Abstract: Disclosed are methods and systems that help identify critical variables for more efficient and robust plan validation process. An example embodiment is a computer implemented method of industrial process control. The example method includes receiving in computer memory a dataset including initial process parameters representing operational data of a subject industrial process, and, using filtering operations and grouping operations on the dataset, identifying a subset of the process parameters indicative of control data for controlling the subject industrial process. The example method further includes automatically applying the identified subset of process parameters controlling the subject industrial process.

Abstract: A computer-implemented method and system for process schedule reconciliation receives a scheduling model and an initial schedule for reconciliation, where the initial schedule includes projected plant data. Current plant data is imported into the system. The current plant data and projected plant data is processed using mathematical modeling techniques to identify event boundaries, stream flowrates associated with tanks and process units. The system builds an optimization model applying identified event boundaries, stream flowrates and pre-determined constraints along a period of time that includes priority slots to reconcile the projected plant data of the initial schedule with the current plant data, and then solves the optimization model to develop a reconciled schedule.

Abstract: Computer tool determines target feedstock for a refinery, process complex, or plant. The tool receives a dataset of market conditions and preprocesses the data based on properties of the plant. Using the preprocessed data and machine learning, the tool trains predictive models. Each predictive model calculates a breakeven value of a candidate feedstock for the given plant under an individual market condition. Different predictive models optimize for different market conditions. A trained predictive model is selected based on a current market condition. The tool applies the selected predictive model and determines whether a candidate feedstock is a target feedstock for the refinery under the current market condition.

Abstract: Computer tool determines target feedstock for a refinery, process complex, or plant. The tool receives a dataset of market conditions and preprocesses the data based on properties of the plant. Using the preprocessed data and machine learning, the tool trains predictive models. Each predictive model calculates a breakeven value of a candidate feedstock for the given plant under an individual market condition. Different predictive models optimize for different market conditions. A trained predictive model is selected based on a current market condition. The tool applies the selected predictive model and determines whether a candidate feedstock is a target feedstock for the refinery under the current market condition.

Abstract: Embodiments control industrial supply chains. An embodiment controls a supply chain formed of multiple nodes by obtaining an input-output model for each node. In response, for each node in the supply chain an equation-oriented model is generated using the obtained input-output model corresponding to the node. The generated equation-oriented models of the multiple nodes are integrated with a linking structure to form an optimization model of the supply chain. The optimization model of the supply chain includes a plurality of variables, e.g., interface variables indicating relationships between the generated equation-oriented models for each node in the supply chain. To continue, the optimization model of the supply chain is solved using a categorization of each of the plurality of variables to determine a value for at least one variable of the plurality. In turn, the method outputs a signal indicating the determined value.

Abstract: Computer implemented methods and systems incorporate physics-based and/or chemistry-based constraints into a model of a chemical, physical, or industrial process. The model is derived from a representative dataset of the subject process. The constrained model provides predictions of process behavior that are guaranteed to be consistent with incorporated constraints such as mass balances, atom balances, and/or energy balances while being less computationally intensive than equivalent first principle models. The constrained model can be constructed by matrix multiplication, namely multiplying the solution of an unconstrained linear model by a matrix that enforces the constraints. Improved process control models result, as well as improved process modeling and simulation models result.

Abstract: A computer-implemented method and system for process schedule reconciliation receives a scheduling model and an initial schedule for reconciliation, where the initial schedule includes projected plant data. Current plant data is imported into the system, and dynamic optimization data representing trends in process data at time-varied values for key process and operation parameters are identified. The current plant data and projected plant data is processed using mathematical modeling techniques to identify event boundaries, stream flowrates associated with tanks and process units. The system builds an optimization model applying identified event boundaries, stream flowrates, dynamic optimization data, key scheduling parameters and pre-determined constraints along a period of time that includes priority slots to reconcile the projected plant data of the initial schedule with the current plant data, and then solves the optimization model to develop a reconciled schedule.

Abstract: A computer-implemented method and system for process schedule reconciliation receives a scheduling model and an initial schedule for reconciliation, where the initial schedule includes projected plant data. Current plant data is imported into the system. The current plant data and projected plant data is processed using mathematical modeling techniques to identify event boundaries, stream flowrates associated with tanks and process units. The system builds an optimization model applying identified event boundaries, stream flowrates and pre-determined constraints along a period of time that includes priority slots to reconcile the projected plant data of the initial schedule with the current plant data, and then solves the optimization model to develop a reconciled schedule.

Abstract: A computer system provides improved classification of operating and scheduling plan data of a process plant. The system finds patterns in cases of the plan data and, based on the patterns, organizes the cases into a hierarchical structure of clusters representing distinct conditions. The system receives a dataset of cases of operating plan data represented by process variables. The system reduces a number of process variables representing operating plan data in the dataset by generating principal component(s) from values of the process variables for each case. The principal component(s) are latent variables generated to capture variation in conditions across the cases. For each case, the system determines a value for each generated principal component in the dataset. Using automated clustering or machine learning techniques, the system iteratively clusters the cases into a hierarchical structure based on the respective determined value of each generated principal component.

