Real Compressor Research Articles

Second law efficiency of air-cooled refrigeration compressors

Isentropic efficiency is a common performance measure for compressors and is useful for modelling the behaviour of real compressors in relation to discharge temperature or required mechanical power input. However, it has a weakness in that a basic assumption in the calculation is that the real compressor is adiabatic. If significant heat transfer from the compressor to the outside environment occurs, then the adiabatic assumption is invalid, but the negative effects of internal entropy generation are reduced, and in principle it is possible to have a measured isentropic ‘efficiency’ greater than 100%, which, while counterintuitive, is highly desirable. We practically demonstrate this effect with an air-cooled swashplate compressor and propose that for cases of intentional compressor cooling, the simple definition for isentropic efficiency is retained as a performance measure but renamed to isentropic performance coefficient (IPC).

Effects of Re on blade load temporal-spatial distribution and correlation with flow stability of transonic compressor under inlet distortion

Compressor blade operating conditions and aerodynamic load change periodically owing to inlet distortion. The effects of low Re are often unavoidable when the inlet distortion is encountered in real compressor operation. Blade load distribution and internal flow field are both affected by variations in Re. In this paper, the effects of Re on the blade load distribution patterns under 60° of total pressure distortion is investigated by numerical simulation. The distortion intensity has been set to 0.05, and four cases with Reynolds number index of 1, 0.3, 0.2 and 0.1 are chosen for further analysis. The findings demonstrate that the variation of Re affects the compressor flow field mainly by modifying the aerodynamic profile of the blade, which also results in the change of the temporal and spatial distribution of the aerodynamic load. Furthermore, the aerodynamic blade profile varies significantly by Reynolds number in different spanwise regions. The blade time constant decreases in the mid-span area, while it increases in the tip. This ultimately leads to a delay in the location of the tip maximum load, changing the starting position of the induced stall.

A data-driven polynomial chaos method for uncertainty quantification of a subsonic compressor cascade with stagger angle errors

The probability-based uncertainty quantification (UQ) methods require a large amount of sampled data to construct the probability distribution of uncertain input parameters. However, it is a common situation that only limited and scarce sampled data are available in engineering applications due to expensive tests. In the present paper, the Data-Driven Polynomial Chaos (DDPC) method is adopted, which can propagate input uncertainty in the case of scarce sampled data. The calculation accuracy and convergence of the self-developed DDPC method are validated by a nonlinear test function. Subsequently, the DDPC method is applied to investigate the uncertain impact of stagger angle errors on the aerodynamic performance of a subsonic compressor cascade. A family of manufacturing error data of stagger angles was obtained from the real compressor blades. Based on the limited measurement data, the DDPC method combined with Computational Fluid Dynamics (CFD) simulation is employed to quantify the performance impact of the compressor cascade. The results show that the performance dispersion under off-design conditions is more prominent than that under design conditions. The actual aerodynamic performance deviating from the nominal performance is not a small probability event, and the probability of deviating from the nominal loss coefficient and exit flow angle by more than 1% can reach up to 47.6% and 36.8% under high incidence i = 7°. Detailed analysis shows that stagger angle errors have a significant effect on the flow state near the leading edge, resulting in variations in separation bubble size and boundary layer thickness.

