Test Rig Data Research Articles

Sparse graph structure fusion convolutional network for machinery remaining useful life prediction

Effective prediction of machinery remaining useful life (RUL) is prominent to achieve intelligent preventive maintenance in manufacturing systems. In this paper, a sparse graph structure fusion convolutional network (SGSFCN) is proposed for more accurate end-to-end RUL prediction of machine. A novel node-level graph structure called time series shapelet distance graph (TSSDG) is designed to convert the time series to node feature. The SGSFCN model is proposed to learn degradation information from the graph structure. In SGSFCN, a sparse graph structure (SGS) layer and a fusion graph structure (FGS) layer preceding the graph convolutional network (GCN) are designed to learn the SGS from node representation and fuse the original graph structure, enabling the graph structure and node update iteratively in subsequent layers. Concurrently, a bidirectional long short-term memory network (BiLSTM) layer is integrated to capture the global temporal dependencies. The method is validated by two test rig data, and results demonstrate that the proposed method offers significantly higher prediction accuracy of RUL compared to several state-of-art methods.

Adaptive mode decomposition method based on fault feature orientation and its application to compound fault diagnosis of planetary gearboxes

Abstract Planetary gearboxes often experience multiple component failures during service, which can accelerate the degradation and failure of industrial equipment. Accurate separation and identification of multiple faults is an important means of ensuring the safe and stable operation of equipment. However, different faults can interact with each other, along with the influence of background noise, making it challenging to accurately extract faults with relatively weak energy among multiple faults. This difficulty leads to the problems of potential misdiagnosis and underdiagnosis. To address this issue, an adaptive mode decomposition method based on fault feature orientation (AMD-FF) is proposed in this paper. Initially, a fault impact indicator (FII) is constructed based on period-weighted kurtosis of envelope spectral and correlated combination negentropy to effectively characterize the impulsiveness and periodicity of fault features. Furthermore, with the objective of maximizing the FII, an adaptive decomposition of the original signal is designed based on blind convolution theory using a finite-impulse response filter group. Subsequently, a variable weight particle swarm optimization is employed to adaptively optimize the key decomposition parameters. Finally, the data of industrial-grade planetary gear transmission test rig are collected to validate the proposed method for compound fault diagnosis of planetary gearboxes. The results indicate that the AFMD-FF can effectively separate and extract compound faults in planetary gearboxes, demonstrating superior fault separation and diagnostic performance compared to the fault mode decomposition (FMD) and adaptive FMD. This method offers a novel approach to diagnosing compound faults in rotating equipment in industrial scenarios.

Integration of STM32 Microcontroller with Arm Cortex Architecture into the 8-Bit Measurement Circuit Customized for Real-Time Data Acquisition on Propeller Shaft to Increase Data Speed and Accuracy

Propeller shafts transmit torque and high-speed rotational movement from the engine to related equipment. Expected propeller shaft functions are calculated and designed by using an analytical and numerical approaches. The verification tests are significant in that they provide feedback to the design and development processes. Therefore, acquiring accurate data from power transmission equipment is becoming even more important. This paper focuses on the study of circuit boards customized for use in the validation tests of the propeller shafts. It should be noted that these measurement circuits were integrated on the propeller shafts. These measurement circuits were used to acquire data such as torque, temperature, etc. in real time. Especially with instantaneous signals such as torque, data loss can occur due to low microcontroller resolution. Therefore, it was aimed to develop to the microcontroller to 32-bit resolution from 8-bit resolution to increase data speed and accuracy. As the first step, an INA125P operational amplifier was installed to amplify the low voltage values gathered from the gages and the sensors. The gain value of the amplifier was calculated in order to provide the highest data resolution and sensitivity. And then, ADS1115 analog to digital converter circuit with a rate of 860 samples per second and 16-bit resolution was used to interpret the analog data. A low pass filter was installed to remove noise from the output signal. Embedded code was also locally developed for the hardware installed. The system was calibrated on a validated test rig. Real-time torque and temperature data were acquired from the propeller shaft. It is observed that compared to the previous 8-bit system, the data accuracy and integrity from the new 32-bit board has increased by a factor of 5 to 100 Hz. The collected data was found to be 99.4% compatible with the validated test rig data.

