Gas liquid cylindrical cyclone flow regime identification using machine learning combined with experimental mechanism explanation

The flow regimes of GLCC with horizon inlet and a vertical pipe are investigated in experiments,and the velocities and pressure drops data labeled by the corresponding flow regimes are collected.Combined with the flow regimes data of other GLCC positions from other literatures in existence,the gas and liquid superficial velocities and pressure drops are used as the input of the machine learning algorithms respectively which are applied to identify the flow regimes.The choosing of input data types takes the availability of data for practical industry fields into consideration,and the twelve machine learning algorithms are chosen from the classical and popular algorithms in the area of classification,including the typical ensemble models,SVM,KNN,Bayesian Model and MLP.The results of flow regimes identification show that gas and liquid superficial velocities are the ideal type of input data for the flow regimes identification by machine learning.Most of the ensemble models can identify the flow regimes of GLCC by gas and liquid velocities with the accuracy of 0.99 and more.For the pressure drops as the input of each algorithm,it is not the suitable as gas and liquid velocities,and only XGBoost and Bagging Tree can identify the GLCC flow regimes accurately.The success and confusion of each algorithm are analyzed and explained based on the experimental phenomena of flow regimes evolution processes,the flow regimes map,and the principles of algorithms.The applicability and feasibility of each algorithm according to different types of data for GLCC flow regimes identification are proposed.

Noninvasive Flow Regime Identification for Wet Gas Flow Based on Flow-induced Vibration

A novel noninvasive approach, based on flow-induced vibration, to the online flow regime identification for wet gas flow in a horizontal pipeline is proposed. Research into the flow-induced vibration response for the wet gas flow was conducted under the conditions of pipe diameter 50 mm, pressure from 0.25 MPa to 0.35 MPa, Lockhart-Martinelli parameter from 0.02 to 0.6, and gas Froude Number from 0.5 to 2.7. The flow-induced vibration signals were measured by a transducer installed on outside wall of pipe, and then the normalized energy features from different frequency bands in the vibration signals were extracted through 4-scale wavelet package transform. A "binary tree" multi-class support vector machine(MCSVM) classifier, with the normalized feature vector as inputs, and Gaussian radial basis function as kernel function, was developed to identify the three typical flow regimes including stratified wavy flow, annular mist flow, and slug flow for wet gas flow. The results show that the method can identify effectively flow regimes and its identification accuracy is about 93.3%. Comparing with the other classifiers, the MCSVM classifier has higher accuracy, especially under the case of small samples. The noninvasive measurement approach has great application prospect in online flow regime identification.

Identification Method of Gas-Liquid Two-phase Flow Regime Based on Image Multi-feature Fusion and Support Vector Machine

The knowledge of flow regime is very important for quantifying the pressure drop, the stability and safety of two-phase flow systems. Based on image multi-feature fusion and support vector machine, a new method to identify flow regime in two-phase flow was presented. Firstly, gas-liquid two-phase flow images including bub- bly flow, plug flow, slug flow, stratified flow, wavy flow, annular flow and mist flow were captured by digital high speed video systems in the horizontal tube. The image moment invariants and gray level co-occurrence matrix texture features were extracted using image processing techniques. To improve the performance of a multiple classifier system, the rough sets theory was used for reducing the inessential factors. Furthermore, the support vector machine was trained by using these eigenvectors to reduce the dimension as flow regime samples, and the flow regime intelligent identification was realized. The test results showed that image features which were reduced with the rough sets theory could excellently reflect the difference between seven typical flow regimes, and successful training the support vector machine could quickly and accurately identify seven typical flow regimes of gas-liquid two-phase flow in the horizontal tube. Image multi-feature fusion method provided a new way to identify the gas-liquid two-phase flow, and achieved higher identification ability than that of single characteristic. The overall identification accuracy was 100%, and an estimate of the image processing time was 8 ms for online flow regime identification.

IDENTIFICATION OF GAS LIQUID FLOW REGIMES IN A HORIZONTAL FLOW USING NEURAL NETWORK

The knowledge of flow regimes is very important in the study of a two phase flow system. A new flow regime identification method based on a Probability Density Function (PDF) and a neural network is proposed in this paper. The instantaneous differential pressure signals of a horizontal flow were acquired with a differential pressure sensor. The characters of differential pressure signals for different flow regimes are analyzed with the PDF. Then, four characteristic parameters of the PDF curves are defined, the peak number (K 1 ), the maximum peak value (K 2 ), the peak position (K 3 ) and the PDF variance (K 4 ). The characteristic vectors which consist of the four characteristic parameters as the input vectors train the neural network to classify the flow regimes. Experimental results show that this novel method for identifying air water two phase flow regimes has the advantages with a high accuracy and a fast response. The results clearly demonstrate that this new method could provide an accurate identification of flow regimes.

Investigation of gas-solids flow hydrodynamics in a cold model of a dual fluidised bed gasifier using electrical capacitance tomography sensors

The flow characteristics of a dual fluidised bed gasifier(DFBG)are more complex than those of a single fluidised bed gasifier.For stable operation and appropriate control,a cold DFBG test facility with both an upper and a lower U-valve was built,and electrical capacitance tomography(ECT)sensors were installed with pressure transducers to investigate the effects of operating conditions on gas-solids flow hydrodynamics.The operating parameters included gas velocities in the riser and in the bubbling fluidised bed,aeration velocity in the lower U-valve,bed material inventory,and particle size.This is the first time that ECT was applied in different flow zones of a dual fluidised bed gasifier system.The experimental results indicated that ECT in the recycle chamber could monitor the performance of the lower U-valve under different operating conditions for early detection of gas shortcut from the riser to the bubbling bed.Three main flow regimes in the riser and the differences between the reactors were identified by two sets of ECT sensors with pressure transducers.Finally,the effects of the operating conditions on the pressure drop in different parts of the DFBG was investigated.

Landmarks in the application of electrical tomography in particle science and technology

Selected milestones in the development and use of electrical tomography in powder conveying, slurry processing and multi-phase flow are highlighted. The ability to map concentration in opaque mixtures under process-realistic conditions was a major innovation for the method and has had far reaching implications. Subsequent developments have enabled velocity information to be abstracted resulting in the ability to measure component flux and motion.