Search

The Poaceae family, the fifth largest family of angiosperms, has always been an important source of cereal crops, medicinal crops, and industrial crops. However, comprehensive studies on the biological activity and chemical composition of many species are lacking. In the present study, we tested the total flavonoid and total phenol contents and antioxidant, α-glycosidase, α-amylase, acetylcholinesterase (AChE), and butyrylcholinesterase (BChE)-inhibitory activities of five Poaceae plants (Echinochloa crus-galli, Miscanthus floridulus, Sporobolus fertilis, Bromus japonicus, and Eleusine indica). As a result, the 95 % ethanol (EtOH) extract of M. floridulus (EMF) was found to possess strongest abovementioned activities among the extracts of the five different species. The IC50 value of the DPPH radical scavenging activity was 23.18 ± 0.36 μg/mL, and those of the anti-AChE, anti-BChE and anti-α-glycosidase activities were 49.92 ± 1.44, 100.37 ± 2.48, and 50.01 ± 2.15 μg/mL, respectively. Forty-four compounds were identified from the EMF through liquid chromatography–hybrid ion trap-time of flight mass spectrometry (LC–IT–TOF–MS) analysis, and the majority of these compounds were flavonoids and organic acids. In general, the obtained results highlight the medicinal value of the five members of the Poaceae family. More importantly, M. floridulus has potential for the development of multitarget anti-Alzheimer’s disease (AD) and antidiabetic agents in the pharmaceutical industry.

Melon seeds have received considerable attention in recent years because of their high protein content, but they have not yet been fully used. The modification of melon seed protein (MSP) using ultrasound-assisted pH-shifting treatment was investigated in this study by analyzing structural characteristics and functional properties. The particle size, free sulfhydryl content, surface hydrophobicity, solubility, secondary structure, water-holding capacity, oil-holding capacity, emulsification activity index, and emulsification stability index of MSP were determined. MSP treated with ultrasound-assisted, pH-shifting had a smaller particle size, lower free sulfhydryl content, higher surface hydrophobicity, and solubility increased from 43.67 % to 89.12 %. The secondary structure of MSP was affected by ultrasonic treatment, manifesting as an α-helix increase and β-helix, β-turn, and random coil content decrease, which may be the reason why the protein structure became more compact after treatment. The water and oil holding capacities of MSP increased from 2.74 g/g and 3.14 g/g in untreated samples to 3.19 g/g and 3.97 g/g for ultrasound-treated samples, and further increased to 3.97 g/g and 5.02 g/g for ultrasound-assisted, pH-shifting treatment at pH 9.0, respectively. The emulsification activity index of MSP was 21.11 m2/g before treatment and reached a maximum of 32.34 m2/g after ultrasound-assisted, pH-shifting treatment at pH 9.0. The emulsification stability of MSP was maximized by ultrasonic treatment at pH 7.0. Ultrasound-assisted, pH-shifting treatment can effectively improve the functional properties of MSP by modifying the protein structure, which improves the potential application of melon seed protein in the food industry.

Accurate time synchronisation is critical in modern communication, navigation, and scientific research. In this context, the eLoran receiver, which is an advanced timing device, has attracted increasing scholarly attention. This study aims to comprehensively analyse the performance and potential applications of eLoran systems for timing monitoring and to specifically explore the relevant indices of two eLoran receivers. To this end, we evaluated the performance of these receivers in receiving time signals through both simulated and empirical data and conducted a regularity analysis to uncover their potential value in practical applications. The findings demonstrate that the eLoran receiver excels at timing monitoring and provides highly accurate time information. An analysis of the timing performance of the eLoran system improved its accuracy and integrity.

Farmland abandonment monitoring is one of the key aspects of land use and land cover research, as well as being an important prerequisite for ecological environmental protection and food security. A Normalized Difference Vegetation Index (NDVI) time series analysis is a common method used for farmland abandonment data extraction; however, extracting this information using high-resolution data is still difficult due to the limitations caused by cloud influence and data of low temporal resolution. To address this problem, this study used STARFM for GF-6 and Landsat 8 data fusion to enhance the continuity of high-resolution and cloudless images. A dataset was constructed by combining the phenological cycle of crops in the study area and then extracting abandoned farmland data based on an NDVI time series analysis. The overall accuracy of the results based on the NDVI time series analysis using the STARFM-fused dataset was 93.42%, which was 15.5% higher than the accuracy of the results obtained using only GF-6 data and 28.52% higher than those obtained using only Landsat data. Improvements in accuracy were also achieved when using SVM for time series analysis based on the fused dataset, indicating that the method used in this study can effectively improve the accuracy of the results. Then, we analyzed the spatial distribution pattern of abandoned farmland by extracting the results and concluded that the abandonment rate increased with the increase in the road network density and decreased with the increase in the distance to residential areas. This study can provide decision-making guidance and scientific and technological support for the monitoring of farmland abandonment and can facilitate the analysis of abandonment mechanisms in the study area, which is conducive to the sustainable development of farmland.

