Artificial Intelligence (AI) techniques have made great advances in assisting antibody design. However, design still heavily relies on isolating antigen-specific antibodies from serum, which is a resource-intensive and time-consuming process. To address this issue, we propose Pre-trained Antibody generative large Language Model (PALM-H3) for the de novo generation of artificial heavy chain complementarity-determining region 3 (CDRH3) with desired antigen-binding specificity, reducing...

Graph learning is widely recognized as a crucial technique in multi-view clustering. Existing graph methods typically involve constructing an adaptive neighbor based on probabilistic neighbors and then consensus to for clustering, however, they are confronted with two limitations. Firstly, often rely Euclidean distance measure similarity when the graph, which proves inadequate capturing intrinsic structure among data points many real-world scenarios. Secondly, most of these focus solely...

Robotic insertion tasks often rely on delicate manual tuning due to the complexity of contact dynamics. In contrast, human is remarkably efficient in these tasks. this context, Programming by Demonstration (PbD) has gained much traction since it shows possibility for robots learn new skills observing demonstration. However, existing PbD approaches suffer from high cost demonstration data collection, and low robustness task uncertainties. order address issues, we propose a PbD-based learning...

Abstract The current ultrasonic milling tool holders are relatively long in structure, and the is susceptible to generating detrimental radial vibrations during high-speed rotation. To achieve high-quality of difficult-to-machine materials, a longitudinal-torsional oscillator suitable for BT40 has been proposed, which can generate large amplitude while maintaining compact length. A longitudinal vibration transducer with an exponential front cover designed using analytical method. Based on...

Abstract Robots are mainly used in mass production applications. Recent advances collaborative robots make it possible for humans and to collaborate on high-mix, low-volume This paper presents a framework human-robot collaboration high-mix Tasks that require delicate or complex operations assigned humans. straightforward robots. Simulation is utilized design the assembly cell by placing sensors cell. Computer vision localize parts cell, automated motion planning generate robot trajectories...

In connection with the multi-variable, strong coupling, nonlinear characteristics of permanent magnet synchronous motor(PMSM) control system, fuzzy method has been introduced to tune PMSM's parameters. Based on model reference adaptive system (MRAS), this paper present a new which combined both PI controller and MRAS model. Some simulations done motor speed system. The simulation results demonstrate feasibility algorithm.

In this paper, a secure voice communication system based on TMS320VC5410 is proposed, which includes the hardware design and software design. Among them, A/D D/A conversions of signals are implemented by MC145480, while data transmission performed TL16C554. A chaotic encryption algorithm combining cat map logistic in system. Experiments both encrypted non-encrypted over short distance have been performed, demonstrates feasibility proposed

Scheduling optimization of prefabricated components production workshop is a prerequisite for the orderly construction construction. Prefabricated are produced under hybrid flow shop mode, which conducive to improving output rate components. Combining multiple constraints actual situation production, established model solved by two-layer coding genetic algorithm. The empirical analysis based on survey data shows that applied and method can be used scheduling give reasonable scheme.

ABSTRACT Generating T-cell receptors (TCRs) with desired epitope-binding properties is a fundamental step in the development of immunotherapies, yet heavily relies on laborious and expensive wet experiments. Recent advancements generative artificial intelligence have demonstrated promising power protein design engineering. In this regard, we propose large language model, termed Epitope-Receptor-Transformer (ERTransformer), for de novo generation TCRs property. ERTransformer built EpitopeBERT...

Traffic flow prediction is a fundamental problem for urban traffic control and management. In practice, incomplete data common challenge due to sparse sensor deployment, loss, hardware failure. this paper, graph convolution recurrent neural network proposed prediction, with considerations of data. The missing complemented node imputation using the Gaussians Mixture Model (GMM) integrated into initial layer network. Then, we utilize parameter learning module capture features individual nodes,...

Abstract To address the issue of piston fracture due to insufficient strength in Down-The-Hole (DTH) hammer. The effect pressure state gas chamber on motion is considered. A model established based steady flow energy equation and working principle piston. finite element method used analyze dynamic response during impact drill bit. RBF neural network employed construct a surrogate for relationship between structural parameters (front-end diameter, front-end length, center bore) performance...

We consider a special type of fast reaction-diffusion systems in which the coefficients reaction terms two substances are much larger than those diffusion while diffusive motion to substrate is negligible. Specifically speaking, rate constants $O(1/\epsilon)$ $O(1)$ where parameter $\epsilon$ small. When become highly large, i.e. tends 0, singular limit behavior such system inscribed by Stefan problem with latent heat, brings great challenges numerical simulations. In this paper, we adopt...

A Kinematics and Compliance facility is a test rig which undertakes quasi-static testing of vehicles to measure the K&C characteristics suspension steering systems. Through simulating bounce roll vehicle body, making up force between tire road, measures alignment parameters, stiffness tire. When body fix, applies load wheels an axle or two with hydraulic actuators. The wheel deflection system fit six degree freedom. steer robot used accomplish angle torque. important tool for design many...

We describe a novel graph analysis that includes multigraph with multiple edges between pair of nodes to model brain functional networks, and introduce three-way correlation BOLD signals from gray matter volumes (nodes) one white bundle define connectivity. Characteristics inter-nodal communication may be derived this analysis. By examining selected databases, we show communications vary ages. integrating fMRI as third component in network analysis, more comprehensive descriptions function obtained.

Fins are extensively employed in heat sinks for high-density flow components. In this study, three distinct objective functions, namely temperature gradient minimization (TO-I type fin), average (TO-II and entransy dissipation (TO-III established topology optimization design of the sink fin structure to effectively minimize source. By conducting a comparative analysis configuration features thermal performance optimized fins with different it is evident that TO-Ⅰ exhibits superior...

Fins are extensively employed in heat sinks for high-density flow components. In this study, three distinct objective functions, namely temperature gradient minimization (TO-I type fin), average (TO-II and entransy dissipation (TO-III established topology optimization design of the sink fin structure to effectively minimize source. By conducting a comparative analysis configuration features thermal performance optimized fins with different it is evident that TO-Ⅰ exhibits superior...

Abstract Sheet-like materials such as composite prepreg fabrics are widely applicable in a variety of industries, especially the automotive and aerospace industries. During manufacturing, several these applications require sheets to be transported from one location another. The use robotic arms suction grippers provides reliable flexible method for transportation operations. Since they compliant, each sheet varying size shape requires unique robot or end effector ensure proper layup. This...

Immune repertoire classification, a typical multiple instance learning (MIL) problem, is frontier research topic in computational biology that makes transformative contributions to new vaccines and immune therapies. However, the traditional instance-space MIL, directly assigning bag-level labels instances, suffers from massive amount of noisy extremely low witness rate. In this work, we propose noisy-label-learning formulation solve classification task. To remedy inaccurate supervision...

A six-degree-of-freedom ultrasonic C-scan imaging inspection system using the water jet method was developed. The motion mechanism mainly relies on six degrees of freedom robotic arm, and focusing transducer is loaded to realize profiling complex curved surfaces. uses amplitude post-process data real-time imaging. experimentally verified be stable reliable, with high detection accuracy. This realizes surface automated provides a new way for detection.