Rotating stall is an important unstable flow phenomenon that leads to performance degradation and limits the stability boundary in centrifugal compressors. The volute is one of the sources to induce the non-axisymmetric flow in a centrifugal compressor, which has an important effect on the performance of compressors. However, the influence of volute on rotating stall is not clear. Therefore, the effects of volute on rotating stall by experimental and numerical simulation have been explored in this paper. It’s shown that one rotating stall cell generates in a specific location and disappears in another specific location of the vaneless diffuser as a result of the distorted flow field caused by the volute. Also, the cells cannot stably rotate in a whole circle. The frequency related to rotating stall captured in the experiment is 43.9% of the impeller passing frequency (IPF), while it is 44.7% of IPF captured by three-dimensional unsteady numerical simulation, which proves the accuracy of the numerical method in this study. The numerical simulation further reveals that the stall cell initialized in a specific location can be split into several cells during the evolution process. The reason for this is that the blockage in the vaneless diffuser induced by rotating stall is weakened by the mainstream from the impeller exit to make one initialized cell disperse into several ones. The volute has an important influence on the generation and evolution process of the rotating stall cells of compressors. By optimizing volute geometry to reduce the distortion of the flow field, it is expected that rotating stall can be weakened or suppressed, which is helpful to widen the operating range of centrifugal compressors.

High performance centrifugal compressors with high pressure ratio are highly applied in turboshaft engines in order to obtain higher power-to-weight ratio and lower fuel consumption. The optimization of the aerodynamic configuration design of splitter blades is one of the effective ways to achieve higher efficiency. An in-house designed single-stage centrifugal compressor with a pressure ratio up to 12.0 is studied in this paper. By using a three-dimensional CFD (computational fluid dynamic) method, this paper investigates influences of the number of splitter blades and their leading edge position on the flow field characteristics and aerodynamic performance of the centrifugal compressor with ultra-high pressure ratio. Results show that three critical flow characteristics lead to severe losses in centrifugal compressor impeller when only full blades are applied. Those flow characteristics include the strong shock wave, the severe tip clearance flow at the inlet region and the severe flow separation at the rear region. Therefore, the inlet blade number should be reduced to decrease the loss caused by strong shock waves and tip clearance flow, while the outlet blade number should be sufficient enough to suppress flow separation. By optimizing the number and the leading edge position of splitters, the performance can be improved under the reduction of combined losses caused by shock waves, tip clearance flow and flow separation. When an aerodynamic configuration with single-splitters is used, numerical results indicate that the leading edge position of splitter blades should be located at 60% of the main blade chord length, and the centrifugal impeller isentropic efficiency with ultra-high pressure ratio can be increased from 82.4% (the aerodynamic configuration with only full blades) to 89.5%; when an aerodynamic configuration with double-splitters is used, the leading edge positions of middle and short splitter blades should be respectively located at 40% and 60% of the main blade chord length, and the impeller isentropic efficiency can be further improved to 90.9%.

Air often flows into compressors with inlet prewhirl, because it will obtain a circumferential component of velocity via inlet distortion or swirl generators such as inlet guide vanes. A lot of research has shown that inlet prewhirl does influence the characteristics of components, but the change of the matching relation between the components caused by inlet prewhirl is still unclear. This paper investigates the influence of inlet prewhirl on the matching of the impeller and the diffuser and proposes a flow control method to cure mismatching. The approach combines steady three-dimensional Reynolds-averaged Navier-Stokes (RANS) simulations with theoretical analysis and modeling. The result shows that a compressor whose impeller and diffuser match well at zero prewhirl will go to mismatching at non-zero prewhirl. The diffuser throat gets too large to match the impeller at positive prewhirl and gets too small for matching at negative prewhirl. The choking mass flow of the impeller is more sensitive to inlet prewhirl than that of the diffuser, which is the main reason for the mismatching. To cure the mismatching via adjusting the diffuser vanes stagger angle, a one-dimensional method based on incidence matching has been proposed to yield a control schedule for adjusting the diffuser. The optimal stagger angle predicted by analytical method has good agreement with that predicted by computational fluid dynamics (CFD). The compressor is able to operate efficiently in a much broader flow range with the control schedule. The flow range, where the efficiency is above 80%, of the datum compressor and the compressor only employing inlet prewhirl and no control are just 25.3% and 31.8%, respectively. For the compressor following the control schedule, the flow range is improved up to 46.5%. This paper also provides the perspective of components matching to think about inlet distortion.

