Search

Recent advances in generative artificial intelligence (GenAI), particularly large language models (LLMs), are profoundly impacting many fields. In chemical engineering, GenAI plays a pivotal role in the design, scale-up, and optimization of chemical and biochemical processes. The natural language understanding capabilities of LLMs enable the interpretation of complex chemical and biological data. Given the rapid developments of GenAI, this paper explores the extensive applications of GenAI in multiscale chemical engineering, spanning from quantum mechanics to macro-level optimization. At quantum and molecular levels, GenAI accelerates the discovery of novel products and enhances the understanding of fundamental phenomena. At larger scales, GenAI improves process design and operational efficiency, contributing to sustainable practices. We present several examples to demonstrate the role of GenAI, including its impact on nanomaterial hardness enhancement, novel catalyst generation, protein design, and the development of autonomous experimental platforms. This multiscale integration demonstrates the potential of GenAI to address complex challenges, drive innovation, and foster advancements in chemical engineering.

As the demand for artificial intelligence (AI) models and applications continues to grow, data centers that handle AI workloads are experiencing a rise in energy consumption and associated carbon footprint. This work proposes a variational quantum computing-based robust optimization (VQC-RO) framework for control and energy management in large-scale data centers to address the computational challenges and overcome limitations of conventional model-based and model-free strategies. The VQC-RO framework integrates variational quantum circuits (VQCs) with classical optimization to enable efficient and uncertainty-aware control of energy systems in AI data centers. Quantum algorithms executed on noisy intermediate-scale quantum (NISQ) devices are used for value function estimation trained with Q-learning, leading to the formulation of a robust optimization problem with uncertain coefficients. The quantum computing-based robust control strategy is designed to address uncertainties associated with weather conditions and renewable energy generation while optimizing energy consumption in AI data centers. This work also outlines the computational experiments conducted at various AI data center locations in the United States to analyze the reduction in power consumption and carbon emission levels associated with the proposed quantum computing-based robust control framework. This work contributes a novel approach to energy-efficient and sustainable data center operation, promising to reduce carbon emissions and energy consumption in large-scale data centers handling AI workloads by 9.8 % and 12.5 %, respectively.

As the enthusiasm for electric vehicles passes the range anxiety and other tests, large-scale transportation electrification becomes a prominent topic in research and policy discussions. In consequence, the public attention has shifted upstream and holistically towards the integration of large-scale transportation electrification to power systems. This paper proposes a method to identify bottlenecks in power transmission systems to accommodate large-scale and stochastic electric vehicles charging demands. First, a distributionally robust chance-constrained direct current optimal power flow model is developed to quantify the impacts of stochastic electric vehicles charging demands. Subsequently, an agent-based model with the trip-chain method is applied to obtain the spatiotemporal distributions of electric vehicles charging demands for both light-duty electric vehicles and medium and heavy-duty electric vehicles. The first two moments of those distributions are extracted to build an ambiguity set of electric vehicles charging demands. Finally, a 121-bus synthetic transmission network for New York State power systems is used to validate the future transportation electrification in New York State from 2025 to 2035. Results show that the large-scale transportation electrification in New York State will account for approximately 13.33 % to 16.79 % of the total load demand by 2035. The transmission capacity is the major bottleneck in supporting New York State to achieve its transportation electrification. To resolve the bottlenecks, we explore some possible solutions, such as transmission capacity expansion and distributed energy resources investment. Wind power shows an advantage over solar energy in terms of total operational costs due to better peak alignment between wind power and electric vehicles charging demand.

Leveraging demand-side flexibility resources (e.g., buildings) is a crucial and cost-effective strategy for addressing the operational and infrastructure-related challenges in power grids pursuing deep decarbonization with high renewable energy penetration. However, the demand flexibility opportunities and financial benefits for residential space heating, which are sizeable demand-side flexibility resources, through emerging building energy management solutions (i.e., smart control and phased change material (PCM) thermal storage) across the US are not fully understood. In this paper, we systematically assess the demand flexibility and cost-saving/revenue potentials in residential space heating through detailed building-level simulations for five consecutive years at a 5-min temporal resolution in 20 metro areas across the high-heating-demand regions of the US. The results show a high degree of synergy between PCM thermal storage and smart control, which enables substantial demand flexibility potential, reaching 98.5% of peak load shifting, and electricity cost-saving/revenue potential, reaching 338.3% of electricity cost reductions, for residential space heating in the US. By achieving such performance, adopting smart control and PCM thermal storage is financially viable in 50% of the tested metro areas. The results reveal that the demand flexibility and cost-saving/revenue potentials of residential space heating in the US are further enhanced by higher volatilities in electricity prices. Active PCM thermal storage has lower energy efficiency but much higher energy flexibility than passive PCM thermal storage. Based on the findings, recommendations for integrating PCM thermal storage and smart control systems within residential space heating are provided.

Providing accurate crop yield estimations at large spatial scales and understanding yield losses under extreme climate stress is an urgent challenge for sustaining global food security. While the data-driven deep learning approach has shown great capacity in predicting yield patterns, its capacity to detect and attribute the impacts of climatic extremes on yields remains unknown. In this study, we developed a deep neural network based multi-task learning framework to estimate variations of maize yield at the county level over the US Corn Belt from 2006 to 2018, with a special focus on the extreme yield loss in 2012. We found that our deep learning model hindcasted the yield variations with good accuracy for 2006-2018 (R2 = 0.81) and well reproduced the extreme yield anomalies in 2012 (R2 = 0.79). Further attribution analysis indicated that extreme heat stress was the major cause for yield loss, contributing to 72.5% of the yield loss, followed by anomalies of vapor pressure deficit (17.6%) and precipitation (10.8%). Our deep learning model was also able to estimate the accumulated impact of climatic factors on maize yield and identify that the silking phase was the most critical stage shaping the yield response to extreme climate stress in 2012. Our results provide a new framework of spatio-temporal deep learning to assess and attribute the crop yield response to climate variations in the data rich era.

