Digital Twin-Based System and Method for Controlling the Operation of a Physical System

Abstract

A method (700) for digital twin-based operational control of a physical system is implemented by at least one processor. The method includes receiving (710) passenger throughput data corresponding to a building that is temperature-regulated by at least one chiller. The method includes estimating (720) a cooling load value as a function of time based on the passenger throughput data to maintain a specified indoor air temperature of the building. The method includes controlling (750) an on / off state of the at least one chiller based on the cooling load value.

Description

[Technical Field]

[0001] This disclosure relates generally to the management and control of systems, and more particularly to digital twin-based operational control of physical systems. [Background technology]

[0002] Modern societies increasingly rely on complex, interconnected infrastructures to function and enable sustainable growth. When this infrastructure functions properly and meets service delivery requirements, societal activities can continue in line with policy incentives, goals, and needs. Conversely, when such infrastructure struggles to meet service requirements due to external adverse factors, disruptions, or deterioration, the impact on societal well-being can be significant. System-level disruptions associated with this infrastructure can cause a variety of adverse effects, including economic losses, harm to human health, reduced social trust, reduced social cohesion, and dangerous environmental impacts. Of major concern are situations in which infrastructure disruptions cause cascading system failures and situations in which infrastructure disruptions permeate society as a whole, causing catastrophic and irreversible consequences. For example, in February 2011, Winter Storm Uri brought extremely cold temperatures (especially eight days of subzero temperatures) to a wide area of ​​North America, including parts of Canada, the United States, and northern Mexico. During Winter Storm Uri, cascading failures of interdependent infrastructure systems in the Texas power grid left millions of people without heat and electricity for extended periods of time.

[0003] Historically, various stakeholders have utilized risk assessment approaches to address the need to protect infrastructure from system-level disruptions. These approaches aim to characterize threats, assess vulnerabilities, and identify the direct and indirect (or unintended) impacts of disruptions. However, many threats to modern infrastructure (e.g., human-borne pathogens or regional extreme weather events) are difficult to predict. Summary of the Invention

[0004] The present disclosure provides digital twin-based motion control of physical systems.

[0005] In a first embodiment, a method for digital twin-based operational control of a physical system is implemented by at least one processor. The method includes receiving passenger throughput data corresponding to a building that is temperature regulated by at least one chiller. The method includes estimating a cooling load value as a function of time based on the passenger throughput data to maintain a specified indoor air temperature of the building. The method includes controlling an on / off state of the at least one chiller based on the cooling load value.

[0006] In a second embodiment, an electronic device for digital twin-based operational control of a physical system is provided. The electronic device includes a processor configured to receive passenger throughput data corresponding to a building that is temperature regulated by at least one chiller. The processor is configured to estimate a cooling load value as a function of time based on the passenger throughput data to maintain a specified indoor air temperature of the building. The processor is configured to control an on / off state of the at least one chiller based on the cooling load value.

[0007] In a third embodiment, a non-transitory computer-readable medium including program code for supporting digital twin-based operational control of a physical system is provided. The computer program includes computer-readable program code that, when executed by a processor of an electronic device, causes the electronic device to receive passenger throughput data corresponding to a building that is temperature-regulated by at least one chiller. The computer-readable program code causes the electronic device to estimate a cooling load value as a function of time based on the passenger throughput data to maintain a specified indoor air temperature for the building. The computer-readable program code causes the electronic device to control an on / off state of the at least one chiller based on the cooling load value.

[0008] Other technical features will be readily apparent to those skilled in the art from the following drawings, descriptions, and claims.

[0009] For a more complete understanding of the present disclosure, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which: [Brief explanation of the drawings]

[0010] [Figure 1] 1 illustrates an electronic device that supports digital twin-based operational control of a physical system according to the present disclosure. [Figure 2] In accordance with the present disclosure, a Social-Ecological Infrastructure System (SEIS) is illustrated. [Figure 3] In accordance with the present disclosure, a Resilience Matrix (RM) is illustrated. [Figure 4] In accordance with this disclosure, we present an exemplary method for understanding how interacting airport systems perform across four stages of a disruption event in various scenarios. [Figure 5A] 1 illustrates a central plant optimization system according to an embodiment of the present disclosure. [Figure 5B] 1 illustrates a user interface showing a timeline of values ​​representing the operation of an MPC process according to an embodiment of the present disclosure. [Figure 6] 1 illustrates an information analysis tool according to an embodiment of the present disclosure. [Figure 7] 1 illustrates a method for digital twin-based motion control of a physical system according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] 1-7 described below, and the various embodiments used within this patent specification to explain the principles of the present invention, are merely exemplary and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that the principles of the present invention may be implemented in any type of suitably arranged device or system.

[0012] As noted above, risk assessment approaches are a legacy solution that is difficult to apply to modern infrastructure. State-of-the-art standards for risk assessment offer significant benefits when "risk objects" (also known as "threats") are well characterized and their behavior is predictable, and when "objects at risk" (also known as "affected systems") are thoroughly mapped and understood based on their vulnerabilities and operational needs. However, when risks or risk objects are poorly characterized or novel, or when affected systems consist of multiple, intertwined, and nested dependencies, risk assessments can often overstate or underestimate the response necessary to address the associated risk objects. Unfortunately, for modern infrastructure, many threats (e.g., human pathogens, regional extreme weather events) are difficult to predict or are relatively novel in terms of scale or complexity. At the same time, modern infrastructure dependencies have become too complex to assess using siloed approaches such as risk assessment. An example of complexity that is too complex for a siloed risk assessment approach is a weather event that causes flight delays and makes airports difficult to access by ground transportation.

[0013] According to this disclosure, airports and airlines are prime examples of this challenge. As complex networks, air travel requires the symbiotic and sustainable operation of various systems and stakeholders, such as airlines (e.g., aircraft and crew), airports (e.g., infrastructure, communications, energy, workforce), security and risk compliance (including multiple state and federal agencies and regulations), and vendors (e.g., food and beverage, luxury goods), to achieve desired outcomes. Each component forms a vast and interconnected network that operates based on predetermined flight plans and strategic schedules driven by passenger and cargo demand.

[0014] These systems can be significantly impacted by adverse factors acting singly or in combination (e.g., climate change and infrastructure fatigue). The COVID-19 pandemic is one example of a global disruption to airport and airline systems. Reduced passenger demand, shifting workforces, and simultaneously shifting priorities to support the transportation of critical medical supplies forced airports and airlines to recover from and adapt to multiple, drastic changes in their operations, often with little advance notice and limited understanding of how society will respond to the pandemic in the long term. For airports, the COVID pandemic has demonstrated the critical importance of comprehensive, systems-driven assessment capabilities that encompass but go beyond traditional risk assessments. Importantly, resilience and sustainability are complementary elements, enabling airports to recover from and adapt to future uncertain disruptions while continuing to meet societal expectations for service and environmental impact and maintaining sustainable financial returns.

[0015] Unfortunately, few resilience-focused tools exist to guide airports and airlines in new ways of dealing with systemic risk. While practitioners have developed tools and capabilities to assess infrastructure resilience, airports are unique in that they must adhere to strict safety and security regulations, not to mention requirements to maintain a pleasant passenger experience, minimize harmful environmental impacts (e.g., emissions, noise, etc.), and maintain financial sustainability (cost per passenger).

[0016] With increased attention to airport operations due to the Bipartisan Infrastructure Investment and Jobs Act (BIL), U.S. airports have a unique opportunity to consider how they can adapt their operations and infrastructure to incorporate resilience as a philosophy and practice. To address this need, this disclosure proposes a three-pronged approach: (a) building and applying a principles-based understanding of resilience approaches in airports, (b) building on this development to develop a quantifiable multi-criteria decision-making framework, and (c) using these quantitative frameworks and airport systems frameworks to assess airport activities and disruptions in real time and under scenarios.

[0017] While this approach cannot predict or prevent all disasters, it will significantly benefit airport stakeholders and the broader society they serve by strengthening airports' ability to absorb, recover, and adapt to a wide range of threats and maintain sustainable operations. Furthermore, as air travel demand recovers from COVID-19 disruptions (the FAA's Aerospace Forecast predicts a 5% annual passenger growth rate from 2021 to 2041) and the range of adverse factors impacting airports becomes more diverse and unpredictable, it is essential that airports systematically invest in their resilience to disruptions. The first step to achieving this is to develop, test, and transfer quantifiable methodologies that enable airport operators to visualize systemic risk, assess airport operational and infrastructure performance under various historical and future scenarios, and comparatively evaluate improvements to mitigate disruptions and maintain cost requirements.

[0018] Previous research in various sciences (e.g., social sciences, ecology, and civil engineering) has advanced our knowledge of key issues in each field. For example, in the field of ecological research, risks and hazards, as well as the ability of systems to adapt to changing conditions, are well understood. The latest report of the Intergovernmental Panel on Climate Change (IPCC) clearly outlines the connection between anthropogenic sources of emissions in Earth system models and the urgency needed to prevent the Earth system from completely collapsing. Over the past five years, there has been an increasing call for interdisciplinary research to better understand the interactions between systems and enable scientists, policymakers, and industry stakeholders to coordinate their actions and realize the shared future outlined in the Brundtland Report. Embodiments of the present disclosure apply to physical systems, which can be small-scale ecosystems such as airports, or large-scale systems such as one or more university campuses, cities, states, or countries. For simplicity, embodiments of the present disclosure are described as applied to airports.

[0019] This disclosure examines existing risk-based approaches and identifies opportunities for defining, analyzing, and implementing systems-driven resilience (the ability of a system to recover and adapt after a disruption) in airports. There is an urgent need to rethink risk management practices in airports. Risks include direct threats, indirect threats, and downstream impacts. Airports are more than just landing and takeoff sites. The list of safety and societal responsibilities is vast and growing. This includes various safety requirements and the Sustainable Development Goals to which they must aspire.

[0020] While traditional approaches (e.g., inventory-based assessment methodologies) have been well-suited to managing past threats, they are not well-suited to address the system-wide challenges facing many airports on their own. Many new threats threaten airport sustainability in new and unpredictable ways that are difficult to remediate. Based on a resilience-based philosophy, the accompanying tools make it possible to quantify system behavior before, during, and after a disruption and evaluate the costs and benefits associated with various recovery strategies.

