Siemens had 60 patents in 3D printing during Q3 2023. One patent describes a computer-implemented method for detecting anomalies in powder-bed-based additive manufacturing. The method involves analyzing image data, clustering the data, comparing the clusters with reference anomalies, and determining anomalies based on the match. Another patent focuses on an active part of an electric machine that is additively printed in layers, with intermediate bodies and contact layers applied between the electrical conductors. The contact layer serves as a diffusion zone and undergoes heat treatment. Additionally, a patent describes a computing system with an access engine and a toolpath reordering engine for 3D printing. The system can reorder toolpath segments to reduce heat impact on the physical part being printed. Lastly, a patent discusses a nickel-based alloy with specific alloy elements, including cobalt, chromium, tungsten, aluminum, hafnium, tantalum, molybdenum, nickel, and impurities. GlobalData’s report on Siemens gives a 360-degreee view of the company including its patenting strategy. Buy the report here.
Siemens grant share with 3D printing as a theme is 35% in Q3 2023. Grant share is based on the ratio of number of grants to total number of patents.
Recent Patents
Application: Computer-implemented, adapted anomaly detection method for powder-bed-based additive manufacturing (Patent ID: US20230260103A1)
Siemens AG has filed a patent for a computer-implemented anomaly detection method in powder-bed-based additive manufacturing of a workpiece. The method involves analyzing image data using Principal Component Analysis (PCA) to compute image clusters. A clustering algorithm is then applied to the analyzed data to compute cluster centroids. These centroids are compared with a set of reference anomaly centroids, and if there is a match, the image data is segmented layerwise into cluster images of a specific anomaly. The segmented images are then transformed into a defined color space, such as Lab color space or greyscale spectrum. By integrating pixel information from the transformed segmented cluster images, a threshold value is computed for the image data set to determine the presence of an anomaly.
The patent claims include various aspects of the method. Claim 1 outlines the steps of the anomaly detection method, including providing image data, applying PCA and clustering algorithms, comparing cluster centroids with reference anomaly centroids, transforming segmented images, and integrating pixel information to compute a threshold value. Claim 2 specifies that the clustering algorithm can be K-Means Clustering, Fuzzy C-Means Clustering, Density-Based Spatial Clustering algorithm, or DBSCAN. Claim 3 mentions the detection of specific anomalies like "hot spots," "cold spots," blob defects, overexposed and/or underexposed regions in the powder bed during manufacturing. Claim 4 states that the image data can be constituted by a stack of layered images, including photographic and/or optical tomography data.
Other claims include using a stack of random and/or test images as a reference for PCA and clustering algorithms, using optical monitoring systems or cameras as input for image data, computing lower and/or upper threshold values for the image data set, using the computed threshold values as input parameters for subsequent anomaly detection or image processing methods, marking anomalies in each layer of the image data set and forming anomaly clusters, determining the location and size of anomaly clusters with reference to the workpiece geometry, storing anomaly cluster information in a report file, correlating anomaly cluster information with real material or manufacturing defects, and applying the anomaly detection method in the manufacturing process of a workpiece.
The patent also includes claims for a data processing apparatus configured to carry out the method, a non-transitory computer-readable medium with instructions for executing the method, and the use of Lab color space or greyscale spectrum as the defined color space. Additionally, there is a claim for forming anomaly clusters using various algorithms like nearest neighbor search, Connected Component Labeling, Proximity or Closest Point Search, Point Location or Point in Triangle Search, or k-Nearest Neighbor algorithm.
To know more about GlobalData’s detailed insights on Siemens, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.