Recent Progress in Reinforcement Learning and Adaptive Dynamic Programming for Advanced Control Applications

Reinforcement learning(RL) has roots in dynamic programming and it is called adaptive/approximate dynamic programming(ADP) within the control community. This paper reviews recent developments in ADP along with RL and its applications to various advanced control fields. First, the background of the development of ADP is described, emphasizing the significance of regulation and tracking control problems. Some effective offline and online algorithms for ADP/adaptive critic control are displayed, where the main results towards discrete-time systems and continuous-time systems are surveyed, respectively.Then, the research progress on adaptive critic control based on the event-triggered framework and under uncertain environment is discussed, respectively, where event-based design, robust stabilization, and game design are reviewed. Moreover, the extensions of ADP for addressing control problems under complex environment attract enormous attention. The ADP architecture is revisited under the perspective of data-driven and RL frameworks,showing how they promote ADP formulation significantly.Finally, several typical control applications with respect to RL and ADP are summarized, particularly in the fields of wastewater treatment processes and power systems, followed by some general prospects for future research. Overall, the comprehensive survey on ADP and RL for advanced control applications has d emonstrated its remarkable potential within the artificial intelligence era. In addition, it also plays a vital role in promoting environmental protection and industrial intelligence.

Removal of Nitrogen Compounds from Fuels by Deep Eutectic Solvent Based on 1-Butyl-3-Methylimidazolium Bromide and Dicarboxylic Acids

1-Butyl-3-methylimidazolium bromide-based ([Bmim]Br-based) deep eutectic solvents (DESs) can be used aseffective extractants for removing nitrogen compounds from fuel oil. Among the DESs studied, the combination of [Bmim]Brand malonic acid (MA) in a 1:1 molar ratio demonstrated good performance for the removal of both basic and non-basic Ncompounds. The DES [Bmim]Br-MA exhibited extraction efficiencies of 98.4% for quinoline and 92.9% for indole after 30min at 30 ℃ with a DES/oil mass ratio of 1/7. Moreover, the extraction efficiencies remained high at 86.4% for quinolineand 85.9% for indole after recycling the DES four times.. In addition, the DES [Bmim]Br-MA could effectively removeN compounds from Fushun shale diesel oil, with extraction efficiencies of 83.5% for total nitrogen and 89.9% for basicnitrogen at a DES/oil mass ratio of 1/1.

Oxidative Desulfurization of Fuel Oil with H_(3)PO_(4)-based Deep Eutectic Solvents

A series of Lewis-acid deep eutectic solvents (DESs) were synthesized by stirring phosphoric acid and zincchloride as raw materials at 80℃ to form H_(3)PO_(4)/nZnCl_(2) (n = 0.1, 0.25, 0.5, 0.75, 1). The DESs were characterized byFourier transform infrared spectrophotometry (FT-IR), thermogravimetry/differential thermogravimetry (TG/DTG), andelectron spray ionization mass spectrometry (ESI-MS). The DESs were used as both extractants and catalysts to removedibenzothiophene from fuels via oxidative desulfurization (ODS). Experiments were performed to investigated the influenceof factors such as composition of DES, temperature, oxidant dosage (molar ratio of O:S), DES dosage (volume ratio ofDES:oil), and number of cycles on desulfurization rate. The results indicated that the removal rate of dibenzothiophene (DBT)was affected by the Lewis acidic DESs, with that of H_(3)PO_(4)/0.25∙ZnCl_(2) reaching 96.4% under optimal conditions (Voil=5 mL,VDES=1 mL, an oxidant dosage of 6, T=50 ℃). After six cycles, the desulfurization rate of H_(3)PO_(4)/0.25∙ZnCl_(2) remained above94.1%. The apparent activation energy of dibenzothiophene (DBT) removal reaction was determined by a pseudo-first orderkinetic equation according to the Arrhenius equation to be 32.34 kJ/mol, as estimated. A reaction mechanism is proposedbased on the experimental data and characterization results.