Research on temperature field of dry clutch assembly based on inverse analysis of thermal boundary conditions

The clutch temperature field is critical for the clutch design and lifetime analysis. In this paper, a new approach for reconstructing the temperature field of the clutch assembly based on the inverse analysis of thermal boundary conditions is present. Three typical types of thermal boundary conditions during the clutch operation were analyzed and integrated into a numerical model with several coefficients assumed. The optimal coefficients are obtained by optimizing the derivation between temperature simulation and rig test based on the reverse problem analysis. The NLPQL optimization algorithm, coupled with the Kriging response surface model, is utilized to improve the computation efficiency. Finally, the optimum parameter combination is conducted and the effectiveness of the proposed method is validated by the rig test data under the continuous engagement and disengagement condition. The results indicate that the clutch temperature distribution and the dynamic mechanism of the clutch contact status using the approach proposed agrees very well with the test data of the rig test. The temperature deviations between the model corrected and the rig test data are within 10 °C, with over 80% of the temperature deviations being less than 5 °C. This study establishes a new clutch temperature field simulation method based on reverse problem analysis and gives a significant insight and foundation for improving the prediction of the temperature distribution of the clutch.

Intelligent Fault Diagnosis of Gearbox Under Variable Working Conditions With Adaptive Intraclass and Interclass Convolutional Neural Network.

The industrial gearboxes usually work in harsh and variable conditions, which results in partial failure of gears or bearings. Accordingly, the continuous irregular fluctuations of gearbox under variable conditions maybe increase the intraclass difference and reduce the interclass difference for the monitored samples. To this end, a new intelligent fault diagnosis method of gearbox based on adaptive intraclass and interclass convolutional neural network (AIICNN) under variable working conditions is proposed. The core of the proposed algorithm is to apply the designed intraclass and interclass constraints to improve the distribution differences of samples. Meanwhile, the adaptive activation function is added into the 1-D convolutional neural network (1dCNN) to enlarge the heterogeneous distance and narrow the homogeneous distance of samples. Specifically, the training sample subset with intraclass and interclass spacing fluctuations under variable conditions is first converted into frequency domain through the fast Fourier transform (FFT), and the designed AIICNN algorithm is employed for model training. Afterward, the testing subset is provided to the trained AIICNN algorithm for fault diagnosis. The experimental data of the planetary gearbox test rig verify the feasibility of the proposed diagnosis method and algorithm. Compared with other methods, this method can eliminate the difference of sample distribution under variable conditions and improve its diagnostic generalization.

Fingerprint feature and dynamic threshold mechanism based on acoustic emission for bearing fault detection

The acoustic emission (AE) technology has emerged as a promising diagnostic tool for the damage detection of bearings. However, the conventional time-domain feature extraction based on AE hits lacks adequate consideration of the periodicity related to faults. To address this issue, taking the outer ring of a bearing as an example and studying the periodic AE hit behavior of bearings, a condition monitoring method, called fingerprint feature (FPF), is established in this study, providing a visual mode for bearing fault detection, along with the ability to track the bearing state. Since the dynamic threshold is essential to the formation of FPF, the dynamic threshold mechanism is investigated in detail, and a threshold coefficient Rt is defined to obtain the adaptive dynamic threshold, which allows instant AE hit extraction. The proposed distributed aggregation index can not only optimize Rt but also help evaluate the damage state of bearings. In addition, based on the FPF, a fingerprint fault spectrum (FFS) is proposed by associating the hit statistics per unit time with the fault characteristic frequency. Finally, the validity of the proposed method is proved by Hilbert envelope demodulation and its capability and practicality are verified using test rig data of rolling bearings and high-speed train bearings that operate close to actual working conditions. This study provides valuable insights into the development of online condition monitoring for bearings in industrial environment.

