Batch Process Research Articles

Inverse design of V-shape feed spacer for batch and semi-batch reverse osmosis

In recent years, batch and semi-batch reverse osmosis (RO) processes have received wide attention because of their internal staging feature. However their performance is severely compromised by salt retention induced by incomplete flushing, which leads to elevated salt concentration, uplifted peak pressure, and extra power consumption. Herein we adopt transient three-dimensional computational fluid dynamics to systematically analyze the effect of the V-shape feed spacer geometries on hydrodynamics and mass transfer characteristics. After a sensitivity analysis of all geometric parameters, a heuristic optimization framework is proposed for inverse design of the feed spacer. Using the optimized feed spacer, flushing efficacies after one displacement volume are 0.936, 0.962, 0.969 and 0.974 respectively for one, three, five and seven membrane elements connected in series (whereas 0.884, 0.926, 0.942 and 0.952 for commercial spacers), which help to mitigate membrane scaling and fouling. Moreover, the optimized spacer has a pressure drop per meter 73 % lower than that of the commercial spacer. If used in conjunction with high permeability membranes, batch RO and semi-batch RO could save 29.9 % and 24.1 % specific energy consumption relative to conventional one-stage seawater RO.

Out-of-order execution enabled deep reinforcement learning for dynamic additive manufacturing scheduling

Additive Manufacturing (AM) has revolutionized the production landscape by enabling on-demand customized manufacturing. However, the efficient management of dynamic AM orders poses significant challenges for production planning and scheduling. This paper addresses the dynamic scheduling problem considering batch processing, random order arrival and machine eligibility constraints, aiming to minimize total tardiness in a parallel non-identical AM machine environment. To tackle this problem, we propose the out-of-order enabled dueling deep Q network (O3-DDQN) approach. In the proposed approach, the problem is formulated as a Markov decision process (MDP). Three-dimensional features, encompassing dynamic orders, AM machines, and delays, are extracted using a ‘look around’ method to represent the production status at a rescheduling point. Additionally, five novel composite scheduling rules based on the out-of-order principle are introduced for selection when an AM machine completes processing or a new order arrives. Moreover, we design a reward function that is strongly correlated with the objective to evaluate the agent’s chosen action. Experimental results demonstrate the superiority of the O3-DDQN approach over single scheduling rules, randomly selected rules, and the classic DQN method. The average improvement rate of performance reaches 13.09% compared to composite scheduling rules and random rules. Additionally, the O3-DDQN outperforms the classic DQN agent with a 6.54% improvement rate. The O3-DDQN algorithm improves scheduling in dynamic AM environments, enhancing productivity and on-time delivery. This research contributes to advancing AM production and offers insights into efficient resource allocation.

The investigation of traffic flow prediction and optimization based on Spark

Spark and flink have made great strides in big data processing in recent years. Although the current spark can process a large amount of data at the same time, it still has a lot of shortcomings in the transparency, credibility, and practicality of the model. This paper provides a comprehensive overview of how to tackle the performance bottlenecks, insufficient model interpretability, and lack of regional adaptability faced by Spark and Flink in big data processing. It discusses the introduction of interpretable algorithms such as SHAP and LIME to enhance the transparency and user trust of neural network models. Then it discusses how to combine time-aware transfer learning and Geographic Information Systems (GIS) technology to enhance the technology generalization and adaptability of Spark machine learning models. Time-aware transfer learning uses historical datas temporal evolution to ensure models perform well in new time periods or scenarios, while GIS technology enables more precise predictions and analyses based on geographical data, enhancing spatial adaptability. Lastly, the study explores hybrid processing strategies by integrating Apache Flink, Kafka Streams, and Spark batch processing frameworks. This approach not only facilitates efficient real-time data processing and detailed analysis but also enhances the models flexibility and processing capabilities in complex data scenarios. By integrating these techniques, it is possible to improve the efficiency and effectiveness of big data processing frameworks in addressing complex real-world challenges, thereby advancing technology and application development in related fields.

