BIO-INSPIRED SELF-ADAPTIVE MANUFACTURING SYSTEM CONTROL ARCHITECTURE

Future manufacturing systems need to cope with frequent changes and disturbances, therefore their control architectures require constant adaptability, agility, stability, self-organization, intelligence, and robustness. Bio-inspired manufacturing system can well satisfy these requirements. For this purpose, by referencing the biological organization structure and the mechanism, a bio-inspired manufacturing cell is presented from a novel view, and then a bio-inspired self-adaptive manufacturing model is established based on the ultra-short feedback mechanism of the neuro-endocrine system. A hio-inspired self-adaptive manufacturing system coordinated model is also established based on the neuro-endocrine-immunity system （NEIS）. Finally, an example based on pheromone communication mechanism indicates that the robustness of the whole manufacturing system is improved by bio-inspired technologies.

WPA1 encodes a vWA domain protein that regulates wheat plant architecture

Plant height,spike,leaf,stem and grain morphologies are key components of plant architecture and related to wheat yield.A wheat(Triticum aestivum L.)mutant,wpa1,displaying temperaturedependent pleiotropic developmental anomalies,was isolated.The WPA1 gene,encoding a von Willebrand factor type A(vWA)domain protein,was located on chromosome arm 7DS and isolated by map-based cloning.The functionality of WPA1 was validated by multiple independent EMS-induced mutants and gene editing.Phylogenetic analysis revealed that WPA1 is monocotyledon-specific in higher plants.The identification of WPA1 provides opportunity to study the temperature regulated wheat development and grain yield.

Evolutionary Neural Architecture Search and Its Applications in Healthcare

Most of the neural network architectures are based on human experience,which requires a long and tedious trial-and-error process.Neural architecture search(NAS)attempts to detect effective architectures without human intervention.Evolutionary algorithms(EAs)for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures.Using multiobjective EAs for NAS,optimal neural architectures that meet various performance criteria can be explored and discovered efficiently.Furthermore,hardware-accelerated NAS methods can improve the efficiency of the NAS.While existing reviews have mainly focused on different strategies to complete NAS,a few studies have explored the use of EAs for NAS.In this paper,we summarize and explore the use of EAs for NAS,as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods.NAS performs well in healthcare applications,such as medical image analysis,classification of disease diagnosis,and health monitoring.EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task.Deep neural network has been successfully used in healthcare,but it lacks interpretability.Medical data is highly sensitive,and privacy leaks are frequently reported in the healthcare industry.To solve these problems,in healthcare,we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection.Moreover,we also point out future research directions for evolutionary NAS.Overall,for researchers who want to use EAs to optimize NNs in healthcare,we analyze the advantages and disadvantages of doing so to provide detailed guidance,and propose an interpretable privacy-preserving framework for healthcare applications.

Exploration of Graduate Student Cultivation Mode of Landscape Architecture under the Background of“Artificial Intelligence+X”

Under the background of“artificial intelligence+X”,the development of landscape architecture industry ushers in new opportunities,and professional talents need to be updated to meet the social demand.This paper analyzes the cultivation demand of landscape architecture graduate students in the context of the new era,and identifies the problems by comparing the original professional graduate training mode.The new cultivation mode of graduate students in landscape architecture is proposed,including updating the target orientation of the discipline,optimizing the teaching system,building a“dualteacher”tutor team,and improving the“industry-university-research-utilization”integrated cultivation,so as to cultivate high-quality compound talents with disciplinary characteristics.