The global energy sector is now transitioning its structure towards carbon neutrality aided by renewable resource use. Despite its immense potential, solar energy contributes minimally to the global energy mix due to its intermittent nature and challenges with power demand fluctuations. Increased use of distributed solar sources alters market dynamics, necessitating conventional power plants to ramp up output during lower renewable energy production times and manage oversupply risks. Concentrated solar power (CSP) can contribute to grid decarbonization, but its high levelized cost of electricity (LCOE) impedes widespread adoption. This study proposes hybridizing CSP and photovoltaic (PV) technologies, aiming to leverage their synergy to maximize economic benefits. We develop a comprehensive process design framework that utilizes a robust multi-objective optimization (MOO) approach, which factors in techno-economic and environmental objectives while accounting for model uncertainty from resource prices and life cycle assessment indicators. Optimization results reveal that in Ivanpah, California, hybrid CSP + PV can reduce 41 % of LCOE and limit environmental impacts compared to standalone CSP plants. This robust framework also identifies design trends, such as a constant dependence on the PV field, and a trade-off between the installed area of the solar concentrators and the backup boiler operation. The optimal unit sizes, less susceptible to future market fluctuations and potential changes in the global warming potential (GWP) of technologies, contribute significantly to robust and sustainable energy planning decisions.

Abstract Computer-aided design of novel molecules and compounds is a challenging task that can be addressed with quantum computing (QC) owing to its notable advances in optimization and machine learning. Here, we use QC-assisted learning and optimization techniques implemented with near-term QC devices for molecular property prediction and generation tasks. The proposed probabilistic energy-based deep learning model trained in a generative manner facilitated by QC yields robust latent representations of molecules, while the proposed data-driven QC-based optimization framework performs guided navigation of the target chemical space by exploiting the structure–property relationships captured by the energy-based model. We demonstrate the viability of the proposed molecular design approach by generating several molecular candidates that satisfy specific property target requirements. The proposed QC-based methods exhibit an improved predictive performance while efficiently generating novel molecules that accurately fulfill target conditions and exemplify the potential of QC for automated molecular design, thus accentuating its utility.

Abstract Automotive electrification holds the promise of mitigating transportation-related greenhouse gas (GHG) emissions, yet at the expense of growing demand for critical metals. Here, we analyze the trade-off between the decarbonization potential of the road transportation sector and its critical metal requirement from the demand-side perspective in 48 major countries committing to decarbonize their road transportation sectors aided by electric vehicles (EVs). Our results demonstrate that deploying EVs with 40–100% penetration by 2050 can increase lithium, nickel, cobalt, and manganese demands by 2909–7513%, 2127–5426%, 1039–2684%, and 1099–2838%, respectively, and grow platinum group metal requirement by 131–179% in the 48 investigated countries, relative to 2020. Higher EV penetration reduces GHG emissions from fuel use regardless of the transportation energy transition, while those from fuel production are more sensitive to energy-sector decarbonization and could reach nearly “net zero” by 2040.

Urban Metabolism (UM) resembles living organisms in a community by consuming necessities and excreting waste. Using a circular economy concept with UM enhances energy and material circulation and closes waste loops. As a campus metabolism case study, this research tracks energy, food waste, and construction materials flows at Cornell University, Ithaca, NY. These sectors are interconnected at the food and energy systems regarding emissions. The study includes environmental and economic evaluations of proposed food waste mitigation and embodied emission estimation of building stocks. The results suggest that transitioning from natural gas to geothermal is a viable renewable option that can increase circularity. Regarding food waste, the proposed digestate composting reduces emissions by 16 to 39 metric tons of CO2e with a financial gain of 90 to 400 kUS$/y compared to traditional composting practices. Lastly, the comparative assessment of operational emissions from the energy sector and embodied emissions from stored building materials reveals a rising trend in the proportion of embodied emissions contribution, from 53.58 % in 2008 to 73.14 % in 2021. To offset the anticipated release of embodied carbon, Cornell requires strategies to manage construction waste and enhance circularity within this sector. To achieve the decarbonization goal, the campus metabolism study advances the understanding of energy and material flows, emissions, and offsets among the three sectors. This study can unlock sustainability pathways and accelerate Cornell’s progress toward carbon neutrality by 2035.

Communities are one of the essential elements of modern cities, significantly contributing to overall energy consumption and carbon emissions. To further support ongoing decarbonization initiatives, this study presents a novel Physically Consistent Deep Learning (PCDL)-based Model Predictive Control (MPC) approach for community energy management. This approach incorporates both building-to-grid interactions and on-site renewable energy resources. The PCDL model, starting with the definition of physics consistency, is constructed in line with the established physical laws that govern community thermal dynamics. Serving as a precise thermal load and indoor climate estimator, the PCDL model is then implemented in a centralized MPC to reduce the community energy cost and maintain comfortable indoor environments for buildings in the community. To verify the effectiveness and control performance of the proposed framework, we use a simulation case study of a student residential hall at Cornell University. The results demonstrate that the PCDL-based MPC is highly effective in maintaining comfortable indoor conditions and contributes to load shifting and shaving through its participation in the demand response service.

This study introduces a novel cyber-physical-biological system for energy management and crop production in plant factories. The goal is to minimize energy consumption while maintaining operational efficiency in sustainable food production. The CPBS integrates various control variables, including temperature, humidity, lighting, and CO2 levels, and accurately captures plant biological dynamics while predicting crop growth within controlled environments. To achieve this, physics-informed deep learning techniques are utilized to develop computationally efficient digital twins of the plant factory's internal microclimate and crop states. By incorporating Artificial Intelligence (AI), the control objectives are fine-tuned to minimize energy usage and resource expenses, ensuring sustainable crop production rates under different daytime scenarios. The results demonstrate an impressive 8.75 % reduction in energy usage compared to alternative control methods, enhancing operational efficiency and promoting sustainability in plant factories. The proposed approach offers a promising solution for achieving sustainable food production with minimized energy consumption in controlled environments.

A novel model predictive control (MPC) framework is proposed to optimize the energy management of integrated rooftop greenhouses and buildings, aiming to reduce control costs and the likelihood of climate violations. The centralized intelligent control approach employed for both the integrated rooftop greenhouse and the building ensures optimal conditions for crops and occupants. The integrated rooftop greenhouse utilizes waste heat and air from the building, resulting in reduced energy and CO2 consumption. The nonlinear dynamic models of temperature, humidity, and CO2 concentration for integrated rooftop greenhouse climate and building are first constructed. An integrated optimization problem is then formulated to acquire the optimal control decisions. The proposed MPC framework is implemented to regulate temperature, humidity, and CO2 level via controlling fans, pad cooling, shades, heating, ventilation and air conditioning systems, CO2 injection, and lighting systems. The indoor climate of an integrated rooftop greenhouse on a building in Brooklyn, New York, is controlled for the case study to show the advantages of the proposed nonlinear model predictive control framework. The results show that the average energy savings from the building to the integrated rooftop greenhouse amount to 15.2 % with the integration of the i-RTG and the host building under the proposed MPC framework.

