Neural networks (NNs) are widely used as surrogate models but they do not guarantee physically consistent predictions thereby preventing adoption in various applications. We propose a method that can enforce NNs to satisfy physical laws that are nonlinear in nature such as enthalpy balances. Our approach, inspired by Picard’s successive approximations method, aims to enforce multiplicatively separable constraints by sequentially freezing and projecting a set of the participating variables. We demonstrate our Picard-KKT-hPINN for surrogate modeling of a catalytic packed bed reactor for methanol synthesis. Our results show that the method efficiently enforces nonlinear enthalpy and linear atomic balances at machine-level precision. Additionally, we show that enforcing conservation laws can improve accuracy in data-scarce conditions compared to vanilla multilayer perceptron.

Fault diagnosis is critical for ensuring the safety and efficiency of chemical processes, as undetected faults can lead to catastrophic consequences. While deep learning-based methods have shown promise in this field, they often require manual hyperparameter tuning, which is not efficient since they heavily rely on expert knowledge and need iterative trial-and-error. This work introduces a novel approach combining a Multiscale Convolutional Neural Network (MSCNN) with Tree-Structured Parzen Estimator (TPE) for automated hyperparameter optimization to enhance the performance of fault diagnosis for chemical processes. The Multi-Scale Module is to capture complex nonlinear features from the fault data, while the TPE efficiently searches for optimal hyperparameters for MSCNN. An experimental study was carried out on the Tennessee Eastman Process (TE Process) dataset, where the proposed method was benchmarked against state-of-the-art models. The results indicate that the MSCNN-TPE method de... [more]

As large language models have shown remarkable capabilities in conversing via natural language, the question arises in which way LLMs could potentially assist chemical engineers in research and industry with domain-specific tasks. We generate dynamic chemical reactor models in Modelica code format from textual descriptions as user input. We fine-tune Llama 3.1 8B Instruct on synthetically generated Modelica code for different reactor scenarios. We compare the performance of our fine-tuned model to the baseline Llama 3.1 8B Instruct model as well as GPT4o. We manually assess the models' predictions regarding the syntactic and semantic accuracy of the generated dynamic models. We find that considerable improvements are achieved by the fine-tuned model with respect to both the semantic and the syntactic accuracy of the Modelica models. However, the fine-tuned model lacks a satisfactory ability to generalize to unseen scenarios compared to GPT4o.

In an era defined by economic competitiveness and environmental awareness, engineering solutions must maximize profitability, efficiency and sustainability, underscoring the relevance of process optimization and the societal impact any contribution in this research field would bring. In chemical reactor engineering, optimization tasks pose significant challenges due to the highly non-linear and non-convex nature of reactor models, often involving differential equations. While conventional approaches have proven to be reliable strategies for solving these complex problems, their application becomes impractical as problem size and complexity increase. This work introduces a novel application of Physics-Informed Neural Networks (PINNs) to address constrained optimization problems in reactor engineering and demonstrates the proposed methodology through two illustrative case studies in chemical reactor design and control. In doing so, we highlight the capability of PINNs to efficiently lear... [more]

To address the limitations of traditional sensing instrumentation in industrial processes, this work explores the use of industrial text data. Given that current instrumentation often fails to capture the full scope of process-related information, text data resulting from operation of industrial settings (for example: maintenance, inspection and incident reports) can provide valuable insights. This study focuses on accessing the effectiveness of natural language processing (NLP) techniques in retrieving critical information from industrial text data. To achieve this, the classification of Process Safety and Containment Events (PSCE) was used as case study. Overall, we found NLP methods are effective in information retrieval from industrial text data. However, the integration of the embeddings into machine learning (ML) approaches poses some challenges. The complexity of the information encoded in the embeddings makes them too disparate and unique samples of a larger domain, making chal... [more]

Many small and medium enterprises (SME) often fail to fully utilize the data they collect due to a lack of technical expertise. The ecoKI platform, a low-code solution that simplifies machine learning application for SMEs, showed a promising answer to the challenge. This study explores the application of ecoKI platform to design process monitoring tools for waste sorting plants. NIR separator data were processed through ecoKI’s building blocks to train two neural network architectures—MLP and LSTM—for predicting NIR separation efficiency. The results showed that the models accurately predicted NIR output and effectively identified regions where NIR separation performance declined, demonstrating the potential of data-driven approaches for real-time performance monitoring. This work highlights how SMEs can leverage existing data for operational efficiency and decision-making, offering an accessible solution for industries with limited machine learning expertise. The approach is adaptable... [more]

