm is used to find optimal combinations of the design parameters of lattice structures under different impact velocities and cell topologies.

Radially and axially graded cellular tibial stems for total knee replacement

In spite of the high success rate of existing total knee replacement implants, the aseptic loosening of components, particularly the tibial one, is a matter of concern. To control the short- and long-term loosening mechanisms, the tibial stem material shall have diverse stiffness properties in different regions and directions, such that it encourages osseointegration and prevents bone resorption. The previous studies in the literature have failed to provide designs which could satisfy these two requirements concurrently. The purpose of this study was to design heterogeneous and anisotropic porous titanium tibial stems, based on radial and axial functionally graded cellular (FGC) structures, with precisely tailored stiffness properties. A 3D voxel-based finite element model of the bone-implant construct was developed to predict the nodal micromotion at the interface, assuming a frictional contact, and the stress-shielding effect in the periprosthetic bone, assuming a perfectly bonded connection. An exponential FGC material model was assigned to the implant along each of the radial and axial directions, with the gradient parameters considered as design variables. A design of experiment (DOE) -based multi-objective optimization procedure was then employed to maximize the osseointegration potential and minimize the bone resorption index. Equivalent FGC lattices, composed of Cartesian and cylindrical arrangements of gyroid unit cells, were produced for the optimal designs and fabricated by selective laser melting (SLM) technique. In comparison with the solid stem, the radial graded design resulted in a substantial improvement in the bone resorption behavior (19.4%), but a moderate reduction in the osseointegration potential (9.4%). The axial graded design, on the other hand, was associated with a moderate improvement in the bone resorption characteristics (12.0%) and slight improvement in the osseointegration index (2.2%). The dimensional accuracy of the SLM-fabricated specimens was reasonably high (strut thickness: 0.032 to 0.076 mm RMSE, pore size: 0.029 to 0.057 mm RMSE). The radial and axial graded specimens behaved as brittle and ductile materials, respectively, in axial compression tests and failed due to buckling and excessive plastic deformation. The cylindrical lattice arrangement demonstrated a lower elastic modulus (radial graded: 1945 vs. 3053 MPa, axial graded: 779 vs. 2324 MPa) and yield strength (radial graded: 77 vs. 114 MPa, axial graded: 17 vs. 43 MPa) than the Cartesian arrangement. It was concluded that an optimal bi-directional axial and radial gradient structure with a cylindrical lattice arrangement would provide the most favorable biomechanical performance for the tibial stem.

Enhancing productivity and efficiency in conventional laser metal deposition process for Inconel 718 - part I: the effects of the process parameters

The sustainable energy transition has spurred the development of technologies that minimize material and energy waste, such as additive manufacturing (AM). Laser metal deposition (LMD) is a promising AM technique, but its complexity and limited automation hinder its implementation in production chains. To enhance productivity, the high deposition rate LMD (HDR-LMD) technology has been developed, requiring advanced equipment and powerful laser sources. In contrast, the conventional LMD (C-LMD) process is simpler and less expensive to implement. This study aims to optimize the productivity and efficiency of C-LMD by adjusting laser power, scan speed, powder feed rate, and standoff distance on Inconel 718 single tracks. An innovative approach eliminates the need for cutting specimens to evaluate single tracks, allowing comprehensive geometric and performance characterization with limited operator involvement, making the analysis quicker and more robust. An extensive experimental campaign was conducted to examine the influence of process parameters on track geometry, productivity, and efficiency. A multi-objective optimization procedure identified parameter combinations maximizing productivity while maintaining high efficiency and desirable clad shape. The study attained deposition rates ranging from 700 to 800 g/h, with powder catchment efficiency ranging between 75 and 90%. These results were achieved using parameters including 1775 W of laser power, scan speeds ranging from 960 to 1140 mm/min, powder feed rates between 810 and 1080 g/h, and standoff distance of 9 mm. The study also clearly indicated that further potential for improving C-LMD process performance may be possible. The findings gathered in this paper are the base for the further optimization presented in the second part of the work, which is focused on multi-pass multi-layer and reaches deposition rates of 1500 g/h, promoting the implementation of C-LMD process at industrial level.

Optimal deep-space heliocentric transfers with an electric sail and an electric thruster

An electric solar wind sail is a propellantless propulsive system that generates thrust by exploiting the electrostatic interaction between solar wind ions and one or more charged tethers. Assuming a realistic near-term design in which the sail is composed of a limited number of tethers, the propulsive acceleration provided to the spacecraft is small. Moreover, according to a recent thrust model, the propulsive acceleration vector is constrained to lie in a cone centered on the outward radial direction with half-angle equal to about 20 degrees, resulting in a small circumferential component of the propulsive acceleration. In order to overcome these issues, a possible strategy consists in combining the electric sail with an electric thruster that provides a low thrust steerable around the circumferential direction. The effectiveness of such a combination is thoroughly analyzed in this work for deep-space heliocentric transfers. Transfer trajectories are obtained as outputs of a multi-objective optimization procedure, in which a suitable linear function of the flight time and the propellant consumption is minimized, considering different relative weights of the two competing requirements. Two exemplary case studies, consisting of Earth-Mars and Earth-Venus ephemeris-free circle-to-circle transfers, are presented to show the effectiveness of the proposed arrangement.

