Model-Based Tool for the Automatic Validation of Rotorcraft Regime Recognition Algorithms
Navy SBIR FY2015.2


Sol No.: Navy SBIR FY2015.2
Topic No.: N152-094
Topic Title: Model-Based Tool for the Automatic Validation of Rotorcraft Regime Recognition Algorithms
Proposal No.: N152-094-0231
Firm: Technical Data Analysis, Inc.
3190 Fairview Park Drive
Suite 650
Falls Church, Virginia 22042
Contact: Subhasis Sarkar
Phone: (703) 226-4069
Web Site: http://www.tda-i.com
Abstract: TDA is proposing to develop a model based technique to automate the verification and validation process of helicopter regime recognition codes in three major steps. In the first step, a list of standardized flight regimes will be developed as this is a pre-requisite for automated verification and validation process, and these may become the recommended standard in the helicopter and HUMS community. In the next step, a helicopter dynamic model will be developed and used to fly various helicopter flight regimes. The parameters of the flight dynamic model will be determined by a nonlinear system identification process utilizing actual HUMS flight data. For all the regimes identified in the first step, appropriate feedback laws will be created to generate the control input time history sequence, which in turn will be given as input to the helicopter dynamic model. The output response from the helicopter model will be fed to the regime recognition codes as simulated HUMS flight data and the predicted regimes will be compared with the known regimes. In the final step, a verification and validation tool will be developed that will compare the regimes predicted by the regime recognition codes with the actual regimes flown by the simulated helicopter model and will output the misclassification rate, confusion matrix and other relevant statistical measures to quantify the accuracy of the regime recognition algorithms. For the purposes of this program, TDA will use the SH-60B helicopter to generate example results and will extend to other helicopter platforms later.
Benefits: The outcome of the SBIR effort primarily improves algorithm improvements in RR codes for correct characterization of flown regimes. Operators can use this tool to validate different RR cods and consider one or a combination of the codes for overall accuracy in regime recognition. This validation tool will be able to identify codes that capture at least 97 percent of maneuvers, or sufficient maneuvers in order to not under-predict the fatigue damage fraction of life-limited parts by more than 0.5 percent. The tool will also alert users to inconsistencies between the outputs of the RR code and the performed maneuvers. The validation tool should use scripted HUMS flights on instrumented aircraft. Furthermore, automating the RR validation process using proposed mode-based tool would expedite the process and reduced costs . Use of accurate RR codes, in turn, will indirectly assures safety of aircraft operations with good definition of service usage spectrum as well as calculation of reliable component damages in service. In addition, these improvements combined would drastically reduce unscheduled maintenance and downtime of rotorcraft.

Return