This study focuses on biomanufacturing case study, i.e. Lovastatin production, employing a hybrid modeling framework that combines mechanistic and data-driven approaches. A time-series dataset was generated using the KT-Biologics I (KTB1) plantwide model, a dynamic simulation of continuous biomanufacturing. The dataset captures critical parameters such as nutrient concentrations and API production. The AI-DARWIN framework was used to develop interpretable machine learning models with constrained functional forms, ensuring both accuracy and clarity. The resulting polynomial-based models reveal key relationships between process variables and system performance, bridging mechanistic insights with data-driven predictions. The models demonstrated reasonable accuracy showing minimal difference between the training and testing errors, highlighting their strong generalization. This work advances hybrid modeling in biomanufacturing by integrating plant-wide mechanistic simulations with interpre... [more]

We propose a methodology that allows communication with Piping and Instrumentation Diagrams (P&IDs) using natural language. In particular, we represent P&IDs through the DEXPI data model as labeled property graphs and integrate them with Large Language Models (LLMs). The approach consists of three main parts: 1) P&IDs are cast into a graph representation from the DEXPI format using our pyDEXPI Python package. 2) A tool for generating P&ID knowledge graphs from pyDEXPI. 3) Integration of the P&ID knowledge graph to LLMs using graph-based retrieval augmented generation (graph-RAG). This approach allows users to communicate with P&IDs using natural language. It extends LLM’s ability to retrieve contextual data from P&IDs and mitigate hallucinations. Leveraging the LLM's large corpus, the model is also able to interpret process information in P&IDs, which could help engineers in their daily tasks. In the future, this work will also open up opportunities in the context of other generative A... [more]

This paper presents state-of-the-art object detection and object identification algorithms for digitizing and interpreting chemical engineering diagrams, including, Block Flow Diagrams (BFDs), Process Flow Diagrams (PFDs), and Piping and Instrumentation Diagrams (P&IDs), using computer vision techniques. These diagrams are essential for visualizing plant processes and equipment but are often stored as image-based PDFs, making manual digitization/interpretation labor-intensive and error-prone. The proposed algorithm automates tasks such as detecting unit operations and identifying them using a set of rule-based and predefined approaches including edge and contour-based rules, spatial arrangement rules, and geometric rules. This method avoids data requirements and computational requirements of deep learning approaches, offering a scalable and efficient solution for preliminary extraction of complex process information.

Phase transition enthalpies, such as those for fusion, vaporization, and sublimation, are vital for understanding thermodynamic properties and aiding early-stage process design. However, measuring these properties is often time-consuming and costly, leading to increased interest in computational methods for fast and accurate predictions. Graph neural networks (GNNs), known for their ability to learn complex molecular representations, have emerged as state-of-the-art tools for predicting various thermophysical properties. Despite their success, GNNs do not inherently obey thermodynamic laws. In this study, we present a multitask GNN designed to predict vaporization, fusion, and sublimation enthalpies of organic compounds. We modified the loss function of the GNN, accounting for the thermodynamic cycle of the three phase transition enthalpies. To train the model, we digitized the extensive Chickos and Acree compendium, which encompasses 32,023 experimental measurements. Two approaches we... [more]

A piping and instrumentation diagram (P&ID) is a central reference document in chemical process engineering. Currently, chemical engineers manually review P&IDs through visual inspection to find and rectify errors. However, engineering projects can involve hundreds to thousands of P&ID pages, creating a significant revision workload. This study proposes a rule-based method to support engineers with error detection and correction in P&IDs. The method is based on a graph representation of P&IDs, enabling automated error detection and correction, i.e., autocorrection, through rule graphs. We use our pyDEXPI Python package to generate P&ID graphs from DEXPI-standard P&IDs. In this study, we developed 33 rules based on chemical engineering knowledge and heuristics, with five selected rules demonstrated as examples. A case study on an illustrative P&ID validates the reliability and effectiveness of the rule-based autocorrection method in revising P&IDs.

As the demand for resources continues to grow, pipelines have become critical for transporting water, fossil fuels, and chemicals. Monitoring pipeline systems is essential, as leaks can lead to severe environmental damage and safety hazards. This study aims to develop a pipeline leakage detection system based on digital twin technology and Physics-Informed Neural Networks (PINNs). By embedding physical principles, such as the continuity and momentum equations derived from the Navier-Stokes equation, into the neural network's loss function, the model can predict pressure and flow dynamics with high accuracy while adhering to physical constraints. PINNs are particularly advantageous as they require minimal data, maintain physical consistency, and provide reliable interpretations, making them well-suited for addressing pipeline safety challenges. The model is designed to simulate fluid dynamics under normal operating conditions, with deviations in prediction errors signaling potential lea... [more]