Multi-objective optimization of inconel 718 turning parameters using PCA-MARCOS and MARCOS-Taguchi

Abstract In recent years, multiple-attribute decision-making (MADM) methods have gathered significant interest from researchers across various scientific fields. In the case of mechanical part machining when faced with conflicting objectives, these methods have become essential in the selection of necessary cutting parameters. The integration of these techniques with multi-objective optimization procedures has proven effective in leading to an efficient solution to a multi-attribute decision-making problem. The present study falls within this context. Its primary objective being the determination of the optimal cutting parameters that reduces tool vibration, minimizes surface roughness along withmaximizing material removal rate during the turning of Inconel 718 using a coated carbide cutting tool grade (GC1105). These various objectives have been achieved through combining the principal component analysis (PCA) method to the measurement of alternatives and ranking according to the compromise solution (MARCOS), and MARCOS method coupled with the Taguchi SN ratio. The results obtained demonstrated that both methods generated the same combination of optimal cutting parameters that achieve the aforementioned objectives and represented by a nose radius r = 0.8 mm, a cutting speed Vc = 70 mm min−1, a feed rate f = 0.08 mm rev−1, and a depth of cut ap = 0.3 mm.

Improving the Performance of Hydrological Model Parameter Uncertainty Analysis Using a Constrained Multi-Objective Intelligent Optimization Algorithm

In the field of hydrological model parameter uncertainty analysis, sampling methods such as Differential Evolution based on Monte Carlo Markov Chain (DE-MC) and Shuffled Complex Evolution Metropolis (SCEM-UA) algorithms have been widely applied. However, there are two drawbacks which may introduce bad effects into the uncertainty analysis. The first disadvantage is that few optimization algorithms consider the physical meaning and reasonable range of the model parameters. The traditional sampling algorithms may generate non-physical parameter values and poorly simulated hydrographs when carrying out the uncertainty analysis. The second disadvantage is that the widely used sampling algorithms commonly involve only a single objective. Such sampling procedures implicitly introduce too strong an “exploitation” property into the sampling process, consequently destroying the diversity property of the sampled population, i.e., the “exploration” property is bad. Here, “exploitation” refers to using good already-existing solutions and making refinements to them, so that their fitness will improve further; meanwhile, “exploration” denotes that the algorithm searches for new solutions in new regions. With the aim of improving the performance of uncertainty analysis algorithms, in this research, a constrained multi-objective intelligent optimization algorithm is proposed that preserves the physical meaning of the model parameter using the penalty function method and maintains the population diversity using a Non-dominated Sorted Genetic Algorithm-II (NSGA-II) multi-objective optimization procedure. The representativeness of the parameter population is estimated on the basis of the mean and standard deviation of the Nash–Sutcliffe coefficient, and the diversity is evaluated on the basis of the mean Euclidean distance. The Chengcun watershed is selected as the study area, and uncertainty analysis is carried out. The numerical simulations indicate that the performance of the proposed algorithm is significantly improved, preserving the physical meaning and reasonable range of the model parameters while significantly improving the diversity and reliability of the sampled parameter population.

Leveraging composition-based energy material descriptors for machine learning models

A comparison of several classifiers is presented, with a focus on the key choice and construction of a minimal set of suitable material features. To this end, an investigation is conducted over a properly selected and high quality database reporting low temperature superconductors, featurized by composition-based descriptors. Fully general strategies to reduce the number of descriptors for material classification are proposed and discussed. The first strategy aims at testing possible invariance of the target material property (here the critical temperature) with respect to (binary) groups of composition-based features in the form xiaxjb,a,b∈R. In addition, a multi-objective optimization procedure for reducing the set of composition-based material descriptors is also suggested and tested on the chosen use case. The latter procedure is then proven to be particularly convenient to be used in combination with Bayesian type classifiers. Finally, by means of the best-performing classification models, an analysis is conducted over all the ∼40,000 inorganic compounds without Ni, Fe, Cu, O in Materials Project (and not in the SuperCon database, here used for model training) and the corresponding predictions are provided. Among those, 41 materials are classified to show Tc≥15K with a probability higher than or equal to 0.6.

A modified back analysis method for deep excavation with multi-objective optimization procedure

Real-time prediction of excavation-induced displacement of retaining pile during the deep excavation process is crucial for construction safety. This paper proposes a modified back analysis method with multi-objective optimization procedure, which enables a real-time prediction of horizontal displacement of retaining pile during construction. As opposed to the traditional stage-by-stage back analysis, time series monitoring data till the current excavation stage are utilized to form a multi-objective function. Then, the multi-objective particle swarm optimization (MOPSO) algorithm is applied for parameter identification. The optimized model parameters are immediately adopted to predict the excavation-induced pile deformation in the continuous construction stages. To achieve efficient parameter optimization and real-time prediction of system behavior, the back propagation neural network (BPNN) is established to substitute the finite element model, which is further implemented together with MOPSO for automatic operation. The proposed approach is applied in the Taihu tunnel excavation project, where the effectiveness of the method is demonstrated via the comparisons with the site monitoring data. The method is reliable with a prediction accuracy of more than 90%. Moreover, different optimization algorithms, including non-dominated sorting genetic algorithm (NSGA-II), Pareto Envelope-based Selection Algorithm II (PESA-II) and MOPSO, are compared, and their influences on the prediction accuracy at different excavation stages are studied. The results show that MOPSO has the best performance for high dimensional optimization task.