Abstract: Method and apparatus for industrial scale production of a suspension for a battery, wherein an input material is processed via ball milling in a rotating chamber of a device that is effected as a continuous process with a continuously controlled addition of the input material and with a continuously controlled delivery of the processed output material, where state parameters of the input material and process parameters of the manufacturing installation are acquired as first parameters during production of the suspension, results of laboratory analyses about the state or quality of the manufactured suspension are acquired as second parameters in a learning phase during production, the first and the second parameters are used in the learning phase for training a model for predicting the state or quality via machine learning, and where the device is open-loop or closed-loop controlled outside the learning phase via the first parameters and the trained model.

Abstract: Various embodiments of the teachings herein include a computer-implemented method of fine-tuning Natural Language Processing (NLP) models. Some examples include: providing a training data set including a multitude of training text documents; providing a NLP model including a Neural Network (NN) based Topic Model (TM) having scalable TM parameters and a parallel large-scale pre-trained Language Model (LM) having scalable LM parameters; and fine-tuning the NLP model by jointly training the NN-based TM and the parallel large-scale pre-trained LM using a projected vector comprising a combination and projection of a document topic proportion generated by the NN-based TM based on the scalable TM parameters from an input training text document of the multitude of training text documents, and of a contextualized document representation generated by the large-scale pre-trained LM based on the scalable LM parameters from the same input training text document.

Abstract: Various embodiments of the teachings herein include a computer-implemented method for a topic modeling with a continuous learning. The method may include: extracting a current topic representation which represents a topic distribution over vocabulary within a current document; adjusting a size of the vocabulary of the current topic representation based on words used in a topic pool, wherein the topic pool includes past topic representations accumulated by each of past documents; regularizing the current topic representation by controlling a degree of topic imitation with past topic representations, based on comparison of the current topic representation and each of the past topic representations; and accumulating the regularized current topic representation into the topic pool.

Abstract: To configure a control agent, predefined training data are read in, which specify state datasets, action datasets and resulting performance values of the technical system. Using the training data, a data-based dynamic model is trained to reproduce a resulting performance value using a state dataset and an action dataset. An action evaluation process is also trained to reproduce the action dataset using a state dataset and an action dataset after an information reduction has been carried out, wherein a reproduction error is determined. To train the control agent, training data are supplied, the trained action evaluation process and the control agent. Performance values output by the trained dynamic model are fed into a predefined performance function. Reproduction errors are fed as performance-reducing influencing variables into the performance function. The control agent is trained to output an action dataset optimising the performance function on the basis of a state dataset.

Abstract: An apparatus for monitoring an actuator system, a method for providing an apparatus for monitoring an actuator system, and a method for monitoring an actuator system where the has at least one actuator and at least one data output signal. An anomaly detector detects anomalies. A suppressing engine determines time periods in which a control intervention has been performed. In a resulting monitoring signal, only anomalies are indicated which do not overlap with time periods in which the control intervention has been performed resulting in less irrelevant alerts and false positives output to a human supervisor monitoring the actuator system. The apparatus for monitoring a system may be provided with a plurality of actuators that may affect one another over time. The apparatus may be applied to a system of submersible pumps, or a system of conveyor belts.

Abstract: An evaluation framework for a generated dataset of a data generation algorithm such as a generative adversarial network is provided. The generated dataset includes a plurality of iterations of multiple instances of generated time series of data points. The evaluation framework provides multiple views. A first view includes at least one distance measure. The at least one distance measure is between the multiple instances of the generated time series and multiple instances of a reference time series, as a function of the plurality of iterations.

Abstract: Provided is a method for the computer-assisted creation of digital rules for monitoring a technical system. In the method, an ontology is used, which contains a plurality of classes including classes of components of the technical system and classes of operating state characteristics of the technical system and contains semantic relations between the classes. By means of a user interface, a user can formulate abstract rules by means of the classes and the semantic relations from the ontology. The abstract rules are converted into concrete rules valid for the specific technical system in an automated manner. The method has the advantage that corresponding rules no longer have to be formulated individually for individual technical systems by the user. Instead, abstract rules only have to be created one time for identical or similar technical systems.

Abstract: Various embodiments include a method for producing a quality test system executing a quality test model with a filter mask and a quality model to determine a quality feature of a battery cell. The system has an electrochemical impedance spectroscopic unit for capturing test data relating to the battery within a frequency range. The method includes: creating the model; and producing the system. Creating the model includes: capturing spectroscopic learning data; creating the filter mask using a first machine learning method with analysis data from part of the frequency range by consulting the filter mask and creating the model using a second machine learning method. The first and the second learning method are coupled based on the learning data. The first machine learning method creates a filter mask determining the analysis data such that the second machine learning method creates a quality model optimized with respect to maximizing the quality.

Abstract: Various embodiments of the teachings herein include methods for amending or adding machine learning capabilities to an automation device in an automation system. The method may include: 1) providing a capability model of the automation device semantically representing capabilities of the device; 2) providing a machine learning model for semantically representing a machine learning functionality and including a semantic model of a neural network; 3) deploying the machine learning model within the automation device; 4) interpreting a semantic part of the machine learning model using a semantic reasoner and matching requirements of the machine learning model with device capabilities inferred by the capability model; and 5) executing the machine learning functionalities on the automation device.

Abstract: An apparatus for monitoring an actuator system, a method for providing an apparatus for monitoring an actuator system, and a method for monitoring an actuator system where the has at least one actuator and at least one data output signal. An anomaly detector detects anomalies. A suppressing engine determines time periods in which a control intervention has been performed. In a resulting monitoring signal, only anomalies are indicated which do not overlap with time periods in which the control intervention has been performed resulting in less irrelevant alerts and false positives output to a human supervisor monitoring the actuator system. The apparatus for monitoring a system may be provided with a plurality of actuators that may affect one another over time. The apparatus may be applied to a system of submersible pumps, or a system of conveyor belts.

