SBIR Phase I: Adaptive Methods for Sensorless Estimation of Induction Motor Efficiency

Veros Systems, Inc., College Station TX

Investigators

Abstract

This Small Business Innovation Research Phase I project will explore and demonstrate the feasibility of obtaining accurate on-line estimates of efficiency in industrial electric motors. The effort will be based on the SMART SENSORLESS (S²) technology that is at the core of the condition monitoring & assessment (CM&A) product currently being developed. The S² technology is sensorless in that only electrical measurements, i.e. three-phase voltages and currents available at the motor control centers (MCCs), are utilized. No mechanical sensors, such as speed, torque, vibration, or temperature are necessary. Consequently, the technology is cost-effective and cost-scalable, while being reliable and effective. The proposed approach to efficiency estimation is based on the existing CM&A framework by utilizing the raw electrical measurements, while augmenting it with adaptive filters for accurate motor speed and shaft torque estimation. The Phase I research plan calls for the development of on-line motor efficiency estimation algorithms based on various machine learning and adaptive filtering approaches. These algorithms will be tested and compared to accurately computed efficiencies from an existing database of staged motor faults and other motor, power quality and load anomalies. Algorithms for the on-line computation of efficiency estimation errors will be investigated. The broader impacts of the project include awareness of the importance of energy efficiency in industrial electric motors, which account for 25% of all electricity sold in the U.S. Widespread adoption of the technology developed through this SBIR project would reduce the total energy consumption by industrial electric motors up to 18%, with an annual savings of up to $6.8 billion in the U.S. alone. The costs of maintaining electric motors, the losses due to inefficient operation and the costs of lost production associated with motor downtime are among the most significant controllable costs in any industrial establishment, and are estimated to cost U.S. industry alone over $50 billion per year. Even a modest adoption of the technology that could result from the proposed innovation would eliminate some fraction of this waste and have a significant impact on the U.S. economy. These cost estimates present a large, unserved market opportunity for new technologies in on-line CM&A and efficiency assessment of electric motors. By addressing energy efficiency in the industrial sector, this effort will enable clients to reduce their energy costs, increase profitability, reduce fuel imports, and lower greenhouse emissions.