Plant factories with artificial lighting (PFALs) are heralded as a potential solution to enhance the resilience of the food production system by amplifying productivity per unit area of land. However, PFALs typically demand higher resource consumption than traditional greenhouses or open-field farming. To optimize resource use in PFALs, we developed a nonlinear, automatic control system utilizing deep reinforcement learning (DRL). The proposed system implicitly learns the intricate dynamics of crop behavior and environmental variables, aiming to curtail resource, particularly energy consumption in PFALs, while ensuring that these environmental variables stay within desired operating parameters. We demonstrate the efficacy of the proposed DRL-based automatic control system through a case study: a shipping container PFAL situated in Ithaca, New York. We evaluate the ability of the proposed DRL system to reduce energy consumption by comparing its energy consumption to that of the conventional control method. Our findings indicate that, in typical Ithaca summer and winter conditions, the DRL-based control system can potentially decrease energy consumption by about 31 % and 23 % compared to the conventional control method.

As the growing world population drives demand for poultry meat and eggs, a surplus of poultry manure is produced, which, when applied to land, can lead to freshwater eutrophication and harmful algae blooms. To sustainably manage poultry manure, several waste-to-energy and soil amendment pathways have been developed, drawing from technologies such as pyrolysis, hydrothermal liquefaction, hydrothermal carbonization, and anaerobic digestion. To evaluate the “cradle-to-gate” life cycle impacts of the pathways, as well as how the introduction of these technologies affects the upstream and downstream markets for products, two different Life Cycle Assessments (LCA) methods are used. In an attributional LCA setting, direct land application has the lowest environmental impacts except prominently in the climate change and eutrophication categories, which are the categories for which its harms are most noticeable. In a consequential LCA setting, due to the consideration of offsets, direct land application only has the lowest energy resource use and ionizing radiation, while other technologies minimize all other impact categories. To meet environmental targets, the choice of management technology for poultry manure is significantly determined by the chosen system boundary, implying that careful consideration of environmental objectives is necessary for the circular economy.

Reshoring silicon photovoltaic manufacturing back to the U.S. improves domestic competitiveness, advances decarbonization goals, and contributes to mitigating climate change.

Transportation electrification plays a crucial role in mitigating greenhouse gas (GHG) emissions and enabling the decarbonization of power systems. However, current research on electric vehicles (EVs) only provides a fragmented examination of their impact on power system planning and operation, lacking a comprehensive overview across both transmission and distribution levels. This limits the effectiveness and efficiency of power system solutions for greater EV adoption. Conducting a systematic review of the effects of EVs on power transmission and distribution systems (e.g., grid integration, planning, operation, etc.), this paper aims to bridge the fragmented literature on the topic together by focusing on the interplay between transportation electrification and power systems. The study sheds light on the interplay between transportation electrification and power systems, delving into the importance of classifying EVs and charging infrastructure based on powertrain design, duty cycle, and typical features, as well as methods of capturing charging patterns and determining spatial-temporal charging profiles. Furthermore, we provide an in-depth discussion on the benefits of smart charging and the provision of grid-to-vehicle (G2V) and vehicle-to-grid (V2G) services for maintaining power system reliability. With the holistic systems approach, this paper can identify the main objectives and potential barriers of power transmission and distribution systems in accommodating transportation electrification at scale. Concurrently, it paves the way for a comprehensive understanding of technological innovation, transportation-power system decarbonization, policy pathways, environmental advantages, scenario designs, and avenues for future research.

Electrification and distributed energy resources (DERs) are vital for reducing the building sector's carbon footprint. However, conventional reactive control is insufficient in addressing many current building-operation-related challenges, impeding building decarbonization. To reduce building carbon emissions, it is essential to consider dynamic grid electricity mix and incorporate the coordination between DERs and building energy systems in building control. This study develops a novel model predictive control (MPC)-based integrated energy management framework for buildings with multiple DERs considering dynamic grid electricity mix and pricing. A linear, integrated high-fidelity model encompassing adaptive thermal comfort, building thermodynamics, humidity, space conditioning, water heating, renewable energy, electric energy storage, and electric vehicle, is developed. An MPC controller is developed based on this model. To demonstrate the applicability, the developed framework is applied to a single-family home with an energy management system through whole-year simulations considering three climate zones: warm, mixed, and cold. In the simulations, the framework reduces the whole-building electricity costs and carbon emissions by 11.9% - 38.3% and 7.2% - 25.1%, respectively, compared to conventional control. Furthermore, the framework can reduce percent discomfort time from 25.7% - 47.4% to nearly 0%, compared to conventional control. The framework also can shift 86.4% - 100% of peak loads to off-peak periods, while conventional control cannot achieve such performance. The case study results also suggest that pursuing cost savings is possible in tandem with carbon emission reduction to achieve co-benefits (e.g., simultaneous 37.7% and 21.9% reductions in electricity costs and carbon emissions, respectively) with the proposed framework.

In light of current energy policies responding to rapid climate change, much attention has been directed to developing feasible approaches for transitioning energy production from fossil-based resources to renewable energy. Although existing studies analyze regional dispatch of renewable energy sources and capacity planning, they do not fully explore the impacts of the energy storage system technology's technical and economic characteristics on renewable energy integration and energy transition, and the importance of energy storage systems to the energy transition is currently ignored. To fill this gap, we propose an integrated optimal power flow and multi-criteria decision-making model to minimize system cost under operational constraints and evaluate the operational performance of renewable energy technologies with multidimensional criteria. The proposed method can identify the most critical features of energy storage system technologies to enhance renewable energy integration and achieve New York State's climate goals from 2025 to 2040. We discover that lead-acid battery requires an additional 38.66 GW capacity of renewable energy sources than lithium-ion battery to achieve the zero carbon dioxide emissions condition. Based on the cross-sensitivity analysis in the multidimensional evaluation, the vanadium redox flow battery performs the best, and the nickel-cadmium battery performs the worst when reaching the zero carbon dioxide emissions target in 2040. The results of the proposed model can also be conveniently generalized to select ESS technology based on the criteria preferences from RE integration and energy transition studies and serve as a reference for ESS configurations in future energy and power system planning.