Intelligent architecture modeling for multilevel fluvial reservoirs based on architecture interface and elements

At present,the architecture modeling method of fluvial reservoirs are still developing.Traditional methods usually use grids to characterize architecture interbeds within the reservoir.Due to the thin thickness of this type of the interlayers,the number of the model grids must be greatly expanded.The number of grids in the tens of millions often makes an expensive computation;however,upscaling the model will generate a misleading model.The above confusion is the major reason that restricts the largescale industrialization of fluvial reservoir architecture models in oilfield development and production.Therefore,this paper explores an intelligent architecture modeling method for multilevel fluvial reservoirs based on architecture interface and element.Based on the superpositional relationship of different architectural elements within the fluvial reservoir,this method uses a combination of multilevel interface constraints and non-uniform grid techniques to build a high-resolution 3D geological model for reservoir architecture.Through the grid upscaling technology of heterogeneous architecture elements,different upscaling densities are given to the lateral-accretion bedding and lateral-accretion bodies to simplify the model gridding.This new method greatly reduces the number of model grids while ensuring the accuracy of lateral-accretion bedding models,laying a foundation for large-scale numerical simulation of the subsequent industrialization of the architecture model.This method has been validated in A layer of X oilfield with meandering fluvial channel sands as reservoirs and B layer of Y oilfield with braided river sands as reservoirs.The simulation results show that it has a higher accuracy of production history matching and remaining oil distribution forecast of the targeted sand body.The numerical simulation results show that in the actual development process of oilfield,the injected water will not displace oil in a uniform diffusive manner as traditionally assumed,but in a more complex pattern with oil in upper part of sand body being left behind as residual oil due to the influences of different levels of architecture interfaces.This investigation is important to guiding reservoir evaluation,remaining oil analysis,profile control and potential tapping and well pattern adjustment.

Long-Term Impacts of Tree Architectures and Branch Configurations on Tree Growth, Yield, Fruit Quality Attributes, and Leaf Minerals in “Aztec Fuji” Apple

Canopy and branch architectures in high-density orchards can be crucial in production and fruit quality. The influence of two canopy orientations (Upright and Tilted) in combination with two arm (branch) architectures (Shortened or Overlapped) on tree growth, yield components, fruit quality, and leaf mineral nutrients in an “Aztec Fuji” apple (Malus domestica Bork.) high-density orchard was studied over five years. Tilted trees with shortened arm configuration (TilShArm) always had significantly larger trunk cross-sectional area (TCSA) than Upright trees with an Overlapped arm configuration (UpOverArm) every year from 2012 to 2016. Trees with a TilShArm system had more cumulative fruit per tree than those with an Upright orientation. Trees with a tilted canopy (TilShArm and TilOverArm) tended to have higher yield per tree and yield per hectare than those with an upright system. Trees with a TilShArm system were more precocious and had more yield per tree than those with an upright canopy orientation in 2012. When values were polled over five years, trees with an upright canopy-shortened arm system (UpShArm) treatment had a lower biennial bearing index (BBI) than those with an upright canopy-overlapped system (UpOverArm). Trees receiving an arm shortening (UpShArm or TilShArm) configuration often had larger fruits than those with overlapped arms (UpOverArm and TilOverArm). Fruit from trees receiving an UpOverArm had higher fruit firmness than those from trees with other canopy-branch arrangements at harvest due to their smaller size. Fruit from trees with a TilShArm and TilOverArm had significantly higher water core and bitter pit but lower sunburn than trees with an upright canopy (UpShArm and UpOverArm). Leaves from trees with an UpOverArm canopy-branch configuration had the lowest leaf Ca but the highest leaf K and Fe concentrations among all treatments.

Air-Ground Collaborative Mobile Edge Computing:Architecture,Challenges,and Opportunities

By pushing computation,cache,and network control to the edge,mobile edge computing(MEC)is expected to play a leading role in fifth generation(5G)and future sixth generation(6G).Nevertheless,facing ubiquitous fast-growing computational demands,it is impossible for a single MEC paradigm to effectively support high-quality intelligent services at end user equipments(UEs).To address this issue,we propose an air-ground collaborative MEC(AGCMEC)architecture in this article.The proposed AGCMEC integrates all potentially available MEC servers within air and ground in the envisioned 6G,by a variety of collaborative ways to provide computation services at their best for UEs.Firstly,we introduce the AGC-MEC architecture and elaborate three typical use cases.Then,we discuss four main challenges in the AGC-MEC as well as their potential solutions.Next,we conduct a case study of collaborative service placement for AGC-MEC to validate the effectiveness of the proposed collaborative service placement strategy.Finally,we highlight several potential research directions of the AGC-MEC.