Abstract: Computer implemented methods and systems incorporate physics-based and/or chemistry-based constraints into a model of a chemical, physical, or industrial process. The model is derived from a representative dataset of the subject process. The constrained model provides predictions of process behavior that are guaranteed to be consistent with incorporated constraints such as mass balances, atom balances, and/or energy balances while being less computationally intensive than equivalent first principle models. The constrained model can be constructed by matrix multiplication, namely multiplying the solution of an unconstrained linear model by a matrix that enforces the constraints. Improved process control models result, as well as improved process modeling and simulation models result.

Abstract: Computer tool determines target feedstock for a refinery, process complex, or plant. The tool receives a dataset of market conditions and preprocesses the data based on properties of the plant. Using the preprocessed data and machine learning, the tool trains predictive models. Each predictive model calculates a breakeven value of a candidate feedstock for the given plant under an individual market condition. Different predictive models optimize for different market conditions. A trained predictive model is selected based on a current market condition. The tool applies the selected predictive model and determines whether a candidate feedstock is a target feedstock for the refinery under the current market condition.

Abstract: System and methods that provide a new paradigm for solving process system engineering (PSE) problems using embedded artificial intelligence (AI) techniques. The approach can facilitate process model building and deployment and benefits from emerging AI and machine learning (ML) technology. The systems and methods can define PSE problems with mathematical equations, first principles and domain knowledges, and physical and economical constraints. The systems and methods generate a dataset of recorded measurements for variables of the process, and reduce the dataset by cleansing bad quality data segments and measurements for uninformative process variables from the dataset. The dataset is then enriched by, for example, applying nonlinear transforms, engineering calculations, and statistical measurements.

Abstract: Disclosed are methods and systems that help identify critical variables for more efficient and robust plan validation process. An example embodiment is a computer implemented method of industrial process control. The example method includes receiving in computer memory a dataset including initial process parameters representing operational data of a subject industrial process, and, using filtering operations and grouping operations on the dataset, identifying a subset of the process parameters indicative of control data for controlling the subject industrial process. The example method further includes automatically applying the identified subset of process parameters controlling the subject industrial process.

Abstract: A computer system provides improved classification of operating and scheduling plan data of a process plant. The system finds patterns in cases of the plan data and, based on the patterns, organizes the cases into a hierarchical structure of clusters representing distinct conditions. The system receives a dataset of cases of operating plan data represented by process variables. The system reduces a number of process variables representing operating plan data in the dataset by generating principal component(s) from values of the process variables for each case. The principal component(s) are latent variables generated to capture variation in conditions across the cases. For each case, the system determines a value for each generated principal component in the dataset. Using automated clustering or machine learning techniques, the system iteratively clusters the cases into a hierarchical structure based on the respective determined value of each generated principal component.

Abstract: A computer system and method optimize feedstock selection planning for an industrial process by evaluating first and second stages at separate intervals throughout the planning process. Evaluating the first stage determines a set of robust feedstocks to procure on long-term contracts. The computer system and method solve, in parallel, multiple simulation cases of a non-linear model generated with different expectation values for uncertain input parameters related to selecting feedstocks to procure on long-term contracts. Probabilistic analyses on the solutions from the simulation cases, including the application of chance-constraints, determine the set of robust feedstocks to procure on long-term contracts. Evaluating the second stage determines a set of robust feedstocks to procure in the spot market, using the information from the first stage.

Abstract: A computer modeling apparatus and method optimize refinery operations. Included are an input module enabling user specification of inventory information including at least one rundown component, and user specification of refinery product commitments, and a processor routine executable by a computer and coupled to the input module. The processor routine, in response to the user specification, sequences refinery operations into a schedule that matches refinery commitments with inventory and unit rundown operations, wherein the refinery operations include refinery operations events, and provides on output a display of the schedule in a manner enabling optimized refinery operations.

Abstract: A computer system and method optimize feedstock selection planning for an industrial process by evaluating first and second stages at separate intervals throughout the planning process. Evaluating the first stage determines a set of robust feedstocks to procure on long-term contracts. The computer system and method solve, in parallel, multiple simulation cases of a non-linear model generated with different expectation values for uncertain input parameters related to selecting feedstocks to procure on long-term contracts. Probabilistic analyses on the solutions from the simulation cases, including the application of chance-constraints, determine the set of robust feedstocks to procure on long-term contracts. Evaluating the second stage determines a set of robust feedstocks to procure in the spot market, using the information from the first stage.

Abstract: A computer modeling apparatus and method optimize refinery operations. Included are an input module enabling user specification of inventory information including at least one rundown component, and user specification of refinery product commitments, and a processor routine executable by a computer and coupled to the input module. The processor routine, in response to the user specification, sequences refinery operations into a schedule that matches refinery commitments with inventory and unit rundown operations, wherein the refinery operations include refinery operations events, and provides on output a display of the schedule in a manner enabling optimized refinery operations.