Non-intrusive Technique Research Articles

WITHDRAWN: Non-intrusive load monitoring technique using deep neural networks for energy disaggregation

Households have increased energy usage in recent years to exceptionally high levels, where it is not feasible anymore. Owing to increased usage of power, there is a desperate need to find a way to use resources more sustainably. One of the key reasons why energy use is not sustainable is because the consumer does not know anything about the energy consumed by intelligent devices (dishwasher, refrigerator, washing machine etc). By letting household users know the energy consumption of intelligent devices. For energy monitoring firms, the energy used by the smart devices in a household needs to be analysed. In the presence of vampire loads such as electric heaters, we use deep neural networks to disaggregate the energy use of the dishwasher in a household. The training times of the deep neural networks are also being attempted. The object of the work is to calculate the efficiency and potentiality of deep neural networks on Non-intrusive load monitoring (NILM). In addition, the essential metrics chosen are RMSE and F1 in order to determine the efficiency of the proposed algorithms. Hence, from our research, we concluded that Gated recurrent Unit (GRU) is, in the presence of vapour loads of households, the best efficiency algorithm in our study, which breaks down the energy consumption of the dishwasher. GRU is however the strongest neural network when the RMSE and F1 scores are known as the metrics. However, if we take training time as the metric, we can take the best training time, that is, 19.34 min, as a simple recurrent neural network.

Stark-Broadening of Ar K-Shell Lines: A Comparison between Molecular Dynamics Simulations and MERL Results

Analysis of Stark-broadened spectral line profiles is a powerful, non-intrusive diagnostic technique to extract the electron density of high-energy-density plasmas. The increasing number of applications and availability of spectroscopic measurements have stimulated new research on line broadening theory calculations and computer simulations, and their comparison. Here, we discuss a comparative study of Stark-broadened line shapes calculated with computer simulations using non-interacting and interacting particles, and with the multi-electron radiator line shape MERL code. In particular, we focus on Ar K-shell X-ray line transitions in He- and H-like ions, i.e., Heα, Heβ and Heγ in He-like Ar and Lyα, Lyβ and Lyγ in H-like Ar. These lines have been extensively used for X-ray spectroscopy of Ar-doped implosion cores in indirect- and direct-drive inertial confinement fusion (ICF) experiments. The calculations were done for electron densities ranging from 1023 to 3×1024 cm−3 and a representative electron temperature of 1 keV. Comparisons of electron broadening only and complete line profiles including electron and ion broadening effects, as well as Doppler, are presented. Overall, MERL line shapes are narrower than those from independent and interacting particles computer simulations performed at the same conditions. Differences come from the distinctive treatments of electron broadening and are more pronounced in α line transitions. We also discuss the recombination broadening mechanism that naturally emerges from molecular dynamics simulations and its influence on the line shapes. Furthermore, we assess the impact of employing either molecular dynamics or MERL line profiles on the diagnosis of core conditions in implosion experiments performed on the OMEGA laser facility.

Testing the Goodness of Model Fit in Tunable Diode Laser Absorption Tomography

Tunable diode laser absorption tomography (TDLAT) has emerged as a popular nonintrusive technique for simultaneous sensing of gas concentration and temperature by making light absorbance measurements. Major challenge of TDLAT imaging is that the measurement data is very sparse. Therefore, precise models are required to describe the measurement process (forward model) and the behavior of the gas flow properites (prior model) to get accurate reconstructions. The sparsity of the measurement data makes TDLAT very sensitive to the accuracy of the models and makes it prone to overfitting. Both the forward and prior models can have systematic errors due to several reasons. So far, substantial amount of work has been done by researchers on developing reconstruction methods and formulating models, forward and prior. Yet, there has not been significant research work done on constructing a metric for goodness of the model fit that can indicate when there is an inaccuracy in the forward or the prior model. In this paper, we present a metric for goodness of model fit that can be used to indicate if the models used in the reconstruction are inaccurate. Results show that our metric can reliably quantify the goodness of model fit for sparese data reconstruction problems such as TDLAT.

