1,720,971 research outputs found
Identification of losses in turbomachinery with machine learning
No full textOne of the issues of handling large CFD datasets and process them to derive important design correlations is the limitation in automating the post-processing of data. Machine learning techniques, developed to process large unlabelled dataset, can play a key role on this subject. In this work an unsupervised approach to isolate different flow features inside a 2D cascade is proposed and validated. The approach relies on machine learning methods and in particular on Exploratory Data Analysis (EDA) and Principal Component Analysis for the pre-processing of the data and on K-means clustering for the post-processing. The K-means algorithm was trained on a Design of Experiments (DoE) of over 140 cases of 2D linear cascade configurations to identify the boundary layer on the profiles and the wake downstream. Validation resulted in a perfect capability of identifying the regions of interest. Then a possible exploitation of this method is presented, to compute pressure losses downstream of the cascade and train an artificial neural network to make a regression able to extend data to all the possible combinations of geometrical and operating parameters of the cascade. The same algorithm was applied to 3D flow cascades of profiles with sinusoidal leading edges to stress its extrapolation capability in case of flow regimes not present in the training DoE
Assessment of multall as cfd code for the analysis of tube-axial fans
No full textThis work deals with the application of the open source CFD code MULTALL to the analysis of tube-axial-fans. The code has been widely validated in the literature for high-speed turbomachine flows but not applied yet to low speed tutbomachines. The aim of this work is to assess the degree of reliability of MULTALL as a tool for simulating the internal flow in industrial axial-flow fan rotors. To this end, the predictions of the steady-state air flow field in the annular sector of a 315mm tube-axial fan obtained by MULTALL 18.3 are compared with those obtained by two state-of-the-art CFD codes and experimental data of the global aerodynamic performance of the fan and the pitch-wise averaged velocity distribution downstream of the rotor. All the steady-state RANS calculations were performed on either fully structured hexahedron or hexadominant grids using classical formulations of algebraic turbulence models. The pressure curve and the trend of the aeraulic efficiency in the stable operation range of the fan predicted by MULTALL show very good agreement with both the experimental data and the other CFD results. Although the estimation of the fan efficiency predicted by MULTALL can be noticeably improved by the more sophisticated state-of-the-art CFD codes, the analysis of the velocity distribution at the rotor exit supports the use of MULTALL as a reliable CFD analysis tool for designers of low-speed axial fans
Development and validation of a novel synthetic blade model for axial flow fans in unsteady CFD
No full textHere we present a Synthetic Blade Model (SBM) for axial flow fans, derived from Actuator Disk and Actuator Line theories. This new approach is able to model the momentum exchange between the fan and the fluid by adding source terms into momentum equation, like an actuator disk. However, the model accounts for the position of the blades, their rotation and the non-uniform distribution of deflection capability in the blade-to-blade passage, like in an actuator line model. This approach is derived, described and validated against available data on a reversible tunnel and metro fan
Challenges of renewable energy communities on small Mediterranean islands. A case study on Ponza island
No full textItaly has reconsidered incentives on Renewable Energy Sources (RES) to implement the EU clean energy transition and Renewable Energy Communities (REC). Legislation on RECs shifted the focus of incentives from production to self-consumptions. In this paper, the REC model is studied in a minor island disconnected from the national grid, with strongly seasonal energy load and water demand. This is a typical Mediterranean scenario where energy demand is covered by Diesel gensets and water supply is provided with tankers. The implementation of RES is investigated, exploiting the current REC model of incentives. Two sub-optimal REC configurations were defined using time-dependent simulations on pyRES, an in-house code for energy systems. Results show the economical unfeasibility of REC when there is a poor mix of users. In contrast, REC can achieve economic profitability including industrial demand of a desalination unit (DES). The increase of self-consumption guaranteed from the DES allows to increase the NPV of the community and results in major cuts in CO2 emissions (60% of those related to water supply and 10% of those from Diesel gensets) and a reduction of fuel costs of 22%. This method can be applied to investigate the performance of a REC in a local environment and help stakeholders in planning the expansions of RECs on the territory
Numerical investigation of CSP air cooled condenser fan
Full text linkCSP power plants are usually installed in desert regions due to the high availability of direct normal irradiance during the year. This usually entails the unavailability of water to be used for condensation in Rankine cycle and the consequent choice of air-cooled condensers (ACCs) [1].