High pressure ratio centrifugal compressors with vaned diffusers are required in modern gas turbine engines. A potential method to further improve the performance of such compressors is to apply splitters to the vaned diffuser. This paper numerically investigates design choices of splitters in a vaned diffuser of a high pressure ratio centrifugal compressor. The splitters are derived from the main vane but with different chord length. The total vane number has been adjusted to maintain the same diffuser flow capability relative to the impeller. When the splitter is designed in the vaned region of the datum diffuser, the flow separation has been successfully suppressed with increasing relative splitter chord length from 0.10 to 0.70. At high mass flow rate conditions, the stage efficiency and the diffuser pressure recovery coefficient increases by 1.0% and 0.04 respectively due to the suppression of the flow separation. However, at design mass flow rate, the stage efficiency and the diffuser pressure recovery coefficient drops slightly by 0.3% and 0.01 respectively because the enhanced friction loss of vane surfaces is now dominant. When the splitter vane reaches the semi-vaneless region by further increasing its chord length from 0.8 to 0.95, the shock structure changes from two parallel shocks to two sequential shocks at first, but turns to the datum shock structure at last. The stage efficiency and the diffuser pressure recovery coefficient of the sequential shock structural case significantly drops by 0.7% and 0.03 respectively due to the increased shock strength and thus the higher shock loss. It is indicated that the application of splitter vanes in the vaned region of the diffuser is beneficial when the induced friction loss from splitter surfaces is less significant or the reduced loss from the suppressed flow separation is more dominant, which is the case that the flow separation happens relatively downstream or the case operating at large incidence conditions.

Tip clearance has great influence on the performance of multistage axial compressors including efficiency, pressure rise, mass flow, as well as matching. This paper reports a study into the influence of tip clearance on the performance and matching of a 5-stage axial compressor by a numerical method. Different tip clearances from 0% to 5.0% span which represents the typical range of tip clearance in modern multistage axial compressors were simulated and analyzed. The results show that as tip clearance increases from 0% to 5.0% span, the choked mass flow decreases by about 21.8%, the peak pressure ratio decreases by about 43.1% and the peak efficiency decreases about 14.3 percents. As tip clearance increases, the efficiency of the whole compressor decreases in a parabolic manner not linearly as previous suggested, which is partially attributable to the cantilevered stators considered in this paper and primarily due to the mismatching of different stages. It is of great importance to control the tip clearance. When tip clearance increases, the front stage tends to work near surge condition and the rear stage tends to work near choke condition, which leads to lower efficiency than in the middle stages. A weight was defined to evaluate each stage’s contribution to the whole compressor’s efficiency deficit caused by the increase of tip clearance. Front and rear stages contribute more to the efficiency deficit than the middle stages, which indicates that more attention should be paid on front and rear stages to improve the performance of multistage axial compressors. In order to evaluate the matching of multistage axial compressors with a quantified method, a new parameter named “Peak Efficiency Deviation (PED)” was defined based on the difference between each stage’s operating efficiency and its peak efficiency. The mass flow of multistage axial compressors should be well considered to make the PED parameter to be close to zero as possible. In the most commonly used range of tip clearance from 0.5% to 3.0% span, the PED varies little within 0.4 percent, which is only about 8.4% of the peak efficiency deficit at 1.5% span tip clearance. So, the PED could be small within a wide range of tip clearances if the matching of the compressor is perfect at design tip clearance.

