Preparation of input and validation data for PZL SW-4 helicopter dynamic model in scope of HELIMARIS project

Abstract

HELIMARIS project ("Modification of an optionally piloted helicopter to maritime mission performance") aims in preparation of maritime operation of PZL SW-4 helicopter. Due to operational and economic issues, it is a reasonable approach to simulate the most hazardous flight stages before proceeding to flight test. Warsaw University of Technology (WUT) developed PZL SW-4 helicopter dynamic model implemented in FLIGHTLAB environment in scope of HELIMARIS project. The model should represent actual performance and dynamics of helicopter in basic flight states (hover, cruise, climb/descent, turn, etc.). Compliance with this requirement allows to predict PZL SW-4 behavior in harsh maritime environment, especially focused on ship approach, in a reliable manner. Presented effort is complementary with analyses and laboratory tests done simultaneously by the other project subcontractor - Ship Design and Research Centre (CTO), which purpose is to obtain ship air wake and helideck motion data to be integrated in FLIGHTLAB environment. Investigation will result in definition of safe and efficient operational procedure for light maritime helicopter. The purpose of the paper is to present each particular phase of required input data preparation done by PZL-?widnik (Original Equipment Manufacturer; OEM). Therefore, it provides necessary background for detailed control algorithms description and regulator system adjustment, performed by Warsaw University of Technology. PZL-?widnik provided definition of mass, inertial and geometrical data set in basic helicopter configuration. This included geometry of main rotor hub, tail rotor hub, stabilizers and landing gear. Position of sensors, indicators and all relevant systems was defined. Main rotor system definition contains also damping characteristics of lag damper. Main rotor and tail rotor blades were defined in terms of necessary properties distribution (mass, inertia, chord). Aerodynamic characteristics of airfoils in entire range of section Mach number were verified and tailored in order to achieve flight test compliance. Static stiffness and strength of landing gear was obtained from stand test results, including also limits for landing conditions. Fuselage was defined in terms of aerodynamics. Due to high predicted angles of attack and sideslip in ship air wake, supplementation of already used characteristics was required. CFD (Computational Fluid Dynamics) ANSYS Fluent solver was employed to obtain missing data. Results were tailored to obtain compliance with existing characteristics in narrow range of inflow angles and with actual power required for flight. Fuselage aerodynamics is to be supplemented by floats once detailed configuration is available. Helicopter control system was defined in terms of kinematic ratio between controls and swashplate position. Kinematics of swashplate was supplemented by longitudinal/lateral feathering coupling and main rotor flap feathering coupling formulation. Dynamic characteristic of hydraulic actuators was also provided. PZL-?widnik calculated vortex ring conditions envelope. Propulsion system definition was a distinct phase of dynamic model development. It included description of kinematic ratio between collective lever and engine control lever position. PZL-?widnik provided detailed kinematics of engine controls and nominal engine control characteristics (nominal output power vs engine control lever position). Fuel system mass flow limits and tank capacity were based on PZL SW-4 Rotorcraft Flight Manual. Simultaneously, set of flight test data was prepared. Dedicated flight test program was prepared and performed. It contained measurements of state parameters relevant for dynamic identification in time domain and validation of the model. First of all, sign convention and measurement system characteristics were provided to obtain compliance and integrity with simulation results. Then, controls input signals were defined for dynamic response investigation. There were two types of inputs - long step and fast doublet. Two groups of dynamic response flight states were established: near-ground maneuvers (hover in-ground effect, hover off-ground effect, directional movements) and forward flights (level flight, climbing/descent, turns). Each contained both long step and fast doublet control input in every control channel (collective, longitudinal cyclic, lateral cyclic, pedals). General description of test helicopter configuration and external conditions were provided. Second stage of flight test was equilibrium conditions and control margins investigation. PZL-?widnik provided detailed controls positon (swashplate pitch and roll), helicopter attitude (fuselage pitch and roll), and rotors collective angles in trimmed level flight. The same data set was prepared for autorotation. Static equilibrium conditions were compared with results obtained from own O50 FORTRAN code. The last group of static tests was in-ground controllability and maneuverability. It contains presentation of controls position vs wind azimuth. Distinct phase was a definition of static performance and dynamic characteristics of propulsion system. Static performance of RR M250-C20R/2 engine was calculated using Rolls-Royce application. Dynamic characteristic includes propulsion system time response in relevant flight states (start-up, vertical take-off, landing from high hover, entry into autorotation, recovery from autorotation). Dedicated on-ground propulsion system stability test was used for engine sub-model calibration. Initial validation of entire model was done with support of selected steady flight test data. Flight test data obtained from landings on a moving platform was used for initial definition of ship landing procedure. Approach and take-off profile were established. Data set contained state parameters measurements correlated with video recording of each particular approach. Additionally, influence of control chain dynamic stiffness and slack of the controls was assessed. Swashplate position calibrated from controls was compared with that calculated kinematically from actuator extension. MATLAB script was employed to calculate transfer function between actuators extension and swashplate position and to compare with flight measurements of actuator forces. Static slack of the controls was defined from stand tests. PZL provided also qualitative and quantitative criteria for dynamic model similarity assessment for both dynamic response and static equilibrium part. Page 2 of 17 Presented at 45th European Rotorcraft Forum, Warsaw, Poland, 17-20 September, 2019 This work is licensed under the Creative Commons Attribution International License (CC BY). Copyright © 2019 by author(s). These were defined in terms of simulation results as follows: response vector signs compliance, attitude deviation from measurement at certain time from input signal beginning, controls position difference in trimmed steady flight. A vital phase of the project is PZL SW-4 autopilot sub-model development. It required detailed definition of sub-system functionality, general architecture, emergency scenarios, requirements and limitations. Autopilot sub-model should allow to perform basic flight states in whole PZL SW-4 operational envelope with Stability Augmentation System (SAS) functionality. Additional automatic flight modes will be tailored to support wide spectrum of maritime missions in both manned and unmanned configuration. The most critical phase is automatic vertical take-off and landing with sea state up to 5. Manual landing procedure will be extensively examined during simulation campaign.