Since the energy sector is the dominant contributor to global greenhouse gas emissions, the decarbonization of energy systems is crucial for climate change mitigation. Two major challenges of energy systems decarbonization are renewable transition planning and sustainable systems operations. To address the challenges, incorporating emerging information and communication technologies can facilitate both the design and operations of future smart energy systems with high penetrations of renewable energy and decentralized structures. The present work provides a comprehensive overview of the applicability of emerging information and communication technologies in renewable transition and smart energy systems, including artificial intelligence, quantum computing, blockchain, next-generation communication technologies, and the metaverse. Relevant research directions are introduced through reviewing existing literature. This review concludes with a discussion of the industrial use cases and demonstrations of smart energy technologies.

This paper proposes a novel data-driven robust model predictive control (MPC) framework for a multi-zone building considering thermal comfort and uncertain weather forecast errors. The control objective is to maintain each zone's temperature and relative humidity within the specified ranges by minimizing the energy usage of the underlying heating system. A state-space model is developed to use a hybrid physics-based and data-driven method for the multi-zone building's temperature and relative humidity. The temperature and humidity RMSEs between the state-space model and the EnergyPlus-based model are less than 0.25 °C and 5.9%, respectively. The uncertainty space is based on historical weather forecast error data, which are clustered by using a k-means clustering algorithm. Machine learning approaches, including principal component analysis and kernel density estimation, are used to construct each basic uncertainty set and reduce the conservatism of resulting robust control action under disturbances. A robust MPC framework is built upon the proposed state-space model and data-driven disjunctive uncertainty set. An affine disturbance feedback rule is employed to obtain a tractable approximation of the robust MPC problem. Besides, the feasibility and stability of the proposed MPC are discussed in detail. A case study of controlling temperature and relative humidity of a multi-zone building in Ithaca, New York, USA, is presented. The results demonstrate that the proposed framework can reduce up to 8.8% of total energy consumption compared to conventional robust MPC approaches. Moreover, the proposed framework can essentially satisfy the thermal constraints that certainty equivalent MPC and robust MPC largely violate.

As greenhouses are being widely adopted worldwide, it is important to improve the energy efficiency of the control systems while accurately regulating their indoor climate to realize sustainable agricultural practices for food production. In this work, we propose an artificial intelligence (AI)-based control framework that combines deep reinforcement learning techniques to generate insights into greenhouse operation combined with robust optimization to produce energy-efficient controls by hedging against associated uncertainties. The proposed control strategy is capable of learning from historical greenhouse climate trajectories while adapting to current climatic conditions and disturbances like time-varying crop growth and outdoor weather. We evaluate the performance of the proposed AI-based control strategy against state-of-the-art model-based and model-free approaches like certainty-equivalent model predictive control, robust model predictive control (RMPC), and deep deterministic policy gradient. Based on the computational results obtained for the tomato crop's greenhouse climate control case study, the proposed control technique demonstrates a significant reduction in energy consumption of 57% over traditional control techniques. The AI-based control framework also produces robust controls that are not overly conservative, with an improvement in deviation from setpoints of over 26.8% as compared to the baseline control approach RMPC.

The clean energy transition plays an essential role in achieving climate mitigation targets. As for the transportation sector, battery and fuel cell electric vehicles (EVs) have emerged as a key solution to reduce greenhouse gasses from transportation emissions. However, the rapid uptake of EVs has triggered potential supply risks of critical metals (e.g., lithium, nickel, cobalt, platinum group metals (PGMs), etc.) used in the production of lithium-ion batteries and fuel cells. Material flow analysis (MFA) has been widely applied to assess the demand for critical metals used in transportation electrification on various spatiotemporal scales. This paper presents a quantitative review and analysis of 78 MFA research articles on the critical metal requirement of transportation electrification. We analyzed the characteristics of the selected studies regarding their geographical and temporal scopes, transportation sectors, EV categories, battery technologies, materials, and modeling approaches. Based on the global forecasts in those studies, we compared the annual and cumulative global requirements of the four metals that received the most attention: lithium, nickel, cobalt, and PGMs. Although major uncertainties exist, most studies indicate that the annual demand for these four metals will continue to increase and far exceed their production capacities in 2021. Global reserves of these metals may meet their cumulative demand in the short-term (2020–2030) and medium-term (2020–2050) but are insufficient for the long-term (2020–2100) needs. Then, we summarized the proposed policy implications in these studies. Finally, we discuss the main findings from the four aspects: environmental and social implications of deploying electric vehicles, whether or not to electrify heavy-duty vehicles, opportunities and challenges in recycling, and future research direction.

ABSTRACT: The objective of this study was to compare the application of iterative linear programming (iteLP), sequential quadratic programming (SQP), and mixed-integer nonlinear programming-based deterministic global optimization (MINLP_DGO) on ration formulation for dairy cattle based on Nutrient Requirements of Dairy Cattle (NRC, 2001). Least-cost diets were formulated for lactating cows, dry cows, and heifers. Nutrient requirements including energy, protein, and minerals, along with other limitations on dry matter intake, neutral detergent fiber, and fat were considered as constraints. Five hundred simulations were conducted, with each simulation randomly selecting 3 roughages and 5 concentrates from the feed table in NRC (2001) as the feed resource for each of 3 animal groups. Among the 500 simulations for lactating cows, 57, 45, and 21 simulations did not yield a feasible solution when using iteLP, SQP, and MINLP_DGO, respectively. All the simulations for dry cows and heifers were feasible when using SQP and MINLP_DGO, but 49 and 11 infeasible simulations occurred when using iteLP for dry cows and heifers, respectively. The average ration costs per animal per day of the feasible solutions obtained by iteLP, SQP, and MINLP_DGO were $4.78 (±0.71), $4.45 (±0.65), and $4.44 (±0.65) for lactating cows; $2.39 (±0.52), $1.48 (±0.26), and $1.48 (±0.26) for dry cows; and $0.98 (±0.72), $0.97 (±0.15), and $0.91 (±0.14) for heifers, respectively. The average computation time of iteLP, SQP, and MINLP_DGO were 0.59 (±1.87) s, 1.15 (±0.62) s, and 58.69 (±68.45) s for lactating cows; 0.041 (±0.070) s, 0.76 (±0.37) s, and 14.84 (±39.09) s for dry cows; and 1.60 (±2.90) s, 0.51 (±0.19) s, and 16.45 (±45.56) s for heifers, respectively. In conclusion, iteLP had limited capability of formulating least-cost diets when nonlinearity existed in the constraints. Both SQP and MINLP_DGO handled the nonlinear constraints well, with SQP being faster, whereas MINLP_DGO was able to return a feasible solution under some situations where SQP could not.

