Abstract: State-of-the-art systems used for plant monitoring and optimization fail to efficiently monitor and improve the performance of blast furnace ironmaking plants due to complexity of such plants. In addition, they attempt optimization without considering the operational stability of the blast furnace. The disclosure herein generally relates to industrial plant monitoring, and, more particularly, to a method and system for ironmaking plant optimization. The system determines an operational stability of the plant in terms of value of a determined Blast Furnace Stability Index (BFSI). Further, if the BFSI or one or more Key Performance Indicators (KPIs) of the plant deviates from corresponding threshold, then the optimization is done.

Abstract: This disclosure relates generally to system and method for molecular property prediction. The conventional methods for molecular property prediction suffer from inherent limitation to effectively encapsulate the characteristics of the molecular graph. Moreover, the known methods are computationally intensive, thereby leading to non-performance in real-time scenarios. The disclosed method overcomes the limitations of typical dynamic neighborhood aggregation (DNA) method by fusing the static edge attributes in determining the self-attention coefficients. In an embodiment, the disclosed method transforms the hidden state of a sink node by utilizing a neural-net function, which takes as input an aggregated single-message vector obtained by the self-attention mechanism and the self-attention mechanism transformed hidden state of the node.

Abstract: This disclosure relates generally to Error! Reference source not found.system and method for molecular property prediction. The conventional methods for molecular property prediction suffer from inherent limitation to effectively encapsulate the characteristics of the molecular graph. Moreover, the known methods are computationally intensive, thereby leading to non-performance in real-time scenarios. The disclosed method includes performing self-attention on the nodes of a molecular graph of different sized neighborhood, and further performing a shared attention mechanism across the nodes of the molecular graphs to compute attention coefficients using an Edge-conditioned graph attention neural network (EC-GAT). The EC-GAT effectively utilizes the edge characteristics in the molecular graph for molecular property prediction.

Abstract: Trajectory optimization is process of designing a trajectory of operating variables that optimizes measure of performance while satisfying a set of constraints, when the system moves from one state to another. It is very necessary to achieve optimization in real time. A system and method for real-time trajectory optimization has been provided. The trajectory optimization of a process can be performed in any dynamical automated system. The system is configured to optimize the trajectory in both online and offline mode. In the online mode, the system optimizes the trajectory of the process in real-time. The system has the ability to handle both machine learning and deep learning based time series models along with first principles based models represented by ordinary/partial differential equation or differential algebraic equation based dynamic models of the process to estimate process variables given the disturbance profile and the actuation profile of manipulated variables.

Abstract: An apparatus and a method for continuous solvothermal synthesis of nanoparticles, are provided. The apparatus includes an inlet section, a reactor section, a flexible quenching unit, and an outlet section. The inlet section separately receives reactants including the solvent and a precursor solution that are allowed to flow into the reactor section. The reactor section includes multiple spiral turns such that each of the spiral turns includes a helical channel followed by a counter-helical channel for enabling mixing of the reactants to cause solvothermal reactions between them. The counter-helical channel changes the direction of flow of reactants upon flow of said reactants from the helical channel to the counter-helical channel. The flexible quenching section enclosing a portion of the reactor section quenches a slurry formed as a result of the solvothermal reactions, wherein the slurry includes the nanoparticles of targeted characteristics. The outlet section facilitates withdrawal of the slurry.

Abstract: This disclosure relates generally to system and method for molecular property prediction. Typically, message-pooling mechanism employed in molecular property prediction using conventional message passing neural networks (MPNN) causes over smoothing of the node embeddings of the molecular graph. The disclosed system utilizes edge conditioned identity mapping convolution neural network for the message passing phase. In message passing phase, the system computes an incoming aggregated message vector for each node of the plurality of nodes of the molecular graph based on encoded message received from neighboring nodes such that encoded message vector is generated by fusing a node information and an connecting edge information of the set of neighboring nodes of the node. The incoming aggregated message vector is utilized for computing updated hidden state vector of each node.

Abstract: This disclosure relates generally to system and method for molecular property prediction using hypergraph message passing neural network (HMPNN). Typical MPNN architectures used for chemical graphs representation learning have limitations, including, inefficiency to learn long-range dependencies for homogeneous graphs, ineffectiveness to model topological properties of graphs taking into consideration the multiscale representations, and lack of anti-smoothing weighting mechanism to address graphs random walk limit distribution. Disclosed method and system HyperGraph attention-driven Hypergraph Convolution. The Hypergraph attention driven convolution, on molecular hypergraph results in learning efficient embeddings on the high-order molecular graph-structured data. By taking into account the transient incidence matrix, the induced inductive bias augments the scope of molecular hypergraph representation learning.