Timing effect of high temperature exposure on the plasticity of internode and plant architecture in maize

The occurrence of high temperature(HT)in crop production is becoming more frequent and unpredictable with global warming,severely threatening food security.The state of an organ’s growth and development is largely determined by the temperature conditions it is exposed to over time.Maize is the main cereal crop,and its stem growth and plant architecture are closely related to lodging resistance,and especially sensitive to temperature.However,systematic research on the timing effect of HT on the sequentially developing internode and stem is currently lacking.To identify the timing effect of HT on the morphology and plasticity of the stem in maize,two hybrids(Zhengdan 958(ZD958),Xianyu 335(XY335))characterized by distinct morphological traits in the stem were exposed to a 7-day HT treatment from the V6 to V17 stages(Vn presents the vegetative stage with n leaves fully expanded)in 2019-2020.The results demonstrated that exposure to HT during V6-V12 accelerated the rapid elongation of stems.For instance,HT occurring at V7 and V12 specifically promoted the lengths and weights of the 3rd-5th and 9th-11th internodes,respectively.Meanwhile,HT slowed the growth of internodes adjacent to the promoted internodes.Interestingly,compared with control,the plant height was significantly increased soon after HT treatment,but the promotion effect became narrower at the subsequent flowering stage,demonstrating a self-adjusting mechanism in the maize plant in response to HT.Importantly,HT altered the plant architectures,including a rising of the ear position and increase in the ear position coefficient.XY335 exhibited greater sensitivity in stem development than ZD958 under HT treatment.These findings improve our systematic understanding of the plasticity of internode and plant architecture in response to the timing of HT exposure.

Auxin-brassinosteroid crosstalk:Regulating rice plant architecture and grain shape

Rice(Oryza sativa)plant architecture and grain shape,which determine grain quality and yield,are modulatedby auxin and brassinosteroid via regulation of cell elongation and proliferation.We review the signaltransduction of these hormones and the crosstalk between their signals on the regulation of rice plantarchitecture and grain shape.

A 48-bp deletion upstream of LIGULELESS 1 alters rice panicle architecture

Panicle architecture is an agronomic determinant of crop yield and a target for cereal crop improvement.To investigate its molecular mechanisms in rice,we performed map-based cloning and characterization of OPEN PANICLE 1(OP1),a gain-of-function allele of LIGULELESS 1(LG1),controlling the spread-panicle phenotype.This allele results from a 48-bp deletion in the LG1 upstream region and promotes pulvinus development at the base of the primary branch.Increased OP1 expression and altered panicle phenotype in chimeric transgenic plants and upstream-region knockout mutants indicated that the deletion regulates spread-panicle architecture in the mutant spread panicle 1(sp1).Knocking out BRASSINOSTEROID UPREGULATED1(BU1)gene in the background of OP1 complementary plants resulted in compact panicles,suggesting OP1 may regulate inflorescence architecture via the brassinosteroid signaling pathway.We regard that manipulating the upstream regulatory region of OP1 or genes involved in BR signal pathway could be an efficient way to improve rice inflorescence architecture.

