Deep Learning-Based Identification of Cavitation Instabilities in Turbopump Inducers Using Raw Visualization Data

This article describes about various types of cavitation instabilities including rotating cavitation and alternate blade cavitation. In section 1, general characteristics of cavitating flow through turbopump inducers are reviewed. In section 2, rotating cavitation in 3- and 4-bladed inducers are discussed focusing on the effect of the number of blades. Section 3 is intended to show the relationship between the operating condition and various types of cavity oscillations in an inducer.

The aim of the present paper is to provide some highlights about the most interesting experimental activities carried out during the years 2000–2004 through the CPRTF (Cavitating Pump Rotordynamic Test Facility) at Centrospazio/Alta S.p.A. After a brief description of the facility, the experimental activities carried out on a NACA 0015 hydrofoil for the characterization of the pressure coefficient on the suction side and evaluation the cavity length and oscillations are presented. Then, the results obtained to characterize the performance and the cavitation instabilities on three different axial inducers are showed: in particular, a commercial three-bladed inducer, the four-bladed inducer installed in the LOX turbopump of the Ariane Vulcain MK1 rocket engine and the “FAST2”, a two-bladed one manufactured by Avio S.p.A. using the criteria followed for the VINCI180 LOX inducer. The most interesting results are related to the effects of the temperature on the cavitation instabilities on hydrofoils and inducers. Experiments showed that some instabilities, like the cloud cavitation on hydrofoils and the surge on inducers, are strongly affected by the temperature, while others seem not to be influenced by the thermal effects. In the final part of this paper, some indications of the main experimental activities scheduled for the next future are provided.

Cavitation instabilities in hydraulic machines, hydro turbines and turbopump inducers, are reviewed focusing on the cause of instabilities. One-dimensional model of hydro turbine system shows that the overload surge is caused by the diffuser effect of the draft tube. Experiments show that this effect also causes the surge mode oscillations at part load. One dimensional model of a cavitating turbopump inducer shows that the mass flow gain factor, representing the cavity volume increase caused by the incidence angle increase is the cause of cavitation surge and rotating cavitation. Two dimensional model of a cavitating turbopump inducer shows that various modes of cavitation instabilities start to occur when the cavity length becomes about 65% of the blade spacing. This is caused by the interaction of the local flow near the cavity trailing edge with the leading edge of the next blade. It was shown by a 3D CFD that this is true also for real cases with tip cavitation. In all cases, it was shown that cavitation instabilities are caused by the fundamental characteristics of cavities that the cavity volume increases with the decrease of ambient pressure or the increase of the incidence angle.

Temperature effects on the critical cavitation number and rotating cavitation in a turbopump inducer have been experimentally investigated in water. Static pressures upstream and downstream of the inducer have been measured to determine the cavitation performance, and cavitation instabilities have been detected using unsteady pressure sensors and a high-speed camera. Two kinds of cavitation instabilities have been identified—rotating cavitation and asymmetric attached cavitation. To quantify temperature effects, nondimensional thermal parameter has been adopted. Increasing water temperature, or increasing nondimensional thermal parameter, lowers the critical cavitation number. Increasing nondimensional thermal parameter also shifts the onset of rotating cavitation to a lower cavitation number and reduces the intensity of rotating cavitation. However, for values larger than 0.540 (340 K, 5000 rpm), the critical cavitation number and the rotating cavitation onset cavitation number become independent of the nondimensional thermal parameter. The onset of the head coefficient degradation correlates with the onset of rotating cavitation regardless of temperature.

The main results of an experimental campaign conducted in the Cavitating Pump Rotordynamic Test Facility at Alta S.p.A. are presented. The experiments have been carried out on two different axial inducers (a three-bladed aluminum-made inducer of simple helical geometry and a prototype of the axial inducer of the Vulcain MK1 engine liquid oxygen turbopump) to characterize the instabilities affecting the pumps in a wide range of flow conditions; some experiments have also been carried out at higher temperatures to investigate the possible influence of the thermal cavitation effects on the observed phenomena. The transparent inlet section of the facility has been instrumented with several piezoelectric pressure transducers located at three axial stations: inducer inlet, outlet, and middle of the axial chord of the blades. For each axial station, at least two transducers were mounted at a given angular spacing to cross correlate their signals for coherence and phase analysis. The most interesting instabilities have been detected on the three-bladed inducer, including a cavitation surge, a rotating stall, and an auto-oscillation leading to a violent surge-mode instability. Some of these instabilities have also been found to be slightly affected by temperature. On the other hand, very few oscillating phenomena have been detected on the MK1 inducer, with a practically flat frequency spectrum at flow coefficients near its nominal operating point.