Abstract: This disclosure relates to method and system for training of deep learning based time-series models based on self-supervised learning. The problem of missing data is taken care of by introducing missing-ness masks. The deep learning model for univariate and multivariate time series data is trained with the distorted input data using the self-supervised learning to reconstruct the masked input data. Herein, the one or more distortion techniques include quantization, insertion, deletion, and combination of the one or more such distortion techniques with random subsequence shuffling. Different distortion techniques in the form of reconstruction of masked input data are provided to solve. The deep learning model performs these different distortion techniques, which force the deep learning model to learn better features. It is to be noted that the system uses a lot of unlabeled data available cheaply as compared to the label or annotated data which is very hard to get.

Abstract: In applications such as adaptive learning of physics-based and data-driven models associated with industrial plants, the models are corrected periodically by taking into consideration the dynamic changes occurring in plant conditions and related data. However, accuracy of adaptive learning depends on accuracy of ground truth data being used as reference data. The disclosure herein generally relates to data preprocessing, and, more particularly, to a method and system for ground truth profile correction and instance selection. The system performs a ground truth profile correction for ground truth profiles having a Profile Deviation Index (PDI) value exceeding a threshold of distortion, to reduce the PDI value, and in turn reduce the distortion in the ground truth profiles. Further, the system performs a data instance selection to identify and remove outliers, and the data that remains after the data instance selection may be then used for applications such as model generation or retuning.

Abstract: This disclosure relates generally to for time lag identification in an industry. The disclosure proposes to monitor an industry continuously at real time to identify one or more parameters from plurality of sources (processes/units/plants) and a time delay or delayed performance or functional impact the identified parameter has on a plurality of Key Performance Indicator (KPI). The proposed time lag identification is performed using one-time lag identification from the proposed plurality of time lag identification techniques that include an individual time lag identification technique, a group-wise time lag identification technique and group-wise/individual time lag identification technique. Further the time lag identification is performed based on domain knowledge as well as data driven techniques.

Abstract: This disclosure relates generally to a method and system for real time monitoring and forecasting of fouling of an air preheater (APH) in a thermal power plant. The system is deploying a digital replica or digital twin that works in tandem with the real APH of the thermal power plant. The system receives real-time data from one or more sources and provides real-time soft sensing of intrinsic parameters as well as that of health, fouling related parameters of APH. The system is also configured to diagnose the current class of fouling regime and the reasons behind a specific class of fouling regime of the APH. The system is also configured to be used as advisory system that alerts and recommends corrective actions in terms of either APH parameters or parameters controlled through other equipment such as selective catalytic reduction or boiler or changes in operation or design.

Abstract: This disclosure relates generally to identification and analysis of regime shift. The identification and analysis of the regime shift includes regime shift identification (RSI), root cause analysis of the identified regime shift and a recommendation unit to rectify the identified regime shift. The disclosure proposes to monitor a system continuously to identify a regime shift at real-time as presence of regime shifts in any system decreases quality of process and products and makes the system less efficient. The regime shift is identified at real-time based on key performance indicators (KPIs), a set of relevant features and real time input data using machine learning techniques. Further the disclosure also proposes techniques for detecting at least one root cause for the identified regime shift and also recommends a rectification action to rectify the identified regime shift based on optimization techniques.

Abstract: Mislabeled data when used for various applications such as training of Machine Learning (ML) models, can cause erroneous results. The state-of-the-art systems performs the mislabel identification with low confidence, and some require manual intervention. The disclosure herein generally relates to data processing, and, more particularly, to a method and system for identifying mislabeled samples using adversarial attacks. The mislabeled sample may refer to a) a data sample that is tagged with a wrong/incorrect label, or b) a distorted/confusing data sample having similarity with multiple classes. The system performs adversarial attack on training data using varying values of adversarial perturbations, and then identifies, for each of the misguided data samples, least value of adversarial perturbation that was required to misguide each of the data samples. Further, the data samples which were misguided by small values of adversarial perturbation, are identified as candidate mislabeled data samples.