Three-dimensional (3D) models provide the most intuitive representation of geological conditions. Traditional modeling methods heavily depend on technicians’ expertise and lack ease of updating. In this study, we introduce a deep learning-based method for 3D geological implicit modeling, leveraging a substantial dataset of geological drilling data. By applying resampling and normalization techniques, we standardize drilling data and significantly expand the dataset, making it suitable for training deep neural networks. Utilizing the characteristics of the sample data, we design and establish the network structure, loss function, and parameter configurations, resulting in the training of a deep neural network with high accuracy and robust generalization capability. Ultimately, we utilize the dataset generated from the network’s predictions to render and construct the 3D geological model. The research in this paper demonstrates the significant promise of deep neural networks in addressing geological challenges. The deep learning-based implicit 3D modeling method surpasses traditional approaches in terms of generalization, convenience, and adaptability.

In the construction of underground engineering, the determination of surrounding rock stresses has important theoretical significance for the design of tunnel structure reliability. In this paper, the nonlinear criterion is introduced to modify and improve the friction resistance in the calculation formula of Bierbaumer’s loose surrounding rock stresses, and the improved application is carried out in view of the situation of composite formation. Due to the variability and discreteness of geotechnical parameters, combined with engineering examples, Monte Carlo random sampling is carried out for various geotechnical parameters in engineering, and the characteristics of loose surrounding rock stresses are analyzed from the perspective of statistical reliability by using the improved shallow tunnel formula. By using the standard regression coefficient, the weight influence of the parameters in various shallow surrounding rock stress formulas is analyzed, and the confidence degree of the calculated results of various formulas is analyzed from a statistical point of view, which verifies the rationality of the improved formula in the statistical sense.

Computing-in-Memory (CIM) is a novel computing architecture that enormously improves energy efficiency and reduces computing latency by avoiding frequent data movement between the computation and memory units. Currently, digital CIM is regarded as more suitable for high-precision operations represented in floating-point arithmetic, as it is not limited by the bit width of ADC/DAC in analog CIM. However, the development of DCIM still faces two problems: On the one hand, mainstream SRAM-based DCIM memory cells introduce large area overheads, which contain at least six transistors per cell. On the other hand, existing DCIM solutions can only support the computing precision up to FP32, failing to meet the demands of high-accuracy application scenarios. To overcome these problems, this work designs a novel SOT-MRAM-based digital CIM macro (MDCIM) with higher area/energy efficiency and achieves double-precision floating-point (FP64) computation with a modified fused multiply–accumulate (FMA) module. The proposed design is synthesized with a 55 nm CMOS technology node, achieving 0.62 mW power consumption, 26.9 GOPS/W, and 0.332 GOPS/mm2 energy efficiency at 150 MHz with 1.08 V supply. Circuit level simulation results show that the MDCIM can achieve higher area utilization compared to previous SRAM-based CIM designs.

Abstract Electrolyte additives have been widely used to solve critical problems of electrodes and electrolytes in various rechargeable batteries. However, due to the lack of well‐directed strategy, the selection for additives have consumed considerable efforts of researchers. In this study, a diversity‐driven electrolyte additive of 5‐bromo‐2‐(difluoro(3,4,5‐trifluorophenoxy)methyl)‐1,3‐difluorobenzene (BDTMD) is directed synthesized, two different substituent groups are designed in the structure. With this newly‐formulated electrolyte, the overpotential during the charging process of Li−CO2 cells can be reduced to 3.8 V without any catalyst. Li metal electrode also shows a stable solid electrolyte interface layer and long cycling life of 1600 h. Based on the synergistic effect between F and Br, this study demonstrates an efficiently and well‐directed strategy to synthesis electrolyte additives.