Analysis of cyclic frosting and defrosting of a vehicle heat pump

Heat pumps are used for energy-efficient heating in electric vehicles. If ambient air is used as a heat source, frost may form on the surface of the exterior heat exchanger at low ambient temperatures. If significant amounts of frost accumulate, the heat exchanger performance degrades and it must be defrosted. The aim of this work is the energy analysis of a heat-release-controlled vehicle heat pump with a direct flat tube evaporator in cyclic frosting and defrosting operation. The effects of fan speed and number of refrosting phases on the heat quantity released per frosting phase are analyzed on a laboratory test rig. The experiments show that the heat quantity released decreases with lower fan speed and increasing number of frosting phases. Defrosting operation parameters are also examined using the test rig and it is found that shorter defrosting times can be achieved at higher compressor speeds and larger expansion valve openings. A numerical model of the heat pump in cyclic frosting and defrosting operation is developed and validated using test rig data. An average relative deviation of less than 7% is determined between the model and experimental data. The heat pump is then integrated in a prototypical vehicle and the model is calibrated with experimental data from test drives. A numerical study of the validated vehicle heat pump model in cyclic frosting and defrosting operation shows a maximum increase of the mean total electric power consumption of 33% between dry air and saturated moist air at an ambient temperature of 0 °C. At an ambient temperature of -10 °C the mean total electric power consumption increases by 17% from dry air to saturated moist air conditions.

Variable-Geometry Rotating Components Modeling Based on Reference Characteristic Curves for the Variable Cycle Engine

The variable cycle engine switches working modes by way of changing variable-geometry components to achieve the dual advantages of high unit thrust and low specific fuel consumption. Due to the lack of a large amount of rig test data and the complex modeling of rotating components, the incomplete characteristics of the variable-geometry rotating components lead to the non-convergence of the component-level model of the variable cycle engine, which makes it difficult to design the follow-up control system. Aiming at this problem, a characteristics modeling method of variable-geometry rotating components for variable cycle engine based on reference characteristic curves is proposed in this paper. This method establishes a neural network estimation model for the offset coefficients of key component operating points based on the characteristic law of the maturely designed variable-geometry rotating component. Combining the neural network model and the reference characteristic curves of the variable-geometry component to be designed, the offset positions of the operating points for positive and negative guide vane angles are determined. Instead of directly connecting operating points to generate characteristic lines, this paper solves the Bezier curve optimization problem based on sequential quadratic programming (SQP) to smoothly fit characteristic lines. Thereby, component characteristics that conform to the actual variable-geometry characteristics can be quickly established in the absence of rig test data. The simulations show that the characteristics of the variable-geometry rotating components established by the proposed method have satisfactory accuracy and reliability, which further improves the operation stability of the component-level model of the variable cycle engine.

Modeling and Prediction of the Aging Process of a Solid Oxide Fuel Cell

This paper presents an approach to analyze the aging process of a solid oxide fuel cell. For this purpose, the current-voltage characteristic is employed as a suitable model to characterize the electrical behavior and the performance of fuel cells. Since this characteristic directly reflects the aging process, a parameter-dependent static ansatz is established that approximates the current-voltage characteristic. A thorough analysis is supported by a visualization of possible parametric changes and their impact on the electrical behavior. Using two exemplary data sets of an existing test rig, time-dependent variations of these parameters are identified to model the aging process properly. Thereby, a prediction becomes possible, which is shown exemplary for available test rig data. The authors want to emphasize that the proposed approach is applicable to other types of fuel cells and larger data sets with more measurement points over time.

A Deep Learning Based Data Recovery Approach for Missing and Erroneous Data of IoT Nodes.