Load-Sharing With Degradation Management in a Compressor Station

Management of compressor degradation is often considered from the perspective of maintenance of the compressor, but most frameworks for the operation of compressors do not take degradation into account. This paper proposes a method for operation of compressors that takes into account the current level of degradation in order to manage further degradation. The algorithm can be used in maintenance planning frameworks, in particular if the timings of maintenance activities are fixed. The algorithm can extend the lifetime of a compressor by mitigating its degradation, or, conversely, can intensify the degradation to reach the maximum level in time for planned maintenance. The performance of the algorithm has been demonstrated in a case study for five compressors. A comparison with equal load approach shows that the new algorithm improves the operation of the system by managing the degradation of selected compressors. Explicit management of degradation allows an extension of the lifetime of selected compressors before maintenance must be performed. Conversely, by ensuring that the desired level of degradation is attained before pre-planned maintenance actions, it contributes to increased efficacy of maintenance actions. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —The paper presents a new framework for load-sharing in a compressor station with compressors subject to degradation. The main innovation of the framework is the use of relationships between custom degradation indicators to manage degradation of the compressors. The results in the paper prove that it is possible to manage degradation in an industrial setting by adjusting the load of each compressor. From a practical perspective, the framework allows more degraded compressors to follow the less degraded compressors (called leaders). The simplicity of the proposed framework enables an intuitive choice of leaders, in particular in compressor stations with more than two compressors. Focused directly on the load sharing, the framework also avoids adjusting lower level control structures, such as speed control or surge protection. The simplicity and the fact that the load-sharing algorithm affects only the loads make the new framework quickly implementable on a high level of the operating system in a compressor station. The limitation of the framework is that the parameters of the framework must be adjusted manually to ensure that the compressors stay within their operating ranges. For instance, the parameters of the algorithm can be chosen in such a way that the load-sharing aims to strike a balance between multiple compressors. The next step would be to give a demonstration of the method on a high-fidelity industrial simulator that acts as a proxy for a real compressor station.

Dynamic causal modeling for nonstationary industrial process performance degradation analysis and fault prognosis

Performance degradation analysis and fault prognosis of industrial processes refer to distinguish different degradation stage and raise warnings before a failure occurs, which are crucial to ensure the operation efficiency and safety. However, processes usually have nonstationarity caused by operating condition switching, which may be confused with the underlying temporal variations in the degradation process. Considering that the degradation process is essentially the changes in short-term temporal information transmission between variables, we propose a dynamic causal graph-driven performance degradation analysis and fault prognosis framework (DCG-PDA), which can reveal the intrinsic connection between causality and degradation mechanism. First, a dynamic causal graph framework is constructed, in which the nonstationarity is removed by transfer entropy-guided symbolization, to extract significant short-term causal information transmission. Second, changes in causality are effectively captured and quantified, and the causal graphs corresponding to some time points with larger changes in causality are selected to characterize the representative causal patterns in the degradation process. Finally, a node anomaly score is designed to consider the accumulation of causal structural changes at anomaly time points and measure variable-wise degradation contributions, so as to realize accurate locations for key variables in the degradation process. The validity of the proposed method is illustrated through a real compressor fouling process.

Performance analysis of a CO2/NH3 cascade refrigeration system with subcooling for low temperature freezing applications

CO2/NH3 cascade refrigeration system is an all-natural refrigeration option which can be specifically designed and gainfully operated for low temperature applications in warm ambient conditions. In the present study, a novel CO2/NH3 cascade refrigeration system with subcooling is proposed and evaluated as an alternative to the HFC-404A refrigeration system, which is typically used for blast freezer applications in the seafood processing industry. The performance of the proposed system is also compared with a conventional CO2/NH3 cascade refrigeration system without any subcooling arrangement. Real compressor specific equations were used to simulate the performance of the systems under various conditions. The proposed system yields 26% higher performance ratio (COP) than the conventional HFC-404A system at the design condition. In addition, the proposed subcooling arrangement enhances COP of the conventional CO2/NH3 cascade unit by around 5%. The system performance is investigated theoretically under various condensing temperatures and subcooling degrees. The annual energy demand and total equivalent warming impact (TEWI) of the proposed system are 15% and 48% lower than for the conventional HFC-404A system which results in significant economic and environmental benefits. Overall, the proposed system exhibits superior energy efficiency and offers a climate and environmental friendly alternative to the conventional environmentally harmful systems within the seafood industry. Adoption of such energy efficient all natural and clean refrigerant technology is crucial for the future phase out of hydrofluorocarbon (HFC) fluids, as ratified in accordance with the Kigali Amendment to the Montreal Protocol in India and other tropical countries.