Quantum computing (QC) holds great promise to open up a new era of computing and has been receiving significant attention recently. To overcome the performance limitations of near-term QC, utilizing the current quantum computers to complement classical techniques for solving real-world problems is of utmost importance. In this article, we develop QC-based solution strategies that exploit quantum annealing and classical optimization techniques for solving large-scale scheduling problems in manufacturing systems. The applications of the proposed algorithms are illustrated through two case studies in production scheduling. First, we present a hybrid QC-based solution approach for the job-shop scheduling problem. Second, we propose a hybrid QC-based parametric method for the multipurpose batch scheduling problem with a fractional objective. The proposed hybrid algorithms can tackle optimization problems formulated as mixed-integer linear and mixed-integer fractional programs, respectively, and provide feasibility guarantees. Performance comparison between state-of-the-art exact and heuristic solvers and the proposed QC-based hybrid solution techniques is presented for both job-shop and batch scheduling problems. Unlike conventional classical solution techniques, the proposed hybrid frameworks harness quantum annealing to supplement established deterministic optimization algorithms and demonstrate performance efficiency over standard off-the-shelf optimization solvers.

Moving conferences from in-person to virtual and hybrid modes may have emissions reductions benefits. Here the authors find that the switch to virtual and hybrid conferencing reduces the carbon footprint by 94% when it comes to the switch to virtual conferencing, and 67% for hybrid conferences with carefully selected hubs.

Due to growing concerns regarding climate change and environmental protection, smart power generation has become essential for the economical and safe operation of both conventional thermal power plants and sustainable energy. Traditional first-principle model-based methods are becoming insufficient when faced with the ever-growing system scale and its various uncertainties. The burgeoning era of machine learning (ML) and data-driven control (DDC) techniques promises an improved alternative to these outdated methods. This paper reviews typical applications of ML and DDC at the level of monitoring, control, optimization, and fault detection of power generation systems, with a particular focus on uncovering how these methods can function in evaluating, counteracting, or withstanding the effects of the associated uncertainties. A holistic view is provided on the control techniques of smart power generation, from the regulation level to the planning level. The benefits of ML and DDC techniques are accordingly interpreted in terms of visibility, maneuverability, flexibility, profitability, and safety (abbreviated as the “5-TYs”), respectively. Finally, an outlook on future research and applications is presented.

Driven by the Paris Agreement for curbing global greenhouse gas (GHG) emissions, extensive research efforts are undertaken in pursuit of a carbon-neutral future. The energy sector is the first to bear the brunt of climate change due to its large share of GHG emissions, and a decarbonization strategy is urgently needed. In compliance with three main lines of Corporate Social Responsibility, this study explores the environmental, economic, and social impacts of hybrid energy systems decarbonization by integrating the environmental sustainability, techno-economic and sociological considerations through a multi-objective optimization framework. The proposed model simultaneously minimizes the total annualized costs, minimizes the scope 1, 2, 3 emissions for environmental justice, and maximizes medium opportunity hours by integrating life cycle assessment and economic input-output analysis. The optimization results provide guidance for sustainable design and operations of energy systems, identify environmental hotspots and associated mitigation strategies, and inform the resulting social benefits. The applicability of the proposed modeling framework is illustrated through case study using Cornell’s main campus.

While the implementation of renewable energy systems and model predictive control (MPC) could reduce non-renewable energy consumption, one challenge to building climate control using MPC is the weather forecast uncertainty. In this work, we propose a data-driven robust model predictive control (DDRMPC) framework to address building climate control with renewable hybrid energy systems under weather forecast uncertainty. The control and energy system configurations include heating, ventilation, air conditioning, geothermal heat pump, photovoltaic panel, and electricity storage battery. Historical weather forecast and measurement data are gathered from the weather station to identify the forecast errors and for the use of uncertainty set construction. The data-driven uncertainty sets are constructed with multiple machine learning techniques, including principal component analysis (PCA) with kernel density estimation (KDE), K-means clustering coupled with PCA and KDE, and Dirichlet process mixture model (DPMM). Lastly, a data-driven robust optimization problem is developed to obtain the optimal control inputs for a building with renewable energy systems. A case study on controlling a building with renewable energy systems located on the Cornell University campus is used to demonstrate the advantages of the proposed DDRMPC framework.

In this study, we develop a novel nonlinear model predictive control (NMPC) framework for climate control of buildings with renewable energy systems to minimize electricity costs. A nonlinear dynamic model of the building climate and renewable energy systems, including temperature, humidity, thermal comfort, geothermal heat pump, and solar panels, is first constructed based on mass and energy balance equations. The nonlinear dynamic model is then integrated into the proposed NMPC framework, which iteratively solves a nonlinear programming problem to generate the optimal control inputs, which minimize energy consumption and carbon footprints for sustainability. A simulation case study on controlling a building located on the Cornell University campus is conducted to demonstrate the capability of renewable energy sources to reduce building energy consumption using the proposed NMPC framework. The results show the NMPC framework could efficiently minimize total electricity cost and constraint violation for thermal comfort to 12.9 % with no more than 0.2 of violation on the predicted mean value index in different seasons. Implementing an electricity storage component could reduce the electricity cost by 19 %. The results indicate better sustainability for the smart building using sustainable energy sources and the NMPC framework.

Various pathways have been adopted for hydrogen production over the years, with technologies like steam methane reforming, often referred to as the 'brown' or 'blue' hydrogen, being the most prevalent. Different colors correspond to different production pathways, each having its life cycle greenhouse gas emissions. Using process simulation results as the major life cycle inventory data, life cycle impact assessment results for brown and blue hydrogen pathways have been presented. The results of the life cycle analysis study were utilized to determine the set of optimum operating conditions, with the global warming potential (GWP) corresponding to the blue hydrogen pathway at the optimal operating conditions being 33 % less than the optimal brown hydrogen counterpart. The operating cost for the brown hydrogen pathway at the optimal operating conditions can be reduced up to 69 % for the studied range of process operating conditions.