A nonwoven supported mixed matrix membrane for CH4/N2 separation

Efficiently enriching low-concentration CH4 is pivotal for enhancing the utilization of unconventional energy sources and mitigating greenhouse gas emissions.This study focuses on modifying the overall performance of CH_(4)/N_(2)separation membranes.A novel mixed matrix membrane(MMM)with a reinforced substrate structure was developed through a straightforward dip-coating technique.This MMM incorporates a polytetrafluoroethylene(PTFE)porous membrane as the supporting framework,while a composite of block polymer(styrene-butadiene-styrene)and metal-organic framework(Ni-MOF-74)forms the selective separation layer.Comprehensive characterization of Ni-MOF-74 and the fabricatedmembranes was conducted using X-rays diffraction,scanning electron microscope,Brunauer-Emmett-Teller analysis,and gas permeance tests.The findings indicate a robust integration of the PTFE porous support with the membrane layer,enhancing the mechanical stability of theMMM.Under optimal conditions,the mechanical strength of the PM20 membrane(containing 20%Ni-MOF-74)was observed to be 37.7 MPa,representing remarkable increase compared to the non-reinforcedMMM.Additionally,thePM20membrane exhibited an impressive CH4 permeation rate of 92 barrer(1 barrer﹦3.35×10^(-16)mol·m·m^(-2)·s^(-1)·Pa^(-1))alongside a CH_(4)/N_(2)selectivity of 4.18.These results underscore the MMM's substantial performance and its promising potential in methane enrichment applications.

Stress corrosion cracking behavior of buried oil and gas pipeline steel under the coexistence of magnetic field and sulfate-reducing bacteria

Magnetic field and microorganisms are important factors influencing the stress corrosion cracking(SCC)of buried oil and gas pipelines. Once SCC occurs in buried pipelines, it will cause serious hazards to the soil environment. The SCC behavior of X80 pipeline steel under the magnetic field and sulfate-reducing bacteria(SRB) environment was investigated by immersion tests, electrochemical tests, and slow strain rate tensile(SSRT) tests. The results showed that the corrosion and SCC sensitivity of X80 steel decreased with increasing the magnetic field strength in the sterile environment. The SCC sensitivity was higher in the biotic environment inoculated with SRB, but it also decreased with increasing magnetic field strength, which was due to the magnetic field reduces microbial activity and promotes the formation of dense film layer. This work provided theoretical guidance on the prevention of SCC in pipeline steel under magnetic field and SRB coexistence.

Hydrocarbon Generation Potential and Organic Matter Enrichment Mechanism of the Cambrian Marine Shale in the Tadong Low Uplift,Tarim Basin

Cambrian shales in China and elsewhere contain abundant oil and gas resources.However,due to its deep burial and limited outcrop,there has been relatively little research conducted on it.The Cambrian shale of the Tadong low uplift in the Tarim Basin of western China,specifically the Xidashan-Xishanbulake Formation(Fm.)and overlying Moheershan Fm.provide a case study through the use of organic petrology,mineralogy,organic and elemental geochemistry,with the aim of analyzing and exploring the hydrocarbon generation potential(PG)and organic matter(OM)enrichment mechanisms within these shale formations.The results indicate that:(1)the Cambrian shale of the Tadong low uplift exhibits relatively dispersed OM that consists of vitrinite-like macerals and solid bitumen.These formations have a higher content of quartz and are primarily composed of silica-based lithology;(2)shale samples from the Xidashan-Xishanbulake and Moheershan formations demonstrate high total organic carbon(TOC)and low pyrolytic hydrocarbon content(S_(2))content.The OM is predominantly typeⅠand typeⅡkerogens,indicating a high level of maturation in the wet gas period.These shales have undergone extensive hydrocarbon generation,showing characteristics of relatively poor PG;(3)the sedimentary environments of the Xidashan-Xishanbulake and Moheershan formations in the Tadong low uplift are similar.They were deposited in warm and humid climatic conditions,in oxygen-deficient environments,with stable terrigenous inputs,high paleoproductivity,high paleosalinity,weak water-holding capacity,and no significant hydrothermal activity;and(4)the relationship between TOC and the paleoproductivity parameter(P/Ti)is most significant in the Lower Cambrian Xidashan-Xishanbulake Fm.,whereas correlation with other indicators is not evident.This suggests a productivity-driven OM enrichment model,where input of landderived material was relatively small during the Middle Cambrian,and the ancient water exhibited lower salinity.A comprehensive pattern was formed under the combined control of paleoproductivity and preservation conditions.This study provides valuable guidance for oil and gas exploration in the Tarim Basin.

