Search

Injection molding of thin-walled plastic parts with minimum deformation in warpage and volume shrinkage is crucial for part quality. Simulation combined Latin hypercube sampling approach was used to research the effects of different process parameters on deformation. Then, random forest regression (RFR) is used to construct the mathematical relationship between process parameters and defects, such as warpage and volume shrinkage. The gaussian process is used as probabilistic surrogate model, while the probability of improvement is used as acquisition function to construct a Bayesian optimization for RFR’s hyperparameters, and the performance of random search is compared. In addition, the gradient boosting regression (GBR) and support vector regression (SVR) were also adopted to establish the prediction models, respectively. Comparing all the above prediction models, it can be found that the Bayesian optimized random forest regression (BO-RFR) has the highest accuracy. The Non-dominated Sorting Genetic Algorithm-II (NSGA-II) is interfaced with the predictive models to find the optimum design parameters for the purpose of effectively predicting and controlling warpage and volume shrinkage. The results show that warpage is reduced by 66.03% while volume shrinkage is 46.20% after optimizing. The final finite element simulation and physical tests indicate that this proposed method can effectively achieve the multi-objective optimization of injection molding.

At present, China's economy keeps developing rapidly. It has begun to vigorously carry out the construction of ecological civilization. As a new format of forest and grass industry, it is necessary to develop forest-based health care. The author puts forward strategies for developing forest-based health care industry on the basis of combing the development process of it.

ECG recorders are precision medical devices, but their thin-walled shells are susceptible to warpage and shrinkage during injection molding production due to the injection molding process, which greatly shortens their service life. To address this problem, a multiobjective optimization method for injection molding process parameters based on a combination of a BP neural network model optimized by an ant colony algorithm (ACO-BP) and an improved non-dominated sorting genetic algorithm (NSGA-II) is proposed. The study takes the warpage deformation amount and volume shrinkage rate of the plastic part as the optimization objectives, and the melt temperature, mold temperature, injection pressure, holding pressure, holding time, and cooling time as the design variables. However, for BP neural networks, it is crucial to choose an appropriate number of hidden layer neurons, so the particle swarm algorithm combined with the grey wolf algorithm (GWO-PSO) is used to solve for the optimal number of hidden layer neurons. Firstly, the number of hidden layer neurons of the BP network model was solved based on the samples obtained from the Box–Behnken experimental design and the GWO-PSO algorithm, and the ACO-BP algorithm was used to build the prediction models for warpage and volume shrinkage, respectively, and then combined with NSGA-II for global optimisation. The pareto optimal solution set was subjected to CRITIC analysis and the optimal process parameters were finally obtained, with a minimum warpage of 0.3293 mm and minimum volume shrinkage of 4.993%, a reduction of 8.93 and 6.95% respectively compared to the pre-optimisation period. At the same time, injection molding tests were carried out on the optimum process parameters, and it was found that the molding quality of the plastic parts was better and met the actual production requirements through measurement. The research in this paper provides a theoretical basis for further improving the quality defects of the thin-walled injection molded parts.

The physical properties of plastic products, such as local strength, wear resistance and electrical properties, can be improved by adding embedded parts in the appropriate position of the products, and the precision of plastic parts can also be improved. However, due to the addition of inserts, the flow and shrinkage around inserts will be affected. Compared with traditional injection molding products, the quality is difficult to predict. To solve this problem, the injection molded parts with inserts (electrostatic test box) was used as an example, according to the product structure, three objectives of volume shrinkage, warpage in the X direction, and warpage in the Z direction were optimized. A generalized regression neural network (GRNN) model was established with molding parameters as input and quality objectives as output. Improved fruit fly optimization algorithm (IFOA) was proposed to select the optimal smoothing parameters dynamically. Through the prediction of samples, the experimental results show that the model is superior to two comparative models. Non-dominated sorting genetic algorithm (NSGA-II) was used to solve the model, and the Pareto-optimal front was obtained. The entropy TOPSIS method was used to evaluate the Pareto-optimal front, and the optimal solution was obtained. The results show that IFOA-GRNN-NSGA is a reliable multi-objective optimization method.

Fiber-reinforced composites are widely used in injection molding processes because of their high strength and high elastic modulus. However, the addition of reinforcing agents such as glass fibers has a significant impact on their injection molding quality. The difference in shrinkage and hardness between the plastic and the reinforcement will bring about warpage and deformation in the injection molding of the product. At the same time, the glass fibers will be oriented in the flow direction during the injection molding process. This will enhance the mechanical properties in the flow direction and increase the shrinkage in the vertical direction, reducing the molding quality of the product. In this study, a test program was developed based on the Box-Behnken test design in the Design-Expert software, using a plastic part as an example. Moldflow software was used for simulation, and data analysis of the experimental data was carried out to investigate the significance of the influence of each injection molding process parameter on the molding quality. In addition to this, a mathematical model between the injection molding process parameters and the quality objectives was established by optimizing the model parameters of the back-propagation (BP) neural network through the ant colony optimization (ACO) algorithm. The established mathematical model is then globally optimized using a multi-objective function optimization based on the non-dominated rank-based sorting genetic algorithm (NSGA-II) to obtain the optimal combination of process parameters. The research in this article provides a theoretical basis for further combining intelligent algorithms to improve injection molding quality.