The existing COVID-19 pandemic has driven personal protective equipment use and consumption surge, leading to plastic pollution as most waste goes to landfills. Biodegradable polypropylene (PP) gowns claimed to have similar production costs as standard PP film ones might be more environmentally friendly due to the degradability after landfilling. The illustration of their sustainable end-of-life waste disposal options is lacking and requires a systematic comparison of their environmental impacts. A holistic life cycle assessment approach based on full-spectrum environmental indicators identifies the environmentally sustainable waste disposal options. Results illustrate the environmental benefits of landfill gas capture and utilization incorporated to landfilling biodegradable gown wastes by reducing 48.81 % land-use, 9.35 % greenhouse gas emissions from fossil sources, and 5.67 % from land-use greenhouse gas emissions, respectively. Despite these environmental advantages, industrial composting embodies lower environmental emissions than sanitary landfills for treating biodegradable gowns. Fossil-based gowns treated by landfills can have lower environmental impacts than composting biodegradable gowns in full-spectrum environmental indicators. The standard gown landfilling is identified as the environmentally sustainable disposal option.

Safe, efficient, and sustainable operations and control are primary objectives in industrial manufacturing processes. State-of-the-art technologies heavily rely on human intervention, thereby showing apparent limitations in practice. The burgeoning era of big data is influencing the process industries tremendously, providing unprecedented opportunities to achieve smart manufacturing. This kind of manufacturing requires machines to not only be capable of relieving humans from intensive physical work, but also be effective in taking on intellectual labor and even producing innovations on their own. To attain this goal, data analytics and machine learning are indispensable. In this paper, we review recent advances in data analytics and machine learning applied to the monitoring, control, and optimization of industrial processes, paying particular attention to the interpretability and functionality of machine learning models. By analyzing the gap between practical requirements and the current research status, promising future research directions are identified. Keywords: Big data, Machine learning, Smart manufacturing, Process systems engineering

As Cornell is transitioning to a carbon-free energy system by 2035, the campus energy system of the future will be based on 100% renewable energy sources. Specifically, the electricity will be mainly sourced from the local electric grid, which is expected to be carbon-free in the next two decades. Earth source heating and lake source cooling will serve as the major source for base-load renewable heating and cooling, respectively. Multiple geothermal wells will be drilled to meet the base-load heating demand. A conventional chiller will continue to provide auxiliary cooling sources for hot summer days in addition to the LSC system. Peak load will be fulfilled by introducing thermal energy storage and green hydrogen. This study addresses the economically optimal future design by developing a multi-period optimization model, to provide insights for the campus energy systems transition. A systematic life cycle assessment is adopted to examine the extent of carbon neutrality based on the optimization results.

This paper proposes a novel robust unit commitment (UC) framework with data-driven disjunctive uncertainty sets for volatile wind power outputs, assisted by machine learning techniques. To flexibly identify the uncertainty space based on wind power forecast error data with disjunctive structures, the uncertainty data are grouped using K-means and density-based spatial clustering of applications with noise following the optimal cluster number determined by the Calinski-Harabasz index. The disjunctive uncertainty sets are constructed accordingly as the union of multiple basic uncertainty sets, including conventional box and budget uncertainty sets, and data-driven uncertainty sets using Dirichlet process mixture model, principal component analysis coupled with kernel density estimation, and support vector clustering. The problem is formulated into a two-stage adaptive robust UC model with data-driven disjunctive uncertainty sets and with a multi-level optimization structure. To facilitate the solution process, a tailored decomposition-based optimization algorithm is developed. The effectiveness of the proposed framework is illustrated using an application on the IEEE 39-bus system. The proposed approach can reduce the price of robustness by 8-38 % compared to the conventional “one-set-fits-all” approaches. Benchmarking with stochastic programming indicates that the proposed framework can achieve the same or better economic performance with over 30 % less computational time.

Consequential life cycle assessment (CLCA) enables evaluating the environmental consequences of market dynamics overlooked by attributional life cycle assessment (ALCA). A consequential life cycle optimization (CLCO) framework is developed in this work to determine the economically optimal and most environmentally sustainable chemical recycling pathway for high-density polyethylene waste. The framework includes CLCA and techno-economic analysis (TEA), and the CLCO problem is formulated as a multi-objective mixed-integer nonlinear fractional programming problem (MINFP) that is effectively solved by an optimization algorithm. This multi-objective optimization problem aims to minimize the unit life cycle environmental impacts with maximum unit net present value (NPV). Environmental assessment results show that the total greenhouse gas (GHG) emissions evaluated by the CLCA approach are 14.22 % lower than those assessed by the ALCA approach. By evaluating the ReCiPe end-point score, CLCA further reduces the total environmental impacts corresponding to particulate matter formation by 17.37 %.

Model Predictive Control (MPC) has gained popularity in recent years and is widely adopted in building control. This study proposes a novel data-driven robust MPC to make the optimal heating plan, specifically for the multi-zone single-floor building. In this study, the room temperature and relative humidity (RH) will be highly valued in the optimisation decision. To better incorporate RH into the state-space model (SSM), the linear relations between RH and other room temperature parameters in the thermal zones are formulated, ensuring the better linear fitting of SSM to the original nonlinear model. Afterward, k-means clustered, principal component analysis (PCA), and kernel density estimation (KDE) based data-driven uncertainty set is constructed and applied to MPC. The other three kinds of MPC’s are compared to our proposed data-driven robust MPC (RMPC), including conventional RMPC, k-means clustered, data-driven RMPC (KM-DDRMPC), PCA and KDE based data-driven RMPC (PKDDRMPC). The results demonstrate that the optimality of our proposed k-means clustered, PCA and KDE based data-driven RMPC (KM-PKDDRMPC), which consumes 9.8 % to 17.9 % less energy in controlling both temperature and RH, compared to other data-driven robust MPC’s, and essentially follow the constraints which certainty equivalent MPC and conventional RMPC cannot conform.

The application of deep learning to a diverse array of research problems has accelerated progress across many fields, bringing conventional paradigms to a new intelligent era. Just as the roles of instrumentation in the old chemical revolutions, we reinforce the necessity for integrating deep learning in molecular systems engineering and design as a transformative catalyst towards the next chemical revolution. To meet such research needs, we summarize advances and progress across several key elements of molecular systems: molecular representation, property estimation, representation learning, and synthesis planning. We further spotlight recent advances and promising directions for several deep learning architectures, methods, and optimization platforms. Our perspective is of interest to both computational and experimental researchers as it aims to chart a path forward for cross-disciplinary collaborations on synthesizing knowledge from available chemical data and guiding experimental efforts.