Abstract Graphdiyne (GDY), an emerging type of carbon allotropes with unique sp−sp2 carbon atoms, is firstly explored for high selectivity direct electrochemical sensing of ultra‐trace level copper ions (Cu2+). It achieves a wide detection range of 0.25–380 μM and a trace level as low as 8 nM, among the best performance of reported carbon‐based sensors. It is found that GDY with unique C≡C possesses both chelating and reducing abilities to efficiently anchor and reduce Cu2+ to Cu+, eventually leading to ultrasensitive detection.

In order to achieve reliability, security, and scalability, the request flow in the Internet of Things (IoT) needs to pass through the service function chain (SFC), which is composed of series-ordered virtual network functions (VNFs), then reach the destination application in multiaccess edge computing (MEC) for processing. Since there are usually multiple identical VNF instances in the network and the network environment of IoT changes dynamically, placing the SFC for the IoT request flow is a significant challenge. This paper decomposes the dynamic SFC placement problem of the IoT-MEC network into two subproblems: VNF placement and path determination of routing. We first formulate these two subproblems as Markov decision processes. We then propose a meta reinforcement learning and fuzzy logic-based dynamic SFC placement approach (MRLF-SFCP). The MRLF-SFCP contains an inner model that focuses on making SFC placement decisions and an outer model that focuses on learning the initial parameters considering the dynamic IoT-MEC environment. Specifically, the approach uses fuzzy logic to pre-evaluate the link status information of the network by jointly considering available bandwidth, delay, and packet loss rate, which is helpful for model training and convergence. In comparison to existing algorithms, simulation results demonstrate that the MRLF-SFCP algorithm exhibits superior performance in terms of traffic acceptance rate, throughput, and the average reward.

The stiffness of static pressure spindles and slides is affected by the phenomenon of fluid-structure interaction (FSI). Many methods have been proposed to calculate the stiffness with FSI considered. However, the influence of FSI on stiffness is still unknown. This paper studies the lateral stiffness of an aerostatic slide. The relationship between the lateral stiffness of the aerostatic slide, the lateral bearings, and the solid structure is deduced. According to the relationship, this paper proposes a theory that reveals the influence mechanism of FSI on the stiffness of static pressure spindles and slides. The proposed theory is also valid for the normal stiffness of the aerostatic slide and the thrust stiffness of an aerostatic spindle.

Underwater target detection plays a vital role in various application scenarios, ranging from scientific research to military and industrial operations. In this paper, a detection method via the Cayley–Klein measure and a prior information of shape is proposed for the issue of hyperspectral underwater target identification. Firstly, by analyzing the data features of underwater targets and backgrounds, a background suppression algorithm based on Cayley–Klein measure is developed to enhance the differentiation between underwater targets and backgrounds. Then, a local peak-based algorithm is designed to discriminate potential underwater target points based on the local peak features of underwater targets. Finally, pseudo-target points are eliminated based on the priori shape information of underwater targets. Experiments show that the algorithm proposed is efficient and can effectively detect underwater targets from hyperspectral images.

With the disruptive carbon cycle being blamed for global warming, the plausible electrocatalytic CO2 reduction reaction (CO2RR) to form valuable C2+ hydrocarbons and feedstock is becoming a hot topic. Cu-based electrocatalysts have been proven to be excellent CO2RR alternatives for high energy value-added products in this regard. However, the selectivity of CO2RR to form C2+ products via Cu-based catalysts suffers from a high overpotential, slow reaction kinetics, and low selectivity. This review attempts to discuss various cutting-edge strategies for understanding catalytic design such as Cu-based catalyst surface engineering, tuning Cu bandgap via alloying, nanocatalysis, and the effect of the electrolyte and pH on catalyst morphology. The most recent advances in in situ spectroscopy and computational techniques are summarized to fully comprehend reaction mechanisms, structural transformation/degradation mechanisms, and crystal facet loss with subsequent effects on catalyst activity. Furthermore, approaches for tuning Cu interactions are discussed from four key perspectives: single-atom catalysts, interfacial engineering, metal-organic frameworks, and polymer-incorporated materials, which provide new insights into the selectivity of C2+ products. Finally, major challenges are outlined, and potential prospects for the rational design of catalysts for robust CO2RR are proposed. The integration of catalytic design with mechanistic understanding is a step forward in the promising advancement of CO2RR technology for industrial applications.