Apical meristem transcriptome analysis identifies a role for the blue light receptor gene GhFKF1 in cotton architecture development

Cotton architecture is determined by the differentiation fate transition of axillary meristem(AM),and influences cotton yield and the efficiency of mechanized harvesting.We observed that the initiation of flowering primordium was earlier in early-maturing than that in late-maturing cultivars during the differentiation and development of AM.The RNA-Seq and expression level analyses showed that genes FLAVIN BINDING,KELCH REPEAT,F-BOX1(GhFKF1),and GIGANTEA(GhGI)were in response to circadian rhythms,and involved in the regulation of cotton flowering.The gene structure,predicted protein structure,and motif content analyses showed that in Arabidopsis,cotton,rapseed,and soybean,proteins GhFKF1 and GhGI were functionally conserved and share evolutionary origins.Compared to the wild type,in GhFKF1 mutants that were created by the CRISPR/Cas9 system,the initiation of branch primordium was inhibited.Conversely,the knocking out of GhGI increased the number of AM differentiating into flower primordium,and there were much more lateral branch differentiation and development.Besides,we investigated that proteins GhFKF1 and GhGI can interact with each other.These results suggest that GhFKF1 and GhGI are key regulators of cotton architecture development,and may collaborate to regulate the differentiation fate transition of AM,ultimately influencing plant architecture.We describe a strategy for using the CRISPR/Cas9 system to increase cotton adaptation and productivity by optimizing plant architecture.

Enhancing MXene-based supercapacitors:Role of synthesis and 3D architectures

MXene has been the limelight for studies on electrode active materials,aiming at developing supercapacitors with boosted energy density to meet the emerging influx of wearable and portable electronic devices.Despite its various desirable properties including intrinsic flexibility,high specific surface area,excellent metallic conductivity and unique abundance of surface functionalities,its full potential for electrochemical performance is hindered by the notorious restacking phenomenon of MXene nanosheets.Ascribed to its two-dimensional(2D)nature and surface functional groups,inevitable Van der Waals interactions drive the agglomeration of nanosheets,ultimately reducing the exposure of electrochemically active sites to the electrolyte,as well as severely lengthening electrolyte ion transport pathways.As a result,energy and power density deteriorate,limiting the application versatility of MXene-based supercapacitors.Constructing 3D architectures using 2D nanosheets presents as a straightforward yet ingenious approach to mitigate the fatal flaws of MXene.However,the sheer number of distinct methodologies reported,thus far,calls for a systematic review that unravels the rationale behind such 3D MXene structural designs.Herein,this review aims to serve this purpose while also scrutinizing the structure–property relationship to correlate such structural modifications to their ensuing electrochemical performance enhancements.Besides,the physicochemical properties of MXene play fundamental roles in determining the effective charge storage capabilities of 3D MXene-based electrodes.This largely depends on different MXene synthesis techniques and synthesis condition variations,hence,elucidated in this review as well.Lastly,the challenges and perspectives for achieving viable commercialization of MXene-based supercapacitor electrodes are highlighted.

Sedimentary architecture of submarine channel-lobe systems under different seafloor topography:Insights from the Rovuma Basin offshore East Africa

Seafloor topography plays an important role in the evolution of submarine lobes.However,it is still not so clear how the shape of slope affects the three-dimensional(3-D)architecture of submarine lobes.In this study,we analyze the effect of topography factors on different hierarchical lobe architectures that formed during Pliocene to Quaternary in the Rovuma Basin offshore East Africa.We characterize the shape,size and growth pattern of different hierarchical lobe architectures using 3-D seismic data.We find that the relief of the topographic slope determines the location of preferential deposition of lobe complexes and single lobes.When the topography is irregular and presents topographic lows,lobe complexes first infill these depressions.Single lobes are deposited preferentially at positions with higher longitudinal(i.e.across-slope)slope gradients.As the longitudinal slope becomes higher,the aspect ratio of the single lobes increases.Lateral(i.e.along-slope)topography does not seem to have a strong influence on the shape of single lobe,but it seems to affect the overlap of single lobes.When the lateral slope gradient is relatively high,the single lobes tend to have a larger overlap surface.Furthermore,as the average of lateral slope and longitudinal slope gets greater,the width/thickness ratio of the single lobe is smaller,i.e.sediments tend to accumulate vertically.The results demonstrate that the shape of slopes more comprehensively influences the 3-D architecture of lobes in natural deep-sea systems than previously other lobe deposits and analogue experiments,which helps us better understand the development and evolution of the distal parts of turbidite systems.