Oxygen-Blown Gasification of Pulp Mill Bark Residues for Synthetic Fuel Production

Synthetic fuel production via gasification of residual biomass streams from the pulp and paper industry can be an opportunity for the mills to enable improved resource utilization and at the same time reduce the production of excess heat. This paper summarizes initial oxygen-blown gasification experiments with two bark residues from a European pulp and paper mill, i.e., a softwood bark and a hardwood bark. The gasification process was characterized by measuring syngas yields and process efficiency to find optimum operating conditions. In addition, impurities in the syngas and ash behavior were characterized. Maximum yields of CO and H2 were obtained from softwood bark and amounted to approximately 29 and 15 mol/kg fuel, respectively. Optimum cold gas efficiency was achieved at an oxygen stoichiometric ratio of λ = 0.40 and was approximately 76% and 70% for softwood bark and hardwood bark, respectively. Increased λ had a reducing effect on pollutants in the syngas, e.g., higher hydrocarbons, NH3, HCl, and soot. The situation for sulfur species was more complex. Evaluation of the bark ashes indicated that slag formation could start already from 800 °C. Furthermore, a non-intrusive laser diagnostics technique gave rapid feedback on the millisecond scale. Measured syngas temperature and water content were in good agreement with the applied reference methods.

Recent Advances in Coherent Anti-Stokes Raman Scattering of Carbon Dioxide for Combustion Diagnostic. (Article Review)

The rising climatic degradation due to the emission of greenhouse gases is leading to emergence of clean combustion technology, oxy-fuel combustion to minimize the emissions of carbon dioxide into the atmosphere in combustion. Nitrogen molecules are used as probe molecule in laser-based combustion diagnostic in nitrogen rich air combustion. However, with the introduction of oxy-fuel combustion, carbon dioxide becomes the dominant molecule and has to be considered as probe molecule in combustion diagnostic. A detailed knowledge about thermodynamic properties: temperature, pressure and species concentration are necessary in optimization of combustion and minimizing the emission into the atmosphere. The non-intrusive spectroscopic measurement techniques are the most accurate methods to determine the combustion properties. The purpose of this review is to provide a brief overview of the recent advances made in application of coherent anti-Stokes Raman scattering of carbon dioxide for development of models for thermometry. However, there is no sufficient empirical data of time-domain S-branch Raman linewidth dependence on temperature that has been determined for pure-rotational coherent anti-Stokes Raman scattering of carbon dioxide and its mixtures for development of models for thermometry.

Recent Advances in Coherent Anti-Stokes Raman Scattering of Carbon Dioxide for Combustion Diagnostic. (Article Review)

The rising climatic degradation due to the emission of greenhouse gases is leading to emergence of clean combustion technology, oxy-fuel combustion to minimize the emissions of carbon dioxide into the atmosphere in combustion. Nitrogen molecules are used as probe molecule in laser-based combustion diagnostic in nitrogen rich air combustion. However, with the introduction of oxy-fuel combustion, carbon dioxide becomes the dominant molecule and has to be considered as probe molecule in combustion diagnostic. A detailed knowledge about thermodynamic properties: temperature, pressure and species concentration are necessary in optimization of combustion and minimizing the emission into the atmosphere. The non-intrusive spectroscopic measurement techniques are the most accurate methods to determine the combustion properties. The purpose of this review is to provide a brief overview of the recent advances made in application of coherent anti-Stokes Raman scattering of carbon dioxide for development of models for thermometry. However, there is no sufficient empirical data of time-domain S-branch Raman linewidth dependence on temperature that has been determined for pure-rotational coherent anti-Stokes Raman scattering of carbon dioxide and its mixtures for development of models for thermometry.

Optimization of Transducer Location for Novel Non-Intrusive Methodologies of Diagnosis in Diesel Engines

The health monitoring has been studied to ensure integrity of design of engine structure by detection, quantification, and prediction of damages. Early detection of faults may allow the downtime of maintenance to be rescheduled, thus preventing sudden shutdown of machines. In cylinder pressure developed, vibrations and noise emissions data provide a rich source of information about condition of engines. Monitoring of vibrations and noise emissions are novel non-intrusive methodologies for which positioning of various transducers are important issue. The presented work shows applicability of these diagnosis methodologies adopted in case of diesel engines. The effects of changing various fuel injection parameters was analyzed. Scope of using non-intrusive technique has been analyzed by changing locations of microphone. Novelty of this worklies in exploring signal processing methods for various locations around the engine test set up. Various frequency ranges of contributing noise and vibration sources were identified. Time-Frequency analysis showed the onset of various cyclic. Based on the identification of various frequency bands, it is possible to device suitable filters in order to extract more information.