[0021] The disclosed system and method not only better protect airports from extreme system-wide disruptions but also provide a more financially secure foundation and infrastructure sustainability to future-proof airports against a wide range of unpredictable threats. This objective is achieved through three aspects of the disclosed system. First, the disclosed system applies a resilience matrix (e.g., RM300 in FIG. 3) to analyze physical systems. The resilience matrix structure includes a physical domain for modeling the state and capacity of physical systems such as large-hub international airports. Airports comprise complex socio-ecological systems with many interdependent functions, drivers, and requirements. Second, the disclosed system measures airport operations and performance in spatial and temporal contexts. In particular, the resilience matrix structure includes an information domain for receiving state data from actuators and measurement data from sensors, which are time-stamped and marked with spatial context data (e.g., physical location and / or position). The spatial and temporal components of the data structure accepted by the system of the present disclosure allow a multi-criteria decision support tool to analyze an airport based on spatial and temporal components. The data structure accepted by the system of the present disclosure also allows a threat-agnostic system to evaluate airport operations and performance against a variety of potential risk targets. As a third aspect, the system of the present disclosure predicts a timeline of physical system behavior (e.g., predicted state data and predicted measurement data as predicted outputs) in response to inputs including resource groupings and adverse agent groupings, using a model trained to make predictions based on training data (e.g., historical time-series data). In this disclosure, the training model may be simply referred to as the DT within the Digital Twin tool (DT) 114 of FIG. 1 . In particular, the selection of scenarios including chronic and acute adverse agents is based on historical data (including spatial and temporal components).), and a model is trained to learn how the physical system behaves in response to resource groupings and adverse factor groupings based on the selection. That is, the training data (and collected real-time data) are divided into resources, adverse factors, and outputs. The training data is marked with resilience stages and resilience domains according to a resilience matrix. In an operational context, the disclosed system can use the trained model to predict the timeline of physical system behavior in response to inputs including real-time resources and adverse factors. In an offline context for preparing for future adverse impact scenarios, the trained model can predict the behavior of the physical system in response to inputs including virtual resources and adverse factors. Thus, the disclosed DT tool includes a resilience-based decision support tool applied to real-time asset-based assessments. The disclosed DT tool enables repeated stress testing of airport functions under changing conditions and asset usage.

[0022] Embodiments of the present disclosure provide a novel resilience-based decision support approach for airports. Methodologically, such a toolkit must be able to evaluate various elements of airport infrastructure, operations, and policies, ranging from core infrastructure (core infrastructure and personnel) to enable takeoff and landing, to financial sustainability, environmental, and social impacts. Accordingly, the methods selected for this task (discussed below) include multi-criteria decision analysis and digital twin assessment (interdependence network analysis and asset management). These decision-making methods enable entities to incorporate uncertain or changing information into their decision-making structures and stress-test alternatives under various management and global driver scenarios. Furthermore, these decision-making tools, which align goals with program governance, provide direct and synergistic benefits to profitability, energy efficiency, ecosystem services, and public relations.

[0023] Cities and airports are microcosms of large-scale social systems, with nuanced behaviors based on the temporal and spatial dynamics of each region. The United Nations predicts that by 2050, approximately 70% of the world's population will live in urban areas, and competition for natural and physical resources will threaten and push interconnected ecosystems to their limits. One unique feature of airports is their governance, which simplifies modeling while providing flexibility for scalability across urban environments. Furthermore, in the United States, airports are critical transportation infrastructure that influences urban form and the structure of urban growth. Cities, in turn, influence airport size and operations. Infrastructure planning policies and actions by airports and airlines cause congestion in the ground transportation infrastructure surrounding airports and affect airline on-time performance. As demand for air travel increases, regional mobility shifts within cities will accelerate, and airports will be subject to rapid changes in new transportation technologies and the resulting infrastructure investments, policies, and revenues. Airport infrastructure provides access for aircraft to take off and land while unlocking the growth potential of local communities. Using ecosystem theory to understand complex interactions and system dynamics is a valuable contribution to integrating human and natural systems in the pursuit of sustainable growth.

[0024] Airports are often described as cities within large cities, a comparison that is important when studying the causal relationship between urbanization and environmental degradation. Theory based on the concept of urban ecosystems holds that sustainable environmental governance requires "...a deeper understanding of the causes and consequences of the complex patterns of interdependence that connect people and ecosystems within and across different scales." Applying this theory to airports will be essential in identifying measures to counter the ecological damage caused by the economic incentives that support urban growth.

[0025] The complexity of the airport ecosystem presents unique challenges in achieving long-term sustainable growth goals. To account for airport complexity, this disclosure analyzes airport components based on economic, administrative (or management system), and functional processes linked to physical, cognitive (or decision-making structure), ecological, and social domains. Economically, airports are expected to maintain competitive operating and maintenance costs to ensure airline costs improve competition in passenger service. From a functional perspective, urban infrastructure enables short-distance travel (i.e., automobile, bus, bicycle, and pedestrian) to meet the needs of residents, workers, and visitors. Meanwhile, airports efficiently connect people and goods over significant distances. To achieve mobility goals, airports must maintain infrastructure for aircraft landing, takeoff, and operational support. From a management perspective, most U.S. airports are government-owned, though substantially privately operated, and are subject to Federal Aviation Administration (FAA) regulations. The FAA also requires airports to obtain certification to ensure the safe, efficient, and environmentally responsible operation of the domestic airport system.

[0026] An airport system and its boundaries are generally characterized by four major components: airside, terminal, landside, and the surrounding community. Airside operations include airspace and airfield operations and are regulated by FAA Part 139. Airports are microcosms of cities, facilitating land development in and within their communities. Airport growth directly corresponds to community growth, as evidenced by increased departing and connecting passengers and increased terminal dwell time. Like cities, airports offer services and amenities such as retail, hotels, infrastructure, and entertainment, and are economic engines for businesses requiring convenient cargo and passenger logistics and operational services. Airports serve as a primary connection point for people, goods, and services, and like cities, economic centers foster new job opportunities and increased revenue streams.

[0027] Airside operations include handling arriving and departing aircraft. Many airports in the United States face similar challenges, including increasing demand for air travel (U.S. Bureau of Transportation Statistics), larger aircraft, and the threat of disruptive technologies to airport business models. Related technological threats include transportation network companies (TNCs) like Uber and Lyft, autonomous vehicles, and unmanned aerial systems (UAS). Airports also face the same federal rules and regulations regarding funding, safety, security, and other issues. However, most airports are managed to maintain sufficient investment in runways, terminals, and other services while maintaining high credit ratings, which airports use to qualify for preferential interest rates through the issuance of municipal bonds.

[0028] Following the deregulation of the aviation industry and the resulting competition among commercial airlines, urgent infrastructure investments have become necessary as a driver of efficiency. Airport operators around the world face increasing pressure from a variety of political, socioeconomic, technological, legal, and environmental factors. While airports' primary function is to connect people and goods globally, operational characteristics and risks have drawn scrutiny from regulators, airlines, and investors, as well as criticism from passengers and the communities they support. The world's population is projected to approach 10 billion by 2050, with the United States (US) one of nine countries expected to absorb more than 50% of the projected growth, with urban areas absorbing nearly all of the growth. The United Nations also predicts that the US population will grow by 78 million to over 400 million by 2060, with the number of people aged 65 and over doubling over the same period. Infrastructure investment remains a key priority in addressing urbanization trends, and the growing need for efficient and reliable airport operations is at the heart of this trend.

[0029] The complexity of decision-making regarding environmental projects is driven by trade-offs based on socio-political, environmental, ecological, and economic factors, each with different priorities and objectives. This perspective is evident when considering the historical influences, policies, and actions that have affected many airports, such as Dallas-Fort Worth (DFW) Airport's current sustainability and environmental compliance program.

[0030] Embodiments of the present disclosure enable the achievement of objectives such as determining the components, networks, and flows of critical and complex social-ecological and infrastructural systems (SEIS) that are essential for measuring and assessing the resilience and operational capacity of physical systems such as airports. Furthermore, embodiments of the present disclosure can identify the performance and decision criteria necessary to adequately describe and manage resilient operational capacity and test various combinations of long-term ecological and social trends and short-term disruptions that are most likely to affect the operational capacity and resilience of airport networks. As a solution to these objectives, the present disclosure provides a DT model of physical systems (e.g., complex airport systems) that serves as a framework for rethinking traditional approaches to topics such as sustainable development and resilience. This DT model incorporates social, ecological, and infrastructural aspects to measure and manage operational capacity and resilience. To address these objectives, the present disclosure provides an annotated list of financial, regulatory, risk, and operational criteria that reflect the interacting decision-making framework used at airports. To address these objectives, the disclosed DT tool identifies systems, key features, and metrics used to predict physical system behavior and feedback / response mechanisms. To address these objectives, the disclosure utilizes a combination of scenario analysis, resilience matrix analysis, and multi-criteria decision analysis to evaluate airport system-level responses to past, present, and future disruptions and visualize complex management tradeoffs under uncertainty. For example, airports offer a unique opportunity to study urban systems, with a rich and easily accessible data system for modeling.

[0031] A current limitation in responding to systemic risk in airport governance is the lack of a toolkit for quantitatively assessing airport resilience and vulnerability. For resilience to be successful as a cultural and philosophical contribution to airport operations, it must be defined and applied in a measurable way by airport managers. Embodiments of this disclosure achieve this goal by applying resilience methods to decision science to evaluate airport performance against disruptions across time (resilience stages) and space (resilience domains).

[0032] Decision science has long been used to assess the performance of infrastructure and ecosystem systems against a variety of uncertain adverse factors. Similarly, using the National Academy of Sciences' definition of resilience (the ability of a system to prepare for, absorb, recover from, and adapt to adverse events) to evaluate infrastructure system performance across physical, cognitive, informational, social, and other domains, decision science has been shown to serve as a platform for assessing the resilience of critical infrastructure. However, significant opportunities remain to adapt this approach, i.e., the "resilience matrix" (e.g., RM300 in Figure 3), and apply it to the unique context of airport infrastructure. Consequently, a decision science approach can facilitate airport resilience assessment by assessing airport operations against legal / safety requirements, financial constraints, and societal expectations.

[0033] In aviation infrastructure management, Airport Collaborative Decision-Making (A-CDM) has been proposed as a useful and transformative framework for integrating multiple regulatory agencies, airlines, and airport operations as an interdependent system. A-CDM is a platform that facilitates traditional risk analysis (threats, vulnerabilities, and consequences) while also including complementary assessments of airport resilience before, during, and after a disruption (resilience stages) and across various airport subsystems and dependencies (resilience domains).

[0034] A risk-based approach alone is insufficient for assessing the interdependencies of airport systems, and its usefulness tends to be limited to clearly defined situations and risks. While risk assessment remains valuable in airport risk management (the loss tolerance or loss absorption capacity of system components), adding a complementary resilience-based approach deepens understanding of how airport systems recover from various disruptions (known and unknown). And, most importantly, how inadequate recovery can lead to systemic and cascading losses in safety, money, and societal value. The DT tool 114 (Figure 1) disclosed here makes trade-offs between airport risk and resilience, thereby guiding airport managers in allocating scarce resources to improving their ability to tolerate or recover from disruptions.