A family of quantum von Neumann architecture

We develop universal quantum computing models that form a family of quantum von Neumann architectures,with modular units of memory,control,CPU,and internet,besides input and output.This family contains three generations characterized by dynamical quantum resource theory,and it also circumvents no-go theorems on quantum programming and control.Besides universality,such a family satisfies other desirable engineering requirements on system and algorithm design,such as modularity and programmability,hence serves as a unique approach to building universal quantum computers.

Dynamic Economic Scheduling with Self-Adaptive Uncertainty in Distribution Network Based on Deep Reinforcement Learning

Traditional optimal scheduling methods are limited to accurate physical models and parameter settings, which aredifficult to adapt to the uncertainty of source and load, and there are problems such as the inability to make dynamicdecisions continuously. This paper proposed a dynamic economic scheduling method for distribution networksbased on deep reinforcement learning. Firstly, the economic scheduling model of the new energy distributionnetwork is established considering the action characteristics of micro-gas turbines, and the dynamic schedulingmodel based on deep reinforcement learning is constructed for the new energy distribution network system with ahigh proportion of new energy, and the Markov decision process of the model is defined. Secondly, Second, for thechanging characteristics of source-load uncertainty, agents are trained interactively with the distributed networkin a data-driven manner. Then, through the proximal policy optimization algorithm, agents adaptively learn thescheduling strategy and realize the dynamic scheduling decision of the new energy distribution network system.Finally, the feasibility and superiority of the proposed method are verified by an improved IEEE 33-node simulationsystem.

Enhanced Differentiable Architecture Search Based on Asymptotic Regularization

In differentiable search architecture search methods,a more efficient search space design can significantly improve the performance of the searched architecture,thus requiring people to carefully define the search space with different complexity according to various operations.Meanwhile rationalizing the search strategies to explore the well-defined search space will further improve the speed and efficiency of architecture search.With this in mind,we propose a faster and more efficient differentiable architecture search method,AllegroNAS.Firstly,we introduce a more efficient search space enriched by the introduction of two redefined convolution modules.Secondly,we utilize a more efficient architectural parameter regularization method,mitigating the overfitting problem during the search process and reducing the error brought about by gradient approximation.Meanwhile,we introduce a natural exponential cosine annealing method to make the learning rate of the neural network training process more suitable for the search procedure.Moreover,group convolution and data augmentation are employed to reduce the computational cost.Finally,through extensive experiments on several public datasets,we demonstrate that our method can more swiftly search for better-performing neural network architectures in a more efficient search space,thus validating the effectiveness of our approach.

Application of Artificial Intelligence Technology in Landscape Architecture Industry

With wide application prospects in landscape industry,artificial intelligence technology plays an important role in improving work efficiency,optimizing design,strengthening construction management,and achieving intelligent maintenance.With the continuous development of technology,the application of artificial intelligence in landscape architecture industry will become more in-depth and extensive,which can provid powerful support for the innovation and development of the industry.It is hoped that the modernization process of the landscape industry can be promoted through the analysis on the application and difficulties of artificial intelligence technology in the landscape industry.