Power system decarbonization is critical for combating climate change, and handling systems uncertainties is essential for designing robust renewable transition pathways. In this study, a bottom-up data-driven multistage adaptive robust optimization (MARO) framework is proposed to address the power systems’ renewable transition under uncertainty. To illustrate the applicability of the proposed framework, a case study for New York State is presented. Machine learning techniques, including a variational algorithm for Dirichlet process mixture model, principal component analysis, and kernel density estimation, are applied for constructing data-driven uncertainty sets, which are integrated into the proposed MARO framework to systematically handle uncertainty. The results show that the total renewable electricity transition costs under uncertainty are 21%-42% higher than deterministic planning, and the costs under the data-driven uncertainty sets are 2%-17% lower than the conventional uncertainty sets. By 2035, on-land wind and offshore wind would be the major power source for the deterministic planning case and robust optimization cases, respectively.

As an organic waste with high nutrient contents, the conventional landfill or incineration of poultry waste has led to growing concerns. This study applies spatial analysis to two scenarios of poultry waste valorization supply chain design in New York State (NYS). Ultimately, the optimal distribution of slow and fast pyrolysis biorefineries within NYS based on the operational and economic parameters, locations of major poultry farms, crude refineries and corn croplands is discerned and mapped. Additionally, the biochar transportation costs are distributed in the form of heat maps. The outcome indicates that building a single centralized biorefinery in NYS for both slow pyrolysis and fast pyrolysis is more economically feasible than building multiple smaller biorefineries for ten slow pyrolysis and eight fast pyrolysis distributed biorefineries in NYS.

With increasing concerns on environmental impacts of retired lithium-ion batteries (LIBs) and supply risk of critical materials, second life and recycling are considered as promising strategies to mitigate the environmental impacts of retired automotive LIBs. In this life cycle assessment (LCA) study, we investigate environmental benefits of second life and recycling methods on three types of widely adopted, commercialized automotive LIBs, including lithium manganese oxide (LMO), lithium nickel manganese cobalt oxide (NCM) and lithium nickel cobalt aluminum oxide (NCA). The results show that when subjected to second life, LIBs with both superior energy storage capacity and energy density generate more environmental benefits than LIBs that is less competitive in either working performance (31 % - 44 %, 16 % - 21 % and 50 % - 56 % reduction in environmental impacts). This is in line with the current trend towards the development of electric vehicles with larger energy storage capacity and longer battery life. We also found that direct cathode recycling outperforms hydrometallurgical and pyrometallurgical recycling regardless of battery chemistry. Among all recycling methods, only pyrometallurgical recycling is carbon intensive. The recovery of cathode active material and metals, including copper and aluminum, are critical to the environmental performance of batteries.

This study examines four prominent thermochemical conversion technologies such as slow pyrolysis (SP), fast pyrolysis (FP), gasification (GA) and hydrothermal liquefaction (HTL), for treating poultry litter in New York State (NYS). Nine cases involving various combinations of the four technologies and different downstream processing options such as bio-oil upgrading, Fischer-Tropsch conversion and combined heat and power generation are chosen based on the product composition and distribution. High-fidelity process simulations for each technological platform are performed to derive the mass and energy balance information. Calculations and breakdown of the equipment costs, capital costs, operating and maintenance costs and utilities are provided and compared extensively for each of the nine cases. The economic performance is further analyzed by calculating and comparing the resultant net present values (NPV), ranging from $10 M to $170 M (SP), $89 M to $314.5 M (FP), $28 M to $196 M (HTL) and $25 M to $234 M (GA). The greenhouse gas emission inventories are also compiled to understand the corresponding impacts of different downstream processing choices (ranging from 217 to 494.5 kg CO2-eq/t feedstock with both the pyrolysis technologies outperforming the others in most cases) and to highlight the trade-off with economic performance. Through sensitivity analysis, the influential factors requiring further investigation are identified and it is found that plant capacity and bio-oil yield are the most important parameters for the fast pyrolysis systems, biochar price for SP is the single most important parameter.

This work proposes a novel data-driven robust model predictive control (DDRMPC) framework for automatic control of greenhouse temperature and CO2 concentration level. The essential concept is to combine dynamic models of greenhouse temperature and CO2 concentration level with data-driven models that identify uncertainty in weather forecast error. By leveraging a machine learning approach, support vector clustering with weighted generalized intersection kernel, data-driven uncertainty sets for ambient temperature and solar radiation are constructed from historical weather data. A training-calibration procedure that tunes the size of uncertainty sets is implemented to ensure that data-driven uncertainty sets attain appropriate performance guarantee. In order to solve the optimization problem in DDRMPC, an affine disturbance feedback policy that provides tractable approximations of optimal control is utilized. A case study of controlling temperature and CO2 concentration level in a greenhouse is carried out. The results show that the proposed DDRMPC framework can prevent the greenhouse climate from becoming harmful to plant and fruit. DDRMPC approach ends up with 20 % less total economic cost than rule-based control strategy. The proposed DDRMPC approach also gives better control performance comparing to certainty equivalent MPC and robust MPC.

Global optimization in sustainable waste high-density polyethylene (HDPE) chemical recycling process is addressed under economic and environmental criteria. In this work, by far the most comprehensive superstructure of the HDPE recycling process with 867 processing routes is developed to produce valuable products from waste HDPE. Using the methodologies of life cycle assessment and techno-economic assessment, the superstructure optimization problem is then formulated as a multi-objective mixed-integer nonlinear fractional programming (MINFP) problem to address the sustainable waste HDPE recycling process with maximum unit net present value (NPV) and minimum levelized ReCiPe end-point score. A tailored parametric algorithm is utilized to efficiently convexify the fractional objective functions. With the help of the piecewise linearization method, nonlinear economic constraints can be linearized and effectively solved. This research proposes to use the ‘e-constraints’ method to obtain the Pareto-optimal curve. Results show that the optimal unit NPV ranges from -65 USD/t HDPE to 170 USD/t HDPE, and the levelized ReCiPe point of the most environmentally friendly design is 0.6 times of that of the most economically competitive design.