[0035] Embodiments of the present disclosure explore the scale and complexity of airport decision-making systems, particularly the evaluation of tradeoffs and consequences. Embodiments of the present disclosure include the following three aspects. One aspect involves providing and evaluating a resilience-based decision support tool that analyzes airports based on spatial and temporal factors. Such analysis enables a threat-agnostic system to evaluate airport operations and performance against a variety of potential risk targets. Another aspect involves designing and adapting a resilience matrix to the unique context of a physical system (e.g., an airport) to evaluate the impacts of decisions across multiple domains (e.g., physical, cognitive, informational, social, and ecological) and the resulting tradeoffs. For example, to help policymakers select from a list of state alternatives (i.e., alternative ways of changing the state of a physical system to address an identified problem), the presently disclosed DT tool accepts a selection of stakeholder criteria to evaluate, predicts the value of each stakeholder criterion, and informs the user of the potential consequences of each listed state alternative. The DT tool uses multiple-criteria decision analysis (MCDM) to determine a rank acceptability index for each listed state alternative. This index represents the extent to which that state alternative satisfies selected stakeholder criteria. The DT tool provides a resilience-based multi-criteria decision support tool that analyzes physical systems (such as airports) based on spatial and temporal requirement factors. Comparing these rank acceptability indexes represents the tradeoffs compared to the simultaneous satisfaction of all selected stakeholder criteria. Third, another aspect involves completing a comparative analysis of current decision-making frameworks in complex airport ecosystems and testing the applicability of a new multi-criteria decision-making method that reflects the interplay and dynamic nature of risk-based versus resilience-based ideologies.For example, a DT tool of the present disclosure can predict a first behavior of a physical system according to a risk-based selection from a list of state alternatives, then predict a second behavior of the physical system according to a resilience-based selection from the list of state alternatives, and then display a comparison of the predicted first and second behaviors. By comparing the predicted results obtained from different hypotheses, the DT tool can quantify the outcomes of the listed state alternatives (i.e., the degree to which selected stakeholder criteria are met) to inform policymakers of the outcomes of the selected stakeholder criteria. These tools provide an up-to-date resource to help airport managers make resilience-related decisions.

[0036] In 2019, the U.S. Department of Homeland Security Cybersecurity and Infrastructure Agency (CISA) published a "Guide to Critical Infrastructure Security and Resilience." In this document, CISA defined critical infrastructure and identified four lifeline functions (i.e., transportation, water, energy, and communications). Regarding lifeline functions, CISA further stated, "Their reliable operation is so critical that the interruption or loss of one of these functions directly impacts the security and resilience of critical infrastructure within and across numerous sectors." Given the recent COVID-19 pandemic, the increasing number of multibillion-dollar disasters across the United States, and the attractiveness of civil infrastructure for terrorist acts, new, systematic, and rigorous approaches must be developed to protect critical infrastructure operations regionally, nationally, and globally.

[0037] A key challenge in resilient airport operations is the ability of an integrated sensor data analytics system (e.g., including sensors 130) to incorporate detailed and complex information streams into a real-time infrastructure decision-making environment. Requirements include evaluating airport operations, personnel, customer, and asset management at the highest granularity and under a variety of scenarios (historical, future, seasonal / normal operations). Using a digital twin for systems-level analysis facilitates the ability to evaluate an airport as a series of interconnected networks. A digital twin is an accurate representation of an asset, providing a robust, reciprocal, and ongoing connection between the physical entity (the airport) and its corresponding virtual copy. Through the digital twin 114 (Figure 1), embodiments of the present disclosure can examine "what if" scenarios to understand how different types of disruptions affect airport performance and whether airport subsystems can recover lost functionality. Importantly, this allows for sensitivity analyses to be performed and systems improvements and adaptations to be more effectively planned when such disruptions occur.

[0038] Applying resilience-based decision support tools to real-time, asset-based assessments can help airports optimize investments in modernizing aging infrastructure while also adapting to climate risks. For example, the disclosed DT tools enable repeated stress testing of airport functions under changing conditions and asset usage. Embodiments of the present disclosure provide one or more digital twins 114 (FIG. 1) that serve as a "gold standard" for experimenting and deploying resilience support tools while strengthening critical infrastructure (physical and cyberinfrastructure). Embodiments of the present disclosure provide a design strategy / framework for using digital twins to address airport resilience analysis and identify the general characteristics and uses of digital twins within and outside the aviation industry, as well as the capabilities they provide.

[0039] The time for action to reverse the course of anthropogenic emissions sources is overdue. Since 2020, the devastating impact of the COVID-19 pandemic has highlighted the need for a revamped approach to resilience in science and practice. Embodiments of the present disclosure leverage the practical application of scientific theory to an airport complex to help advance human understanding of how to study and improve the critical infrastructure systems we rely on to deliver consistent, reliable performance. A central plant is an example of an airport infrastructure system designed to operate in manual mode and regulate the indoor air temperature within a building based on the on / off status of a series of chillers. In manual mode, a fixed, predetermined cooling demand is set based on an expected environment (e.g., an outdoor temperature of 95°F throughout the day across five terminal buildings), and the cooling demand correlates to running multiple chillers at full cooling capacity. Due to the significant reduction in heat sources and personnel within each airport building due to COVID-19, the predetermined cooling demand was fixed and could not be adjusted to match the building's occupancy. To solve this problem, embodiments of the present disclosure provide an automated-control interface (AIC) that allows the central plant to operate in an automated control mode. Embodiments of the present disclosure provide an electronic device that implements the method to adjust cooling demand based on passenger throughput data, control the central plant using the adjusted cooling demand, and turn on N chillers for a specified period of time and turn off the remaining chillers for that period.

[0040] International airports are heavily regulated due to the potential for catastrophic failure. Furthermore, the COVID-19 pandemic has highlighted the vital role airports play in safely, efficiently, and quickly transporting people and cargo around the world as part of nationally recognized critical infrastructure and in their respective communities. However, to meet the needs of growing communities, airports must strengthen their efforts in resilience decision-making regarding operational and infrastructure investments. Embodiments of the present disclosure provide novel systems and methods for measuring and monitoring airport resilience. Furthermore, the present disclosure offers a unique perspective on operational indicators representing critical functions, other operational metrics, and new decision-support tools for accounting for the direct and indirect impacts of critical trade-offs, particularly in the ecosystem domain.

[0041] Embodiments of the present disclosure demonstrate the potential for digital twin tools to conduct more robust stress testing and enhance resilience-by-design strategies for critical infrastructure. Some embodiments of the present disclosure include a system in which electronic devices control physical components of a physical system to maintain a specified indoor air temperature for a building based on passenger throughput data.

[0042] FIG. 1 illustrates an exemplary electronic device 100 that supports digital twin-based operational control of a physical system in accordance with the present disclosure. The embodiment of the electronic device 100 illustrated in FIG. 1 is for illustrative purposes only. Other embodiments may be used without departing from the scope of the present disclosure. The electronic device 100 of FIG. 1 may be used, for example, in a social-ecological-infrastructural-system (SEIS) 200 of FIG. 2 to interact with and control the operation of a management system 218, an information component 224, and / or a technology system 252.

[0043] 1, electronic device 100 includes at least one processing device 102, at least one storage device 104, at least one communication unit 106, and at least one input / output (I / O) unit 108. Processing device 102 may execute instructions that may be loaded into memory 110. Processing device 102 may include any number of processors or other processing devices of any type in any suitable arrangement. Examples of types of processing device 102 include one or more microprocessors, microcontrollers, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or discrete circuits. In some embodiments, processing device 102 is a programmable logic controller (PLC) with user-selectable parameters.

[0044] Memory 110 and persistent storage 112 are examples of storage device 104 and represent any structure capable of storing and facilitating the retrieval of information (e.g., temporary or persistent data, program code, and / or other suitable information). Memory 110 may represent random access memory or other suitable volatile or non-volatile storage devices. Persistent storage 112 may include one or more components or devices that support longer-term data storage, such as read-only memory, a hard drive, a flash memory, or an optical disk. Persistent storage 112 is an example of a non-transitory computer-readable medium. Storage device 104 stores at least one digital twin (DT) tool 114 of a physical system; details regarding the DT tool 114 (referred to simply as a "digital twin") are described throughout this disclosure. In certain embodiments, the at least one DT tool 114 includes multiple DT tools 114, each for a different physical system. In certain embodiments, the DT tool 114 is a software platform running on a group of servers.

[0045] The DT tool 114 includes one or more physical models, such as the physical model of FIG. 5, and digital three-dimensional (3D) models of components within the physical system, which may include computer-aided drawing (CAD). The DT tool 114 includes one or more predictive models, such as model predictive control 502 of FIG. 5. To predict the behavior of a physical system, the DT tool 114 includes rules and relationships for the interactions of components within the physical system. For example, a valve allows fluid to pass when the valve's open state is open and prevents fluid from passing when the valve's closed state is closed. In some embodiments, the DT tool 114 includes the ability to automatically respond to demand response signals from the power grid, thereby adapting the physical system to provide demand response capabilities and enabling grid-interactive efficient buildings (GEBs).

[0046] The communication unit 106 supports communication with other systems or devices. For example, the communication unit 106 may support communication with an external system that provides information to the electronic device 100 for use in digital twin-based operational control of a physical system. The communication unit 106 may support communication via any suitable physical or wireless communication link, such as the network 120 or dedicated communications. For example, the communication unit 106 may be controlled by the processing device 102 to receive performance measurements, such as performance measurements from various components of the SEIS 200 of FIG. 2 . The various components of the SEIS 200 of FIG. 2 may be communicatively coupled to the communication unit 106 via the network 120. The performance measurements received by the communication unit 106 are input to the DT tool 114.

[0047] The I / O unit 108 enables the input and output of data. For example, the I / O unit 108 may provide a connection for user input via a keyboard, mouse, keypad, touchscreen, or other suitable input device. The I / O unit 108 may also send output to a display or other suitable output device. The I / O unit 108 may further support communication with various components of the SEIS 200 of FIG. 2. For example, the input and output data of the I / O unit 108 may be received from or output to one or more sensors 130, one or more databases 132, one or more controllers 134 of physical components, or IoT and other connected data sources 136 within the SEIS 200 of FIG. 2. The sensors 130 may include road cameras that provide the speed of vehicles on a roadway or toll gates that provide the number of vehicles passing through a vehicle lane or the open / closed status of the vehicle lane. The database 132 may store relevant data, such as the on / off status of a controller 134, which is related to temperature measurements from the sensors 130 that control the controller. As described in more detail below, the SEIS 200 of FIG. 2 may include various components 130 , 132 , 134 , 136 that are connected to the electronic device 100 at the I / O unit 108 or via the network 120 .

[0048] 1 depicts one example of an electronic device 100 that supports digital twin-based operational control of a physical system, although various modifications may be made to FIG. 1. For example, there are many different configurations of computing devices and devices, and FIG. 1 does not limit the disclosure to any particular computing or communication device or system.

[0049] FIG. 2 illustrates an exemplary Social-Ecological-Infrastructural System (SEIS) according to the present disclosure. The embodiment of the SEIS 200 shown in FIG. 2 is for illustrative purposes only. Other embodiments may be used without departing from the scope of the present disclosure. The SEIS 200 of FIG. 2 may include and be used in conjunction with the electronic device of FIG. 1 to support digital twin-based operational control of a physical system. For example, the electronic device 100 of FIG. 1 may identify important functional and domain-centric (e.g., physical, informational, ecological, or social) performance metrics of the SEIS 200 of FIG. 2 and monitor and analyze the operation and feedback / response mechanisms of the SEIS 200. However, it should be noted that the SEIS 200 of FIG. 2 may be used in conjunction with any other suitable devices and systems.