To improve vehicle engine overall performance, an off-design analysis method of centrifugal compressor is presented. Mismatching degree is involved in the off-design analysis method. The mismatching degree of compressor flow fields of different working conditions is investigated, and flow control method of the compressor is proposed to decrease the mismatching degree and improve performance of the off-design point of the compressor. As an example, a diesel engine turbocharger centrifugal compressor is analyzed and optimized. The maximum mismatching degree of different working conditions locates at the leading edge of the main blades near shroud, and vaned shroud is presented to decrease the mismatching degree. Simulation shows that the maximum mismatching degree of the centrifugal compressor with vaned shroud decreases significantly, and experimental results show that the performance of the centrifugal compressor with vaned shroud is improved and pressure ratio is increased by 3.1% and efficiency by 1.7% at 80,000 RPM near surge.

As the result of strict emission regulations and the demand of fuel reduction, an increasing number of engines are highly boosted with complex two-stage turbocharging systems. In a two-stage turbocharging system, an additional intercooler between LP compressor and HP compressor is usually used to reduce the inlet temperature of the HP compressor for extending its life and reducing the compression work. Spatial restrictions of the turbocharging system lead to reduced distances between different components of the two-stage turbocharging system. Consequently, each component’s performance is influenced by the upstream and downstream ones. The high pressure compressor performance is affected by the inlet flow condition, which is significantly distorted by the intercooler. In this paper, the characteristics of flow distortion caused by the intercooler and its evolution process are analyzed. Performances of the downstream high pressure compressor under uniform inlet condition and distorted inlet condition caused by the intercooler are analyzed and compared. The results show that the inlet distortion caused by intercooler deteriorates the compressor pressure ratio, flow range, and efficiency.

A fluid-solid coupled approach is presented for investigating turbine blade fatigue life in a vehicle turbocharger under the parallel computation of aeroelasticity. The unsteady flow field of rotor is studied in a CFD model of the turbine flowpath, coupled with a finite element model for the impeller to identify modal shapes, natural frequencies, vibratory stress and the structural deformations. Based on the interface exchange of information between the fluid and structural domains, interactions between fluid and structure are calculated in a coupled manner. The numerical simulation is proceeded under different angle of attack and back pressure. In addition the influence of inlet total pressure fluctuation is considered. The inlet total pressure fluctuation is the function of 1 st and 3 rd natural frequency of impeller. The result shows that the maximum stress is less than the yield stress. The fatigue life will overpass ten million cycles.

A Brayton cycle waste heat recovery (WHR) system for turbocharged diesel engines was described and the performance of a diesel engine integrated with this system was investigated. The waste heat recovery system is integrated with the turbocharging system of the diesel engine, with the turbocharger compressor working as the Brayton cycle compressor simultaneously. The combined cycle of diesel cycle and Brayton cycle was simulated using the engine cycle simulation code GT-Suite 7.0, and the performance of the Brayton cycle WHR engine was investigated. The turbocharging turbine and the Brayton cycle turbine were designed and their performances were simulated with turbine through-flow model. The mass-flow rates of the diesel cycle and the Brayton cycle have a great influence on their power outputs, which are determined by the turbocharger performance greatly. The influence of the charging turbine geometric parameters on the performance of the Brayton cycle WHR system was discussed. Results show that there is a tradeoff in performance between high and low engine-speed operating conditions with the investigated parameters variation, and different geometric dimensions should be selected when different common operating conditions are considered.