Audio-Text Multimodal Speech Recognition via Dual-Tower Architecture for Mandarin Air Traffic Control Communications

In air traffic control communications (ATCC), misunderstandings between pilots and controllers could result in fatal aviation accidents. Fortunately, advanced automatic speech recognition technology has emerged as a promising means of preventing miscommunications and enhancing aviation safety. However, most existing speech recognition methods merely incorporate external language models on the decoder side, leading to insufficient semantic alignment between speech and text modalities during the encoding phase. Furthermore, it is challenging to model acoustic context dependencies over long distances due to the longer speech sequences than text, especially for the extended ATCC data. To address these issues, we propose a speech-text multimodal dual-tower architecture for speech recognition. It employs cross-modal interactions to achieve close semantic alignment during the encoding stage and strengthen its capabilities in modeling auditory long-distance context dependencies. In addition, a two-stage training strategy is elaborately devised to derive semantics-aware acoustic representations effectively. The first stage focuses on pre-training the speech-text multimodal encoding module to enhance inter-modal semantic alignment and aural long-distance context dependencies. The second stage fine-tunes the entire network to bridge the input modality variation gap between the training and inference phases and boost generalization performance. Extensive experiments demonstrate the effectiveness of the proposed speech-text multimodal speech recognition method on the ATCC and AISHELL-1 datasets. It reduces the character error rate to 6.54% and 8.73%, respectively, and exhibits substantial performance gains of 28.76% and 23.82% compared with the best baseline model. The case studies indicate that the obtained semantics-aware acoustic representations aid in accurately recognizing terms with similar pronunciations but distinctive semantics. The research provides a novel modeling paradigm for semantics-aware speech recognition in air traffic control communications, which could contribute to the advancement of intelligent and efficient aviation safety management.

Multilevel carbon architecture of subnanoscopic silicon for fast‐charging high‐energy‐density lithium‐ion batteries

Silicon(Si)is widely used as a lithium‐ion‐battery anode owing to its high capacity and abundant crustal reserves.However,large volume change upon cycling and poor conductivity of Si cause rapid capacity decay and poor fast‐charging capability limiting its commercial applications.Here,we propose a multilevel carbon architecture with vertical graphene sheets(VGSs)grown on surfaces of subnanoscopically and homogeneously dispersed Si–C composite nanospheres,which are subsequently embedded into a carbon matrix(C/VGSs@Si–C).Subnanoscopic C in the Si–C nanospheres,VGSs,and carbon matrix form a three‐dimensional conductive and robust network,which significantly improves the conductivity and suppresses the volume expansion of Si,thereby boosting charge transport and improving electrode stability.The VGSs with vast exposed edges considerably increase the contact area with the carbon matrix and supply directional transport channels through the entire material,which boosts charge transport.The carbon matrix encapsulates VGSs@Si–C to decrease the specific surface area and increase tap density,thus yielding high first Coulombic efficiency and electrode compaction density.Consequently,C/VGSs@Si–C delivers excellent Li‐ion storage performances under industrial electrode conditions.In particular,the full cells show high energy densities of 603.5 Wh kg^(−1)and 1685.5 Wh L^(−1)at 0.1 C and maintain 80.7%of the energy density at 3 C.

An Interpretable Light Attention-Convolution-Gate Recurrent Unit Architecture for the Highly Accurate Modeling of Actual Chemical Dynamic Processes

To equip data-driven dynamic chemical process models with strong interpretability,we develop a light attention–convolution–gate recurrent unit(LACG)architecture with three sub-modules—a basic module,a brand-new light attention module,and a residue module—that are specially designed to learn the general dynamic behavior,transient disturbances,and other input factors of chemical processes,respectively.Combined with a hyperparameter optimization framework,Optuna,the effectiveness of the proposed LACG is tested by distributed control system data-driven modeling experiments on the discharge flowrate of an actual deethanization process.The LACG model provides significant advantages in prediction accuracy and model generalization compared with other models,including the feedforward neural network,convolution neural network,long short-term memory(LSTM),and attention-LSTM.Moreover,compared with the simulation results of a deethanization model built using Aspen Plus Dynamics V12.1,the LACG parameters are demonstrated to be interpretable,and more details on the variable interactions can be observed from the model parameters in comparison with the traditional interpretable model attention-LSTM.This contribution enriches interpretable machine learning knowledge and provides a reliable method with high accuracy for actual chemical process modeling,paving a route to intelligent manufacturing.