Siemens Aktiengesellschaft

Marcel Hildebrandt

Serghei Mogoreanu

Mitchell Joblin

Thomas Hubauer

Ingo Thon

Stephan Grimm

Mohamed Khalil

Martin Ringsquandl

Christoph Heinrich

Swathi Shyam Sunder

Sudev Nair

Christoph Bergs

Dominik Dold

Chandra Sekhar Akella

Anna Himmelhuber

Akhil Mehta

Yushan Liu

A database stores a set of items, with each item having technical attributes, and with each item representing a module that can be used in an engineering project of a first user, u. A feature encoder embeds each item based on its technical attributes into a low-dimensional vector space. Then, in a second step, a graph neural network pools over these item embeddings to compute an updated user embedding for the first user A decoder mapping then addresses the recommendation task by outputting recommendation scores for each item. That means, heuristically speaking, that the method and system lift the recommendation task to the level of technical attributes to overcome the sparsity problem caused by item sets that are not overlapping between user groups. Thus, when matching similar users, the method does not rely on users configuring exactly the same modules but rather on configured modules that are similar from a technical point of view.

Method and system for semi-automatic completion of an engineering project

For anomaly detection in a network, a temporal knowledge graph represents the network including interactions between network modules with a set of entities, a set of relations, and a set of timestamps. In a first step, temporal random walks are sampled from the temporal knowledge graph. These are transformed in a second step into temporal logical rules. After observing an event in the network—or in a different network—the observed event is classified in a third step regarding an anomaly, using the temporal logical rules. The temporal knowledge graph is used as a stream-based data structure to extract rules that identify typical temporal behavior of the network and is used to identify anomalies in a human-interpretable way. The anomaly detection task is framed as a quadruple classification problem, using the temporal logical rules and their respective groundings in the temporal knowledge graph to support the classification.

Method and system for anomaly detection in a network

Provided is a recommendation engine to provide automatically recommendations for the completion of an engineering project, the recommendation engine including: a first artificial intelligence, AI, module adapted to provide latent representations of a sequence of selected items; and a second artificial intelligence, AI, module adapted to process the latent representations of the sequence of selected items provided by the first artificial intelligence, AI, module to generate at least one sequence of complementary items required to complement the sequence of selected items to provide a complete sequence of items output via an interface as a recommendation to complete the engineering project.

Method and apparatus for providing recommendations for completion of an engineering project

A computer-implemented method for efficient processing of pooled data shared by users of a cloud platform, the method includes the steps of uploading at least one dataset by a client device of a user to said cloud platform; calculating similarity scores indicating a degree of similarity between the current uploaded dataset and datasets previously uploaded by client devices of other users; and performing a procedure selected by a user on the cloud platform based on pooled data including the current dataset of the respective user and the datasets previously uploaded from client devices of other users stored in a database of the cloud platform having calculated similarity scores in relation to the current uploaded dataset of the respective user exceeding a configurable similarity score threshold, is provided.

Cloud platform and method for efficient processing of pooled data

A plurality of basic simulations independent of one another are carried out, which determine respective remaining service life predictions for the machine. The remaining service life predictions and characteristic data are fed to a neural network, which outputs weights for the remaining service life predictions. A final prediction is calculated from the remaining service life predictions by weighting the remaining service life predictions relative to one another. A hybrid model is produced, which results from the combination of the basic simulations with the neural network. The remaining service life can be predicted not only for a small number of machines for which a specific simulation model has been manually created. The hybrid model enables condition monitoring for any further types and configurations of machines that merely belong to the same machine class. The basic simulations can therefore also be applied to previously unknown machines.

Assembly and method for predicting the remaining service life of a machine

Growing community of inventors

Munich, Germany

Marcel Hildebrandt