Abstract: Machine-learned networks provide generative design. Rather than emulate the typical human design process, an inverse model is machine trained to generate a design from requirements. A simulation model is machine trained to recover performance relative to the requirements for generated designs. These two machine-trained models are used in an optimization that creates further designs from the inverse model output design and tests those designs with the simulation model. The use of machine-trained models in this loop for exploring many different designs decreases the time to explore, so may result in a more optimal design or better starting designs for the design engineer.

Abstract: A bimetal thermo mechanical actuator (10) includes a multi-layer bimetal structure (22) that includes a plurality of bimetal structures (12). Each bimetal structure (12) includes a pair of metals, a structural metal (14) and a driving metal (16) that are bonded together along a bonded interface (18) with the pair of metals forming one layer. A sliding interface (20) is between each pair of metals. The multi-layer bimetal structure (22) has a shape that has at least one arch. The multi-layer bimetal structure (22) includes a first end (24) and a second end (26), an inner edge (40) and an inner radius (42), an outer edge (44) and an outer radius (46). A first pivot head (28) is connected to the first end (24) and a second pivot head (34) is connected to the second end (26) of the multi-layer bimetal structure (22). Each pivot head includes a through-hole (30) located approximately center. The multi-layer bimetal structure (22) expands and contracts with temperatures changes.