Visible‐Light‐Induced Diastereoselective Cascade Cyclization to Construct Polycyclic Spiroindolines

AbstractWe herein reported, a visible‐light‐induced K2S2O8 mediated cascade reaction of 3‐(2‐isocyanoethyl)indoles with α‐oxocarboxylic acids leading to diastereoselective synthesis of polycyclic spiroindolines bearing an N‐formyl unit derivative utilising the alkali metal salt K2S2O8 as a mediator under mild reaction conditions. This study offers an illustration of α‐oxocarboxylic acids and 3‐(2‐isocyanoethyl)indoles‐based reactions as it shows the involvement of glyoxolate ions without decarboxylation to access polycyclic spiroindolines bearing an N‐formyl unit analogue. The batch process can be extended to a continuous flow system using a glass tube loaded with K2S2O8 placed between the PFA capillary reactor, which can greatly advance the reaction efficiency and can even be promoted to gram‐scale synthesis.

Application research on vector coherent frequency‐domain batch adaptive line enhancement in deep water

AbstractThe low frequency line spectrum noise radiated by ships has strong stability and is difficult to eliminate, which is the key information required for passive signal detection. A vector coherent frequency‐domain batch adaptive line enhancement method is proposed to address the issue of insufficient detection capability of traditional scalar adaptive line enhancement (ALE) algorithms for ship characteristic line spectra in complex deep‐sea environments. This method not only introduces the idea of frequency‐domain batch processing, but also uses synchronously collected sound pressure and particle velocity as dual input, fully utilising the coherence characteristics between vector channels to output high gain line spectrum signals and improve computational efficiency. In simulation and sea trial data validation, compared with the time‐domain vector coherent adaptive line enhancement algorithm, this method has shorter time consumption, higher efficiency, and can improve the detection ability of line spectrum signals under low signal‐to‐noise ratio conditions. The bearing estimation results output by this algorithm is also more accurate.

Agricultural low-cost waste adsorption of methylene blue and modelling linear isotherm method versus nonlinear prediction

AbstractThis study shows that geographically marked wheat hull, named Siyez, rice hull Sarı Kılçık, and Taşköprü Garlic stalk were used as agricultural waste to potential adsorbent materials for removing methylene blue from aqueous solution. Experimental data were evaluated in both equilibrium batch process and kinetic studies. In addition, the factors that affect the adsorption capacities, such as pH solutions, methylene blue concentration, contact time, and temperatures, were also investigated. Obtained data were subject to two constant adsorption models of Langmuir, Freundlich, Temkin, and Dubinin−Radushkevich. The kinetic models (pseudo-first-order, pseudo-second-order, intra-particle diffusion and film diffusion) and thermodynamic parameters were evaluated. The adsorption isotherms, characterized by an excellent fit with the Langmuir model (R2 = 0.99) across all adsorbents, underscore the prevalence of monolayer adsorption of methylene blue, in contrast to the Freundlich equation. Adsorption kinetics of the methylene blue onto the adsorbents followed pseudo-second-order kinetic model. According to high regression coefficient (R2) and minimal values of nonlinear error functions like RMSE; the maximum monolayer adsorption capacities of wheat hull, rice hull and garlic stalk were found to be 62.50 (mg/g), 54.94 (mg/g), and 370.37 (mg/g), respectively. The results indicated that these proposed adsorbents could be low-cost and effective adsorbents for water purification and have adsorption capacity as much as comparable with the literature. In batch equilibrium studies, the adsorption of methylene blue dye onto the wheat hull, rice hull, and garlic stalk exhibited a significant correlation with temperature, contact time, and initial concentration of methylene blue dye and Adaptive Neuro-Fuzzy Inference System algorithm for forecasting overall the system parameter well fitted with these findings with the accuracy of outputs (R2 about 0.99 for each). Consequently, the thermodynamic analysis revealed that the adsorption process takes place in bulk diffusion by liquid phase mass transfer and occurred spontaneously with endothermically except garlic stalk. Adsorption thermodynamic studies show that the adsorption of methylene blue onto the garlic stalk was spontaneous and exothermic. Graphical Abstract

Data-Driven Batch Process Monitoring for Continuous Annealing of Cold-Rolled Strip Steel