Internet of things (IoT) nodes are deployed in large-scale automated monitoring applications to capture the massive amount of data from various locations in a time-series manner. The captured data are affected due to several factors such as device malfunctioning, unstable communication, environmental factors, synchronization problem, and unreliable nodes, which results in data inconsistency. Data recovery approaches are one of the best solutions to reduce data inconsistency. This research provides a missing data recovery approach based on spatial-temporal (ST) correlation between the IoT nodes in the network. The proposed approach has a clustering phase (CL) and a data recovery (DR) phase. In the CL phase, the nodes can be clustered based on their spatial and temporal relationship, and common neighbors are extracted. In the DR phase, missing data can be recovered with the help of neighbor nodes using the ST-hierarchical long short-term memory (ST-HLSTM) algorithm. The proposed algorithm has been verified on real-world IoT-based hydraulic test rig data sets which are gathered from things speak real-time cloud platform. The algorithm shows approximately 98.5% reliability as compared with the other existing algorithms due to its spatial-temporal features based on deep neural network architecture.

Automated Model Calibration for Urea-SCR Systems Using Test-Rig Data

Automated Model Calibration for Urea-SCR Systems Using Test-Rig Data

Fan Stability With Leading Edge Damage: Blind Prediction and Validation

Abstract This paper considers the impact of a damaged leading edge (LE) on the stall margin and stall inception mechanisms of a transonic, low-pressure ratio fan. The damage takes the form of a squared-off leading edge over the upper half of the blade. Full-annulus, unsteady CFD simulations are used to explain the stall inception mechanisms for the fan at low- and high-speed operating points. A combination of steady and unsteady simulations shows that the fan is predicted to be sensitive to leading edge damage at low-speed, but insensitive at high-speed. This blind prediction aligns well with rig test data. The difference in response is shown to be caused by the change between subsonic and supersonic flow regimes at the leading edge. Where the inlet relative flow is subsonic, rotating stall is initiated by the growth and propagation of a subsonic leading edge flow separation. This separation is shown to be triggered at higher mass flow rates when the leading edge is damaged, reducing the stable flow range. Where the inlet relative flow is supersonic, the flow undergoes a supersonic expansion around the leading edge, creating a supersonic flow patch terminated by a shock on the suction surface. Rotating stall is triggered by the growth of this separation, which is insensitive to the leading edge shape. This creates a considerable difference in sensitivity to damage at low- and high-speed operating points.

Performance analysis of auxiliary entrainment ejector used in multi-evaporator refrigeration system

• Effect of diffuser angle on auxuiliary ejector performance is identified. • Effect of diffuser angle on auxiliary inlet position, angle and, ER is investigated. • An optimal parameters (θ = 4°, LR = 1.9, and γ = 45°) of the new-type ejector exists. • An optimal operating scheme exists under different evaporating temperatures. In this paper, a numerical method is used to maximize the performance of an auxiliary entrainment ejector used in multi-evaporator refrigeration system developed for refrigerated trucks in tropical and humid climates. The Computational Fluid Dynamics model based on the previous study is developed to discuss the effects of key design parameters of an auxiliary entrainment ejector. The simulation model is validated against experimental data of an R134a test rig. The numerical models agree fairly well with the experiments. In order to maximize the efficiency of the auxiliary inlet, the diffuser angle is considered critically. The influence of diffuser angle on the optimal position and angle of the auxiliary inlet is then discussed. As well, the effect of secondary flow and auxiliary flow at varying temperatures on the efficiency of auxiliary entrainment ejector is analyzed. The main results are given as follows: (1) optimal auxiliary inlet position as well as angle and entrainment ratio depend on diffuser angle, and optimum parameters combination (length ratio of 1.9, inlet angle of 45°, and diffuser angle of 4°) of the new-type ejector exists; (2) with the secondary flow working temperature of −10 °C, the entrainment ratio reaches its maximum value of 0.658, and the corresponding entrainment ratio at secondary inlet is 2.60; (3) the auxiliary entrainment ejector can only function if the auxiliary flow working temperature is between −30 °C to −22 °C; (4) it is best to connect the secondary inlet to the medium temperature evaporator, with the auxiliary inlet to the low temperature evaporator, when using the secondary and auxiliary inlet connections to separate evaporators with varied operating temperatures.

