BAE Systems had one patent in artificial intelligence during Q3 2023. BAE Systems Plc has filed a patent for a method of estimating the properties of an article manufactured using additive manufacturing (AM) techniques such as directed energy deposition (DED) or powder bed fusion (PBF). The method involves obtaining a set of in-process parameters with positional information, inferring attributes of the article based on these parameters, and estimating the property of the article using the inferred attributes. The method is implemented using a computer with a processor and memory. GlobalData’s report on BAE Systems gives a 360-degreee view of the company including its patenting strategy. Buy the report here.
BAE Systems grant share with artificial intelligence as a theme is 100% in Q3 2023. Grant share is based on the ratio of number of grants to total number of patents.
Application: Structural simulation of additively manufactured components (Patent ID: US20230264264A1)
The patent filed by BAE Systems Plc describes a method for estimating the mechanical properties of an article manufactured using additive manufacturing (AM) techniques such as directed energy deposition (DED) and powder bed fusion (PBF). The method involves obtaining a set of in-process parameters of the AM process, which have positional information associated with them. These parameters can include optical, thermal, and acoustic emissions monitored during the manufacturing process. The method then infers a set of attributes of the article based on the in-process parameters, where the attributes are spatially resolved mechanical properties of the article. Finally, the method estimates the mechanical property of the article based on the inferred set of attributes. The mechanical properties that can be estimated include brittleness, bulk modulus, coefficient of restitution, compressive strength, creep, density, ductility, durability, elasticity, elastic limit, fatigue life, fatigue limit, flexibility, flexural modulus, flexural strength, fracture toughness, friction coefficient, hardness, malleability, mass diffusivity, plasticity, Poisson's ratio, resilience, shear modulus, shear strength, slip, specific modulus, specific strength, specific weight, stiffness, surface roughness, tensile strength, toughness, ultimate tensile strength, viscosity, yield strength, and Young's modulus.
The patent also includes claims that further specify the method. These claims include the use of trained machine learning algorithms to infer the attributes of the article, the use of finite element analysis (FEA) to estimate the mechanical property, and the application of the method to determine the mechanical properties of an assembly of articles. Additionally, the patent describes a method for training a machine learning algorithm using sets of in-process parameters and corresponding mechanical properties of a set of articles. The training data can be obtained through non-destructive testing (NDT) or destructive testing (DT) of the articles.
In summary, the patent filed by BAE Systems Plc describes a method for estimating the mechanical properties of articles manufactured using additive manufacturing techniques. The method involves obtaining in-process parameters, inferring attributes of the article based on these parameters, and estimating the mechanical property of the article. The method can utilize trained machine learning algorithms and finite element analysis. The patent also includes claims for determining the mechanical properties of an assembly of articles and training machine learning algorithms using sets of in-process parameters and corresponding mechanical properties.