The continuous annealing process (CAP) is a crucial process of steel production, which has a significant impact on the uniformity and stability of mechanical properties. A novel batch monitoring process based on kernel dissimilarity (KDISSIM) and Kmeans++ is proposed in this paper, focusing on problems such as unequal sample lengths between batches and nonlinearity between variables. First, KDISSIM is used to describe the dissimilarity between batches. Secondly, Kmeans++ is employed to improve the accuracy of clustering tasks based on historical batches. The largest cluster is considered to be at a relatively stable control level, and these batches are further used as training data. Then, the center batch and boundary batch of the training set are used as the reference batch and monitoring threshold for the monitoring model, respectively. Finally, the effectiveness of the proposed method is verified via the actual CAP data, providing a feasible solution for CAP batch monitoring.

Expressed Protein Ligation in Flow.

The development of a flow chemistry platform for the generation of modified protein targets via expressed protein ligation (EPL) is described. The flow EPL platform enables efficient ligation reactions with high recoveries of target protein products and superior reaction rates compared to corresponding batch processes. The utility of the flow EPL technology was first demonstrated through the semisynthesis of the tick-derived chemokine-binding protein ACA-01 containing two tyrosine sulfate modifications. Full-length, sulfated ACA-01 could be efficiently assembled by ligating a recombinantly expressed C-terminal protein fragment and a synthetic sulfopeptide thioester in flow. Following folding, the semisynthetic sulfoprotein was shown to exhibit potent binding to a variety of pro-inflammatory chemokines. In a second modified protein target, we employed an in-line flow EPL-photodesulfurization strategy to generate both unmodified and phosphorylated forms of human β-synuclein by fusing a recombinant protein thioester, generated through cleavage of an intein fusion protein, and a synthetic (phospho)peptide. The semisynthetic proteins were assembled in 90 min in flow, a significant improvement over corresponding batch protein assembly, and enabled access to tens of milligrams of high purity material. Flow EPL has the potential to serve as a robust technology to streamline access to homogeneously modified proteins for a variety of applications in both academia, as well as in the pharmaceutical and biotechnology sector.

Revitalizing the single batch environment: a ‘Quest’ to achieve fairness and efficiency

In the realm of computer systems, efficient utilization of the CPU (Central Processing Unit) has always been a paramount concern. Researchers and engineers have long sought ways to optimize process execution on the CPU, leading to the emergence of CPU scheduling as a field of study. In this research, we have analyzed the single offline batch processing and investigated other sophisticated paradigms such as time-sharing operating systems and wildly used algorithms, and their shortcomings. Our work is directed toward two fundamental aspects of scheduling: efficiency and fairness. We propose a novel algorithm for batch processing that operates on a preemptive model, dynamically assigning priorities based on a robust ratio, employing a dynamic time slice, and utilizing periodic sorting to achieve fairness. By engineering this responsive and fair model, the proposed algorithm strikes a delicate balance between efficiency and fairness, providing an optimized solution for batch scheduling while ensuring system responsiveness.

Learning optimal edge processing with offloading and energy harvesting

Modern portable devices can execute increasingly sophisticated AI models on sensed data. The complexity of such processing tasks is data-dependent and has relevant energy cost. This work develops an Age of Information Markovian model for a system where multiple battery-operated devices perform data processing and energy harvesting in parallel. Part of their computational burden is offloaded to an edge server which polls devices at given rate. The structural properties of an optimal policy for a single device-server system are derived. They permit to define a new model-free reinforcement learning method specialized for monotone policies, namely Ordered Q-Learning, providing a fast procedure to learn the optimal policy. The method is oblivious to the devices’ battery capacities, the cost and the value of data batch processing and to the dynamics of the energy harvesting process. Finally, the polling strategy of the server is optimized by combining this policy improvement technique with stochastic approximation methods. Extensive numerical results provide insight into the system properties and demonstrate that the proposed learning algorithms outperform existing baselines.