Bearing signal models and their effect on bearing diagnostics

Signal models of rolling-element bearings have been a fundamental tool for the development and justification of new bearing signal processing techniques. A cyclostationary model proposed twenty years ago is still the most common reference for physically-justified bearing diagnostics. Surprisingly, a pseudo-cyclostationary bearing model proposed just a few years later has been almost entirely neglected. Moreover, the validity of the two models has not been tested against actual bearing signals, and there is limited knowledge about the quantitative effect of their parameters. This paper therefore aims at discussing the physical motivations of the two models, developing a single generalised model which includes the properties of both, and assessing the impact of key model parameters, corresponding to physical properties of the bearing, on the signal in the time and frequency domains. This work’s main novelties are analytical and include an assessment of truly- vs pseudo-cyclostationarity in rolling element bearing signals and of the relative importance of first- and second-order (pseudo-) cyclostationary fault symptoms. The first aspect sheds light on the spectral spread of bearing fault harmonics in both the spectrum and envelope-spectrum, while the second provides useful information for the selection of the best diagnostic indicator. Application of the model is illustrated on test-rig data using a bearing with an outer race fault.

Experimental Studies on Outer Race Fault Diagnosis of Rolling Element Bearings

Rolling element bearing condition monitoring and fault diagnosis can predict the condition of bearing and its components such as outer race, inner race, rolling elements and cage at any given instant. Online condition monitoring is more advanced and preferred over conventional methods. Advanced signal processing techniques and statistical analysis techniques aid in the component level fault prediction, its severity, interaction of different defect frequencies and remaining life of the component in a given system. However, the basic concept in most of the techniques is vibration data acquisition and identification of amplitude of vibration at defect frequencies. In this paper, the basic concepts of bearing defect frequencies are discussed and shown experimentally the presence of these frequencies. In experimentation 50mm diameter bearings with different faults are mounted on a simple test rig and vibration data is acquired under different loads at 2500 rpm. A simple frequency domain has been carried out to identify the defect.

Improvement of Turboshaft Restart Time Through an Experimental and Numerical Investigation

Abstract Inflight shutdown of one engine for twin-engine helicopters has proven beneficial for fuel consumption. A new flight mode is then considered, in which one engine is put into sleep mode while the second engine runs at nominal load. The ability to restart the engine in sleep mode is then critical for safety reasons. Indeed, the certification of this flight mode involves ensuring a close-to-zero failure rate for in-flight restarts and a fast restart capability of the shutdown engine (focus of this paper). Fast restart capability is necessary in case of a failure of the operating engine. Indeed, there is no more power available, and the helicopter can lose up to 15–20 meters per second during autorotation. The restart time becomes a critical parameter to limit the loss of altitude. The aim of the paper is to assess the potential restart time saving using an approach combining test rig data analysis and numerical results generated by a thermodynamic model able to simulate at low rotational speed. It is important to understand the detailed phenomenology of the startup process and the various subsystems involved, first to highlight the influencing parameters and then to establish an exhaustive listing of the possible time optimizations. The results of this study show that a fast restart going from sleep mode to max power speed can be up to 60% faster than a conventional restart going from sleep mode to idle speed, which is significantly faster.