Numerical Prediction of Long-Term Droplet Erosion and Washing Efficiency of Axial Compressors Through the Use of a Discrete Mesh Morphing Approach

Abstract Online water washing represents an operation strategy commonly used to reduce compressor performance deterioration due to blade fouling. Since this kind of washing is applied when the machine operates close to full load conditions, injected droplets are strongly accelerated and consequently impact the rotor blades at high velocity, thus inducing undesirable phenomena like erosion. Here, we present a novel technique to study long-term water droplets erosion by also considering the geometry modification caused by droplets impacts. Two-phase unsteady numerical simulations were carried out, considering the injection of water droplets and their transport across the fluid flow in the first part of a real compressor, which is modeled in the region extending from the inlet to the rotor blades of the first stage. Simulations are performed on the whole machine to account for the asymmetric distribution of the spray injectors, the machine struts, inlet guide vanes (IGVs), and rotor blades. The k−ɛ realizable turbulence model with standard wall functions was coupled with the discrete phase model to track injected droplets motion. Droplets-wall interaction is modeled following the Stanton–Rutland approach aiming at detecting the effect of droplet impact (deposit, rebound, and splashing) depending on the impact conditions. Moreover, a semi-empirical erosion model developed by the authors was used to evaluate the erosion induced by the droplets injection. Material removal due to erosion is converted into nodal mesh displacement that is used by a secondary routine to implement the mesh morphing scheme. The mesh modification is applied at discrete steps to reduce the computational load. This technique is adopted to account for the blades geometry modification due to water droplet erosion leading to performance losses. Moreover, an estimation of the compressor operating life before maintenance operations is given and the water washing efficiency during the whole life of the machine is evaluated by means of proper indices. At the end of the simulation workflow, erosion phenomena are observed in all the compressor regions, especially in the rotor where erosion peaks are reached at the hub of the blades leading edge. The rotor blades wet surface was found to remain almost constant at around 50% during compressor water washing. Erosive phenomena were proved to evolve nonlinearly with time indicating the need to account for the geometry modification for obtaining an accurate prediction of the long-time process.

Transient rotordynamic thermal bow (Morton effect) modeling in flexure-pivot tilting pad bearing systems

The Morton effect (ME) is a thermally induced rotodynamic instability problem frequently reported in the fluid film bearing system. The temperature difference across the journal circumference may cause large spiral, synchronous vibration. Earlier ME modeling approaches rarely focus on the flexure-pivot bearing (FPB), significantly disproportionate to its wide application. Hereby, the transient ME prediction algorithm is proposed based on a high-fidelity thermo-elasto-hydrodynamic FPB model. A hybrid finite element analysis is adopted to evaluate the entire rotor thermal bow profile including the rotor midspan. The algorithm is benchmarked with experiments and applied to a realistic compressor model. Simulations confirm the proneness of the FPB system to the ME instability, and reveal the ME sensitivity to the bearing-rotor configuration and operation conditions.

Identification of Mistuning Based on Forced Response Measurements and Assessment for a Real Compressor Blisk

Abstract We develop a novel method for the identification of mistuned blisks. The method relies on forced response measurements in a certain target frequency band. Modal stiffness deviations and an overall structural damping coefficient are obtained using a least-squares fit, fully consistent with a component-mode-based reduced-order model. To achieve a high frequency resolution in a very short test, we carry out rapid frequency sweeps and use a noise-robust, wavelet-based estimation of the steady-state response. The forcing is eliminated from the equations governing the mistuning parameters, and it is identified separately to reconstruct the forced response if requested. We first validate the method numerically and then apply it experimentally to a compressor blisk with an alternating (B-A) mistuning pattern. While most previous works on mistuning identification are limited to the lowest 1-2 mode families, we assess the method for the seventh (1C) and eighth (3T) mode family. Measured and reconstructed frequency responses are in good to very good agreement. We conclude that the developed method is useful to obtain high-quality estimates of mistuning parameters, which can then be used for vibration prediction.

Applying the phononic crystal concept to the intake muffler of a refrigeration compressor

Many previous works explore the creation of wave stop bands (bandgaps) using periodic structures. Aiming at improving the transmission loss of a muffler used in a refrigerant compressor, this work proposes the redesign of an existing intake muffler using the concept of wave attenuation bands caused by Bragg scattering. The challenge consists in designing a phononic crystal (PnC) with bandgaps in the audio frequency range with the size constraints of a practical muffler application. The first part of the research focuses on the development and validation of numerical and experimental techniques to provide guidelines for periodic muffler design. The second part consists in applying the developed concepts and tools to a real compressor muffler following the constraints of the product specifications. A shape optimization technique is used to enlarge the bandgap while keeping the unit cell length in the order of tens of millimeters. Two designs were 3D-printed and tested on the real compressor, reaching 7 dB reduction on the sound power level (SPL) in the aimed frequency band and 4 dB reduction on the overall compressor noise level.

