Abstract: By accurately predicting industrial aging processes (IAP), such as the slow deactivation of a catalyst in a chemical plant, it is possible to schedule maintenance events further in advance, thereby ensuring a cost-efficient and reliable operation of the plant. So far, these degradation processes were usually described by mechanistic models or simple empirical prediction models. In order to accurately predict IAP, data-driven models are proposed, comparing some traditional stateless models (linear and kernel ridge regression, as well as feed-forward neural networks) to more complex stateful recurrent neural networks (echo state networks and long short-term memory networks). Additionally, variations of the stateful models are discussed. In particular, stateful models using mechanistical pre-knowledge about the degradation dynamics (hybrid models).

Abstract: The present invention relates to a computer-implemented method for activity based chemical substance formulation management of a chemical substance formulation comprising (i) receiving input data, preferably via an input unit (10), of at least one storage segment data defined by at least temperature and storage duration and an initial chemical substance activity value of said chemical substance formulation; (ii) determining, specifically calculating, a remaining activity value of the chemical substance formulation based on the storage segment data and the initial chemical substance activity value via a processing unit (20); (iii) providing a remaining activity value for the chemical substance formulation, preferably via an output unit (30), and (iv) managing said chemical substance formulation based on the remaining activity value of step (iii), said managing preferably comprising at least one of —providing a dosage recommendation based on the remaining activity value of the chemical substance formulation, pr

Abstract: By accurately predicting industrial aging processes (IAP), such as the slow deactivation of a catalyst in a chemical plant, it is possible to schedule maintenance events further in advance, thereby ensuring a cost-efficient and reliable operation of the plant. So far, these degradation processes were usually described by mechanistic models or simple empirical prediction models. In order to accurately predict IAP, data-driven models are proposed, comparing some traditional stateless models (linear and kernel ridge regression, as well as feed-forward neural networks) to more complex stateful recurrent neural networks (echo state networks and long short-term memory networks). Additionally, variations of the stateful models are discussed. In particular, stateful models using mechanistical pre-knowledge about the degradation dynamics (hybrid models).

Abstract: The present invention relates to a computer-implemented method for activity based enzyme formulation management of an enzyme formulation comprising (i) receiving input data, preferably via an input unit (10), of at least one storage segment data defined by at least temperature and storage duration and an initial enzyme activity value of said enzyme Predicted degradation formulation; (ii) determining, specifically calculating, a remaining activity value of the enzyme formulation based on the storage segment data and the initial enzyme activity value via a processing unit (20); (iii) providing a remaining activity value for the enzyme formulation, preferably via an output unit (30), and (iv) managing said enzyme formulation based on the remaining activity value of step (iii), said managing preferably comprising at least one of—providing a dosage recommendation based on the remaining activity value of the enzyme formulation, preferably via an output unit (30);—providing a residual shelf life indicator for said e

Abstract: In order to predict the future evolution of a health-state of an equipment and/or a processing unit of a chemical production plant, e.g., a steam cracker, a computer-implemented method is provided, which builds a data-driven model for the future key performance indicator based on the key performance indicator of today, the processing condition of today, and the processing condition over a prediction horizon.