Energy Efficient Material Processing Through Automated Process Monitoring and Controls
Project Lead: Virginia Tech
Partners: University of Virginia, Pennsylvania State University, Arconic, Commonwealth Center for Advanced Manufacturing
Member % Cost Share: 30%
CESMII % Cost Share: 70%
Duration: 21 Months
- Wireless sensor nodes impacted by electromagnetic, vibration, and thermal noise
- High volume of data raises challenges related to data transfer and processing
- Complex, nonlinear, dynamic mfg systems – challenges in real time decision making
- Existing sensor based decision making models are computationally complex
Demonstrate energy efficient metal material processing at Arconic facility through advanced sensing, automated process monitoring and model based controls.
Self-powered wireless sensor nodes and deployment of energy harvesting approaches; Efficient computational framework to acquire, post-process, and synthesize large quantities of sensory data in real time; In-process monitoring capability with offline big-data analysis techniques; Closed-loop system to enable real-time, model-based control for energy consumption optimization.
Delivered Functional Requirements Documentation and Conceptual Architecture for a Data System Supporting Data Acquisition, Simulation, and Analysis of Honeywell’s Chemical Vapor Infiltration (CVI) Process.
Devised recommendations for additional sensor technologies applicable to the Chemical Vapor Infiltration process monitoring.
Delivered complete KPI (carbon gain) prediction and energy consumption forecast.
Delivered a generalized framework that finds inputs to an externally provided forecasting model to maximize a value derived from the forecasting model’s predictions.
Delivered Maintenance Optimization Tool that reads historical sensor data, clusters similar runs, and graphically displays the output.
Key Performance Indicator prediction Apps.
Energy consumption monitoring Apps.
Process optimization and control algorithms for CVI furnace applications.
Control system that uses analyzed data to enable real-time, model-based control for energy consumption optimization of chemical manufacturing processes.
- Intelligently monitoring and controlling manufacturing processes for a relevant testbed, the usage of 800,000 to 1,000,000 kW of power could be affected
- A reduction of only 15% could save $100,000 annually for a single process
- Widespread adoption throughout the world increases the cascade of possible resource and cost saving processes
- Contribution to the SM Platform™ core technologies in process monitoring and control
- Sensor instrumentation, computational models, and analytics package will be available to other CESMII members with similar processes.
- Contribution to the SM Platform™ core technologies in process monitoring and control.