Generating axial compressor maps to zero speed

Gas turbine performance models typically rely on component maps to characterize engine component performance throughout the operational regime. For the sub-idle case, the lack of reliable rig test data or inability to run design codes far from design conditions entails that component maps have to be generated from the extrapolation of existing data at higher speeds. This undermines the accuracy of whole-engine sub-idle performance models, at times impacting engine development and certification of aviation engines and the accuracy of start-up performance prediction in industrial gas turbines. One of the main components driving this issue is the core compression system, which can present operability concerns during light-up and which also sets the combustor airflow required for ignition. This paper presents, discusses, and draws on previous approaches to describe a method enabling the creation of sub-idle compressor maps from analytical and physical grounds. The method relies on the calculation of zero-speed and torque-free lines to generate a map down to zero speed along with analytical interpolation. A method for the interpolation process is described. A sensitivity study is carried out to assess the effects that different elements of the map generation process may have on the accuracy of the resulting performance calculation. Overall, a method for the generation of accurate, consistent maps from limited geometry data is identified.

Experimental analysis of a micro-scale organic Rankine cycle system retrofitted to operate in grid-connected mode

This work presents the retrofitting of a micro-scale organic Rankine cycle system originally designed for off-grid operation, modified to operate in grid-connected mode. Quantitative and qualitative performance comparisons of the grid-connected and off-grid connected organic Rankine cycle system are made based on experimental data of an organic Rankine cycle test rig. The operating strategies of the two systems are discussed in detail. The organic Rankine cycle test rig has a nominal power output of 1 kW with R245fa as the working fluid, a scroll expander as the expansion machine and steam as the heat source. Initially, the experiments were performed in off-grid operation mode with a self-excited alternating current generator. Subsequently, the organic Rankine cycle test rig was retrofitted for grid-connected mode, replacing the alternating current generator with an induction motor, regenerative variable frequency drive and grid connection. The rig modification was carefully done in such a way that the expander rotational speed was not fixed by the grid frequency, offering an additional control parameter for operational control optimisation. The modified system is able to supply 1.162 kW of gross electric power to the grid while a net electric output of 0.967 kW is measured. The same system is able to generate 1.016 kW (gross) and 0.838 kW (net) electric output in grid-connected mode. The net thermal-to-electric conversion efficiency at peak power generation is 7.36% and 4.66% in grid-connected and off-grid operating mode, respectively.

Research on Modeling of a Micro Variable-Pitch Turboprop Engine Based on Rig Test Data

Exact component characteristics are required for establishing an accurate component level aeroengine model. When component characteristics is lacking, the dynamic coefficient method based on test data, is suitable for establishing a single-input and single-output aeroengine model. When it is applied to build multiple-input, multiple-output aeroengine models, some parameters are assumed to be unchanged, which causes large error. An improved modeling method based on rig data is proposed to establish a double-input, double-output model for a micro variable-pitch turboprop engine. The input variables are fuel flow and pitch angle, and the output variables are rotational speeds of the core engine and the propeller. First, in order to gather modeling data, a test bench is designed and rig tests are carried out. Then, two conclusions are obtained by analyzing the rig data, based on which, the power turbine output is taken as the function of the core speed and the propeller speed. The established model has the property that the input variables can vary arbitrarily within the defined domain, without any restriction to the output variables. Simulation results showed that the model has a high dynamic and steady-state accuracy. The maximum error was less than 8%. The real-time performance was greatly improved, compared to the component level model.

The benefit of droplet injection on the performance of an ejector refrigeration cycle working with R245fa

Abstract This work presents a combined experimental and thermodynamic approach used to investigate the influence of droplet injection on the performance of an ejector-based refrigeration cycle developed for Heating, Ventilation and Air-Conditioning (HVAC) applications. Thermodynamic models have been developed within MATLAB for each component of the cycle. They are validated against experimental data of an R245fa test rig installed at Hydro-Quebec's Energy Technologies Laboratory (LTE). The numerical models agree fairly well with the experiments. The results show that injecting R245fa droplets at the end of the ejector diffuser with different glycol temperatures ranging from 20 to 26 °C has a significant impact on the performance of the ejector itself and more interestingly on the performance of the whole cycle. It reduces especially and significantly the temperature and the pressure of the overheated vapors at the outlet of the ejector leading to a significant improvement of the thermodynamic coefficient of performance (COP) by up to 20%.