Extraction of the key infrared radiation temperature features concerning stress and crack evolution of loaded rocks

The infrared radiation temperature(IRT)variation concerning stress and crack evolution of rocks is a critical focus in rock mechanics domain and engineering disaster warning.In this paper,a methodology to extract the key IRT features related to stress and crack evolution of loaded rocks is proposed.Specifically,the wavelet denoising and reconstruction in thermal image sequence(WDRTIS)method is employed to eliminate temporal noise in thermal image sequences.Subsequently,the adaptive partition temperature drift correction(APTDC)method is introduced to alleviate temperature drift.On this basis,the spatial noise correction method based on threshold segmentation and adaptive median filtering(OTSU-AMF)is proposed to extract the key IRT features associated with microcracks of loaded rocks.Following temperature drift correction,IRT provides an estimation of the thermoelastic factor in rocks,typically around 5.29×10^(-5) MPa^(-1) for sandstones.Results reveal that the high-temperature concentrated region in cumulative thermal images of crack evolution(TICE)can elucidate the spatiotemporal evolution of localized damage.Additionally,heat dissipation of crack evolution(HDCE)acquired from TICE quantifies the progressive failure process of rocks.The proposed methodology enhances the reliability of IRT monitoring results and provides an innovative approach for conducting research in rock mechanics and monitoring engineering disasters.

Multi-Objective Optimization of Multi-Product Parallel Disassembly Line Balancing Problem Considering Multi-Skilled Workers Using a Discrete Chemical Reaction Optimization Algorithm

This work investigates a multi-product parallel disassembly line balancing problem considering multi-skilled workers.A mathematical model for the parallel disassembly line is established to achieve maximized disassembly profit and minimized workstation cycle time.Based on a product’s AND/OR graph,matrices for task-skill,worker-skill,precedence relationships,and disassembly correlations are developed.A multi-objective discrete chemical reaction optimization algorithm is designed.To enhance solution diversity,improvements are made to four reactions:decomposition,synthesis,intermolecular ineffective collision,and wall invalid collision reaction,completing the evolution of molecular individuals.The established model and improved algorithm are applied to ball pen,flashlight,washing machine,and radio combinations,respectively.Introducing a Collaborative Resource Allocation(CRA)strategy based on a Decomposition-Based Multi-Objective Evolutionary Algorithm,the experimental results are compared with four classical algorithms:MOEA/D,MOEAD-CRA,Non-dominated Sorting Genetic Algorithm Ⅱ(NSGA-Ⅱ),and Non-dominated Sorting Genetic Algorithm Ⅲ(NSGA-Ⅲ).This validates the feasibility and superiority of the proposed algorithm in parallel disassembly production lines.

Machine learning method to predict dynamic compressive response of concrete-like material at high strain rates

Machine learning(ML)methods with good applicability to complex and highly nonlinear sequences have been attracting much attention in recent years for predictions of complicated mechanical properties of various materials.As one of the widely known ML methods,back-propagation(BP)neural networks with and without optimization by genetic algorithm(GA)are also established for comparisons of time cost and prediction error.With the aim to further increase the prediction accuracy and efficiency,this paper proposes a long short-term memory(LSTM)networks model to predict the dynamic compressive performance of concrete-like materials at high strain rates.Dynamic explicit analysis is performed in the finite element(FE)software ABAQUS to simulate various waveforms in the split Hopkinson pressure bar(SHPB)experiments by applying different stress waves in the incident bar.The FE simulation accuracy is validated against SHPB experimental results from the viewpoint of dynamic increase factor.In order to cover more extensive loading scenarios,60 sets of FE simulations are conducted in this paper to generate three kinds of waveforms in the incident and transmission bars of SHPB experiments.By training the proposed three networks,the nonlinear mapping relations can be reasonably established between incident,reflect,and transmission waves.Statistical measures are used to quantify the network prediction accuracy,confirming that the predicted stress-strain curves of concrete-like materials at high strain rates by the proposed networks agree sufficiently with those by FE simulations.It is found that compared with BP network,the GA-BP network can effectively stabilize the network structure,indicating that the GA optimization improves the prediction accuracy of the SHPB dynamic responses by performing the crossover and mutation operations of weights and thresholds in the original BP network.By eliminating the long-time dependencies,the proposed LSTM network achieves better results than the BP and GA-BP networks,since smaller mean square error(MSE)and higher correlation coefficient are achieved.More importantly,the proposed LSTM algorithm,after the training process with a limited number of FE simulations,could replace the time-consuming and laborious FE pre-and post-processing and modelling.