LASER-BASED DIAGNOSTICS OF SLUG FLOW BOILING IN A HORIZONTAL PIPE

We present results from an experimental investigation on flow boiling, in the slug flow regime, of refrigerant R245fa through a 12.6-mm inner diameter horizontal plain pipe using particle image velocimetry (PIV) and an interface detection method. The study is supplemented by an overview of the state-of-the-art in experimental research of two-phase dispersed pipe flows and the development of modern optical and laser-based full-field nonintrusive measurement techniques as applied to these flows. We consider different flow conditions, with heat fluxes over the range 5.3 to 7.9 kW/m2 and mass fluxes from 300 to 460 kg/m2 s. Significant disturbances in the instantaneous velocity fields are revealed in both the noses and tails of slugs, with their values being two times higher behind vapor bubbles. The slug passage frequency is determined based on the results of the interface detection method. The vapor bubble velocity is found to increase linearly with the interfacial velocity of the two-phase mixture, while its gradient grows with the heat flux. Moreover, at increased heat fluxes the bubbles may move even faster than the mixture itself, which implies that they must significantly enhance local turbulence, thereby additionally intensifying heat transfer. This research demonstrates the applicability of laser-based diagnostic techniques to boiling slug flows, to which such experimental methods have not been widely applied in macroscale geometries. In addition to the conclusions, we provide practical recommendations for possible future research in this particular field of fluid mechanics and the further development of sophisticated laser-based measurement techniques for boiling, and similar, flows.

Review of Filtered Rayleigh Scattering technique for mixing studies in supersonic air flow

Filtered Rayleigh Scattering or FRS is a non-intrusive laser diagnostic technique that is utilized to observe and quantify various flow properties in high-speed gas flows. In this technique, a laser beam is used to irradiate a portion of the gas under investigation. The receiving gas molecules scatter the incident laser beam in all the directions. The scattered radiation from a single molecule stems from the induced oscillations of the electron cloud that acts as a classical dipole radiator. The intensity of the net scattered radiation from all the molecules that are in random motion depend on the spatio-temporal fluctuations of the dielectric permittivity of the gas. The thermodynamic properties of the gas govern the nature of these variations and the resulting spectra of the scattered radiation is referred to as Rayleigh Brillouin spectrum. In high-speed gas flows applications, the bulk velocity of the gas molecules causes the entire scattered spectrum to be Doppler shifted from the lasing frequency. The extent of this shift is very significant in supersonic flows and this enables the isolation of the scattered spectrum from the laser reflections using a molecular notch filter. Measurement of flow properties can be made based on the intensity of the RBS signal that is collected by a charged coupled device. The FRS technique is also utilized to perform mixing measurements in binary mixture of gases in high-speed flows. In non-reacting mixing studies, helium is often used as a passive scalar and simulant gas for hydrogen fuel. Helium concentration measurements in supersonic flows based on the FRS technique require two independent experiments. The first experiment involves injection of helium gas and the second separate experiment requires air injection. The helium mole-fraction is retrieved from the FRS imaging data under the key assumptions that the total number density profiles and the extent of the Doppler shift that is associated with each one of the two independent experiments at the measuring plane are identically the same. In this review, the theory governing Rayleigh scattering and the FRS technique to retrieve helium mole-fraction in supersonic flow will be discussed in detail. Measurement results from recent FRS studies on complex vortex interactions in supersonic flow will be also presented and discussed as well as the impact on the results of a departure from the key assumptions in FRS mixing studies will be analyzed for a canonical parallel injection using strut injector in supersonic flow.

Information Fusion of Infrared Images and Vibration Signals for Coupling Fault Diagnosis of Rotating Machinery

Rotating machinery has a complicated structure and interaction of multiple components, which usually results in coupling faults with complex dynamic characteristics. Fault diagnosis methods based on vibration signals have been widely used, however, these methods are intricate when identifying coupling faults, especially in the situation where coupling faults share similar patterns. As a noncontact and nonintrusive temperature‐measuring technique, methods by infrared images can recognize multiple faults through temperature variations; however, it is not effective if the faults are temperature‐insensitive. In this paper, an improved machinery fault diagnosis technique based on information fusion of infrared images and vibration signals is studied, to have better utilization of multisource sensors and to solve the problems when one single type of data is separately used. Firstly, data enhancement for infrared images and data visualization for vibration are performed on the dataset by using the principle of graphics and Short‐Term Fourier Transform, which increases the diversity of the dataset and enhances the generalization ability of the model. Then, a multichannel convolution neural network‐based method is constructed to achieve data‐level information fusion and improve the fault diagnosis accuracy. The effectiveness of the presented method is validated by the experimental studies on a rotor test stand, the results illustrate that the coupling faults can be effectively identified by the information fusion method, and the fault diagnosis accuracy is higher in comparison with the method by a signal from single‐source sensors.