Abstract: Provided is a computer-implemented method for inter-sententially determining a semantic relationship between a first entity and a second entity in a natural language document, comprising at least the steps of: generating a first dependency parse tree, DPT, for a first origin sentence of the document which comprises the first entity, wherein each DPT comprises at least a root node; generating a second DPT for a second origin sentence of the document which mentions the second entity; linking the root nodes of the first DPT and the second DPT so as to create a chain of words, COW; determining for each word in the COW a subtree; generating for each word in the COW a subtree embedding vector cw which is based at least on word embedding vectors xw of the words of the subtree; generating a representation vector pw for each word in the COW; and classifying, using a recurrent neural network, the semantic relationship between the first entity and the second entity, based on the input representation vectors pw.

Abstract: Provided is a computer-implemented method for inter-sententially determining a semantic relationship between a first entity and a second entity in a natural language document, comprising at least the steps of: generating a first dependency parse tree, DPT, for a first origin sentence of the document which comprises the first entity, wherein each DPT comprises at least a root node; generating a second DPT for a second origin sentence of the document which mentions the second entity; linking the root nodes of the first DPT and the second DPT so as to create a chain of words, COW; determining for each word in the COW a subtree; generating for each word in the COW a subtree embedding vector cw which is based at least on word embedding vectors xw of the words of the subtree; generating a representation vector pw for each word in the COW; and classifying, using a recurrent neural network, the semantic relationship between the first entity and the second entity, based on the input representation vectors pw.

Abstract: Provided is a method for the computer-assisted creation of digital rules for monitoring a technical system. In the method, an ontology is used, which contains a plurality of classes including classes of components of the technical system and classes of operating state characteristics of the technical system and contains semantic relations between the classes. By means of a user interface, a user can formulate abstract rules by means of the classes and the semantic relations from the ontology. The abstract rules are converted into concrete rules valid for the specific technical system in an automated manner. The method has the advantage that corresponding rules no longer have to be formulated individually for individual technical systems by the user. Instead, abstract rules only have to be created one time for identical or similar technical systems.

Abstract: An assistance system for assisting the planning of automation systems includes a configuration database that has configuration datasets of automation systems, where a respective configuration dataset in each case has the configuration data of a predefined automation system, includes a vectorization component for structuring and adjusting configuration datasets, where the vectorization component is configured to convert the configuration datasets of the configuration database into vectorized configuration datasets, and includes an AI component for processing the vectorized configuration datasets using artificial intelligence, where the processing of the vectorized configuration datasets by the artificial intelligence (AI) component entails utilization of a neural network that has a deep learning architecture.

Abstract: A method for computer-assisted determination of usage of electrical energy produced by a power generation plant such as a renewable power generation plant is provided. The method uses a plurality of neural networks having a different structure or being learned differently for calculating future energy amounts produced by a power generation plant. To do so, the energy outputs of the power generation plant forecasted by the plurality of the neural networks are used to build histograms. Based on the histograms, energy amounts for different confidence levels describing the likelihood of the availability of the energy amount are determined, and different uses are assigned to different energy amounts. Energy amounts having a higher likelihood of availability in the future are sold at higher prices than other energy amounts.

Abstract: The embodiments relate to a method for the computer-aided control and/or regulation of a technical system, particularly a power generation installation. The actions to be performed in the course of regulation or control are ascertained using a numerical optimization method (e.g., a particle swarm optimization). In this case, the numerical optimization method uses a predetermined simulation model that is used to predict states of the technical system and, on the basis thereof, to ascertain a measure of quality that reflects an optimization criterion for the operation of the technical system.

Abstract: The embodiments relate to a method for the computer-aided control and/or regulation of a technical system, particularly a power generation installation. The actions to be performed in the course of regulation or control are ascertained using a numerical optimization method (e.g., a particle swarm optimization). In this case, the numerical optimization method uses a predetermined simulation model that is used to predict states of the technical system and, on the basis thereof, to ascertain a measure of quality that reflects an optimization criterion for the operation of the technical system.

Abstract: A method for computer-assisted determination of usage of electrical energy produced by a power generation plant such as a renewable power generation plant is provided. The method uses a plurality of neural networks having a different structure or being learned differently for calculating future energy amounts produced by a power generation plant. To do so, the energy outputs of the power generation plant forecasted by the plurality of the neural networks are used to build histograms. Based on the histograms, energy amounts for different confidence levels describing the likelihood of the availability of the energy amount are determined, and different uses are assigned to different energy amounts. Energy amounts having a higher likelihood of availability in the future are sold at higher prices than other energy amounts.

Abstract: A method for the treatment of waste paper to produce a finished product in several process stages, comprises the steps of for at least one quality parameter, prescribing a set value for the finished product, wherein ahead of and/or following at least two of the process stages a value is determined by measurements of the at least one quality parameter, establishing the efficiency of a process stage with regard to the improvement of the at least one quality parameter in this process stage, and dynamically balancing in a process management system the individual process stages taking into account the overall efficiency of the process.

Abstract: A decision tree clustering procedure is provided which employs a unified approach to extracting both the decision tree and (preferably fuzzy) clusters. The decision tree is built by subsequent clustering of single dimensions or features, and the choice of the winning separation is based on cluster validity. In one embodiment, the clustering employs a fuzzy c-means (FCM) model and the partition coefficient (PC) to determine the selected separations.