Uncertainty analysis of global and local performance impact of inflow and geometric uncertainties using sparse grid-based non-intrusive polynomial chaos

Flow variations at the inlet boundary due to the compressor operational condition changes and geometric variations of the realistic compressor blades due to the manufacturing variability cannot be absolutely avoided, the global and local performance impact of which requires to be considered in the mechanism study of performance change and the aerodynamic shape design. In this paper, a method to analyze the simultaneous impact of the inflow Mach number, inlet incidence and geometric uncertainties was proposed. To make the uncertainty modeling of geometric variations faster and closer to engineering practice, a parametric mathematical model based on scanning points on the blade was introduced to describing the profile and torsion errors in the method specially. Meanwhile, a sparse grid-based Non-Intrusive Polynomial Chaos (NIPC) was used for uncertainty quantification and uncertainty sensitivity analysis to alleviate the computational burden. Then, the method was combined with a loss source calculation method to estimate the global and local aerodynamic loss changes of a controlled diffusion compressor blade in a reference flow state of high inflow Mach number and large positive incidence, and the response performance of sparse grid-based NIPC was verified. The results show that inlet incidence and torsion error have a significance uncertainty effect on the boundary-layer separation above the suction surface, which is main reason for the fluctuation of global aerodynamic loss. The uncertainty effect of profile error on the boundary-layer separation is relatively weak, but profile error could have a certain uncertainty effect on the leading edge separation. The boundary-layer separation is insensitive to Inflow Mach number. Furthermore, a stochastic aerodynamic analysis in different reference inflow states was investigated, which reveals some laws that the uncertainty of the aerodynamic losses and the sensitivity of the inflow and geometric uncertainties change with reference inflow states.

Evaluation of water washing efficiency and erosion risk in an axial compressor for different water injection conditions

Gas turbines performance losses are mainly due to the deposition of dirt on the compressor blades that needs to be periodically removed. This is the reason motivating the presence of water washing systems (WWS) in most of the compressor gas turbines. Water washing is generally achieved by installing a number of nozzles on the compressor casing and spraying water that clean the dirty surfaces of the compressor. The side effect of such a technique is the rising risk of erosion due to the impact of water droplets on the compressor blades which is even more pronounced when dealing with online water washing systems that is done while the unit is at normal load. The design of these systems must balance benefits and disadvantages associated to the process itself. The benefits can be measured in terms of water washing efficiency that is a quantity not uniquely defined. In previous works, the authors introduced some indices useful to evaluate the spatial cleaning coverage (the wet to the total surface) and the quantity of water mass actually impacting the dirty surfaces (the impacted to injected mass). On the other hand, water washing erosion is a complex phenomenon depending on several parameters, such as the mechanical properties of the blade material, the impact velocity and angle and the droplet diameter. For this reason, the WWS are strongly influenced by the adopted nozzles and by the injection conditions. The present paper aims at assessing water washing for six different injection conditions in the first stage of a real axial compressor. Two-phase CFD simulations are carried out with Ansys Fluent where a User Defined Function implemented by the authors is used to properly model water droplet erosion mechanism and to obtain all the quantities needed to evaluate the washing quality. Results confirm the strong influence of the injection conditions on the main features of the washing system. The study is part of an ongoing partnership between Baker Hughes and Sapienza University of Rome aiming at maximizing the washing of the compressor blades while maintaining the erosion under specific thresholds.

