Crashworthiness Optimization Of Energy Absorbing Structures For Landing System

Landing system has been widely used in aircraft, aerospace, military and other industries. The cushioning characteristics of a landing system mainly depend on the energy absorption capacity of its energy absorbing structures. It is significant to design an energy absorbing structure with excellent energy absorption capacity, highly light weight and satisfaction with the paractical cushioning requirement. The commonly used energy absorbing structures in landing system are metal thin-walled structures and airbag structures. Based on the finite element analysis (FEA), we mainly investigate the energy absorption characteristics of honeycomb structure and honeycomb-filled tubular structure as well as multi-chamber airbag structure using theoretical and experimental methods in this dissertation. In addition, we optimize the energy absorbing capacity of these energy absorbing structures using the suitable optimization method. The detailed research works of this dissertation are listed as follows:(1) A crashworthiness optimization design method for metal honeycomb structure based on metamodel and multi-objective particle swarm optimization (MOPSO) algorithm is proposed. In order to design a metal hexagonal honeycomb structure with maximum energy absorbing capacity among all kinds of hexagonal honeycomb structures, we multi-optimize the energy absorbing capacity of the hexagonal honeycomb structures. In our study, the cell wall width, the cell wall thickness and the branch angle are choosed as the design variables while the specifical energy absorption (SEA) and the peak crushing stress (PCS) are selected as the optimization objectives. During the optimization process, the optimal Latin hypercube design (OLHD) method is used for sampling design points in design space firstly. Then, the SEA and PCS of the corresponding honeycomb structure of the sampling points are obtained by employing the nonlinear finite element code LS-DYNA. Based on the informations of these sampling points, the Polynomial functions, Radial Basis Functions (RBF), Kriging, Multivariate Adaptive Regression Splines (MARS) and Support Vector Regression (SVR) are utilized to formulate the metamodels of the two optimal objectives SEA and PCS, respectively. By comparing the accuracy of these metalmodels, we find that the quadratic and cubic polynomial functions are the most accurate ones among these metamodels for predicting SEA and PCS. According to the accurate metamodels of SEA and PCS, the crashworthiness of the metal hexagonal honeycomb structure is multi-optimized by utilizing MOPSO algorithm. A series of optimal designs of metal hexagonal honeycomb structures with the PCSs constrained under different values are finally obtained.(2) The mean crushing stresses of three commonly used honeycombs (regular hexagonal honeycomb, reinforced regular hexagonal honeycomb and flex honeycomb) are calculated by using the Simplified Super Folding Element (SSFE) theory. Based on the theoretical solutions of the mean crushing stresses, the SEAs of three pre-crushed commonly used honeycombs are calculated, respectively. In our following optimization, the cell wall width and the cell wall thickness are choosed as the design variables while the specifical energy absorption (SEA) and the peak crushing stress (PCS) are selected as the optimization objectives. Then, the crashworthiness of three commonly used pre-crushed honeycombs is optimized by using MOPSO algorithm. According to the multiobjective optimization results, the optimal designs of three commonly used honeycombs with the PCSs constrained under different values can be obtained. In addition, we find that the energy aborption capcity of regular hexagonal honeycomb structure is the most excellent one among that of three kinds of honeycombs.(3) A six-level judgement method for crashworthiness evaluation and a crashworthiness optimization design method for honeycomb-filled polygonal tubes based on metamodel and MOPSO algorithm are proposed. After ranking the levels of the SEA and crash load efficiency (CLE) of HSPTs and HBPTs under axial loading based on the six-level judgement method, it is found that the honeycomb-filled single and honeycomb-filled bitubular tubes with enneagonal configuration have comparative most excellent energy absorption characteristics among the considered cases. Next, the HSPTs and HBPTs with enneagonal configuration are optimized by adopting MOPSO algorithm to achieve maximum SEA capacity and minimum peak crushing force (PCF). During the process of optimization, accurate metamodels of SEA and PCF of the HSPTs and HBPTs with enneagonal configuration are established. After optimization, we obtain the Pareto fronts of SEA and PCF for the HSPTs and HBPTs with enneagonal configuration. According to the Pareto fronts, the optimal designs of HSPTs and HBPTs with enneagonal configuration are obtained when their PCFs are constrained under different values. By comparing the Pareto fronts of two kinds of honeycomb-filled tubes, we find that the energy absorption capability per unit mass of the optimized honeycomb-filled single enneagonal tubes is more powerful than that of the honeycomb-filled bitubular enneagonal tubes while the PCF constrained under the same level. This indicates that the honeycomb-filled single enneagonal tube is superior to the honeycomb-filled bitubular enneagonal tube in the considered cases.