[0050] SEIS 200 includes a physical system 202 and its associated metabolic components 204, stressors and risks 206, and system outputs 208. By way of example only, physical system 202 is represented as a real-world airport, e.g., Dallas-Fort Worth (DFW) International Airport, which is part of the United States (US) critical infrastructure. Physical system 202 includes multiple domains, such as a physical domain 210, an information domain, an ecological domain 212, a social domain 214, an infrastructure domain 216, and a management system domain 218. Physical system 202 may have porous boundaries, represented by dashed lines surrounding physical domain 210.

[0051] Ecological domain 212 can affect social domain 214 and / or infrastructure domain 216, as represented by arrow 219A present in ecological domain 212. Arrows 219B and 219BB represent the impacts social domain 214 may have on other domains. Arrow 219C represents the impacts infrastructure domain 216 may have on other domains. Ecological domain 212 includes land, water, soil, wildlife, nutrients, and energy. Social domain 214 includes passengers, workers, designers, builders, operators, and policymakers. Infrastructure domain 216 includes airport infrastructure, terminal infrastructure, and ground infrastructure. Airport infrastructure includes airspace and airfields. For example, airport infrastructure includes parking spaces assigned to arrival / departure gates for aircraft 216A. Terminal infrastructure includes security checkpoints (e.g., access points to secure areas), airline business activities, and retail, food, and beverage outlets within one or more terminal buildings 216B or other buildings. The ground infrastructure includes roads, parking facilities, curbside areas, and pre-security areas. The management system domain 218 includes information and communications as a subset of the resilience matrix 300 in Figure 3. The social domain 214 and the management system domain 218 are domains where dynamic decisions are made, typically based on cost. Note that once trained, the DT 114 can be used to determine the tradeoffs that may arise if a proposed change is implemented in a particular aspect of the physical system 202.

[0052] Metabolism component 204 includes resources consumed (or required) by physical system 202 to respond to stressors / risks 206 that occur. Metabolism component 204 includes energy component 220, water component 222, information component 224, human component 226, and transportation component 228. Energy component 220 can be categorized as renewable energy 230 or non-renewable energy 232, and these categories of energy are input to physical system 202. In particular embodiments, energy component 220 is monitored by electronic device 100 of FIG. 1 to ascertain the amount of each type of energy input to physical system 202. In the example shown, 67% renewable energy and 33% non-renewable energy are input to physical system 202. An electrical power grid is physically coupled to the infrastructure of physical system 202 to provide power to operate the various components of infrastructure domain 216. For example, the power grid supplies electricity to terminal buildings for lights and fans, toll booths for cameras and gates, and a central plant for chillers, valves, and pumps. The power grid can be owned by a utility company and can transmit renewable energy 230 and non-renewable energy 232 alike.

[0053] Water component 222 can be classified as potable water 234 or reclaimed water 236, and these classifications of water are input to physical system 202. In certain embodiments, water component 222 is monitored by electronic device 100 of FIG. 1 to ascertain the amount of each type of water input to physical system 202. A water meter is an example of a sensor 130 that measures the water received by physical system 202 as potable water 234 or reclaimed water 236.

[0054] Information components 224 include regulatory information component 238, operational information component 240, and preference information component 242. In particular embodiments, electronic device 100 of FIG. 1 monitors various types of information components 238, 240, and 242 input to physical system 202. For example, regulatory information component 238 can be received from an external device connected to network 120, such as a computer belonging to the Federal Aviation Administration (FAA) sending a ground stop command to electronic device 100 belonging to an airport, requesting the airport to suspend operations and hold flights at their origin. Operational information component 240 can include operational states (e.g., on, off, out of order, under maintenance) of and measurements from physical components within physical system 202, such as a toll booth reader that outputs the number of vehicle crossings per unit time when on, and a ledger of vehicles entering and exiting physical system 202 through a toll booth. The ledger can include license plate numbers, entry and exit times, and toll booth lane IDs. The preference information component 242 can be user input received from an external device belonging to the person 226 indicating their preferences, such as a preference that the smartphone prefers to connect to a paid, secure Wi-Fi service provided by an airport's communications network rather than a free, unsecured Wi-Fi service.

[0055] In particular embodiments, people component 226 monitors to ascertain the number of people of each category entering physical system 202. In the example shown, 60,000 of the people entering physical system 202 are workers 244, and 75 million of the people entering physical system 202 are passengers 246. In particular embodiments, electronic device 100 of FIG. 1 monitors various categories of people entering physical system 202. Specifically, an employee badge can be scanned at a workplace to indicate to electronic device 100 where that particular worker has arrived, and a passenger ID can be scanned at an airline self-check kiosk to indicate to electronic device 100 that a passenger with a particular flight number is within a particular area of ​​the terminal building.

[0056] The transportation component 228 includes vehicles that carry passengers by road, vehicles that carry passengers by rail, and freight vehicles. Freight vehicles include light-, medium-, and heavy-duty vehicles that travel by land. Furthermore, freight vehicles include light-, medium-, and heavy-duty vehicles that pass through airports. Similarly, freight vehicles include aircraft. In certain embodiments, the electronic device 100 of FIG. 1 monitors various land-side vehicles 248 entering the land-side of the infrastructure domain 216 while carrying passengers and / or cargo. Similarly, the electronic device 100 of FIG. 1 monitors various airspace-side vehicles 250 entering the airspace of the infrastructure domain 216 as aircraft or ground support equipment (GSE). In a particular embodiment, the airspace side of the infrastructure domain 216 is modeled as a first airspace side DT tool 114, the land side of the infrastructure domain 216 is modeled as a second land side DT tool 114, and the terminals of the infrastructure domain 216 are modeled as a third building DT tool 114, which are implemented as software platforms, running on a group of servers, and interconnectable with each other.

[0057] Inputs 230-250 from metabolism component 204 to physical system 202 are measured by sensors 130 or reported over time by IoT and other connected data sources 136. To generate time-series data, sensor measurements and data from connected data sources 136 may be stored in database 132 in association with contextual information, such as the day of the week, date, or time of day. Data sources 136 may also be updated to include time-series data, such as the amount of inputs 230-250 between metabolism component 204 used or consumed by physical system 202 at a particular point in time or over a particular period of time. For example, periodically (e.g., daily, weekly, or monthly), the amount of each of energy components 220—electricity, natural gas, diesel, gasoline, and propane—used or consumed by physical system 202 may be added to database 132 as time-series data. Specifically, in certain embodiments, the amount of electricity consumption is received from sensors 130, such as a power meter, that are communicatively coupled to DT 114 (e.g., via network 120). In another embodiment, the amount of electricity consumption is obtained from a monthly bill document that has been subjected to object character recognition (OCR) to electronically recognize / extract the electricity consumption value, which updates the corresponding field in database 132. Similarly, electricity charges (e.g., utility bills) can be obtained from the monthly bill, and statistics can be calculated by obtaining utility charges for various months. Similarly, data source 136 can be configured to include time-series data, such as the amount of stressors / risks 206 that impose risks 260-266 on and affect the operation of components of physical system 202. This includes the impact on the amount of metabolism component 204 required by physical system 202 at a particular point in time or over a particular period of time. With respect to non-renewable energy 232, a fluid meter is an example of a sensor 130 that measures natural gas, diesel, gasoline, and propane received by physical system 202.

[0058] Stressors and risks 206 include technology 252, environmental degradation and climate change 254, and population growth 256. Within the physical system 202, stressors and risks 206 result in “fast” risks 262, “slow” risks 264, and unknown risks 266. An example of technology 252 as a risk factor is the commercialization of new 5G wireless technology, which prompted government regulations requiring buffers (e.g., Radius) to reduce potential signal interference. Another example of technology 252 is the commercialization of ride-sharing mobile applications, which cause human congestion and idling vehicles on the roadside. Increased exhaust fumes from idling vehicles also contributed to the deterioration of air quality within the physical system 202. An example of environmental degradation is extreme temperatures. An example of a fast risk 262 that is also a technology risk 252 is a cyberattack. Inputs to the physical system 202 from stressors 206 can be measured by sensors 130 or reported over time by IoT and other connected data sources 136. For example, data sources 136 may include predicted and measured time-based temperature values ​​from weather services, and known risks 260 may include precipitation in sub-freezing temperatures increasing demand for energy 220 for heating, or icy roads impeding routes for workers 244 and passengers 246 to reach airport physical system 202, resulting in reduced land-side traffic 248 entering through the boundary.

[0059] System outputs 208 leaving physical system 202 include benefits 268 and disadvantages 270. Benefits 268 include economic growth, jobs, tax revenue, and ecosystem services. Disadvantages 270 include air pollution, water pollution, and noise pollution.

[0060] While Figure 2 depicts one example of SEIS 200, various modifications may be made to Figure 2. For example, various components of Figure 2 may be combined, further subdivided, duplicated, omitted, or rearranged, and additional components may be added according to particular needs. As a specific example, the multiple domains of physical system 202 may include different or additional domains, such as a cognitive domain and an informational domain. As another specific example, physical system 202 of SEIS 200 may be a city, metropolitan area, or other transportation hub (e.g., a train station, bus station, or helicopter transportation hub).

[0061] Resilience is an important system characteristic for complex systems, and this work provides a foundation for implementing resilience with respect to airport operations. According to embodiments of the present disclosure, we leverage digital twin 114 (FIG. 1), AI, and other advanced technologies that provide visualization and analytical means for data in complex systems. Digital twin 114 allows for interfacing with other digital twins and integrates resilience measurements with respect to physical systems (e.g., airport operations).

[0062] Social-ecological infrastructure systems (SEIS), such as cities and transportation hubs, exhibit complex characteristics, including unpredictability, nonlinearity, interconnectedness, hierarchy, and "emergency." Furthermore, frequent disruptions due to a combination of rapid stressors, such as cyberattacks, extreme weather, and terrorism, and slower stressors, such as climate change and urbanization, hinder the primary objective of SEIS: to move people and cargo globally and meet performance targets, such as safety, efficiency, and cost. As urbanization trends continue, modern societies increasingly rely on complex, interconnected infrastructures to support sustainable growth. Furthermore, ports (e.g., airports and seaports) rely on sophisticated coordination between interconnected and interdependent domains (e.g., physical, information, and social) to achieve mutually agreed-upon outcomes.

[0063] International airports prioritize safe, efficient, and environmentally sound operations using traditional risk assessment methods that require risk identification and quantitative assessment of vulnerabilities and impacts. However, these traditional risk management methods are inadequate for addressing emerging threats (i.e., in terms of scale and complexity) and rapidly mitigating direct, indirect, and cascading impacts. As a result, the ability to develop resilient airport infrastructure is essential to meeting these multi-criteria functional objectives and the increasing demands of disaster response. Embodiments of the present disclosure provide resilience-focused tools and guidance to guide airports and airlines in implementing this new way of addressing systemic risk.

[0064] In some embodiments, digital twin tools 114 can be used in industry and academia due to their ability to analyze, visualize, and control complexity. Note that DTs 114 share three common characteristics: (a) a digital replica of a physical system, (b) bidirectional data exchange, and (c) connectivity across the entire lifecycle. Furthermore, with the rapid evolution of time-series data availability through the Internet of Things (IoT) and other connected data sources, DTs 114 can now analyze and visualize asset-level and system-level performance, and diagnose and minimize cascading failures. As a result, integrated DTs 114 can serve as a new decision support toolkit to help airports develop resilient decision-making and operational capabilities to prepare for, absorb, recover from, and adapt to adversity.