Chemical process models often involve high non-linearity due to thermodynamic and kinetic relationships, with non-convex bilinear terms adding complexity to process optimization. Recently, data-driven models, particularly artificial neural networks (ANNs), have gained traction for representing chemical processing units. The predictive accuracy of ANNs depends on data quality, variable interactions, and network architecture, the latter being an optimization challenge itself. This study proposes and evaluates two strategies to optimize ANN architecture for modeling a propylene glycol production process from glycerol. The process includes a reactor and two distillation columns, with training data generated through simulation in Aspen Plus by varying design and operating variables. Two approaches are compared: random ANN structure generation and architecture optimization using the ant colony algorithm, a method suitable for discrete problems. Decision variables include the number of hidden... [more]

Artificial neural networks (ANNs) have gained popularity in the last years as tools to develop data-driven models of chemical process units. However, representing a system only with such artificial intelligence models may lead to a loss in the comprehension of the occurring phenomena. Hybrid models allow combining the predictive capabilities of ANNs with the foundational knowledge of rigorous models. This study explores the impact of hybridization in designing and optimizing shell-and-tube heat exchangers, comparing a full ANN-based model with a hybrid model. The hybrid model incorporates ANN predictions for highly nonlinear components, such as heat transfer coefficients, while other calculations are performed using the rigorous Bell-Delaware model. To generate the necessary data, the rigorous model is solved under randomly selected conditions. Using Python, one ANN predicts the exchanger's cost, while another predicts the heat transfer coefficients. Both models are optimized using the... [more]

Detailed assessment of fuel production processes at an early stage of a project is crucial to identify potential technical challenges, optimize efficiency and minimize costs and environmental impact. While process simulations often are either very rigid and accurate or very flexible and unprecise, informed decision making can only be maintained by establishing a detailed process model as early as possible within the project lifecycle while keeping relevant aspects of the process flexible enough. In this work, we present the development of a framework based on a dynamic interface between AspenPlus® process simulations and Python, enabling enhanced flexibility and automation for process modeling and optimization. This integration leverages the powerful simulation capabilities of AspenPlus® with the versatility of Python for data analysis and optimization, delivering significant improvements in workflow efficiency and process control. By utilizing the dynamic simulation data exchange with... [more]

Process synthesis is a fundamental step in process design. The aim is to determine the optimal configuration of unit operations and stream flows to enhance key performance metrics. Traditional methods provide just one optimal solution and are strongly dependent on user-defined technologies, stream connections, and initial guesses for unknown variables. Usually, a single solution is not sufficient for adequate decision-making, especially, when properties such as flexibility or reliability are considered in addition to the process economics. Wastewater Treatment network synthesis and design is a complex problem that demands innovative approaches in design, retrofits, and maintenance strategies. Considering this, an enhanced framework for improving reliability in wastewater transportation networks based on graph theory and machine learning is presented. Machine learning models were developed to predict failure probability, where the XGBoost model provided the best predictions. To select t... [more]

While bilevel optimization is gaining prominence across various domains, the field lacks standardized tools for generating test problems that can effectively evaluate and guide the development of efficient solution algorithms. We define the term "trivial bilevel optimization problem" as a bilevel problem whose high-point relaxation solution is also feasible. These easy-to-solve problems frequently arise in naïve implementations of random bilevel optimization problem generators, significantly impacting the evaluation of bilevel solution algorithms. However, this problem has not been addressed in the literature to our best knowledge. This work introduces a non-trivial mixed-integer bilevel optimization problem generator, NT-BMIPGen, and a problem library, designed to eliminate the generation of trivial bilevel problems. Through analysis of the bilevel problem structure, we identify key factors contributing to problem triviality. Particularly, the upper to lower variable ratios and the nu... [more]

The electrification of industrial heat is a promising pathway for decarbonization, yet challenges persist in balancing capital costs, operating costs, and emissions reduction. While previous studies have assessed electrification through heat integration and graphical methods, these approaches do not inherently determine the optimal hybrid technology configuration. This study introduces an optimization-based framework that systematically evaluates the cost-optimal allocation of electrified and conventional heating technologies. Formulated as a Mixed-Integer Linear Programming (MILP) model and implemented in Gurobi, the framework minimizes Total Annualized Cost (TAC) while satisfying heat demand, technology constraints, and emissions targets. Applied to an industrial case study, the model compares three scenarios: a fully conventional system relying on steam boilers and fired heaters, a fully electrified system utilizing high-temperature heat pumps, electrode boilers, and electric heater... [more]