The present work was devoted to developing rhodamine-like chemosensing systems for cysteine (Cys) optical recognition. Aiming at low background light and minimal photobleaching effect, up-converting nanocrystals were firstly synthesized and latterly coated by α-cyclodextrin, and finally used as an exciting host. An energy transfer procedure from these nanocrystals and rhodamine sensors was established via their spectroscopic analysis and emissive decay dynamics comparison. The binding dynamics of our chemosensors for Cys were revealed to have uncomplicated recognition with a stoichiometric ratio of 1 vs. 1. The addition of cysteine increased the emission intensity of the chemosensors. As a consequence, the luminescence off-on effect with sensing selectivity and linear sensing behavior for Cys was demonstrated. Sulfur modification on our chemosensors was shown to be effective in improving their selectivity and photostability.

Recently, frequent data movement between computing units and memory during floating-point arithmetic has become a major problem for scientific computing. Computing-in-memory (CIM) is a novel computing paradigm that merges computing logic into memory, which can address the data movement problem with excellent power efficiency. However, the previous CIM paradigm failed to support double-precision floating-point format (FP64) due to its computing complexity. This paper presents a novel all-digital CIM macro-DCIM-FF to complete FP64 based fused multiply-add (FMA) operation for the first time. With 16 sub-CIM cells integrating digital multipliers to complete mantissa multiplication, DCIM-FF is able to provide correct rounded implementations for normalized/denormalized inputs in round-to-nearest-even mode and round-to-zero mode, respectively. To evaluate our design, we synthesized and tested the DCIM-FF macro in 55-nm CMOS technology. With a minimum power efficiency of 0.12 mW and a maximum computing efficiency of 26.9 TOPS/W, we successfully demonstrated that DCIM-FF can run the FP64-based FMA operation without error. Compared to related works, the proposed DCIM-FF macro shows significant power efficiency improvement and less area overhead based on CIM technology. This work paves a novel pathway for high-performance implementation of an FP64-based matrix-vector multiplication (MVM) operation, which is essential for hyperscale scientific computing.

In this article, we reported the synthesis and characterization of gold nanoclusters (AuNCs) with a diameter of ∼2 nm. A simple method of microwave-assisted reaction was applied here, with L-cysteine as both reducing agent and stabilizer. The resulting AuNCs were analyzed by means of TEM, XPS, DLS, and IR. Their photophysical performance was then analyzed in detail, including UV-vis absorption, emission, quantum yield, and lifetime. Efficient red emission was observed from these AuNCs, originating from ligand-to-metal nanoparticle core charge transfer (LMNCT). This red emission was found quenchable by Fe(III) cations. The corresponding quenching curve and sensing performance were discussed. An effective working region of 0–80 μM with an LOD of 3.9 μM was finally observed. Their quenching mechanism was revealed as Fe(III) energy competing for the LMNCT process. The novelty and advancement of this work is the simple synthesis and impressive sensing performance, including wide working region, good linearity, and selectivity.

At present, static text passwords are still the most widely-used identity authentication method. Password-generation technology can generate large-scale password sets and then detect the defects in password-protection mechanisms, which is of great significance for evaluating password-guessing algorithms. However, the existing password-generation technology cannot ignore low-quality passwords in the generated password set, which will lead to low-efficiency password guessing. In this paper, a password-generation model based on an ordered Markov enumerator and critic discriminant network (OMECDN) is proposed, where passwords are generated via an ordered Markov enumerator (OMEN) and a discriminant network according to the probability of the combination of passwords. OMECDN optimizes the performance of password generation with a discriminative network based on the good statistical properties of OMEN. Moreover, the final password set is formed by the selected passwords with a higher score than the preset threshold, which guarantees the superiority of the hit rate of almost all ranges of combinations of passwords over the initial password set. Finally, the experiments show that OMECDN achieves a qualitative improvement in hit rate metrics. In particular, regarding the generation of 107 passwords on the RockYou dataset, the matching entries of the password set generated by the OMECDN model are 25.18% and 243.58% higher than those generated by the OMEN model and the PassGAN model, respectively.

Large bone defects caused by tumors and traumas are difficult clinical problems. For its treatment, autogenous bone transplantation is the “gold standard”. However, there are some shortcomings of this treatment, such as limited supply and extra trauma from bone removal. In recent years, orthopedic implants and bone tissue engineering materials have developed rapidly, adding impetus to the solution of this problem. At present, the main orthopedic implants include medical metal materials, medical polymer materials, bone cement, ceramic materials, artificial bone materials, etc. Among which medical polymer materials have become the research hotspot and achieved outstanding results in recent years, especially the in-depth study of polyether-ether-ketone (PEEK) materials has great promising to solve this medical problem. PEEK materials have the advantages of non-toxicity, high-temperature resistance, corrosion resistance, abrasion resistance, high strength, high toughness, X-ray radiolucency, and excellent sterilization performance. PEEK materials have been successfully applied in clinical practice and have achieved excellent clinical efficacy and wide recognition. This review overviews the research progress of the performance requirements, material development, and material surface modification of PEEK as an orthopedic implant, and discusses prospects for the advance of medical PEEK materials.