Study the application of polydopamine preparation technology in the interface modification of phthalonitrile composite

Polydopamine (PDA), recognized for its exceptional adhesion and high-temperature resistance, serves as an effective sizing agent in demanding composite material production. This study explores the application of PDA as an interface adhesive for carbon fiber-reinforced high-temperature phthalonitrile (PN) composites (CFRPN), assessing the effects of two PDA synthesis methods—electrochemical and solution oxidation—on interfacial modification. The electrochemical method uniformly fills grooves on CF surfaces, creating an optimal interface layer, yet it encounters challenges in bulk, multi-layer fiber processing. Conversely, the solution oxidation method, despite less uniform PDA distribution, exhibits greater consistency in batch processing and significantly enhances the interlaminar shear strength of CFRPN, outperforming the electrochemical method in mechanical enhancements. Given its simplicity and reduced equipment requirements, the solution oxidation method offers considerable potential for broader engineering applications.

Multi-agent deep reinforcement learning for dynamic reconfigurable shop scheduling considering batch processing and worker cooperation

Reconfigurable manufacturing system is considered as a promising next-generation manufacturing paradigm. However, limited equipment and complex product processes add additional coupled scheduling problems, including resource allocation, batch processing and worker cooperation. Meanwhile, dynamic events bring uncertainty. Traditional scheduling methods are difficult to obtain good solutions quickly. To this end, this paper proposes a multi-agent deep reinforcement learning (DRL) based method for dynamic reconfigurable shop scheduling problem considering batch processing and worker cooperation to minimize the total tardiness cost. Specifically, a dual-agent DRL-based scheduling framework is first designed. Then, a multi-agent DRL-based training algorithm is developed, where two high-quality end-to-end action spaces are designed using rule adjustment, and an estimated tardiness cost driven reward function is proposed for order-level scheduling problem. Moreover, a multi-resource allocation heuristics is designed for the reasonable assignment of equipment and workers, and a batch processing rule is designed to determine the action of manufacturing cell based on workshop state. Finally, a strategy is proposed for handling new order arrivals, equipment breakdown and job reworks. Experimental results on 140 instances show that the proposed method is superior to scheduling rules, genetic programming, and two popular DRL-based methods, and can effectively deal with various disturbance events. Furthermore, a real-world assembly and debugging workshop case is studied to show that the proposed method is applicable to solve the complex reconfigurable shop scheduling problems.

Scheduling parallel batch processing machines: A case study in the semiconductor industry

This paper presents a mathematical programming-based solution approach for the scheduling problem of batch-processing parallel machines with eligibility constraints. A case study has been presented in the semiconductor industry, where the ovens are scheduled for the underfill cure operation of products. The case includes constraints, such as oven-product eligibility restrictions, loading constraints for the batching of products for ovens, daily production requirements, and oven capacity constraints. In this study, we also assess the difference between creating batches of a single product type or different product types to be allocated to the ovens. The case study results have shown that the proposed models, in comparison to the current situation, increase the occupancy rate of ovens. The execution of the models aids the company in gaining visibility on the scheduling of ovens and successfully managing the production plan and order commitment. The proposed models have been effective and supportive of the semiconductor company.

A Unified Adsorption Kinetic Model Inspired by Epidemiological Model: Based on Adsorbates "Infect" Adsorbents.

Adsorption is a unit operation used in various fields, including the environmental, chemical, and pharmaceutical industries. Understanding the adsorption kinetics is crucial to designing efficient adsorption systems. However, existing empirical adsorption models are limited in providing insights into the mass transfer mechanisms. Additionally, the absence of a unified adsorption kinetic model hampers the effective comparison of different adsorption systems. Here, we viewed the adsorption as an "infectious process of adsorbates by adsorbents" akin to epidemiology. In epidemiology, individuals can be divided into susceptible, infected, and recovered compartments, ignoring the complexities of movement among individuals. Analogously, we have categorized the adsorbates as adsorbable, adsorbed, and removed compartments. Thus, we proposed a unified adsorption kinetic model (the monolayer-multilayer-adsorbable-adsorbed-removed model) that accommodates monolayer/multilayer adsorption. The model was designed to encompass diverse adsorption setups, including continuous and batch processes with fixed/dispersed adsorbents. The versatility and applicability of the model were demonstrated through validation using a diverse set of experimental data. This validation underscored its effectiveness in water/wastewater treatment, salt reduction, metal recovery, and drug purification. A MATLAB-based program for solving this model was made available to researchers for their utilization and further investigations. Overall, this study developed a versatile adsorption kinetic model that offers a comprehensive and unified understanding of adsorption kinetics across various applications.