C_(9)H_(10)O_(2):0.5ZnCl_(2)/SG as a High-Efficiency Catalyst for Desulfurization of Model Oil

A C_(9)H_(10)O_(2):0.5ZnCl_(2)/SG catalyst was synthesized using a one-step sol-gel method with silica gel(SG)as the carrier and C_(9)H_(10)O_(2):0.5ZnCl_(2)deep eutectic solvent(DES)as active component.The structure of the supported catalyst was characterized by FT-IR,XRD,SEM,and N2 adsorption-desorption,and the DES was found to have successfully permeated the SG through its pores.The removal of dibenzothiophene(DBT)in model diesel was studied using C_(9)H_(10)O_(2):0.5ZnCl_(2)/SG as a catalyst and H_(2)O_(2)as an oxidant.The influence of loading dose of DES,reaction temperature,catalyst dosage,O/S molar ratio,and sulfide type on the desulfurization rate was investigated.The removal rates of DBT,4,6-dimethyldibenzothiophene(4,6-DMDBT),and benzothiophene(BT)under optimal reaction conditions were 99.4%,96%,and 78.2%,respectively.C_(9)H_(10)O_(2):0.5ZnCl_(2)/SG catalyst could be recycled five times with a little decrease of oxidative desulfurization activity,and the adsorption-oxidation desulfurization mechanism was examined.

Design and Development of Highly Efficient Ni/ZnO Catalysts Based on Semiconductor Metal Oxide Catalysis Theory

In the context of reactive adsorption desulfurization,the development of an efficient Ni/ZnO desulfurizer has attracted increasing attention.In the work reported here,a novel Ni/Mn-ZnO composite nanowire desulfurizer is designed on the basis of the catalytic theory of semiconductor metal oxides and the characteristics of one-dimensional nanomaterials.X-ray diffraction,scanning electron microscopy,N_(2) adsorption-desorption,and X-ray photoemission spectroscopy demonstrate that Mn doping changes the crystal structure and morphology of the Ni/ZnO desulfurizer,increases the number of quasi-free electrons in the ZnO,and promotes H_(2)S adsorption.The Ni/Mn-ZnO composite nanowire desulfurizer exhibits good desulfurization performance when used with gasoline as the raw material.

Carbonate-activated binder modified by supplementary materials for mine backfill and the associated heavy metal immobilization effects

Cemented paste backfill(CPB)is one of the effective methods for resource utilization of tailings,but the high cost of ordinary Portland cement(OPC)limits its utilization.Considering the poor performance of Na_(2)CO_(3)-activated binders,in this work,supplementary materials,including CaO,MgO,and calcined layered double hydroxide(CLDH),were used to modify their properties with the aim of finding an alternative binder to OPC.Isothermal calorimetry,X-ray diffraction,and thermogravimetric analyses were conducted to explore the reaction kinetics and phase assembles of the binder.The properties of the CPB samples,such as flowability,strength development,and heavy metal immobilization effects,were then investigated.The results show that the coupling utilization of MgO and CLDH showed good performance.The strength of the Mg_(2)-CLDH_(3) sample was approximately 2.94 MPa after curing for 56 d,which was higher than that of the OPC-based sample.Moreover the cost of the modified Na_(2)CO_(3)-activated binder was lower than that of the OPC-based binder.Modified sample showed satisfactory heavy metal immobilization effects.These findings demonstrate that carbonate-activated binder modified by supplementary materials can be suitable in CPB.

Spatiotemporal phase change materials for thermal energy long-term storage and controllable release