This paper presents a novel fault detection method based on key variable forecasting models. The approach integrates future forecasts of key variables into a time window, allowing for early fault detection without modifying the offline training phase of the existing fault detection model. By incorporating predicted data into the detection process, the proposed method significantly improves fault detection rates and reduces detection delays. Experiments using the Continuous Stirred Tank Heater (CSTH) system demonstrate the superiority of our method over traditional approaches, showing the advantages of forecasting in enhancing detection performance. However, our results also highlight the dependency of the method's effectiveness on the quality of the forecasting model, suggesting the need for more advanced time-series forecasting techniques. Additionally, the current point forecasting method may not be sufficient in real-world applications, where probabilistic modeling of key variables... [more]

Large Language Models have demonstrated potential in accelerating scientific discovery, but they face challenges when making inferences in rapidly evolving and niche domains like Process Systems Engineering (PSE). To address this, we propose a Graph-based Retrieval-Augmented Generation (RAG) pipeline specifically designed for PSE papers. Our pipeline includes custom document parsing, knowledge graph construction, and refinement to enhance retrieval accuracy. We evaluate the effectiveness of our approach using an automatically generated benchmark consisting entirely of PSE-related questions. The results show that our pipeline outperforms both non-RAG and vanilla RAG implementations in terms of relevant document retrieval and overall answer quality. Additionally, our implementation is fully customizable, allowing users to select the papers most relevant to their specific tasks. This framework is openly available, providing a flexible solution for those working in PSE or similar domains.

Robust optimisation is a powerful approach for addressing uncertainty ensuring constraint satisfaction for all uncertain parameter realisations. While convex robust optimisation problems are effectively tackled using robust reformulations and cutting plane methods, extending these techniques to non-convex problems remains largely unexplored. In this work we propose a method that is based on a parallel robustness and optimality search. We introduce a novel spatial Branch-and-Bound algorithm integrated with robust cutting-planes for solving non-convex robust optimisation problems. The algorithm systematically incorporates global and robust optimisation techniques, leveraging McCormick relaxations. The proposed algorithm is evaluated on benchmark pooling problems with uncertain feed quality, demonstrating algorithm stability and solution robustness. The computational time for the examined case studies is within the same order of magnitude as state-of-the-art. The findings of this work hig... [more]

This work addresses the process design of Power-to-Ammonia plants (i.e. ammonia from renewable-powered electrolysis) by a novel methodology based on the multi-objective optimization of the “Three pillars of sustainability”: economic, environmental, and social. Specifically, we developed a tool estimating the installed capacities of every main process section typically featured by Power-to-Ammonia facilities (e.g., the renewable power plant, the electrolyzer, energy and hydrogen storage systems, etc.) to maximize the plant’s “Global Sustainability Score”.

Pressure swing adsorption (PSA) is an energy-efficient technology for gas separation, while the multi-objective optimization of PSA is a challenging task. To tackle this, we propose a hybrid optimization framework, which integrates three steps. In the first step, we establish surrogate models for the constraints using Gaussian processes (GPs) and employ multi-objective Bayesian optimization to search for feasible points that satisfy the constraints. In the second step, we establish surrogate models for the objective function and constraints using GPs and utilize constrained multi-objective Bayesian optimization to search for an approximate Pareto front. In the third step, we perform a local search based on the approximate Pareto front. By employing the trust region filter method, we construct quadratic models for each constraint and objective function and refine the Pareto front to achieve local optimality. This framework demonstrates the efficiency of Bayesian optimization and the loc... [more]

Individual-based modeling (IbM) has emerged as a powerful approach for studying microbial populations, offering a bottom-up framework to simulate cellular behaviors and their interactions. Unlike continuum-based models, IbM explicitly captures the heterogeneity and emergent dynamics of microbial communities, making it invaluable for studying spatially structured phenomena such as nutrient competition, biofilm formation, and colony interactions. However, IbM faces significant computational challenges, particularly in resolving spatial overlaps during simulations of large microbial populations. Traditional approaches, such as pairwise comparisons or kd-trees, are computationally expensive and scale poorly with population size. The Discretized Overlap Resolution Algorithm (DORA) introduces a novel grid-based solution to overcome these limitations. By encoding spatial information into an occupancy matrix, DORA achieves a time complexity of O(N), enabling efficient resolution of overlaps wh... [more]

Bayesian classification (BC) is a powerful supervised machine learning method for modelling the relationship between a set of continuous variables and a set of discrete variables that represent classes. BC has been successful in engineering and medical applications, including feasibility analysis and clinical diagnosis. Gaussian process (GP) models are widely used in BC methods to model the probability of assigning a class to an input point, typically through an indirect approach: a GP predicts a continuous function value based on Bayesian inference, which is then transformed into class probabilities using a nonlinear function like a sigmoid. The final class labels are assigned based on these probabilities. In this commonly used workflow, the uncertainty associated with the class prediction is usually evaluated as the uncertainty in the GP function values. A disadvantage of this approach is that it does not consider the uncertainty directly associated with the decision-making. In this... [more]