Evaluation of a rapid multi-attribute combinatorial high-throughput UV-Vis/DLS/SLS analytical platform for rAAV quantification and characterization

Recombinant adeno-associated virus (rAAV)-based gene therapies are expanding in their application. Despite progress in manufacturing, current analytical methods for product quantification and characterization remain largely unchanged. Although critical for product and process development, in-process testing, and batch release, current analytical methods are labor-intensive, costly, and hampered by extended turnaround times and low throughput. The field requires more efficient, cost-effective analytical techniques capable of handling large sample quantities to accelerate product and process development. Here, we evaluated Stunner from Unchained Labs for quantifying and characterizing rAAVs and compared it with established analytical methods. Stunner is a combinatorial analytic technology platform that interpolates ultraviolet-visible (UV-Vis) absorption with static and dynamic light scattering (SLS/DLS) analysis to determine capsid and genomic titer, empty and full capsid ratio, and assess vector size and polydispersity. The platform offers empirical measurements with minimal sample requirements. Upon testing hundreds of rAAV vectors, comprising various serotypes and transgenes, the data show a strong correlation with established analytical methods and exhibit high reproducibility and repeatability. Some analyses can be applied to in-process samples from different purification stages and processes, fulfilling the demand for rapid, high-throughput analysis during development. In sum, the pipeline presented streamlines small- and large-batch analytics.

Scaling the polymerization of polyaldehydes through continuous flow synthesis

A plug flow reactor was constructed to scale the synthesis of metastable, phthalaldehyde-based polymers to achieve production rates of 1–2 kg per day. The flow-induced mixing and in-line polymerization quench and precipitation sequences resulted in improved polymer purity and long-term stability compared to the same materials made from a conventional batch process. Cryogenic rheology was used to probe the complex fluid dynamics encountered during the polymerization of PPA homopolymers. It is envisioned that this continuous flow manufacturing approach could be extended to other low ceiling temperature or aldehyde monomer systems to help implement and support a plastic circular economy.Graphical abstract

Efficient Bio-Oxidation of Cellobiose with Engineered Gluconobacter oxydans to Provide Highly Concentrated Cellobionic Acid

Cellobionic acid (CBA) can be obtained through the oxidation of cellobiose, the monomer of cellulose. CBA serves as a plant-based alternative to its stereoisomer lactobionic acid, which is used in the pharmaceutical, cosmetic, and food industries. Gluconobacter oxydans is a well-established whole-cell biocatalyst with membrane-bound dehydrogenases (mDH) for regio-specific oxidations. As G. oxydans wildtype cells show low cellobiose oxidation activities, the glucose mDH from Pseudomonas taetrolens was overexpressed in G. oxydans BP9, a multi mDH deletion strain. Whole-cell biotransformation studies were performed with resting cells of the engineered G. oxydans in stirred tank bioreactors. Initial biomass specific cellobionate formation rates increased with increasing cellobiose concentrations up to 190 g L−1, and were constant until the solubility limit. The maximal volumetric CBA formation rates and the oxygen uptake rates increased linearly with the concentration of engineered G. oxydans. This enables the estimation of the maximum biocatalyst concentration limited by the maximum oxygen transfer rate of any bioreactor. Thus, 5.2 g L−1 G. oxydans was sufficient to produce 502 g L−1 CBA with >99% yield in a simple aerobic batch process. The highly concentrated CBA will reduce downstream processing costs considerably after cell separation.