The small scale Organic Rankine Cycle (ORC) using a turbine as expanders is considered as one of the most efficient ways to convert the waste heat energies of automobile engine into electrical energy, in the power range from several kW up to dozens of kW. In general, two important factors must be taken into account when designing the ORC turbine: the real gas effects of the organic fluid and the high expansion ratio presented in the machinery due to thermodynamic and efficiency factors. The characteristic parameters of R245fa have great difference compared with air. When designing an ORC turbine, the initial problem is the influence of the working fluid characteristic parameters on the turbine performance. Similarity analysis method is used to analyze this problem. When using the similarity theory to design two similar operating conditions, gas constant R and dynamic viscosity μ are easy to be taken into account as they have a dimension; specific heat ratio κ is hard to be taken into consideration. In this paper, a high expansion ratio ORC turbine is generated for a small scale ORC system working with the organic fluid R245fa, and the influence of κ on the turbine performance parameters analyzed. Firstly, the turbine geometry is designed in an iterative process using the commercial design tool Concepts-NREC Rital program, and the designed operating condition is determined. Secondly, the similarity criterion numbers are deduced with no consideration of the specific heat ratio κ, and the similar operating condition is calculated by the deduced similarity criterion numbers. The results achieved by the one-dimensional design software Rital shows that the deviation of similarity criterion numbers is small, which indicates that the two operating conditions are similar. Then, to verify the conclusion above, the flow field simulation using one channel model is conducted based on the R245fa operating condition and air operating condition using the computational fluid dynamics (CFD) code NUMECA-FineTurbo. However, deviations of the similarity criterion numbers from the CFD results are more than twice to those from the one-dimensional results. The larger deviation may indicate that the influence of the specific heat ratio κ cannot be ignored. After that, the distributions of the total pressure, total temperature, the relative Ma number and entropy in the nozzle and rotor channel are compared. These results show that except the total temperature the distributions of the other three parameters are very similar. The similar distribution may indicate that the specific heat ratio κ has little influence on the efficiency and expansion ratio, but κ has greater influence on the temperature and output power. All of these conclusions are made according to the simulation results, the accuracy should be verified through the test results.

Engine turbo-compounding (TC) is an effective technology to improve the engine efficiency and cut down CO 2 emission. A power turbine downstream of the conventional turbocharger turbine is used to recover the exhaust energy and transfer it into mechanical work. However, the performance of turbo-compound system deteriorates seriously at engine off-design points. It is mainly caused by the fact that the designs of turbocharger turbine and power turbine are independent from each other, failing to consider the effects of upstream swirls on the downstream power turbine performance. Analysis showed that the off-design conditions, off-design incidence loss, transition duct loss and non-uniformity inlet loss contributed to the deterioration performance of power turbine at off-design conditions. Among these reasons, the incidence loss played significant effect on power turbine’s performance. The current research aims to improve the turbo-compounding performance at engine off-design conditions by counter-rotating turbine (CRT) configuration. CRT consists of a radial-flow turbocharger turbine and an axial-flow power turbine with counter-rotating direction. The main purpose is to decrease the incidence loss and turning loss in the power turbine stator by CRT. Performance comparisons between the counter- and co-rotating turbines have also been conducted. First, meanline analysis was carried out to investigate the influences of design parameters on the velocity triangles and turbine performance. Analysis shows that counter-rotating turbine requires the design of high reaction radial turbine. Then, the 1D preliminary design starts and it is followed by 3D detail design. Finally, the computational fluids dynamic (CFD) method was used to evaluate the counter-rotating turbines performance. Results show that the designed counter-rotating turbine improves the off-design performance effectively, with 3.8% increase of power turbine efficiency points at 1200rpm condition. Further analysis on the flow field was conducted and it was found that the flow angle distribution at upstream turbine exit was highly non-uniformity along the span. To weaken the non-uniformity may be a potential way to further improve the performance of turbo-compounding systems.