Study on the noise reduction methods for a semi-hermetic variable frequency twin-screw refrigeration compressor

Abstract The semi-hermetic twin-screw refrigeration compressors have been widely applied in middle size refrigeration and heat pump systems. The operation noise of the twin-screw refrigeration compressor is becoming a crucial limitation for their applications. Two noise reduction methods for a semi-hermetic variable frequency twin-screw refrigeration compressor were developed. One is the end face attenuation passage and the other is the discharge pipe damping. The former reduces the gas pulsation amplitude in the discharge chamber which is the main noise source. The discharge pipe damping is comprised of several Helmholtz resonators and is installed in the discharge pipe within the compressor. Methods for the flow field and sound field simulation of the compressor were further proposed to verify the noise reduction effect of each method. Moreover, an experimental study was carried out to evaluate the general effect when the two method were applied in a real compressor. Both simulation results and experimental results demonstrated that both noise reduction methods had good effect. According to the experimental results, the actual operation noise level was no more than 85.5 dBA and the mean noise reduction value was in the range of 5.0 dBA–10.0 dBA when the rotational speed increased from 3000 rpm to 5100 rpm.

Impinging oil separator for compressors

Abstract Oil separation is commonly needed in air conditioning and refrigeration systems to reduce the oil circulation rate and keep the oil inside the compressor. For compactness, the oil separation structure integrated into the compressor is more and more popular than traditional external oil separators. However, a quantitative guideline for the design of oil separation structure is missing for irregular geometry and realistic flow condition at the compressor discharge. This paper presents impinging separation mechanism, one of the basic mechanisms of droplet separation. The separating structures are studied by flow visualization and experimental measurement under realistic compressor discharge conditions. The video of oil mist flowing through the baffles and plates is captured by a high-speed camera and analyzed quantitatively. The effect of the oil flow condition and the effect of separator geometry design are investigated based on the experimental data. With the help of flow visualization, the flow details are analyzed, and a semi-empirical model is proposed to predict the separation efficiency and pressure drop of a certain wave-plates structure. The results and conclusions from this study give useful guidelines about how to reduce the oil circulation ratio by designing oil separators.

A new method for multivariable nonlinear coupling relations analysis in complex electromechanical system

Coupling relations analysis of monitoring variables in complex electromechanical system is a powerful and useful means for abnormal detection, false monitoring information identification and fault diagnosis. However, due to the characteristics of multivariable, nonlinear and non-stationarity in the actual production process, it is difficult to achieve multivariable coupling modeling of complex electromechanical systems. In this paper, a new coupling modeling method is proposed by combining causality analysis with RBF neural network. First, considering the multivariate and nonlinearity of complex system, the conditional Granger and nonlinear Granger is combined to analyze the causal relations of process variables and obtain the cause variable set of any variable in complex system. Second, the RBF neural network is applied to achieve the nonlinear fitting and the parameter is optimized to ensure the fitting precision. Finally, the effectiveness of the proposed method is verified by an analysis of one case study of real compressor groups data set in chemical production system. This new approach can handle general coupling modeling problems and obtain a quantitative nonlinear coupling model by determine the dependent and independent variables in system and the functional relationship between them. Which dedicated to studying quantitative functional relationship of variable coupling, not just considering the direction or strength of coupling as in the existing, and does not need any prior knowledge about the physical structure. Thus, the proposed method can be effectively used in coupling modeling of complex electromechanical systems and formulate the foundation of anomaly detection, information quality assessment, and failure propagation mechanism, as well as other engineering applications.

Using Machine Learning Tools to Predict Compressor Stall

Abstract Clean energy has become an increasingly important consideration in today’s power systems. As the push for clean energy continues, many coal-fired power plants are being decommissioned in favor of renewable power sources such as wind and solar. However, the intermittent nature of renewables means that dynamic load following traditional power systems is crucial to grid stability. With high flexibility and fast response at a wide range of operating conditions, gas turbine systems are poised to become the main load following component in the power grid. Yet, rapid changes in load can lead to fluid flow instabilities in gas turbine power systems. These instabilities often lead to compressor surge and stall, which are some of the most critical problems facing the safe and efficient operation of compressors in turbomachinery today. Although the topic of compressor surge and stall has been extensively researched, no methods for early prediction have been proven effective. This study explores the utilization of machine learning tools to predict compressor stall. The long short-term memory (LSTM) model, a form of recurrent neural network (RNN), was trained using real compressor stall datasets from a 100 kW recuperated gas turbine power system designed for hybrid configuration. Two variations of the LSTM model, classification and regression, were tested to determine optimal performance. The regression scheme was determined to be the most accurate approach, and a tool for predicting compressor stall was developed using this configuration. Results show that the tool is capable of predicting stalls 5–20 ms before they occur. With a high-speed controller capable of 5 ms time-steps, mitigating action could be taken to prevent compressor stall before it occurs.