Abstract: One of the key challenges in healthcare and personal care industries is arriving at an optimum delivery vehicle and formulation which can deliver a specified active molecule to an intended site of action with minimal or no side effects. This disclosure relates to method of designing and testing of a vehicle and formulation for delivery of an active molecule. A plurality of inputs is processed to generate plurality of drug delivery routes. The vehicle associated with formulation is designed based on plurality of parameters associated with the active molecule and the plurality of drug delivery routes. The designed vehicle associated with the formulation on an in-silico model of corresponding chosen drug delivery route is tested to obtain data associated with delivery of the active molecule. The data associated with delivery of the active molecule is reiteratively processed to obtain a desired data associated with the delivery of the active molecule.

Abstract: Performance optimization of power plants is one of the major challenges. Several machine learning based techniques are available which are used for optimization of the power plants. Coal selection and blending is critical to ensuring optimum operation of thermal power plants. The present disclosure provides a system and method for optimum coal selection for the power plant and power plant optimization. The system mainly comprises two components. First, a coal usage advisory module providing coal usage and blending ratio advice to the operators based on the available coal. The optimization is with respect to the entire power plant operation including its components. And second, a performance optimization advisory module provides operation instruction for boiler, SCR, APH and other power plant equipment based on the implemented coal blend in real-time.

Abstract: The disclosure relates to anomaly detection in an industrial environment including multiple industrial units and systems, generating huge volume of data. The conventional methods rely only on sensor data alone. The techniques of handling missing data plays a crucial role in determining the performance of industrial anomaly detection system. Further, imputation of missing data could cause error in computation, thus affecting the accuracy of the industrial anomaly detection system. The present disclosure addresses the problems associated with missing data by utilizing a masking technique. Further, the present disclosure utilizes quantitative and qualitative metadata associated with industrial system along with the sensor data to improve anomaly detection performance. Furthermore, the present disclosure includes a model recommendation system which provides transfer learning based utilization of existing models for similar industrial systems.

Abstract: Process control of continuous production of biomolecules is a major challenge due to complex nature of processes and time scales of operations involved. Availability of key process variables in real-time is one of main requirements. This disclosure relates to a processor implemented method of controlling a continuous bioprocessing plant which includes at least one of: receiving, an input data associated with one or more equipments; generating, by a recipe builder, a sequence of unit operations to determine at least one job order based on the at least input data; obtaining, a control decision associated with a control parameter based on the at least one job order; communicating, via the middleware, the control decision associated with the control parameter to the PLC; and executing, by a control system of the PLC, the control decision on a unit equipment to control: (i) a continuous bioprocessing train, and (ii) an individual unit operation.

Abstract: In traditional systems, every time a digital twin of a component needs to be generated, behavioral as well as operational data specific to the component needs to be fetched, which has practical difficulties owing to complex nature of processes/equipment the component is associated with. The disclosure herein generally relates to building digital twins, and, more particularly, to a method and system for building digital twin by leveraging existing knowledge. The system determines extent of similarity between two components, and based on the determined extent of similarity, uses different approaches to retrain a ANN data-driven model to obtain a desired accuracy for features of the component for which the digital twin is being generated.

Abstract: In an industrial plant, various equipment are used to handle processing of raw materials. Considering complexities involved in the processes and the equipment, constant monitoring is required to obtain desired results. The disclosure herein generally relates to industrial process and equipment monitoring, and, more particularly, to data analysis for Just In Time (JIT) characterization of raw materials in any process industry. The system collects real-time plant data among other inputs, and performs characterization of raw materials being used in the plant. The characterization involves categorizing the raw materials into different classes. The class information is further used to predict performance of the industrial plant, and in turn to generate recommendations for optimization of the industrial plant.

Abstract: Agglomeration process in agglomeration plants is quite sensitive to changes in input feed material characteristics. End-to-end optimization of the agglomerate process by combining all the units is difficult due to unique complexities and challenges associated with combining the individual process outputs. A method and system for optimizing the operation of an agglomeration plant has been provided. The system performs real time optimization on integrated wet agglomeration and thermal agglomeration process which subsequently increases the plant productivity and agglomerate quality and minimizes the operating cost and emissions from the plant. The optimization process involves various steps such as receiving data, pre-processing of data, prediction using physics-based and data-driven models of agglomeration plant, and optimization execution and configuration. The process also involves continuous monitoring of model performance and self-learning of the models in case of a performance drift.