Phase change materials(PCMs)have attracted much attention in the field of solar thermal utilization recently,due to their outstanding thermal energy storage performance.However,PCMs usually release their stored latent heat spontaneously as the temperature below the phase transition temperature,rendering thermal energy storage and release uncontrollable,thus hindering their practical application in time and space.Herein,we developed erythritol/sodium carboxymethylcellulose/tetrasodium ethylenediaminetetraacetate(ERY/CMC/EDTA-4Na)composite PCMs with novel spatiotemporal thermal energy storage properties,defined as spatiotemporal PCMs(STPCMs),which exhibit the capacity of thermal energy long-term storage and controllable release.Our results show that the composite PCMs are unable to lose latent heat due to spontaneous crystallization during cooling,but can controllably release thermal energy through cold crystallization during reheating.The cold-crystallization temperature and enthalpy of composite PCMs can be adjusted by proportional addition of EDTA-4Na to the composite.When the mass fractions of CMC and EDTA-4Na are both 10%,the composite PCMs can exhibit the optical coldcrystallization temperature of 51.7℃ and enthalpy of 178.1 J/g.The supercooled composite PCMs without latent heat release can be maintained at room temperature(10-25℃)for up to more than two months,and subsequently the stored latent heat can be controllably released by means of thermal triggering or heterogeneous nucleation.Our findings provide novel insights into the design and construction of new PCMs with spatiotemporal performance of thermal energy long-term storage and controllable release,and consequently open a new door for the development of advanced solar thermal utilization techniques on the basis of STPCMs.

Synthesis of Bimodal Mesoporous TiO_(2)-PTA/BMMS and Its Enhanced Performance in the Photocatalytic Oxidative Desulfurization

With the bimodal mesoporous silica(BMMS)acting as the support and the composite of TiO2 with phosphotungstic acid(PTA)functioning as the active constituent,TiO_(2)-PTA/BMMS was synthesized by the two-step impregnation route.This catalyst was applied in the photocatalytic oxidative desulfurization(PODS)process,with dibenzothiophene serving as the model sulfur compound.PODS proceeds in one pot,in which H_(2)O_(2) acts as the oxidant and methanol plays the role of the solvent.TiO_(2)-PTA/BMMS was characterized by XRD,N_(2) adsorption and desorption,XRF,FTIR,UV-vis,SEM,EDS and TEM techniques.It showed that the introduction of PTA contributes higher order,higher surface area and pore volume to the bimodal mesoporous support.With TiO_(2)-PTA/BMMS used as the catalyst under the UV irradiation,the desulfurization rate can reach 99.6%.This result is obviously higher than that achieved by TiO_(2)/BMMS.The catalyst also has no significant drop in catalytic activity after eight runs of reusing.In such catalytic system,the synergistic effects of this photocatalytic oxidation and the extraction with the methanol serving as the solvent played an indispensable role.

Influence of B4C particle size on microstructure and damping capacities of(B_(4)C+Ti)/Mg composites

To study the influence of B4C particle size on the microstructure and damping capacities of(B_(4)C+Ti)/Mg composites,in situ reactive infiltration technique was utilized to prepare Mg-matrix composites.The microstructure,produced phases and damping capacities of the composites prepared with different particle size of B4C were characterized and analyzed.The results show that the reaction between B4C and Ti tends to be more complete when finer B_(4)C particle was used to prepare the composites.But the microstructure of the as-prepared composites is more homogenous when B4C and Ti have similar particle size.The strain-dependent damping capacities of(B_(4)C+Ti)/Mg composites improve gradually with the increase of strain amplitude,and composites prepared with coarser B4C particles tend to have higher damping capacities.The temperature-dependent damping capacities improve with increasing the measuring temperatures,and the kind of damping capacities of the composites prepared with 5mm B4C are inferior to those of coarser particles.The dominant damping mechanism for the strain-damping capacity is dislocation damping and plastic zone damping,while that for the temperature-damping capacity is interface damping or grain boundary damping.

Removal of Basic Nitrogen Compounds from Diesel Fraction with NMP-0.5ZnCl_(2) Coordinated Ionic Liquid

An inexpensive coordinated ionic liquid NMP-0.5ZnCl_(2) was synthesized by reacting N-methyl-pyrrolidone with anhydrous ZnCl_(2),and its structure was characterized with FT-IR spectroscopy.The performance of IL for removing basic nitrogen compounds from model oil containing quinoline and actual coker diesel was studied.Experimental results showed that the IL,NMP-0.5ZnCl_(2),exhibited a good denitrogenation performance,which can be attributed to its low viscosity and unoccupied orbitals of Zn ion,while obtaining a 99.68%quinoline removal efficiency under conditions covering a temperature of 50℃,an IL/model oil mass ratio of 1:2,and a reaction time of 30 min.In the case of coker diesel,above a 91%basic N-removal efficiency(with N-content reduced from 536μg/g to 47μg/g)was realized by the IL after 5-stage extraction.Moreover,the quinoline extraction efficiency could still reach 96.73%during four recycles of the IL.