[0065] Embodiments of the present disclosure integrate resilience analysis using a resilience matrix (e.g., RM300 in FIG. 3) and related methods (e.g., method 400 in FIG. 4) with a DT architecture (e.g., DT114 in FIG. 1) to visualize the operational characteristics of critical functions, systems, and subsystems under various threat scenarios. Furthermore, the integrated model can help provide a systems perspective (essential for developing resilience capabilities). As a result, DT114 can facilitate the revitalization of the aviation sector by developing threat-agnostic resilience capabilities (e.g., extreme weather, terrorism, cyberattacks, urbanization, technological disruption).

[0066] FIG. 3 depicts an example resilience matrix (RM) 300 according to the present disclosure. Specifically, the RM 300 includes mapping system domains 302 across the event management cycle of resilience functions 304. In particular, the domains 302 include physical, informational, cognitive, and social. The RM cells provide guidelines for developing resilience indicators that, when combined, can measure the resilience of the entire system.

[0067] For example, using metrics derived from FIG. 3, the electronic device 100 can understand how interacting airport systems perform across four phases of a disruptive event for various scenarios (e.g., extreme weather, pandemic). Following a previous event management cycle of the resilience function 304, another event management cycle of the resilience function 304 is initiated. For example, the four phases of a disruptive event include a planning / preparation phase 306, an absorption phase 308, a recovery phase 310, and an adaptation phase 312. That is, each event management cycle of the resilience function 304 includes four phases 306-312.

[0068] Although Figure 3 depicts an example RM 300, various modifications may be made to Figure 3. For example, various components of Figure 3 may be combined, further subdivided, duplicated, omitted, or rearranged, and additional components may be added according to particular needs.

[0069] 4 depicts an example method 400 for understanding how interacting airport systems function across the four stages 306-312 (shown in FIG. 3) of a disruptive event for various scenarios in accordance with the present disclosure. For ease of explanation, method 400 is described as including the use of processing device 102 executing DT 114 of electronic device 100 of FIG. 1. However, method 400 may include the use of any other suitable device in any other suitable system.

[0070] 4, at block 402 of method 400, the processing device 102 configures the time series data and geospatial representation. That is, the electronic device 100 obtains a formatting standard for exporting the time series data and geospatial representation to DT software and configures the time series data and geospatial representation of the physical system 202 according to the formatting standard. In some embodiments, the formatting standard for exporting data is defined by the DT software to which such export is intended.

[0071] After constructing the time-series data and geographic 3D / 2D airport ecosystem representation with the DT software, the processing device 102 enables observation of interactions between systems and subsystems under various stressors (at block 404) and reveals or identifies important interconnections and dependencies (at block 406). For example, at block 404, the stressors 206 can be received as input, and in response, the processing device 102 can generate outputs representing the behavior of systems within domains 210-218 in the physical system 202, as well as finer-scale (or higher fidelity) outputs representing the behavior of subsystems within a particular domain. To identify important interconnections and dependencies, at block 408, the processing device 102 compares the behavior of the physical system 202 in response to a first group of stressors 206 (e.g., the 2011 Texas Freezing Storm extreme weather scenario) with the behavior of the physical system 202 in response to a second group of stressors 206 (e.g., the 2021 Winter Storm Uri extreme weather scenario). That is, in block 406, processing device 102 compares the behavior of the airport during extreme weather scenarios, such as the 2011 Texas freezing rain and the 2021 freezing rain (e.g., a "cold snap"), to identify critical interconnections and dependencies. Specifically, the ability of physical system 202 (e.g., an airport) to perform critical functions (e.g., clearance for takeoff and landing) is determined based on the critical interconnections and dependencies. For example, airport physical system 202 may shut down airside operations of infrastructure domain 216 if workers 244 performing work essential to the airport's critical functions cannot arrive at work because roads external to physical system 202 are impassable, preventing land-side vehicles 248 from entering physical system 202 (e.g., unable to enter airport toll booths). In this example, important interconnections and dependencies can be identified by sensors measuring the number of vehicles passing through toll booths, employee ID badge scans at work, information component 224 indicating external road closures, and DT 114 trained to compare normal ranges of toll booth passage, employee ID badge scans, and other applicable measurements.

[0072] From a discrete set of time series data, the model in DT 114 can be trained to learn how the physical system 202 actually behaved in response to a particular group of stressors (e.g., stressors 260-266) over a particular time period when a particular group of inputs (e.g., 230-250 from metabolism component 204) was received. Once trained, DT 114 can predict a timeline of the behavior of the physical system 202 based on the newly received group of inputs and the received group of stressors.

[0073] The DT 114 includes a system-level performance indicator. In certain embodiments, the system-level performance indicator includes multiple system-level performance indicators. In certain embodiments of method 400, at block 410, a user of the DT 114 selects a system-level performance indicator, and the DT 114 receives and utilizes the user-selected indicator. In certain embodiments of method 400, the system-level performance indicator is predetermined; for example, a designer of the DT 114 may select a system-level performance indicator that the DT 114 is then configured to utilize automatically (i.e., without user selection of an indicator). Method 400 may include a combination of these two embodiments; for example, a first system-level performance indicator is predetermined and a second system-level performance indicator is selected by a user. In certain scenarios, the physical system that the DT 114 models is an airport system, and thus the system-level resilience performance is the resilience performance of the airport system. The value of the system-level performance indicator can be the number of takeoffs and landings per unit time (e.g., per day per year).

[0074] In block 412, using modern data science techniques, the processing device 102 can analyze historical data from the DT software to identify or investigate how anomalies in different data sources may impact key performance indicators (KPIs) and resilience metrics (e.g., measurements of system-level performance indicators) in the physical system 202 (e.g., airport). An example of a resilience indicator is an indicator of system-level performance of the DT 114.

[0075] An additional step includes identifying, at block 414, which KPIs may be predictors (or early warning signals) of impending operational failures (i.e., critical events). For example, employees not reporting to the airport or passengers who purchased tickets not showing up at the airport are examples of anomalies in disparate data sources affecting takeoffs and landings, each of which may be early warning signals indicating that a reduction in takeoffs and landings may be imminent. Another similar example of a KPI is load factor, which indicates whether a takeoff and landing disruption is imminent. Load factor is the percentage of utilization of an aircraft's number of seats and / or amount of cargo space. If utilization falls below an expected normal range, a takeoff and landing disruption may be imminent. Generally, it is more economically rational to fly a single aircraft with a high load factor (e.g., a full load) than to fly multiple aircraft with low load factors.

[0076] At block 416, by visualizing and correlating different data streams, the processing device 102 can identify outliers or correlated patterns in the time series data that can potentially act as predictors of future events. The processing device 102 can identify these outliers or correlated patterns by utilizing pattern recognition algorithms on big data (e.g., the time series data in database 132). The computed features and statistical summaries of the time series data are used to analyze how these indicators correlate with important events of interest. Additionally, other state-of-the-art time series analysis, decomposition, anomaly detection, and forecasting techniques can be considered to provide more detailed insights and improve airport decision-making.

[0077] At block 418, the processing device 102 measures a system-level performance indicator, such as the number of takeoffs and landings over time. This indicator of system-level performance may be compared to a normal range, which may vary depending on the season, time of day, day of the week, etc. If the processing device 102 detects that the measurement of the system-level performance indicator is outside of the normal range for that indicator, the processing device 102 may use the RM 300 to determine what steps are applicable to the current behavior of the system-level performance indicator. Further, at block 420, the processing device 102 outputs the measurement of the system-level resilience performance indicator using the I / O unit 108 for output to an output device. For example, the measurement of the system-level resilience performance indicator may be output via a user interface displayed on a display. The user interface may display the measurement of the system-level resilience performance indicator with respect to any of the following: a first group of stressors, a second group of stressors, another group of stressors, and a group of inputs (e.g., 230-250 of the metabolism component 204).

[0078] While Figure 4 represents one example of a method 400 for understanding how interacting airport systems function across four stages of a disruptive event for various scenarios, various modifications may be made to Figure 4. For example, although shown as a sequence of steps, the various steps in Figure 4 may overlap, occur in parallel, occur in a different order, or occur any number of times.

[0079] As outlined in the IPCC's Sixth Climate Assessment Report, released in 2021, humanity is not on track to limit global warming to 1.5°C and calls for anthropogenic CO2 removal and net-zero global CO2 emissions. This call to action involves identifying strategies to transition away from fossil fuels and reduce current anthropogenic sources of emissions, while recognizing the importance of adapting critical infrastructure to the impacts of climate change extremes.

[0080] Historically, understanding, managing, and understanding related extreme events has been problematic and challenging. More specifically, the complex and contingent nature of related extreme events results in several attributes that differ from those associated with isolated or univariate extreme events. These include a high sensitivity to small changes in average climate conditions and poorly characterized data availability on key physical and social characteristics. These factors combine to heighten the risk of crossing unknown tipping points in response capacity. Because the linkages between extreme events are highly dependent on contextual factors such as season, location, and affected groups, careful impact-focused analysis, the use of high-order indicators, and the collection of high-quality, high-resolution impact data are essential components of progress in addressing them. This is an area where the power of emerging computational and communication technologies is likely to be strongly felt.

[0081] However, according to embodiments of the present disclosure, DT114 (Figure 1) can be a catalyst for strengthening airport decarbonization and resilience capabilities by defining the temporal and spatial characteristics of systemic risks and synergistic opportunities. This newly developed Integrated Resilience Matrix + DT tool (e.g., DT114) improves scientists' ability to analyze airport disruptions across a range of interdependent critical functions and ecosystem domains. Furthermore, as society transitions to renewable energy (e.g., RE230), DT can help understand cross-scale impacts and opportunities to adjust energy loads to take advantage of renewable energy intermittency (i.e., abundant nighttime wind and daytime sunlight versus daytime peak energy loads). Together, supporting outcomes can facilitate airport decision-making at the ecosystem level, more accurately forecast traffic and congestion, and optimize airport transportation networks based on aircraft movements and weather, while reducing energy consumption and associated emissions.

[0082] FIG. 5A illustrates a central plant optimization system 500 according to an embodiment of the present disclosure. The central plant optimization system 500 includes an infrastructure 501, such as the terminal infrastructure of the infrastructure domain 216 of FIG. 2, and optimizes the operation of the infrastructure 501. The infrastructure 501 in this embodiment is a central plant and is referred to as the central plant 501 for ease of explanation. The embodiment of the central plant optimization system 500 shown in FIG. 5A is for illustrative purposes only, and other embodiments may be used without departing from the scope of the present disclosure. In a particular embodiment, the central plant optimization system 500 of FIG. 5A includes a DT 114 of FIG. 1, such as the third building DT 114.