Electrochemical deposition (ED) dates back to Faraday and is a well-established approach for producing coatings. This review summarizes unconventional ED approaches, where nanostructured films are deposited, starting with nanomaterials as building blocks, instead of molecular or ionic species. This “nano to nano” deposition concept has a significant advantage of directly transferring nano-objects from the dispersion to the coating, which maintains their unique nanostructures. It is achieved via destabilizing the nano-objects in the dispersions by applying electrochemical potential or current, which diminishes the interparticle repulsion through two different mechanisms: (a) Direct redox induced deposition; (b) Indirect pH and ionic strength induced deposition. With electrochemical reactions as a driving force, the deposition process is selective to conductive and electrochemically active substrates and can be manipulated by potential/current and time, yielding films with thickness from nanometers to a few microns. Notably, electrochemically inactive and even non-conductive nanomaterials may also be electrochemically deposited via the “nano to nano” approach using an indirect mechanism. Challenges and prospects are also discussed.

Fenton reaction is a powerful technology for pollutants’ removal from water. However, the cost of H2O2 becomes one of the major stumbling blocks in its application. H2O2 has a relatively high price and is easily decomposed during transportation and use; therefore, in situ synthesis of H2O2 could improve economic benefits effectively. In this study, a Fe/Ni/Pd ternary metal-doped graphitic carbon nitride (FeNi-Pd@CN) is prepared, and in situ H2O2 generation using formic acid as hydrogen sources for organics removal was proved. The catalyst is advantageous, as H2O2 could accumulate to 1.69 mmol/L in 150 min when pumping air rather than oxygen gases in other studies. Furthermore, 92.0% of Acid Red 73 (200 mg/L) and 93.2% of tetracycline hydrochloride (10 mg/L) could be removed in 150 min without any pH adjustment. Characterization results show that the catalyst has good stability in metal leaching and reuse tests. It is proved that •OH and •O2− are the main reactive oxygen species, and a synergistic effect between Fe and Ni exists that enhances ROS generation for organics degradation. This work offers a promising method to remove refectory organic contaminants from industrial wastewater.

Asphalt mixture is a type of textured composite material made of aggregates and mastic part. Overall strength and failure behavior in such materials depends on the texture or heterogeneity of the mixture. In particular, the crack growth mechanism from the tip of the pre-crack is significantly affected by the texture of the asphalt composite and environmental conditions. The crack can extend through the soft mastic, tight aggregates or interface of the mastic/aggregates. In this research, by performing some fracture tests on a typical asphalt mixture with different test specimens under mode I, mixed mode I/II and mixed mode I/III, the fracture resistance and trajectory of propagating crack is studied at two low and medium temperatures (i.e., −15 and +15 °C). The load bearing capacity and the fracture resistance of the tested asphalt samples increases by decreasing the temperature. It is also shown that a significant part of fracture plane passes through the soft mastic and boundary of aggregates (i.e., the interface of aggregates and mastic) and only about 10–15% of the fracture surface of the propagating crack passes via the tight aggregates by breaking them. This percentage decreases for mode II and III loading conditions and higher testing temperatures. Compared to brittle and isotropic materials, the fracture path of the asphalt mixture shows more deviation, and this deviation increases for those mixtures containing coarser aggregates in the ligament and tested under medium temperature conditions.