(4) A crashworthiness optimization design method for cushioning honeycomb in the legged lander based on Wierzbicki’s mean crushing stress (MCS) expression for honeycomb is proposed. According to Wierzbicki’s expression for mean crushing stress (MCS) of honeycomb, we calculate the MCS and SEA of the honeycomb energy absorbing structure in the legged lander. Based on these theoretical solutions, we optimize the crashwothiness of the honeycomb energy absorbing structure in the legged lander using MOPSO algorithm while the cell wall with and cell wall thickness of the honeycomb structur are selected as the design variables. By using the energy equivalent principle, we can obtain the minimum length of the honeycomb structure according to the previous optimal results. Thus, the optimal solution to the cell wall width, cell wall thickness and the length of the honeycomb energy absorbing structure can be obtained. Then, we optimize the aluminium honeycomb structure in the primary strut of the legged lander with four landing legs. Finally, the optimal design is investigated using finite element analysis with finite element model validated by experiment. The numerical results show that the optimal design not only impoves the energy absorbing capacity but also meets the practical engineering demand very well.(5) A crashworthiness optimization design method based on metamodel and MOPSO algorithm for multi-chamber airbag cushioning system is proposed. For the optimization model of the multi-chamber airbag cushioning system, the volume, initial pressure, vented pressure and vented area of each airbag are chosen as the design variables and the acceleration of the centre of gravity (CG) of the cushioned landing equipment and the SEA of the airbag system are selected as the optimization objectives. Then, the crashworthiness of multi-chamber airbag cushioning system is multi-optimized using MOPSO. By using this proposed multi-chamber airbag cushioning system optimization method, we optimize the non-connected multi-chamber airbag cushioning system for airdropping heavy equipment and the connected multi-chamber airbag cushioning system for Mars landing. During the optimization process of the non-connected multi-chamber airbag cushioning system for airdropping heavy equipment, the initial pressure, vented pressure and vented area of the each single airbag are chosen as the design variables. And the acceleration of the centre of gravity (CG) of the cushioned landing equipment and the SEA of the airbag system are selected as the optimization objectives. In order to establish the metamodel of the optimization objectives, we sample the design points using OLHD method and obtain the objective function values using FEA through LS-DYNA. Then, the simplified quartic polynomial function metamodels of the optimization objectives of non-connected multi-chamber airbag system are established by employing the error reduction ratio (ERR) structure-selection techniques. Finally, non-connected multi-chamber airbag cushioning system is optimized on the basis of the metamodels by adopting the MOPSO algorithm to achieve minimum peak acceleration as well as minimum peak rebound velocity of the CG of the landing equipment. Based on the Pareto front obtained by the optimization, we can get the optimal design of non-connected multi-chamber airbag cushioning system for airdropping heavy equipment. During the optimization process of the connected multi-chamber airbag cushioning system for Mars landing, the initial pressure, volume of the single airbag and the connected area between airbag subsystems are chosen as the design variables. And the acceleration of the centre of gravity (CG) of the cushioned landing equipment and the SEA of the airbag system are selected as the optimization objectives. In order to establish the metamodel of the optimization objectives, we sample the design points using the full factorial design method and obtain the objective function values using FEA through LS-DYNA. Then, the polynomial function metamodels of the optimization objectives of connected multi-chamber airbag system are established. Finally, the connected multi-chamber airbag cushioning system is optimized on the basis of the metamodels by adopting the MOPSO algorithm to achieve minimum peak acceleration of the CG of the landing equipment as well as maximum SEA of the airbag system. Based on the Pareto front obtained by the optimization, we can get the optimal design of the connected multi-chamber airbag cushioning system for Mars landing. In additon, we implemented the numerical simulation for the optimal design of the Mars landing connected multi-chamber airbag using the finite element model, which has been validated by experiment. The numerical results show that the energy absorbing capacity of the optimal design of the multi-chamber airbag cushioning system is improved when meeting the demand of the overload of the landing equipment in the practical engineering.