[0083] The central plant optimization system 500 includes a model predictive control (MPC) 502, which includes an optimization algorithm 504, an objective 506, a set of constraints 508, and a model 510 of the physical components of the central plant 501. The MPC 502 may be implemented by the electronic device 100 of FIG. 1. The MPC 502 in this example optimizes the sequencing of chillers and thermal energy storage systems within the central plant 501. In some embodiments, the optimization algorithm 504 is implemented by an external device, such as an optimization server system, which is communicatively coupled to the MPC 502 to receive as inputs the objective 506, constraints 508, and the physical model 510, process these inputs through the MPC process, including the optimization algorithm, and return cooling load values ​​to the MPC 502.

[0084] The MPC 502 sends control signals 514 to the central plant 501 to control its operation. The central plant 501 can operate in an automated control mode, in which an automated-control interface (AIC) 515 receives the control signals 514 from the MPC 502, converts them into a format that the physical components of the central plant 501 are configured to function in, and provides the converted control signals to the physical components of the central plant 501 to switch the on / off states of chillers, points, and other physical components of the central plant 501, control setpoints, or control other functions or settings of the physical system. Alternatively, the central plant 501 can operate in a manual mode, in which an operation control system of the central plant 501 receives the control signals 514 from the MPC 502 and causes output devices to output instructions for a user to operate controllers built into the console in accordance with the control signals 514. The operation control system can be associated with a control room console (e.g., including the information analysis tool 600 of FIG. 6 ). A user of the console can be a person authorized to operate controllers in the control room. The instructions may be output via a visual indicator, a visually displayed user interface, or an audio / voice user interface. For ease of explanation, FIG. 5A is described as if the central plant 501 were operating in an automatic control mode. The control signals 514 may control the operation of the central plant 501 by updating one or more control parameters for operating the physical components of the central plant 501. For example, the control signals 514 may initially establish, maintain, or switch the state of the physical components of the central plant 501.

[0085] MPC 502 generates control signals 514 based on inputs 516 and 518 and plant information feedback 520. Stressor inputs 518 to MPC 502 can be the same as inputs 522 to central plant 501 and represent stressors such as weather, occupancy, flight information, and lighting loads or other loads that affect the operation of central plant 501. Optimization algorithm 504 can generate control signals 514 based on inputs 516 and 518 and plant information feedback 520, and, for example, if optimization algorithm 504 is implemented as an optimization server system, based on objectives 506, constraints 508, and physics model 510. In particular, MPC 502 obtains (e.g., retrieves from storage device 104 or receives from an external data source) usage charge inputs 516 and stressor inputs 518 (e.g., representing stressors 206 of FIG. 2 ). The usage charge input 516 includes the price per unit of power (e.g., kilowatt-hour of energy and / or kilowatts of power) received from the utility company's power grid. The usage charge input 516 can be received from user input based on communication from the utility company or can be downloaded from the utility company's API for the subscriber.

[0086] The MPC 502 automatically responds to demand response signals received from the utility company's power grid. The usage charge input 516 can be or include the demand response signal. In response to the demand response signal, the MPC 502 can vary the control signal 514 to change energy consumption within the central plant 501. For example, physical components of the central plant 501 configured with demand response capabilities reduce or change energy consumption based on receiving the demand response signal via the power grid. Here, the utility company may be contractually permitted to send a demand response signal during a power generation shortage. As another example, a utility company's power grid operator may send a message (phone call or email) requesting a demand response within a specified time frame, and such a demand response signal may be received from a user input within a control room console. User input of the demand response signal may include typing a specified time frame, activating the demand response capability at the start of the specified time frame, or deactivating the demand response capability at the end of the specified time frame.

[0087] In some embodiments, instead of waiting to receive a demand response signal, the MPC 502 can help maintain grid resilience by reducing the load on the grid when the difference between available generating capacity and grid load is within a specified reserve margin. At the same time, instead of charging an energy storage device (e.g., an on-site battery or a thermal energy storage system (TES)) every time the state of charge drops to a low level, the MPC 502 can enhance grid resilience by increasing the load on the grid when available generating capacity far exceeds the grid load. In particular, the MPC 502 receives grid status data (e.g., input 516) corresponding to the grid to which it is physically coupled to power the M chillers of the central plant 501. The grid status data includes the generating capacity available to the grid and the grid load. The grid load can be a measurement, for example, measured and published periodically (e.g., every 5 or 10 minutes) by a power system operator. The grid load can be a forecast, for example, published as a five-minute short-term load forecast or a day-ahead forecast by a power system operator. Based on a determination that the grid load is outside the reserve margin for available generating capacity, the MPC 502 determines a charging window for charging an energy storage device (e.g., a TES) physically coupled to supply at least some electricity to the physical system 202. The reserve margin can be received from the power system operator or can be determined by the MPC 502 as a value greater than the reserve margin set by the power system operator. Similarly, the MPC 502 determines a power conservation window for discharging an energy storage device based on a determination that the grid load is within the reserve margin for available generating capacity. During the power saving time frame, the MPC 502 selects a discharge state of an energy storage discharge device (e.g., a valve that discharges fluid from the TES) to discharge energy so that the energy storage device at least partially maintains the building's specified indoor air temperature, and also reduces the period during which at least some of the M chillers operate in an on state.During the charging window, the MPC 502 selects a charging state for the energy storage charging device (the valve that draws temperature-conditioned fluid from the TES) so that the energy storage device does not release energy to at least partially maintain the building's specified indoor air temperature.

[0088] Stressor inputs 518 include real-time measurements and forecasts of weather, occupancy, flight information, and lighting and other loads. Stressor inputs 518 may be received from sensors 130 or data sources 136 of FIG. 1. Stressor inputs 518 may include various risks 260, 262, 264, and 266 from FIG. 2. Examples of stressor inputs 518 include indoor temperature and precipitation sensors located within the boundary of airport physical system 202, indoor temperature sensors located in one or more buildings at the airport, or forecast weather values ​​as a function of time provided by a weather service, such as minimum / maximum temperatures, precipitation amount, and probability of precipitation.

[0089] Examples of stressor inputs 518 include occupancy levels, e.g., measurements from a carbon dioxide (CO2) detector correlated with the occupancy level of a nearby space or the number of people breathing. Flight information is another example of stressor input 518 received from an external device that may be connected via network 120. For example, flight information provided by an airline may provide the number of passengers and crew boarding and disembarking each aircraft, the arrival and departure times of each aircraft, etc. This flight information is an example of passenger throughput data, which shows a timeline of the number of people entering or expected to enter a terminal building (e.g., through an arrival gate, entrance door, or inter-terminal rail) and the number of people exiting or expected to exit a terminal building (e.g., through a departure gate, exit door, or inter-terminal rail). People entering and exiting a building not only arrive and depart via aircraft arriving and departing from the building according to a flight schedule, but people may also depart and board ground transportation vehicles that may or may not operate according to a vehicle schedule. A flight schedule is an example of a vehicle schedule, and each aircraft is an example of a vehicle carrying passengers and crew.

[0090] Stressor inputs 518 include other loads, such as lighting loads. Light bulbs operating in a building generate and radiate heat. Similarly, electric or gas appliances that generate heat while operating in a building increase the heat gain value associated with the building. Examples of such appliances include coffee makers, ovens, fryers, stoves, griddles, refrigerators, hand dryers, computers, etc.

[0091] One or more of the terminal buildings of airport physical system 202 are temperature-regulated by central plant 501. Within central plant 501, physical components include a thermal energy storage system (designated "TES"), chillers (designated CH1-CH6), preconditioned air (PCA), tunnels (designated S. Tunnel and N. Tunnel), and EP HVAC. The TES can be a fluid thermal storage tank with a specified storage capacity, and fluid cooled overnight can be used (e.g., released from the TES) to cool physical system 202 during peak hours (times of high demand on the power grid or when physical system 202 has high electrical demand) during the day. The TES helps reduce the electrical demand of physical system 202. When the TES is filled with fluid to its specified storage capacity, the state of charge is 100%, but when an outlet valve is opened to allow fluid to be released (e.g., flow out) from the TES, the state of charge decreases. Similarly, the state of charge of the TES increases when the inlet valve opens, allowing cooled fluid from the chiller to flow into the TES (charging the TES). The physical components of the central plant 501 also include multiple P-pumps 530, multiple S-pumps 552, and valves 554 that control fluid flow between the various physical components. The physical components of the central plant 501 are connected to each of the terminal buildings so that temperature-conditioned fluid (e.g., water) is pumped from the central plant 501 by the S-pumps 552 through supply piping in the terminal building's HVAC system to maintain a specified indoor air temperature in the terminal building. The physical components of the central plant 501 receive return fluid from the return piping in the terminal building's HVAC system, where the fluid absorbs heat from the terminal building's indoor air before being returned to the TES and / or chiller. The number of P-pumps can be the same as the number of chillers. In some embodiments, multiple P-pumps 530 are connected in parallel with each other as a first parallel group, and multiple chillers are connected in parallel with each other as a second parallel group, with the first parallel group connected in series with the second parallel group.In some embodiments, each P-pump is connected to one of the chillers. These physical components operate in response to receiving control signals 514 and output plant information feedback 520. In some embodiments, multiple S-pumps 552 are connected in parallel with each other as a third parallel group, and a fourth parallel group consisting of the PCA, S. Tunnel, N. Tunnel, and EP HVAC connected in parallel with each other, with the third parallel group connected in series with the fourth parallel group.

[0092] The MPC 502 operates based on an objective 506, which can be prioritizing cost reduction or prioritizing energy reduction. In some embodiments, the objective 506 is adjusted based on a demand response signal or a usage rate input 516. The objective 506 selects to prioritize reducing the consumption of electricity from the power grid during each designated period of demand response or whenever demand response capability is activated. The objective 506 can be set by default to prioritize reducing operational costs, such as when demand response is disabled or outside of designated periods of demand response. To prioritize reducing energy consumption, the objective 506 can adjust the optimization algorithm 504 to prioritize discharging the TES to avoid consuming power over consuming power to charge the TES.

[0093] Constraints 508 provide operating limits to optimization algorithm 504 to ensure that control signals 514 operate the central plant within the operating limits given as constraints 508. Constraints 508 include the operating states of each of the physical components in the central plant, such as whether a chiller is on or off, or whether a TES is charging or discharging. Constraints 508 include a state change of the TES, indicating the percentage of the TES storage device full. Constraints 508 include one or more specified indoor air temperatures for the terminal buildings. Within constraints 508, each building can have its own specified indoor air temperature, or a group of buildings can share a common specified indoor air temperature. Constraints 508 include the flow rate of valve 554 and the electricity consumption rate of the chiller. Constraints 508 can be modified by plant information 520 received from the physical components. For example, if a particular chiller (e.g., CH3) is in a fault state, plant information 520 may modify constraint 508 to prevent MPC 502 from generating control signal 514 that is configured to activate the particular chiller that is in a fault state.