Different noise generation mechanisms of a controlled diffusion aerofoil and their dependence on Mach number

The self-noise of an isolated controlled-diffusion aerofoil is investigated using direct noise computations. The high-order finite difference solver HiPSTAR is used to conduct four large eddy simulations and one direct numerical simulation. The motivation is to investigate the multiple sources of aerofoil self-noise on a realistic compressor blade geometry, the physical mechanisms leading to the generation of sound, and the influence of compressibility effects. The angle of attack is 8°, the chord-based Reynolds number is 105, and results obtained for four values of the free-stream Mach number [0.2, 0.3, 0.4, 0.5] are compared. For those flow parameters, the pressure side is fully laminar, whereas a separation bubble is present on the suction side close to the leading edge that promotes transition to turbulence. The size of the separation bubble is found to increase with the Mach number. Two noise sources are observed, one at the trailing edge and one in the leading edge transition/reattachment region. The first has a broadband, low frequency spectrum, while the second displays a tone whose frequency depends on the local Mach number. Because the leading edge separation bubble is very small, the associated tone frequency is high and requires a significantly finer grid to faithfully resolve the acoustic propagation than what is typically deemed sufficient. Finally, cross correlations between the surface pressure and the far-field pressure reveal that the pressure fluctuations reaching the trailing edge are initially generated in the transition/reattachment region, which indicates that the trailing edge noise is a consequence of the pressure fluctuations generated by the separation bubble.

Development of a condition monitoring system for compressor equipment with neural network data analysis

Currently, one of the most widely used and effective types of technological equipment is screw compressor equipment. Along with the fact, that such equipment has a number of advantages that determine its high efficiency, it is characterized by increased wear of important structural elements. This can lead to reduced compressor efficiencies and malfunctions that can result in emergencies. In this regard, the paper presents the results of developing a scheme for continuous monitoring of the technical condition of screw compressor units. Variants of installing vibration sensors that provide data collection of vibration diagnostics are determined. In order to automate the analysis of the collected data, it is proposed to use the method of data mining based on neural networks to recognize the technical condition. The results of testing the neural network data method of a real compressor unit are presented.

Wavy plates as impinging oil separating structure at compressor discharge

Oil separation is commonly needed in air conditioning and refrigeration systems to reduce the oil circulation rate and keep the oil inside the compressor. For compactness, the oil separation structure integrated into the compressor is more and more popular than traditional external oil separators. This paper presents the impinging separation mechanism, one of the basic mechanisms of droplet separation. As a representative of the impinging separator, wave-plates structures are studied by flow visualization and experimental measurement under realistic compressor discharge conditions. The video of oil mist flowing through the baffles and plates is captured by a high-speed camera and analyzed quantitatively. The effect of the oil flow condition and the effect of separator geometry design are investigated based on the experimental data. With the help of flow visualization, the flow details are analyzed.

The Analysis of Leading Edge Deformations on Turbomachine Blades

With recent advancements in the development of material and manufacturing technology, the leading edge geometry of turbomachine blades has attracted widespread attention. “Sharp” leading edges always have a better aerodynamic performance, though it is prone to deformations easily. Thus, flat plates and real compressor cascades with different leading edge deformations were investigated to study the influence, which is applicable for thin blades at low speeds. Different boundary layer characteristics, including the velocity profile, transition process, and loss, are compared. The results show that there are several kinds of contradictory influence mechanisms and that the final phenomenon is closely related to the condition of the original boundary layer. In low turbulence, with large and laminar separation, the deformations can suppress separation and decrease loss. In high turbulence, with short and transitional separation, deformations can promote the transition process and increase the loss. The sensitivities of different the original leading edge shapes are also compared. This indicates that a good design always has a better robustness at low turbulence values, while it is worse at high turbulence values. The cascade experiment and simulation show that the deformation influence is similar to flat plates and that it is enlarged near the hub, which affects the corner separation.