The thermodynamic effect on a cavitating inducer is investigated from joint experiments in cold water and Refrigerant 114. The analysis is focused on leading edge cavitation and cavitation instabilities, especially on alternate blade cavitation and supersynchronous rotating cavitation. The cavity length along cylindrical cuts at different radii between the hub and casing is analyzed with respect to the local cavitation number and angle of attack. The similarity in shape of the cavity closure line between water and R114 is examined and deviation caused by thermodynamic effect is clarified. The influence of rotation speed on cavity length is investigated in both fluids and analyzed on the basis of a comparison of characteristic times, namely, the transit time and a thermal time. Thermodynamic delay in the development of leading edge cavities is determined and temperature depressions within the cavities are estimated. Thresholds for the onset of cavitation instabilities are determined for both fluids. The occurrence of cavitation instabilities is discussed with respect to the extent of leading edge cavitation. The thermodynamic delay affecting the occurrence of cavitation instabilities is estimated and compared with the delay on cavity development.

The exponentially increasing volumes of health data of different types is a challenge for reaching robust conclusions. The Health Vision platform overcomes this issue by providing advanced and multi-scaling data analysis and visualisation. The platform functions through user defined workflows, each organised into four distinct layers: 1. Raw Data, 2. Data Analytics, 3. Decision Support and 4. Data visualisation. In the Raw Data Layer, data from medical records, healthcare platforms, smart homes, monitoring sensors, and medical devices, clinical features from Electronic Health Records (EHRs), etc. can be considered, while the Data Analytics Layer provides data analysis services for data mining. The Decision Support Layer facilitates the selection of appropriate visualisation methods taking into consideration the specifications of the raw data and the output of the data analysis. Finally, the Data visualisation Layer provides intelligent visualisation methodology in order for the user to interact with the data and gain meaningful results. Visualisation is supported by a reactive workflow mechanism, allowing the composition of analysis and visualisation services in a custom application-specific data flow. We demonstrate and evaluate this novel approach in several use cases, where different types of data are involved, including: Clustering of Sjogren's syndrome patients, visualisation of Parkinson's disease patients, mutation association analysis in Chronic Lymphocytic Leukemia patients, behavioral analysis of asthma patients and clinical analysis based on clinical similarities of older adults.

Cavitation dynamics, such as cavitation surge and rotating cavitation, are natural performance limits for turbopump inducers and pose a significant risk to the development and operation of liquid rockets. In the last 63 years, progress has been made in understanding cavitation instabilities and their relationship to cavitation number, inlet flow coefficient, rotating speed, inducer size, and thermal effects. However, we are still unable to predict the cavitation behavior of a real inducer installed in rocket engines or contribute much to the design of a more cavitation-resistant machine. This paper comprehensively reviews findings from computational fluid dynamics simulations, reduced-order modeling, and experiments in cryogenic fluids and water, focusing on the effects of compressibility on multiphase flow calculations, the assessment of cavitation compliance and mass flow gain factor, and their influences on cavitation stability analysis, as well as the transfer function for cavitating inducers. This review will provide insights for future research to achieve more accurate performance predictions, improve flow modeling, and enhance the interpretation of experimental results.

Data visualization is basically a field that deals with graphical representation of information from raw data. Data visualization tools provide areachable way to see and understand the pattern, outliers and trends in data by using data tools like charts, graphs and maps. It makes decision based on knowing rather than guessing. Large amount of data can be accessible because of Data visualization. Data Visualization also helps in telling stories that highlight trends, pattern, and outliers in organization. Thispaper basically focuses on how data visualization is important in sector of education and also the role of dashboard in improvement of students performance and institutional growth.

Identification of inducer cavitation instabilities using high-speed visualization

Interactive visualization of APT data at full fidelity

This study aims to analyze pharmaceutical sales performance by leveraging big data and advanced information technology tools, providing insights for improved business strategies. Using the Exploratory Data Analysis (EDA) method, the study processes raw sales data through spreadsheet applications to identify key patterns and trends. The findings are then visualized with Looker Studio on an intelligent dashboard tailored to the needs of the marketing team. The dashboard enables quick, data-driven decision-making by displaying performance metrics and trends relevant to sales and marketing strategies. The results reveal critical insights into sales behavior, including high-performing products, regional sales variations, and temporal sales trends. The analysis equips the marketing team with actionable data to refine their strategies, prioritize product focus, and adjust marketing efforts in response to identified patterns. In conclusion, the use of EDA and data visualization provides a structured approach to big data interpretation, allowing the company to optimize business and marketing performance. The study underscores the potential of data-driven dashboards to enhance strategic planning, suggesting broader applications in various data-intensive business environments.

