Purpose -To investigate the feasibility of using single/multi variable optimisation techniques with vibration measurements in solving the inverse crack identification problem. Design/methodology/approach -The finite element method is used to solve the forward crack problem with a special nodal crack force approach. The multi-variable optimisation approach is reduced to a much more efficient single-variable one by decoupling the physical variables in the problem. Findings -It is shown that, for the crack identification problem, global optimisation algorithms perform much better than other algorithms relying heavily on objective function gradients. Simultaneous identification of crack size and location proved to be difficult. Decoupling of the physical variable is introduced and proved to provide efficient results with single-variable optimisation algorithms.Research limitations/implications -Need for improving the reliability and accuracy of the procedure for smaller crack sizes. Need for developing and investigation more rigorous and robust multi-variable optimisation algorithm. Practical implications -Any information about approximate crack size and location provides significant aid in the maintenance and online monitoring of rotating equipment. Originality/value -The paper offers practical approach and procedure for online monitoring and crack identification of slow rotating equipment.

Blade vibrations resulting in alternating stresses are often the critical factor in determining blade life. Indeed, many of the failures experienced by turbomachinery blades occur due to high-cycle fatigue caused by blade vibrations. These vibrations can arise either through self-excited oscillations known as flutter or through aerodynamic forcing of the blades from factors such as periodic wakes from up and/or downstream vanes or unsteady flow phenomena such as compressor surge. The current paper deals with the design and the analytical and experimental verification of the axial blading for a new generation of industrial compressors, a hybrid axial compressor that combines the advantages of conventional industrial compressors — broad operating range and high efficiency — with the advantages of gas turbine compressors — high power-density and high stage pressure ratios. Additionally, the surge robustness of this novel compressor blading has been greatly improved. During the development phase extensive efforts were made to ensure safe operation for future service life. This was achieved by designing blades that will not flutter, do not have high resonance amplitudes throughout their entire operating range and are extremely robust against surge. This strongly increased robustness of the new compressor blading was achieved by the implementation of a “wide-chord” blade design in all rotor blade rows in combination with a proper tuning of resonance frequencies throughout the entire operating range. For the verification of the new blading well-established methods accepted by industry were used such as CFD and FEA. Furthermore, coupling of the two into a method referred to as Fluid Structure Interaction (FSI) was used to more closely investigate the interaction of flow and structural dynamics phenomena. These analytical techniques have been used in conjunction with extensive testing of a scaled test compressor, which was operated at conditions of dynamic similitude (matching of scaled blade vibration frequencies, flow conditions, and Mach number) with full-scale operational conditions. Strain gauges placed on the blades and a state of the art technique known as “tip timing” were used to verify blade vibrations over a wide range of combinations of guide vane positions and rotational speeds. No propensity was found of any of the blades to develop high vibration amplitudes at any of the operating conditions investigated in the rig tests. The comparison of non-linear forced response analyses and the rig test results from strain gauges and tip timing showed close agreement, verifying the analysis techniques used. In conclusion it can be stated that the blade design exhibits a very high level of safety against vibrations within the entire operating range and during surge.

An industrial axial compressor has to meet a wide range of operation requirements and therefore must run within the whole compressor map without restrictions at an overall high level of efficiency. Additionally a robust design is required allowing a continuous operation of up to five years under industrial boundary conditions without inspection. These requirements led the industrial turbomachinery market to be generally conservative and sensitive to every single change through modern compressor development. The consequence for industrial compressor designs are, that these have made only moderate development steps during the last 50 years. This paper deals with a novel hybrid axial flow compressor, which combines the advantages of an conventional industrial compressor, such as good operating range and efficiency, with the advantages of gas turbine compressors, mainly the higher power density resulting in a higher stage pressure ratio. Furthermore, the surge robustness of the novel compressor blading has been strongly improved. Starting from scratch, the development began with comprehensive matrix studies in all areas of the design, taking into account aerodynamics, mechanics, rotor dynamics and power density in order to ascertain the overall optimum for this new hybrid generation. State of the art CFD analysis has been intensively used to optimize the compressor blading as well as the flow behavior of inlet and exit for the specified requirements and different compressor control mechanisms. The novel hybrid compressor is designed for a volume flow of 930 000 m3/h and allows a scaling from 100 000 up to 1 500 000 m3/h of air. To verify the design, a rig — downscaled by the factor of 3 — was tested. The rig was intensively instrumented with thermocouples and pressure probes, a torquemeter, strain gauges, tip-timing probes, and transient pressure transducers. Besides the measurement of blading performance, inlet and exit flange-to-flange instrumentation has been used to collect performance data under a variety of industrial operating conditions. The compressor behavior will be presented with a focus on aerodynamic aspects. The analytical and experimental data will be discussed in detail.

The turbine blade is one of the most critical components of a steam turbine. The high thermal loads and large centrifugal forces cause extreme stresses on the blade, especially on its root. This paper focuses on improving the double-T root of a turbine blade of the control stage by decreasing the root’s peak equivalent von-Mises stress. An 18% reduction was achieved in the peak stress by changing the convexity of the contact surface between the root and the groove. The equivalent von-Mises stress was determined in a static structural analysis of a three dimensional finite element model (3D FEM-model) using ANSYS Workbench. This numerical model was developed to include one blade and the associated part of the shaft, whereas the complete circle of blades was considered by applying cyclic symmetry. Furthermore, this paper includes a modal analysis comparing the natural frequencies of the initial FEM-model with the frequencies of the optimized one. The results were established by an investigation of the influence of the FEM-model’s parameters, its material properties, thermal effects, and an additional damping wire in the shroud.

Bearings are a key factor in achieving a good rotor dynamics performance for turbo machinery. Large compressors, steam and gas turbines for industrial applications are generally equipped with journal bearings either as tilting pad or multi-lobe bearing type. Here bearing parameters such as bearing geometry, bearing load or oil viscosity significantly alter bearing behavior and influence the rotor dynamics of the entire rotor-bearing system. In order to find an optimal set of bearing parameters for a given rotor-bearing system a nonlinear parameter optimization approach is employed. The rotor-bearing system is parameterized using bearing width, clearance and preload as design variables, since they represent design parameters that can be modified without significantly influencing the rotor design as a whole. The set of design variables is further constraint to stay within feasible limits of bearing design. The objective function is defined as a quantitative measure of rotor dynamic performance evaluating the distance from required separation margins with respect to rotor critical speeds based on API 617 7th Ed. In order to compute the objective function based on the design variables the bearing code ALP3T, solving Reynolds equations for the bearing fluid film, is used to compute the required stiffness and damping coefficients as input to the rotor dynamics program. The rotor dynamics performance is then evaluated using the rotor dynamics code SR3 based on the transfer matrix method. Both programs have been developed by the University of Braunschweig and are defacto industry standard within the German turbo machinery industry. The two programs are coupled and the nonlinear constraint optimization problem is solved using MATLAB’s optimization toolbox. The feasibility of this method is discussed based on an example of an axial flow compressor using two-lobe bearings. It is shown that a significant improvement in rotor dynamic performance can be achieved when compared to previous bearing selections for similar compressor designs and that the approach is suitable for a real-life engineering environment.