[0094] The physics module 510 includes a geographic information systems (GIS) model of the physical system 202, a building information management (BIM) model of the central plant 501, and BIM models of each of the terminal buildings in the infrastructure domain 216. For the central plant 501, the physics model 510 includes a three-dimensional (3D) model of the geometry of the central plant building and its physical components, including chillers, pumps, TES, etc. For each of the terminal buildings, the corresponding BIM model may include a 3D model of the building's geometry and a 3D model of the HVAC system within the building. In a specific example, the physics model 510 may include a 3D model of six chillers CH1-CH6, pipes connected to the chiller outlets and the inlets of pumps 552, and control valves that open or close to allow chilled fluid to flow out of the chillers or prevent chilled fluid from flowing out of the chillers. The physics model may include sensors within the central plant and visually represent measurements using a color spectrum. For example, a sensor measuring chilled fluid may display the outlet pipe from the chiller in a dark blue color, while a sensor measuring warmer return fluid may display the inlet pipe to the chiller in a different color. As another example, sensor readings may be displayed numerically on a user interface that displays a digital model.

[0095] MPC 502 utilizes physical model 510 to generate and output instrument curves 524 that are displayed on user interface 526. In this example, instrument curves 524 show CCP / CDP values ​​(Y-axis) versus part load ratios (X-axis) for corresponding physical components in central plant 501.

[0096] MPC 502 estimates a cooling load value as a function of time to maintain a specified indoor air temperature in a building (e.g., one or more terminal buildings at an airport) based on passenger throughput data corresponding to the building being temperature-controlled by central plant 501. To estimate the cooling load value, MPC 502 calculates a building occupancy based on the passenger throughput data of input 518, estimates a first heat gain value corresponding to the building occupancy, and estimates a cooling load value based on the first heat gain value. The building occupancy increases based on people entering and decreases based on people leaving. The first heat gain value may be estimated based on an assumption that each person has a heat generation rate, e.g., a heat generation rate corresponding to a normal human body temperature (e.g., approximately 97.6 degrees Fahrenheit). The cooling load value may be estimated based on the first heat gain value and the second heat gain value. The MPC 502 estimates a second heat gain value corresponding to at least one of solar radiation through semi-transparent surfaces of the building (e.g., glass windows, glass doors, skylights), heat conduction through the building's exterior surfaces (e.g., exterior walls and floors in contact with the ground or outdoor air), or outdoor air intrusion (e.g., opening and closing doors and windows, and ventilation).

[0097] The MPC 502 controls the on / off state of at least one chiller in the central plant 501 based on the estimated cooling load value. In embodiments in which the cooling load value is received from an external device, the MPC 502 controls the on / off state of the at least one chiller in the central plant 501 based on the received cooling load value. The MPC 502 controls the on / off state of the at least one chiller in the central plant 501 by determining N chillers to activate from a total of M chillers forming the at least one chiller based on the cooling capacities of the N chillers being greater than or equal to the cooling load value. In an example scenario, central plant 501 includes a total of M=6 chillers, each with a cooling capacity of 5,500 tons and an estimated cooling load value of 10,000 tons. MPC 502 can output control signals 514 to N=2 chillers to switch them on or maintain them in an on state, and to the remaining chillers (M=4 chillers) to switch them off or maintain them in an off state. Valves that allow fluid to flow out of the chillers can be switched to the same on / off state for each chiller. If constraint 508 indicates that a particular chiller (e.g., CH4) has failed or is out of service for maintenance, MPC 502 can control the P-pump and valves corresponding to the failed chiller to be in an off state. Alternatively, the P-pump and valves corresponding to the failed chiller can automatically switch to an off state without control from control signal 514 and transmit the off state as plant information 520 to allow the physics system to modify constraint 508. In other words, components within the central plant 501 that are configured to change state based on environmental conditions can send plant information 520 to update the MPC 502 according to the changed conditions.

[0098] MPC 502 utilizes optimization algorithm 504 to generate and output a timeline of values ​​representing the operation of MPC process 528, which is displayed in user interface 550 (also shown enlarged as FIG. 5B). Optimization algorithm 504 powers the functionality of MPC 502.

[0099] 5B, a user interface 550 includes a current time t, a past time before the current time, and a future time after the current time. The user interface 550 is divided to show a first timeline 532 and a second timeline 534. Both timelines 534 and 536 include a time slot 538 from the current time t to a first time t+1, but the time slot 538 is displayed as future in the first timeline 532 and as past in the second timeline 534. The first timeline 532 displays an input 540 (u k ) and the predicted output 542 (y k ) and the predicted output 542(y k ) may represent a desired output level. Second timeline 534 includes input 546 and output 548 that converges to function r(t) 544. In some embodiments, predicted output 542 may represent a cooling supply corresponding to manipulated inputs 540, which may each represent a cooling load value converted to a number of chillers that are on. Resilience is improved when predicted output 542 for future time slot 538 is determined at or before current time t, for example, when optimization algorithm 504 provides such input 540 to ACI 515 via control signal 514. Predicted output 542 develops and converges to function r(t) 544 more quickly and efficiently compared to output 548.

[0100] FIG. 6 illustrates an information analysis tool 600 according to an embodiment of the present disclosure. The information analysis tool 600 generates an output user interface (e.g., a display device associated with a facility manager) that shows a predicted outcome that may occur if a user selects a particular alternative selected from a set of alternatives. The information analysis tool 600 can receive a control signal 514 from the MPC 502 of FIG. 5A and cause an output device to output instructions for the user to operate a controller embedded in a control room console in accordance with the control signal 514. The control room console is associated with or is a component of an operational control system from a manufacturer of the physical components of the central plant 501. The control room console may be housed in a secure control room associated with the operational control system.

[0101] 7 illustrates a method 700 for digital twin-based operational control of a physical system according to an embodiment of the present disclosure. The embodiment of method 700 illustrated in FIG. 7 is for illustrative purposes only, and other embodiments may be used without departing from the scope of the present disclosure. Method 700 is performed by an electronic device or server system including at least one processor, such as electronic device 100 of FIG. 1. For ease of explanation, method 700 is described as being performed by electronic device 100 with processing device 102 executing MPC 502 of FIG. 5A.

[0102] At block 710, the electronic device 100 receives passenger throughput data corresponding to a building temperature-controlled by at least one chiller. The passenger throughput data includes vehicle schedules for vehicles arriving and departing from the building, such as schedules for airplanes, trains, buses, or ships arriving and departing from an airport, train station, bus station, or seaport. In some embodiments, the passenger throughput data includes vehicle schedules for automobiles (including personal automobiles), for example, when an individual's schedule indicates a worker in the terminal building or when a worker uses a parking reservation software system to specify a specific time when a parking space will be occupied for work in the terminal building. Further, the passenger throughput data includes at least one of a passenger load factor corresponding to the vehicle and a respective passenger load factor corresponding to each vehicle in the vehicle. At block 712, the electronic device 100 receives time-based weather data corresponding to the building. At block 714, the electronic device 100 receives power grid status data corresponding to a power grid physically coupled to power the M chillers that form the at least one chiller. The grid status data includes grid load and grid available generating capacity.

[0103] At block 720, the electronic device 100 estimates a cooling load value as a function of time to maintain a specified indoor air temperature for the building based on the passenger throughput data. In some embodiments, to estimate the cooling load value, the electronic device 100 calculates (at block 722) a building occupancy rate based on the passenger throughput data, estimates (at block 724) a first heat gain value corresponding to the building occupancy rate, estimates (at block 726) a second heat gain value, and estimates (at block 728) a cooling load value based on the first and second heat gain values. The second heat gain value corresponds to at least one of solar radiation through a semi-transparent surface of the building, heat conduction through an exterior surface of the building, or outdoor air infiltration. Although block 728 indicates that the cooling load value is estimated based on both the first and second heat gain values, it is understood that the electronic device 100 can also estimate one heat gain value (e.g., the first heat gain value) and estimate the cooling load based on the one heat gain value.

[0104] As indicated by block 730, in some embodiments, electronic device 100 estimates a cooling load value by providing the passenger throughput data as input to a model predictive control (MPC) process. The MPC process calculates a building occupancy rate based on the passenger throughput data, estimates a first heat gain value corresponding to the building occupancy rate, and estimates the cooling load value based on the first heat gain value. As an example, to provide the passenger throughput data as input to the MPC process, electronic device 100 transmits the passenger throughput data via a network connection to an external server system configured to process the input through the MPC process. As indicated by block 732, in such embodiments, electronic device 100 estimates a cooling load value by obtaining a cooling load value from the MPC process. As an example, to obtain a cooling load value from the MPC process, the electronic device receives a cooling load value from the external server system.

[0105] At block 750, the electronic device controls the on / off state of at least one chiller based on the cooling load value. Electronic device 100 controls the on / off state of at least one chiller by determining which N chillers to activate from among M chillers constituting the at least one chiller based on the cooling capacities of the N chillers being equal to or greater than the cooling load value. In embodiments including an automatic control mode, electronic device 100 controls the on / off state of at least one chiller by automatically controlling the operation control system to output control signals to N chillers to switch to or remain in an on state and to switch to or remain in an off state the remaining chillers of the at least one chiller. In embodiments including a manual mode, electronic device 100 controls the on / off state of at least one chiller by outputting instructions to a user via an output device associated with the operation control system to switch N chillers to an on state and switch the remaining chillers of the at least one chiller to an off state, or optionally to control other settings.

[0106] At block 760, the electronic device 100 determines a charging window for charging the energy storage device based on the determination that the power grid load is outside a margin for available power generation capacity. Also at block 760, the electronic device determines a power saving window for discharging the energy storage device based on the determination that the power grid load is within a margin for available power generation capacity.

[0107] In block 770, the electronic device 100 controls the charge / discharge states of the energy storage charging / discharging devices based on the charging time frame and the power saving time frame. In particular, during the power saving time frame, the electronic device 100 selects the discharge state of the energy storage discharging devices to discharge energy so that the energy storage devices at least partially maintain the specified indoor air temperature of the building, and reduces the period during which at least some of the M chillers operate in an on state. Furthermore, during the charging time frame, the electronic device 100 selects the charge state of the energy storage discharging devices to not discharge energy so that the energy storage devices at least partially maintain the specified indoor air temperature of the building.

[0108] 7 depicts an example method 700 for digital twin-based operational control of a physical system, although various modifications may be made to FIG. 7. For example, although shown as a sequence of steps, the various steps in FIG. 7 may overlap, occur in parallel, occur in a different order, or occur any number of times.

[0109] The above flowcharts represent examples of methods that may be implemented in accordance with the principles of the present disclosure, and various modifications may be made to the methods represented in the flowcharts herein. For example, while shown as a sequence of steps, the various steps in each figure may overlap, occur in parallel, occur in a different order, or occur any number of times. In other examples, steps may be omitted or replaced with other steps.

[0110] In some embodiments, various functions described in this patent document are implemented or supported by a computer program formed from computer-readable program code and embodied in a computer-readable medium. The phrase "computer-readable program code" includes any type of computer code, including source code, object code, and executable code. The phrase "computer-readable medium" includes any type of medium accessible by a computer, such as read-only memory (ROM), random-access memory (RAM), hard disk drive, compact disc (CD), digital video disc (DVD), or other type of memory. "Non-transitory" computer-readable medium excludes wired, wireless, optical, or other communication links that carry transient electrical or other signals. Non-transitory computer-readable media include media capable of permanently storing data and media on which data can be stored and subsequently overwritten, such as rewritable optical disks or erasable memory devices.