Ce_(2)(MoO_(4))_(3) as an efficient catalyst for aerobic oxidative desulfurization of fuels

Ce_(2)(MoO_(4))_(3)was synthesized by a simple reflux method using cerium nitrate hexahydrate and ammonium molybdate as reactants.The as-prepared Ce_(2)(MoO_(4))_(3)was characterized by Fourier transform infrared spectroscopy(FT-IR),X-ray diffraction(XRD),Scanning electron microscope(SEM),and X-ray photoelectron spectroscopy(XPS).The removal of dibenzothiophene(DBT)in model oil was studied using Ce_(2)(MoO_(4))_(3)as catalyst and oxygen as oxidant.The reaction factors such as reaction temperature,amount of catalyst,and sulfide type on sulfur removal were researched.The results prove that both Ce3+and MoO42-play significant role in the conversion from DBT to DBTO2.The Ce_(2)(MoO_(4))_(3)catalyst has an excellent performance for the sulfur removal of DBT.Under the optimum reaction conditions,sulfur removal of 99.6%was obtained.After recycling five times,no significant loss in catalyst activity of Ce_(2)(MoO_(4))_(3).Mechanism of aerobic oxidative desulfurization was proposed based on the experiment of free radical capture and infrared characterization.

A one-for-all strategy of polyimide coating layer for resolving the comprehensive issues of phosphorus anode

Phosphorus is a promising anode with high capacity (2596 mAh g^(-1)and 6075~6924 mAh cm^(-3)),low lithium-ion diffusion barrier (0.08 e V),and appropriate lithiation potential (~0.7 V vs Li+/Li).However,it faces the problems of huge volume expansion (~300%),low electronic conductivity (10^(-14)~10^(2)S cm^(-1)),soluble intermediates (lithium polyphosphides,Li_(x)Ps),degradation in air,and low thermal stability.In this work,phosphorus/carbon nanotube composites were coated with a polyimide layer,which plays the roles of a buffer layer to relieve the volume expansion of phosphorus,an obstruct layer to confine LixPs,an inert layer to prevent the degradation of phosphorus in air,a heat resistant layer to improve the thermal stability of the anode.The resulting composites (P/CNT@PI) display high capacity retention of798.1 m Ah g^(-1)after 150 cycles at 1 A g^(-1),achieving 17 times as much as the control sample (P/CNT).

One-step Synthesis and Photocatalytic Degradation Performance of Sulfur-doped Porous g-C_(3)N_(4)Nanosheets

In this study,the sulfur-doped porous g-C_(3)N_(4) nanosheets(CN-T-U 1.75)were synthesized successfully by onestep calcination utilizing urea and thiourea as precursors.Under visible light irradiation,CN-T-U 1.75 showed remarkable photocatalytic activity for Rhodamine B(RhB)degradation with a kinetic reaction rate constant of 0.01838 min^(-1).The characterization analysis indicated that CN-T-U 1.75 had a higher specific surface area and the doping altered the energy band structure.This work offers a new viewpoint on modifying the band structure of a photocatalyst using a doping strategy,as well as new insights into the generation routes of active species involved in the photocatalytic process.

Drop impact analysis of TSV-based 3D packaging structure by PSO-BP and GA-BP neural networks

Particle swarm algorithm(PSO) and genetic algorithm(GA) were used to optimize the back propagation(BP) artificial neural network for predicting the dynamic responses of the through silicon via(TSV) based three-dimensional packaging structures.A finite element model of the TSV packaging structure with a strain-rate dependent constitutive model for solder joints was created to simulate the drop impact due to a free fall of 0.8 m to the rigid ground to investigate the structural dynamic responses during the whole impact process.The spatial coordinates of the solder joints were used as the input parameters of the hybrid neural network model for the drop impact responses,while the maximum Von Mises stress and PEEQ(plastic strain) values are identified the output parameters.The correlation coefficient(R),the mean absolute percentage error(MAPE) and the training time were used as the measures to validate and compare the proposed PSO-BP and GA-BP neural networks.The results show that both the PSO-BP model and the GA-BP model can achieve high accuracy predictions with strong generalization capability.Apparently,both optimized algorithms outperform the original BP model,but the PSO-BP model is slightly more superior than the GA-BP model.It is also demonstrated that the proposed optimized algorithms efficiently predict the drop impact responses of TSV packaging structures by greatly saving the computational and experimental cost of drop impact tests.