A multi-biomarker micronucleus assay using imaging flow cytometry

Genetic toxicity testing assesses the potential of compounds to cause DNA damage. There are many genetic toxicology screening assays designed to assess the DNA damaging potential of chemicals in early drug development aiding the identification of promising drugs that have low-risk potential for causing genetic damage contributing to cancer risk in humans. Despite this, in vitro tests generate a high number of misleading positives, the consequences of which can lead to unnecessary animal testing and/or the abandonment of promising drug candidates. Understanding chemical Mode of Action (MoA) is vital to identifying the true genotoxic potential of substances and, therefore, the risk translation into the clinic. Here we demonstrate a simple, robust protocol for staining fixed, human-lymphoblast p53 proficient TK6 cells with antibodies against ɣH2AX, p53 and pH3S28 along with DRAQ5™ DNA staining that enables analysis of un-lysed cells via microscopy approaches such as imaging flow cytometry. Here, we used the Cytek® Amnis® ImageStream®X Mk II which provides a high-throughput acquisition platform with the sensitivity of flow cytometry and spatial morphological information associated with microscopy. Using the ImageStream manufacturer’s software (IDEAS® 6.2), a masking strategy was developed to automatically detect and quantify micronucleus events (MN) and characterise biomarker populations. The gating strategy developed enables the generation of a template capable of automatically batch processing data files quantifying cell-cycle, MN, ɣH2AX, p53 and pH3 populations simultaneously. In this way, we demonstrate how a multiplex system enables DNA damage assessment alongside MN identification using un-lysed cells on the imaging flow cytometry platform. As a proof-of-concept, we use the tool chemicals carbendazim and methyl methanesulphonate (MMS) to demonstrate the assay’s ability to correctly identify clastogenic or aneugenic MoAs using the biomarker profiles established.

High-Throughput Covalent Modifier Screening with Acoustic Ejection Mass Spectrometry.

Interests in covalent drugs have grown in modern drug discovery as they could tackle challenging targets traditionally considered "undruggable". The identification of covalent binders to target proteins typically involves directly measuring protein covalent modifications using high-resolution mass spectrometry. With a continually expanding library of compounds, conventional mass spectrometry platforms such as LC-MS and SPE-MS have become limiting factors for high-throughput screening. Here, we introduce a prototype high-resolution acoustic ejection mass spectrometry (AEMS) system for the rapid screening of a covalent modifier library comprising ∼10,000 compounds against a 50 kDa-sized target protein─Werner syndrome helicase. The screening samples were arranged in a 1536-well format. The sample buffer containing high-concentration salts was directly analyzed without any cleanup steps, minimizing sample preparation efforts and ensuring protein stability. The entire AEMS analysis process could be completed within a mere 17 h. An automated data analysis tool facilitated batch processing of the sample data and quantitation of the formation of various covalent protein-ligand adducts. The screening results displayed a high degree of fidelity, with a Z' factor of 0.8 and a hit rate of 2.3%. The identified hits underwent orthogonal testing in a biochemical activity assay, revealing that 75% were functional antagonists of the target protein. Notably, a comparative analysis with LC-MS showcased the AEMS platform's low risk of false positives or false negatives. This innovative platform has enabled robust high-throughput covalent modifier screening, featuring a 10-fold increase in library size and a 10- to 100-fold increase in throughput when compared with similar reports in the existing literature.

Green adsorption of quinoline as basic N-compounds from liquid fuel using mesoporous and Fe-functionalized walnut shell-activated carbon

Removing nitrogenous compounds from liquid fuels is essential, and adsorption de-nitrogenation is a new process for this goal. This work has the purpose of investigating quinoline adsorption using Iron functionalized Activated carbon (IFAC) mesoporous nanocomposite. The mesoporous nanocomposite was synthesized in a one-step process by chemical activation of the walnut shell using Iron chloride as the activation and modification agent. The IFAC mesoporous nanocomposite was utilized to remove quinoline from isooctane as the liquid fuel via a discontinuous (batch) process by examining the effect of initial concentration, contact time, and temperature. The effect of contact time on the adsorption ability of IFAC for quinoline was well described by the pseudo-second-order model. The Langmuir model described the equilibrium manner of the process well, and the adsorbent could retain 344.82 mg/g of quinoline (37.37 mg N/g). The thermodynamic parameters show the endothermic (ΔH°= +30.070 kJ/mol) and spontaneity of the adsorption process. The regeneration experiments indicated that functionalized AC’s adsorption ability for quinoline adsorption decreased by 23% after five cycles. The obtained data indicate that the functionalization of activated carbon surface with iron particles leads to improving its adsorption ability toward N-containing compounds from liquid fuels.