Owing to the influence of the injection molding process, warpage and volume shrinkage are two common quality defects for products manufactured by glass fiber-reinforced plastic (GFRP) injection molding. To minimize these two defects, an extreme learning machine optimized with a genetic algorithm (GA-ELM), multi-objective firefly algorithm (MOFA), and a multi-objective decision-making method (GRA-TOPSIS) were implemented in this study. All of the experiments, based on Latin hypercubic sampling (LHS), were conducted using Moldflow software to obtain the results for warpage and volume shrinkage. The prediction accuracy of the defect-prediction models based on the extreme learning machine (ELM) and GA-ELM algorithms were compared. The results show that the GA-ELM models can better predict the defect values. Finally, MOFA was used to find the Pareto optimal front, and the GRA-TOPSIS method was used to find the optimum solution from the Pareto optimal front. According to the results of the simulation verification, the warpage and volume shrinkage were effectively reduced by 12.25% and 6.11%, respectively, compared with before optimization, which indicates the effectiveness and reliability of the optimization method.

Mesoporous silica nanoparticles (MSNs) loaded with doxorubicin and then modified with hyaluronic acid (HA) and nanobody (DOX@MSNs-HA-11C12) is a pH- and redox-responsive drug delivery system, resulting in over 90% of LoVo cell apoptosis within 48 h.

Graph neural networks (GNNs) have many variants for graph representation learning. Several works introduce PageRank into GNNs to improve its neighborhood aggregation capabilities. However, these methods leverage the general PageRank to perform complex neighborhood aggregation to obtain the final feature representation, which leads to high computational cost and oversmoothing. In this paper, we propose simple hierarchical PageRank graph neural networks (SHP-GNNs), which first utilize the simple PageRank to aggregate different neighborhood ranges of each node, and then leverage a jumping architecture to combine these aggregated features to enable hierarchical structure-aware representation. In this case, first, the simple PageRank turns the neighborhood aggregation process to no-learning, thereby reducing the computational complexity of the model. Then, the jumping structure combines the aggregation features of each node's different hierarchy (neighborhood range) to learn more informative feature representation. Finally, the successful combination of the above methods alleviate the oversmoothing problem of deep GNNs. Our experimental evaluation demonstrates that SHP-GNNs achieve or match state-of-the-art results in node classification tasks, text classification tasks, and community prediction tasks. Moreover, since SHP-GNNs' neighborhood aggregation is a no-learning process, SHP-GNNs are successfully extended to node clustering tasks.

Recently, higher-order topological insulators have been attracting extensive interest. Unlike the conventional topological insulators that demand bulk gap closings at transition points, the higher-order band topology can be changed without bulk closure and exhibits as an obstruction of higher-dimensional boundary states. Here, we report the first experimental realization of three-dimensional surface-obstructed topological insulators with using acoustic crystals. Our acoustic measurements demonstrate unambiguously the emergence of one-dimensional topological hinge states in the middle of the bulk and surface band gaps, as a direct manifestation of the higher-order band topology. Together with comparative measurements for the trivial and phase-transition-point insulators, our experimental data conclusively evidence the unique bulk-boundary physics for the surface-obstructed band topology. That is, the topological phase transition is determined by the closure of the surface gap, rather than by closing the bulk gap. Our study might spur on new activities to deepen the understanding of such elusive topological phases.

Exploring new topological phases and phenomena has become a vital topic in condensed matter physics and materials sciences. Recent studies reveal that a braided colliding nodal pair can be stabilized in a multi-gap system with $$PT$$ P T or $${C}_{2z}T$$ C 2 z T symmetry. This exemplifies non-abelian topological charges beyond the scope of conventional single-gap abelian band topology. Here, we construct ideal acoustic metamaterials to realize non-abelian braiding with the fewest band nodes. By emulating the time with a sequence of acoustic samples, we experimentally observe an elegant but nontrivial nodal braiding process, including nodes creation, braiding, collision, and repulsion (i.e., impossible to annihilate), and measure the mirror eigenvalues to elucidate the braiding consequence. The latter, at the level of wavefunctions, is of prime importance since essentially braiding physics aims to entangle multi-band wavefunctions. Furthermore, we experimentally unveil the highly intricate correlation between the multi-gap edge responses and the bulk non-abelian charges. Our findings pave the way for developing non-abelian topological physics that is still in its infancy.