Centrifugal compressors are deemed to have a wide operating range. Recirculation devices, which comprise a bleed slot, an upstream slot and an annular cavity connecting them, are often used particularly compressors for turbochargers. They remove low energy fluid at the inducer and improve the incidence angle of the impeller leading edge, i.e. the blade loading of the inducer, at low flow rates due to the recirculation flow supplied to the compressor inlet. The impeller of a centrifugal compressor is often housed in a volute. Since the geometry of the volute is not axisymmetric, the impeller might be surrounded by an asymmetric flow field, hence there is the potential to enlarge the compressor operating range and improve efficiency using a recirculation device with an asymmetrically-distributed bleed slot, referred to here as a non-axisymmetric recirculation device. The authors [1] applied non-axisymmetric recirculation devices to a compressor with a vaneless diffuser. The results showed the effectiveness of a non-axisymmetric recirculation device with a bleed slot partially channeled in the circumferential direction. They also showed that the surge line of the compressor characteristics, which is the line connecting the operational points of the smallest flow rates on all peripheral Mach numbers, was significantly affected by the change in the circumferential position of the bleed slot relative to the volute tongue. The tested compressor was originally designed to feature a vaned diffuser [2]. Enhancing the compressor operating range is the key for marine use turbochargers, integrally geared compressors, multistage compressors and gas turbines as well as automotive turbochargers. These compressors normally use vaned diffusers. In this study the authors tried to apply non-axisymmetric recirculation devices developed in their previous study [1] to compressors with a vaned diffuser. Moreover different circumferential positions of the bleed slot relative to the volute tongue were tested as well as the vanelss diffuser case. The change in the surge line of the compressor characteristics was much smaller compared to the compressor with the vaneless diffuser. The circumferential static pressure distributions in the compressors in combination with vaneless and vaned diffusers were measured to determine the above reason as well as conducting unsteady calculations with a simplified outlet boundary condition. These measurements and calculations showed that the impeller with the vaned diffuser was surrounded by a less distorted static pressure field than that with the vaneless diffuser. These results implied that the vaned diffuser depresses the spread of the circumferential static pressure non-uniformity effect caused by a volute to the impeller.

The asymmetric flow field induced by the volute has considerable influence on the performance of a turbocharger centrifugal compressor, especially through its effect on the stability. In this paper, a novel asymmetric vaneless diffuser treatment with a circumferentially non-axisymmetric diffuser width distribution was firstly developed to enhance the stability of a centrifugal compressor for turbocharger. Design principle of the asymmetric diffuser was proposed based on the asymmetric flow in the compressor. The objective of the asymmetric vaneless diffuser design is to alleviate the flow asymmetry in the diffuser, which requires that the phase of the maximal diffuser width is coincident with the phase of the minimum static pressure in the diffuser. The results of the numerical simulation showed that the designed asymmetric diffuser was able to decrease the magnitude of the pressure distortion induced by the volute and therefore alleviated the negative effect of the volute on compressor stability. Experimental results showed that the designed asymmetric diffuser extended the stable flow range by 28.3% at designed speed compared to the prototype with symmetric diffuser.

As the result of increasingly strict emission regulations and demand of fuel reduction, current light and medium duty engines are being highly boosted with complex two-stage turbocharging systems. The purpose of this work is to investigate the influence of two-stage turbocharging system parameters on the engine performance and the optimization of these parameters. An analytical pre-design model of the series two-stage turbocharging system for an internal combustion engine was developed, which builds the relationship between total pressure ratio, total expansion ratio and other two-stage turbocharging system parameters. Considering total expansion ratio as a function of expansion ratio between HP and LP turbine, minimum total expansion ratio can be determined using this model. The ratio of total pressure ratio to total expansion ratio, engine brake thermal efficiency and total heat exchange of coolers are considered as the parameters for engine performance evaluation. Influence of two-stage turbocharging system parameters, such as efficiency of compressors and turbines, cooling water temperature, cooler efficiency, pressure loss of coolers, EGR rate and bypass gas rate of wastegate, etc., on engine performance was analyzed respectively. Results show that the performance of a two-stage turbocharging engine is impacted mainly by LP turbocharger efficiency, intercooler performance and air filter performance.