Silver vanadates (SVOs) have been widely investigated as cathode materials for high-performance lithium-ion batteries (LIBs). However, similar to most vanadium-based materials, SVOs suffer from structural collapse/amorphization and vanadium dissolution from the electrode into the electrolyte during the Li insertion and extraction process, causing poor electrochemical performance in LIBs. We employ ultrathin Al2O3 coatings to modify β-AgVO3 (as a typical example of SVOs) by an atomic layer deposition (ALD) technique. The galvanostatic charge-discharge test reveals that ALD Al2O3 coatings with different thicknesses greatly affected the cycling performance. Especially, the β-AgVO3 electrode with ~10 nm Al2O3 coating (100 ALD cycles) exhibits a high specific capacity of 271 mAh g−1, and capacity retention is 31%, much higher than the uncoated one of 10% after 100 cycles. The Coulombic efficiency is improved from 89.8% for the pristine β-AgVO3 to 98.2% for Al2O3-coated one. Postcycling analysis by cyclic voltammetry (CV), cyclic voltammetry (EIS), and scanning electron microscopy (SEM) disclose that 10-nm Al2O3 coating greatly reduces cathode-electrolyte interphase (CEI) resistance and the charge transfer resistance in the β-AgVO3 electrode. Al2O3 coating by the ALD method is a promising technique to construct artificial CEI and stabilize the structure of SVOs, providing new insights for vanadium-based electrodes and their energy storage devices.

As a new type of advanced electrical power generation system, free-piston engines can be applied to many fields, such as electric vehicles and communication base stations. In this paper, a novel tubular moving magnet linear generator system is designed for free-piston engines. A three-dimensional finite element model of the designed linear generator is presented using a transient computing method with a motion solver. The electromagnetic force, electromotive force, and generating efficiency of the designed linear generator are analyzed in detail. A prototype is manufactured and tested under laboratory conditions, and then the main performance of the prototype is evaluated using the validated finite element model. The electromagnetic analysis in this paper indicates that a generating efficiency of 91.4% can be obtained by the moving magnet linear generator under the rated working frequency of 50 Hz. Compared with the existing moving coil linear generator prototype, the power volume density and the power mass density of the new designed moving magnet linear generator prototype are increased by 150% and 35%.

Camless valvetrains have become a promising direction to improve the performance of internal combustion engines. In this paper, an electric load simulator is proposed to simulate and implement gas force on the exhaust valve for camless valvetrains under semi-physical conditions. According to test data, the 1D gas-dynamic model was established to get boundary conditions and initial values for 3D finite element simulation. The 3D finite element simulation model was solved to obtain the gas force characteristics of the exhaust valve for camless valvetrains. The electromagnetic actuator was designed according to the system scheme and performance requirements of the electric load simulator. The PID (Proportion Integration Differentiation) algorithm was designed to control the output force of the electric load simulator and reproduce the gas force characteristics of the exhaust valve. It was found that the output force of the electric load simulator could follow the variation of the target gas force and meet the performance requirements of the electric load simulator based on simulation results and experimental results.

The camless electromagnetic valve train (EMVT), as a fully flexible variable valve train, has enormous potential for improving engine performances. In this paper, a new valve strategy based on the electromagnetic intake valve train (EMIV) is proposed to achieve variable cylinder deactivation (VCD) on a four-cylinder gasoline engine. The 1D engine model was constructed in GT-Power according to test data. In order to analyze the VCD operation with the proposed valve strategy, the 1D model was validated using a 3D code. The effects of the proposed valve strategy were investigated from the perspective of energy loss of the transition period, the mass fraction of oxygen in the exhaust pipe, and the minimum in-cylinder pressure of the active cycle. On the premise of avoiding high exhaust oxygen and oil suction, the intake valve timing can be determined with the variation features of energy losses. It was found that at 1200 and 1600 rpm, fuel economy was improved by 12.5⁻16.6% and 9.7⁻14.6%, respectively, under VCD in conjunction with the early intake valve closing (EIVC) strategy when the brake mean effective pressure (BMEP) ranged from 0.3 MPa to 0.2 MPa.

A quantitative structure-property relationship (QSPR) model is proposed to explore the relationship between the pKa of various compounds and their structures. Through QSPR studies, the relationship between the structure and properties can be obtained. In this study, a novel chaos-enhanced accelerated particle swarm algorithm (CAPSO) is adopted to screen molecular descriptors and optimize the weights of back propagation artificial neural network (BP ANN). Then, the QSPR model based on CAPSO and BP ANN is proposed and named the CAPSO BP ANN model. The prediction experiment showed that the CAPSO algorithm was a reliable method for screening molecular descriptors. The five molecular descriptors obtained by the CAPSO algorithm could well characterize the molecular structure of each compound in pKa prediction. The experimental results also showed that the CAPSO BP ANN model exhibited good performance in predicting the pKa values of various compounds. The absolute mean relative error, root mean square error, and square correlation coefficient are respectively 0.5364, 0.0632, and 0.9438, indicating the high prediction accuracy. The proposed hybrid intelligent model can be applied in engineering design and the prediction of physical and chemical properties.