When dealing with ACCs the main mechanical component is the fan, that drives fresh air towards the heat exchangers driving all the process. The main characteristic of this fan is its size that spans from 7 to 13 m of external diameters, thus resulting in a series of technical problems to be addressed. Such high-volume and low speed fans present critical problem during the testing phase [2]. This dissertation focuses on the certification of fan performance, that follows ISO 5801 [3]
CFD analysis of ventilation systems for gas turbine enclosures
No full textThe fans performance are assessed in configurations according, traditionally, to international standards such as ISO-5801. Similar reference to the so-called standardised airways is used in the design process of industrial fans which provide the ISO test-bench. In recent years, the objective of fan optimization has, in most cases, been re-focused on the optimization of the fan-system coupling in a view to solve for the dynamics of such coupling (either mechanically or aerodynamically). This has been true when ventilating systems are equipped with components such as bended inlets, spinner cones, gravity dampers, that could be responsible for influencing the fan aerodynamic response. In all these cases, when using CFD, accounting for a real fan geometry inside the domain would be extremely expensive from a computational point of view and too slow for industrial purposes. When the objective under investigation is the whole ventilation system, it is necessary to account for the single components by means of synthetized methodologies. In particular, fans can be accounted for by simple pressure rise or actuator disks that synthetize their aerodynamic response in the systems of ducting by means of body forces inside the momentum equation. In the paper the methodology is briefly presented and applied to resolve the fluid-dynamics behaviour of two gas turbine ventilation systems
Assessment of a machine-learnt adaptive wall-function in a compressor cascade with sinusoidal leading edge
No full textNear-wall modelling is one of the most challenging aspects of CFD computations. In fact, integration-to-the-wall with low-Reynolds approach strongly affects accuracy of results, but strongly increases the computational resources required by the simulation. A compromise between accuracy and speed to solution is usually obtained through the use of wall functions, especially in RANS computations, which normally require that the first cell of the grid to fall inside the log-layer (50 < y+ < 200) [1]. This approach can be generally considered as robust, however the derivation of wall functions from attached flow boundary layers can mislead to non-physical results in presence of specific flow topologies, e.g. recirculation, or whenever a detailed boundary layer representation is required (e.g. aeroacoustics studies) [2]. In this work, a preliminary attempt to create an alternative data-driven wall function is performed, exploiting artificial neural networks (ANNs). Whenever enough training examples are provided, ANNs have proven to be extremely powerful in solving complex non-linear problems [3]. The learner that is derived from the multi-layer perceptron ANN, is here used to obtain two-dimensional, turbulent production and dissipation values near the walls. Training examples of the dataset have been initially collected either from LES simulations of significant 2D test cases or have been found in open databases. Assessments on the morphology and the ANN training can be found in the paper. The ANN has been implemented in a Python environment, using scikit-learn and tensorflow libraries [4][5]. The derived wall function is implemented in OpenFOAM v-17.12 [6], embedding the forwarding algorithm in run-time computations exploiting Python3.6m C_Api library. The data-driven wall function is here applied to k-epsilon simulations of a 2D periodic hill with different computational grids and to a modified compressor cascade NACA aerofoil with sinusoidal leading edge. A comparison between ANN enhanced simulations, available data and standard modelization is here performed and reported
Aeroacoustic assessment of leading edge bumps in industrial fans
No full textStall control is a key technology for industrial fans. In tunnel and metro applications this technologies are regarded as a key assets, as regulatory and operational conditions limit the viable design choices. In an ongoing investigation we are assessing the stall control capabilities of biomimicry-derived sinusoidal leading edges. In the present paper we focus on the prediction of acoustic emissions of blades with sinusoidal leading edges and compare the acoustic performance with that of the datum blade with straight leading edge. To this aim we performed URANS calculation with the non-linear low-Reynolds model of Lien et al. and applied SPL prediction from Fukano et al. The paper demonstrate an overall decrease of SPL and a redistribution of Powell’s sound sources along the blade surface
Exploration of Axial Fan Design Space with Data-Driven Approach
Full text linkSince the 1960s, turbomachinery design has mainly been based on similarity theory and empirical correlations derived from experimental data and manufacturing experience. Over the years, this knowledge was consolidated and summarized by parameters such as specific speed and diameters that represent the flow features on the meridional plane, hiding however the direct correlations between all the actual design parameters (e.g., blade number or hub-to-tip ratio). Today a series of statistical tools developed for big data analysis sheds new light on correlations among turbomachinery design and performance parameters. In the following article we explore a dataset of over 10,000 axial fans by means of principal component analysis and projection to latent structures. The aim is to find correlations between design and performance features and comment on the capabilities of this approach to give new insights on the design space of axial fans
- …