The present paper reports the experimental characterization of the cavitation induced flow instabilities of a high head three-bladed inducer at design condition detected simultaneously by means of piezoelectric pressure transducers located at different axial and azimuthal stations on the casing of the pump and through strain gages mounted on the pressure side of each blade. The purpose of this experimental campaign is a deeper insight of the phenomena associated to the flow instabilities obtained by the simultaneous analysis performed in the stationary and rotating frames. Both the sensors set-ups proved to be able to detect the main flow instabilities such as synchronous and sub-synchronous rotating cavitation, cavitation surge and tip vortex instability. Moreover, the simultaneous analysis of the results supported by high speed movie has allowed for assessing the mutual relations between the instabilities.

Across scientific disciplines, there is a rapidly growing recognition of the need for more statistically robust, transparent approaches to data visualization. Complementary to this, many scientists have called for plotting tools that accurately and transparently convey key aspects of statistical effects and raw data with minimal distortion. Previously common approaches, such as plotting conditional mean or median barplots together with error-bars have been criticized for distorting effect size, hiding underlying patterns in the raw data, and obscuring the assumptions upon which the most commonly used statistical tests are based. Here we describe a data visualization approach which overcomes these issues, providing maximal statistical information while preserving the desired 'inference at a glance' nature of barplots and other similar visualization devices. These "raincloud plots" can visualize raw data, probability density, and key summary statistics such as median, mean, and relevant confidence intervals in an appealing and flexible format with minimal redundancy. In this tutorial paper, we provide basic demonstrations of the strength of raincloud plots and similar approaches, outline potential modifications for their optimal use, and provide open-source code for their streamlined implementation in R, Python and Matlab ( https://github.com/RainCloudPlots/RainCloudPlots). Readers can investigate the R and Python tutorials interactively in the browser using Binder by Project Jupyter.

Monitoring the CO2 plume evolution is essential for ensuring geologic storage security and integrity. Traditional numerical simulation-based data assimilation workflow is computationally expensive, and this is further complicated by the fact that geologic uncertainty must be incorporated for robust performance prediction. Therefore, reservoir simulation and model calibration accounting for geologic uncertainty are not amenable to real time monitoring of the CO2 plume evolution for large-scale applications. We propose a deep learning-based approach which enables near real time CO2 plume visualization and rapid data assimilation incorporating multiple geological realizations for predicting future CO2 plume evolution and area of review (AOR) determination. The proposed deep learning-based data assimilation framework considers geological uncertainty utilizing multiple plausible models for training data generation. Rather than utilizing all available geologic realizations, a representative subset is sampled using dissimilarity measures in flow patterns computed via multidimensional scaling (MDS) and streamline time-of-flight. The approach substantially reduces training data generation cost while preserving the uncertainty inherent in the original ensemble of geomodels. The CO2 plume evolution is represented using ‘onset time’ images, depicting the calendar time when the CO2 saturation exceeds a prespecified threshold value at a given location. The use of a single CO2 onset time image instead of multiple CO2 saturation snapshots across different timesteps significantly reduces the dimensionality of the problem, making the deep learning model robust and scalable for large-scale field applications. A variational autoencoder encodes the onset time images into latent variables, which are predicted by another neural networks using the available monitoring data. The power and efficacy of the proposed method are demonstrated through application to a large-scale field case, the Illinois Basin-Decatur Carbon Sequestration Project. The available monitoring data consists of bottom-hole pressure at the injector, distributed pressure data and CO2 saturation log data at the monitoring well. Out of 200 geostatistical realizations, 10 representative models are selected by the MDS while preserving diversity of the geologic model. Additional calibration parameters including transmissibility and pore volume multipliers are applied to the selected realizations for generating a comprehensive training dataset. The trained ML model is then employed for reservoir model calibration, significantly accelerating the calibration process and enabling real time CO2 plume imaging from the monitoring data. The trained deep learning model achieves history matching of both pressure and saturation responses in seconds. The calibrated models are then used for forecasting future CO2 plume migration and the AOR assessment. The deep learning-based data assimilation approach enables near real time monitoring and verification of field-scale CO2 sequestration projects while accounting for geologic uncertainty. Utilizing the trained deep learning model, reservoir model calibration and prediction of CO2 plume evolution is performed within seconds, orders of magnitude faster compared to traditional history matching.