Systems, methods, and computer programming products for self-learning order dressing rules applied to manufacturing products in accordance with received product specifications. The translation from commercial characteristics to manufacturing characteristics of the product being manufactured are learned and adjusted to meet the specifications for quality required by the provided commercial characteristics. Reinforcement learning models learn from the quality characteristics of produced products by applying positive scores when the commercial to manufacturing characteristic translation is on-specification, otherwise a penalty is applied when an off-spec product is produced. Digital twins of manufacturing equipment, simulated in real time, provide insight and recommendations for achieving correct quality characteristics. Sensors in each device or within the surrounding environment help digital twins to measure operational performance and lifecycle of the manufacturing equipment against historical baselines. Reinforcement models dynamically adjust equipment settings for producing products to account for equipment performance degradation over time and changes in operation performance.

Methods and systems are disclosed for simulating a fabrication process based on real time sensor measurements obtained during the process. In one embodiment, a first simulation of the process computes a set of predicted physical responses based on a first set of assumed boundary conditions, and then, during the fabrication process sensor measurements are obtained and used to compute a second set of boundary conditions. A second simulation, based on the second set of boundary conditions, can then be performed to compute an updated set of predicted physical responses that can be compared to the previously computed set of physical responses. The difference(s) can be used to determine line, surface or volumetric response distribution from point, line or surface boundary conditions respectively, whether and how to modify the fabrication process (or other processes) and how to take additive and other manufacturing process decisions real-time using simulation. Other examples are also described.

To manage a generation history of a dynamic model for performing a dynamic simulation of a plant to efficiently generate the dynamic model, an apparatus, a method, and a program are provided, the apparatus including a model acquiring unit configured to acquire a first dynamic model, a second dynamic model and a third dynamic model that are models calculating dynamic states of a plant, a first difference extracting unit configured to extract at least one first difference being at least one difference between the first dynamic model and the second dynamic model, and a second difference extracting unit configured to extract at least one second difference being at least one difference between the first dynamic model and the third dynamic model.

A plasma cutting system for measuring or monitoring the voltage between a plasma torch and the material being cut to determine a voltage or voltage signature and comparing that measurement against predetermined values to indicate that an initial pierce of the material is complete, and based on the measurement, moving the torch or the material to a different location for additional cutting. The system further provides a Fix Drawing Tool, which will automatically detect and fix gaps or overlaps in a drawing that are very difficult to find visually. These gaps and overlaps become a problem when trying to make a proper toolpath because a CAM program requires a clean, closed shape. The system also provides a Dynamic Corner Looping system, which automatically adjusts with the feed-rate and accelerations of the toolpath and plasma machine, eliminates unwanted dross, sharpens corners and minimizes material loss. A pendant tethering system is also disclosed for managing control of a CNC machine remotely. Additional disclosed functionality includes a data collection system, a manual hand wheel with 3D simulation and a multiple fabrication head management system.

A CNC machining device (1) comprises a control console (3) and a CNC machine (5). A training assembly (33) for training operation of the CNC machining device includes a training control console (35) substantially identical to the control console (3) of the CNC machining device (1); a digital twin (36) of the CNC machine (5), comprising a simulator (37) configured to simulate the effect of commands from the training control console (35) on the CNC machine (5); and a display device (39) configured for a trainee to view the current state of the simulator (37).

A plasma cutting system for measuring or monitoring the voltage between a plasma torch and the material being cut to determine a voltage or voltage signature and comparing that measurement against predetermined values to indicate that an initial pierce of the material is complete, and based on the measurement, moving the torch or the material to a different location for additional cutting. The system further provides a Fix Drawing Tool, which will automatically detect and fix gaps or overlaps in a drawing that are very difficult to find visually. These gaps and overlaps become a problem when trying to make a proper toolpath because a CAM program requires a clean, closed shape. The system also provides a Dynamic Corner Looping system, which automatically adjusts with the feed-rate and accelerations of the toolpath and plasma machine, eliminates unwanted dross, sharpens corners and minimizes material loss. A pendant tethering system is also disclosed for managing control of a CNC machine remotely. Additional disclosed functionality includes a data collection system, a manual hand wheel with 3D simulation and a multiple fabrication head management system.

A method and apparatus for machining parts with variable stiffness includes determining, by a controller, a chatter-lobe plot of a cutter assembly. A preliminary tool path is developed by the controller. Virtual machining of a blank part using the preliminary tool path is performed by the controller. A chatter-lobe plot of the virtually machined part is determined by the controller. A dynamic chatter-lobe plot using the chatter-lobe plot of the cutting tool assembly and the chatter-lobe plot of the virtually machined part is determined by the controller. A chatter-free rotational speed of the cutting tool from the dynamic chatter-lobe plot is determined by the controller. A machining apparatus, controlled by the controller, uses the determined chatter-free rotational speed of the cutting tool to machine a blank part.

Methods and systems are disclosed for simulating a fabrication process based on real time sensor measurements obtained during the process. In one embodiment, a first simulation of the process computes a set of predicted physical responses based on a first set of assumed boundary conditions, and then, during the fabrication process sensor measurements are obtained and used to compute a second set of boundary conditions. A second simulation, based on the second set of boundary conditions, can then be performed to compute an updated set of predicted physical responses that can be compared to the previously computed set of physical responses. The difference(s) can be used to determine line, surface or volumetric response distribution from point, line or surface boundary conditions respectively, whether and how to modify the fabrication process (or other processes) and how to take additive and other manufacturing process decisions real-time using simulation. Other examples are also described.

Systems, devices, and methods for managing and optimizing a machining process for a computer numerical control (CNC) machine tool with a virtual machine-aware kernel. The virtual machine-aware kernel may predict steps for the manufacturing of a virtual machine part by constructing a virtual model of a CNC machine to mimic the CNC machine. The virtual machine-aware kernel may receive as input virtual data to simulate the real life conditions of the CNC machine milling processes for manufacturing of a machine part. The virtual machine-aware kernel may check, improve, and optimize the program data using the digital representation as a predictive model. Therefore, the virtual machine-aware kernel may allow for intelligent, real-time decision making to avoid defective manufacturing of a machine part.

Systems, methods, and computer programming products for self-learning order dressing rules applied to manufacturing products in accordance with received product specifications. The translation from commercial characteristics to manufacturing characteristics of the product being manufactured are learned and adjusted to meet the specifications for quality required by the provided commercial characteristics. Reinforcement learning models learn from the quality characteristics of produced products by applying positive scores when the commercial to manufacturing characteristic translation is on-specification, otherwise a penalty is applied when an off-spec product is produced. Digital twins of manufacturing equipment, simulated in real time, provide insight and recommendations for achieving correct quality characteristics. Sensors in each device or within the surrounding environment help digital twins to measure operational performance and lifecycle of the manufacturing equipment against historical baselines. Reinforcement models dynamically adjust equipment settings for producing products to account for equipment performance degradation over time and changes in operation performance.