The conventional wastewater treatment facilities (WWTFs) are designed with the prime directive of simply purifying sewage by solids, pathogens, and nutrients removal. However, this ill-considered design philosophy stands chance to render the WWTFs unfavorably energy intensive and wasteful of nutrients. A promising route to raise the energy and nutrients utilization efficiencies of WWTFs is to retrofit the existing systems with smart upgrades for resource recovery. In this article, promising retrofits based on the benchmark design of WWTFs are proposed by virtue of process modifications and biosolids upcycling. This study directly assesses the economics and life cycle environmental impacts of the proposed retrofits of WWTFs, which are in turn compared to the reference design. Emerging thermochemical technologies and more complicated bioreactors are considered to maximize the biogas production, nutrient recovery, and biosolids upcycling. The results show the retrofit with hydrothermal liquefaction and secondary anaerobic digester demonstrates the highest net present value of - $ 97.93 MM, despite the highest annual production costs. The life cycle assessment results identify chlorination as the major contributor for most impact categories.

This work develops a novel dynamic material flow analysis (MFA)-based optimisation modelling framework for sustainable design of shale gas energy systems. This dynamic MFA-based framework provides high-fidelity modelling of complex material flow networks with recycling options, and it enables detailed accounting of time-dependent life cycle material flow profiles. Moreover, by incorporating a dimension of resource sustainability, the proposed modelling framework facilitates the sustainable supply chain design and operations with a more comprehensive perspective. The resulting optimisation problem is formulated as a mixed-integer linear fractional program and solved by an efficient parametric algorithm. To illustrate the applicability of the proposed modelling framework and solution algorithm, a case study of Marcellus shale gas supply chain is presented. The optimisation results help to identify clear trade-offs among economic, environmental, and resource performances in the shale gas energy system, and the MFA-based framework offers an effective way to find an optimal solution with well-balanced sustainability perfor

This paper aims to leverage the big data in shale gas industry for better decision making in optimal design and operations of shale gas supply chains under uncertainty. We propose a two-stage distributionally robust optimization model, where uncertainties associated with both the upstream shale well estimated ultimate recovery and downstream market demand are simultaneously considered. In this model, decisions are classified into first-stage design decisions as well as second-stage operational decisions. A data-driven approach is applied to construct the ambiguity set based on principal component analysis and first-order deviation functions. By taking advantage of affine decision rules, a tractable mixed-integer linear programming formulation can be obtained. The applicability of the proposed modeling framework is demonstrated through a case study of Marcellus shale gas supply chain. Comparisons with deterministic optimization models are investigated as well.

The petrochemical industry is an integral component of the global today. The petrochemical supply chain itself is a worldwide undertaking, and the final products that are offered to consumers will often travel thousands of miles from oil well to gas station pump. In the crude oil supply chain, there are three main stages: crude oil extraction, oil separation and refinement, and final product delivery. Within this supply chain, various companies will compete and attempt to maximize their profits by exploiting the demands and needs of the other companies within the supply chain. Each company or player in the crude oil industry has its own objective, and it will compete against other companies trying to pursue their individual objectives. In this work, the crude oil supply chain from oil well to refinery is modelled as a mixed-integer bilevel linear program (MIBLP) which accounts for conflicting objectives and interactions between different stakeholders. The composition, pricing, transportation distances, and environmental impacts of the different oils are taken into consideration when constructing the model. The resulting model is then applied to a U.S.-based refinery that purchases oil from various countries surrounding the Persian Gulf. The case study produces a pareto-optimal curve that displays the trade-off between environmental impact and profit. One intermediate point on the curve is selected to show how a 4.4 % decrease in profit can lead to a 3.0 % decrease in total kg CO2-eq produced.

Sustainability has become a key concern in process systems engineering. Renewable energy, such as biofuels, and renewable materials, such as bioproducts, could replace their non-renewable, petroleum-based counterparts. However, there remain many challenges in producing biofuels and bioproducts economically and efficiently. There are many different biomass feedstocks, processes to convert them, and many different possible biofuels and bioproducts to produce. Furthermore, prices and demands of biofuels and bioproducts are uncertain. The variation of price or demand of one bioproduct could influence price or demand of another, further complicating the problem. An approach that can identify economical, efficient, and sustainable biofuel and bioproduct production processes from the myriad possible options while also considering correlated and uncorrelated price and demand uncertainties of the final bioproducts is required. In this work, a data-driven decision-making framework is proposed for biomass processing network design that directly integrates machine learning with robust optimization. Principal component analysis (PCA) is used to identify latent uncertainties behind observed uncertainty data. A kernel density estimation approach captures probability distributions of the projected uncertainty data extracted from PCA. This uncertainty data analysis approach is applied to a bioconversion product and process network to identify cost-effective and environmentally-friendly biofuels and bioproducts production pathways. Our approach identifies a total annualized cost of $ 18.3M/y, 6 % lower than the cost found with conventional adaptive robust optimization.

In this paper, we address the sustainable design and operations of shale gas supply chains by proposing an integrated hybrid life cycle optimization (LCO) modelling framework. Unlike the traditional process-based LCO that suffers system truncation, the integrated hybrid LCO supplements the truncated system with a comprehensive economic input-output system. Meanwhile, the integrated hybrid LCO retains the precision in modelling major unit processes within the well-to-wire system boundary compared with the economic input-output-based LCO models. With the help of the integrated hybrid LCO framework, we can automatically identify the optimal sustainable alternatives in the design and operations of shale gas supply chains. To demonstrate the applicability, we present a case study of a well-to-wire shale gas supply chain located in the UK. According to the optimization results, the lowest levelized cost of electricity generated from shale gas is £51.8 /MWh, and the optimal life cycle GHG emissions are 473.5 kg CO2-eq/MWh.

This article addresses the optimal design of a geothermal binary trigeneration system for simultaneous cooling, heating, and power generation. A new processing network with alternative processing pathways is proposed. The major processing sections include geofluid production, Organic Rankine Cycle (ORC) evaporation, power generation, heat redistribution, cooling generation, condensing, and geofluid injection. Based on the superstructure, a cradle-to-gate life cycle analysis and techno-economic analysis are integrated with multiobjective optimization to simultaneously optimize the life cycle environmental impacts and the economic performance. To this end, a mixed integer nonlinear programming (MINLP) model is proposed to select the optimal processing pathway in terms of both greenhouse gas emissions and total annualized cost. A case study is also presented to cope with the seasonal demand of cooling, heating, and power to illustrate the application of the proposed modelling and solution methods.