[0111] It may be advantageous to provide definitions of certain words and phrases used throughout this patent document. The terms "application" and "program" refer to one or more computer programs, software components, sets of instructions, procedures, functions, objects, classes, instances, associated data, or portions thereof, adapted for implementation in suitable computer code (including source code, object code, or executable code). The term "communicate" and its derivatives encompass both direct and indirect communication. The terms "comprise" and "have" and their derivatives mean an open-ended inclusion. The term "or" is inclusive, meaning "and / or." The phrase "associated with" and its derivatives may mean "comprise," "included in," "interconnect with," "contain," "contain in," "connect to," "communicable with," "coordinate with," "interleave," "parallel," "adjacent to," "coupled to," "having," "characterized by," "related to," and the like. The phrase "at least one of," when used in conjunction with a list of items, means that different combinations of one or more of the listed items may be used, and that only one item in the list may be required. For example, "at least one of A, B, and C" includes any of the following combinations: A, B, C, A and B, A and C, B and C, A, B, and C.

[0112] Nothing in this patent document should be read as implying that any particular element, step, or function is essential or critical to inclusion in the scope of any claim. Furthermore, no claim is intended to invoke 35 U.S.C. § 112(f) with respect to any of the appended claims or claim elements unless the precise terms "means for" or "step for" are expressly used following a participial phrase identifying that function. The use of terms such as (but not limited to) "mechanism," "module," "device," "unit," "component," "element," "member," "apparatus," "machine," "system," "processor," "processing device," or "controller" in the claims is understood and intended to refer to structures known to those skilled in the relevant art, as further modified or enhanced by the features of the claims themselves, and is not intended to invoke 35 U.S.C. § 112(f).

[0113] While this disclosure has described particular embodiments and generally associated methods, modifications and permutations of these embodiments and methods will be apparent to those skilled in the art. Accordingly, the above description of exemplary embodiments does not define or constrain the disclosure. Other modifications, substitutions, and alterations are also possible without departing from the spirit and scope of the disclosure, as defined by the following claims.

Claims

1. 1. A method implemented by at least one processor, comprising: receiving passenger throughput data corresponding to a building temperature regulated by the at least one chiller; estimating a cooling load value as a function of time to maintain a specified indoor air temperature for the building based on the passenger throughput data; controlling an on / off state of the at least one chiller based on the cooling load value; A method having the following.

2. the passenger throughput data includes at least one of a vehicle schedule for vehicles departing from and arriving at the building and a passenger load factor corresponding to the vehicles or a respective passenger load factor corresponding to each vehicle in the vehicle schedule; The estimating of the cooling load value includes: calculating a building occupancy rate based on the passenger throughput data; estimating a first heat gain value corresponding to the building occupancy; estimating the cooling load value based on the first heat gain value; Including, The method of claim 1.

3. The estimating of the cooling load value includes: providing the passenger throughput data as input to a model predictive control (MPC) process that calculates the building occupancy based on the passenger throughput data, estimates the first heat gain value corresponding to the building occupancy, and estimates the cooling load value based on the first heat gain value; obtaining the cooling load value from the MPC process; Further comprising: The method of claim 2.

4. Providing the passenger throughput data as the input to the MPC process comprises: transmitting the passenger throughput data over a network connection to an external server system configured to process the inputs by the MPC process; Obtaining the cooling load value from the MPC process includes: further comprising receiving the cooling load value from the external server system. The method of claim 3.

5. receiving time-based weather data corresponding to the building; further estimating the cooling load value by estimating a second heat gain value corresponding to at least one of solar radiation through a semi-transparent surface of the building, heat conduction through an exterior surface of the building, or outdoor air infiltration, and estimating the cooling load value based on the first heat gain value and the second heat gain value. The method of claim 2 further comprising:

6. Controlling the on / off state of the at least one chiller comprises: determining N chillers to be activated from among the M chillers constituting the at least one chiller based on the cooling capacities of the N chillers that are equal to or greater than the cooling load value; automatically controlling an operation control system to output control signals to the N chillers to switch to or remain in an on state and to the remaining chillers of the at least one chiller to switch to or remain in an off state, or outputting instructions to a user via an output device associated with the operation control system to switch the N chillers to an on state and switch the remaining chillers of the at least one chiller to an off state; Further comprising: The method of claim 1.

7. receiving power grid status data corresponding to a power grid physically coupled to power M chillers forming the at least one chiller, the power grid status data including a power grid load and a generating capacity available to the power grid; determining a charging window for charging an energy storage device based on a determination that the power grid load is outside a margin for the available power generation capacity; and determining a power conservation window for discharging an energy storage device based on a determination that the power grid load is within a margin relative to the available power generation capacity; and selecting a discharge state of an energy storage discharge device to discharge energy such that the energy storage device at least partially maintains the specified indoor air temperature of the building during the power saving time frame, thereby reducing a period of time during which at least some of the M chillers operate in an on state; selecting a state of charge of the energy storage and discharge device such that the energy storage device does not discharge energy during the charging time frame to at least partially maintain the specified indoor air temperature of the building; The method of claim 1 further comprising:

8. at least one processor, the at least one processor receiving passenger throughput data corresponding to a building that is temperature regulated by at least one chiller; estimating a cooling load value as a function of time to maintain a specified indoor air temperature for the building based on the passenger throughput data; Controlling the on / off state of the at least one chiller based on the cooling load value. It is configured as follows: Electronic devices.

9. the passenger throughput data includes at least one of a vehicle schedule for vehicles departing from and arriving at the building and a passenger load factor corresponding to the vehicles or a respective passenger load factor corresponding to each vehicle in the vehicle schedule; To estimate the cooling load value, the at least one processor: calculating a building occupancy rate based on the passenger throughput data; estimating a first heat gain value corresponding to the building occupancy; The cooling load value is estimated based on the first heat gain value. further configured as follows: The electronic device of claim 8 .

10. To estimate the cooling load value, the at least one processor: providing the passenger throughput data as input to a model predictive control (MPC) process that calculates the building occupancy based on the passenger throughput data, estimates the first heat gain value corresponding to the building occupancy, and estimates the cooling load value based on the first heat gain value; Obtain the cooling load value from the MPC process further configured as follows:

10. The electronic device of claim 9.

11. To provide the passenger throughput data as the input to the MPC process, the at least one processor: further configured to transmit the passenger throughput data over a network connection to an external server system configured to process the inputs by the MPC process; To obtain the cooling load value from the MPC process, the at least one processor: further configured to receive the cooling load value from the external server system. The electronic device of claim 10.

12. The at least one processor receiving time-based weather data corresponding to the building; and estimating the cooling load value by estimating a second heat gain value corresponding to at least one of solar radiation through a semi-transparent surface of the building, heat conduction through an exterior surface of the building, or outdoor air infiltration, and estimating the cooling load value based on the first heat gain value and the second heat gain value. further configured as follows:

10. The electronic device of claim 9.

13. To control the on / off state of the at least one chiller, the at least one processor: determining N chillers to be activated from among the M chillers constituting the at least one chiller based on the cooling capacities of the N chillers that are equal to or greater than the cooling load value; automatically controlling an operation control system to output control signals to the N chillers to switch to or remain in an on state and to the remaining chillers of the at least one chiller to switch to or remain in an off state, or outputting instructions to a user via an output device associated with the operation control system to switch the N chillers to an on state and switch the remaining chillers of the at least one chiller to an off state. further configured as follows: The electronic device of claim 8 .

14. The at least one processor receiving power grid status data corresponding to a power grid physically coupled to power the M chillers forming the at least one chiller, the power grid status data including a power grid load and a generating capacity available to the power grid; determining a charging window for charging an energy storage device based on a determination that the grid load is outside a margin for the available generation capacity; determining a power conservation window for discharging an energy storage device based on a determination that the power grid load is within a margin for the available power generation capacity; selecting a discharge state of an energy storage discharge device to discharge energy such that the energy storage device at least partially maintains the specified indoor air temperature of the building during the power saving time frame, thereby reducing a period of time during which at least some of the M chillers operate in an on state; selecting a state of charge of the energy storage and discharge device such that the energy storage device does not discharge energy during the charging time frame to at least partially maintain the specified indoor air temperature of the building; further configured as follows: The electronic device of claim 8 .

15. A non-transitory computer-readable medium embodying a computer program, comprising: The computer program includes computer-readable program code that, when executed by a processor of an electronic device, causes the electronic device to: receiving passenger throughput data corresponding to a building temperature regulated by the at least one chiller; estimating a cooling load value as a function of time to maintain a specified indoor air temperature for the building based on the passenger throughput data; controlling an on / off state of the at least one chiller based on the cooling load value; Execute Non-transitory computer-readable medium.

16. the passenger throughput data includes at least one of a vehicle schedule for vehicles departing from and arriving at the building and a passenger load factor corresponding to the vehicles or a respective passenger load factor corresponding to each vehicle in the vehicle schedule; The program code, which when executed causes the electronic device to estimate the cooling load value, when executed causes the electronic device to: calculating a building occupancy rate based on the passenger throughput data; estimating a first heat gain value corresponding to the building occupancy; estimating the cooling load value based on the first heat gain value; further comprising program code to execute 16. The non-transitory computer-readable medium of claim 15.

17. The program code, which when executed causes the electronic device to estimate the cooling load value, when executed causes the electronic device to: providing the passenger throughput data as input to a model predictive control (MPC) process that calculates the building occupancy based on the passenger throughput data, estimates the first heat gain value corresponding to the building occupancy, and estimates the cooling load value based on the first heat gain value; obtaining the cooling load value from the MPC process; further comprising program code to execute 17. The non-transitory computer-readable medium of claim 16.

18. The program code, which when executed causes the electronic device to provide the passenger throughput data as the input to the MPC process, when executed causes the electronic device to: and program code for transmitting the passenger throughput data over a network connection to an external server system configured to process the inputs by the MPC process. The program code, which when executed causes the electronic device to obtain the cooling load value from the MPC process, when executed causes the electronic device to: further comprising program code for receiving the cooling load value from the external server system.

20. The non-transitory computer-readable medium of claim 17.

19. The program code, when executed, further causes the electronic device to: receiving time-based weather data corresponding to the building; further estimating the cooling load value by estimating a second heat gain value corresponding to at least one of solar radiation through a semi-transparent surface of the building, heat conduction through an exterior surface of the building, or outdoor air infiltration, and estimating the cooling load value based on the first heat gain value and the second heat gain value. Execute 17. The non-transitory computer-readable medium of claim 16.

20. The program code, which when executed causes the electronic device to control the on / off state of the at least one chiller, when executed causes the electronic device to: determining N chillers to be activated from among the M chillers constituting the at least one chiller based on the cooling capacities of the N chillers that are equal to or greater than the cooling load value; automatically controlling an operation control system to output control signals to the N chillers to switch to or remain in an on state and to the remaining chillers of the at least one chiller to switch to or remain in an off state, or outputting instructions to a user via an output device associated with the operation control system to switch the N chillers to an on state and switch the remaining chillers of the at least one chiller to an off state; further comprising program code to execute 16. The non-transitory computer-readable medium of claim 15.

Images