A review on the applied techniques of exhaled airflow and droplets characterization.

In the last two decades, multidisciplinary research teams worked on developing a comprehensive understanding of the transmission mechanisms of airborne diseases. This article reviews the experimental studies on the characterization of the exhaled airflow and the droplets, comparing the measured parameters, the advantages, and the limitations of each technique. To characterize the airflow field, the global flow‐field techniques—high‐speed photography, schlieren photography, and PIV—are applied to visualize the shape and propagation of the exhaled airflow and its interaction with the ambient air, while the pointwise measurements provide quantitative measurements of the velocity, flow rate, humidity and temperature at a single point in the flow field. For the exhaled droplets, intrusive techniques are used to characterize the size distribution and concentration of the droplets' dry residues while non‐intrusive techniques can measure the droplet size and velocity at different locations in the flow field. The evolution of droplets' size and velocity away from the source has not yet been thoroughly experimentally investigated. Besides, there is a lack of information about the temperature and humidity fields composed by the interaction of the exhaled airflow and the ambient air.

Toward Water Volume Fraction Calculation in Multiphase Flows Using Electrical Capacitance Tomography Sensors

Real-time monitoring of water volume fraction in multiphase flows is an important problem for a number of industrial applications. The water phase in the multiphase flows may correspond to either the dispersed phase or the continuous phase. In the past, several low-cost and nonintrusive techniques based on the electrical capacitance tomography (ECT) has been developed to image and monitor in real-time multiphase flows containing water. However, such monitoring becomes increasingly challenging for high salinity levels, and no reliable ECT-based method is presently available which could work for obtaining water volume fraction in multiphase flows for all water salinity levels. In this paper, we propose a new approach based on the Hanai’s formula for complex dielectric constant and taking advantage of the Maxwell-Wagner-Sillars effect to obtain, to a good approximation, water volume fractions in multiphase flows containing water as either dispersed or continuous phase.

Three-dimensional particle size determination in a laminar diffusion flame by tomographic laser-induced incandescence

Non-intrusive measurement techniques are required to gain a comprehensive understanding about the processes of soot formation, growth and oxidation. Time-resolved laser-induced incandescence (TiRe-LII), commonly performed 0D or 2D within a flame, has proven to be a very suitable tool for the in situ sizing of soot primary particles. In this work, the technique is expanded to the third dimension by employing volumetric illumination and coupling it with a tomographic approach, which allows to computationally gain 3D information from 2D images taken at various angles. To minimize experimental cost, an approach using nine fiber bundles arranged in a semicircle around the flame and imaging the light onto a single camera is used. The technique is demonstrated on an ethene diffusion flame on a standard burner, providing spatially resolved 3D particle sizes. One focus of this work is to reveal the influence of input parameters such as the local bath gas temperature, which we measured by two-color pyrometry, and local laser fluence, which are both required for an accurate evaluation of the local particle size. It is shown that the assumption of an average temperature may result in a wrong picture even of qualitative soot size evaluation. In the end, a concept is proposed for a simultaneous determination of the 3D distribution of particle sizes through TiRe-LII and the required bath gas temperature via two-color pyrometry using a tomographic approach with only three cameras.

2D experiments and numerical simulation of the oscillatory shallow flow in an open channel lateral cavity

Steady shallow flows past an open channel lateral cavity can induce the excitation of an eigenmode of a gravity standing wave inside the cavity, called seiche, which may be coupled with the shedding of vortices at the opening of the cavity. The presence of the seiche is of fundamental interest as it enhances the mass exchange between the main channel and the cavity. Measurements of the time evolution of the water surface are not often found in the literature for this type of flows. In this work, an experimental and numerical study of a shallow flow past a channel lateral cavity is carried out. The main novelty is the use of a pioneering non-intrusive experimental technique to measure the water surface at the channel-cavity region. This optical technique offers high resolution 2D data in time and space of the water surface evolution, allowing to determine the relevant features of the seiche oscillation. Such data are supplemented with Particle Image Velocimetry measurements. Furthermore, the experiments are numerically reproduced using a high-resolution depth-averaged URANS shallow water model, under the assumption that shallow water turbulence is mainly horizontal. The experimental and numerical results are analyzed in the frequency domain. High-resolution two-dimensional amplitude oscillation maps of the seiche phenomenon, as well as velocity fields, are presented. The high quality of the experimental data reported in this work makes this data set a suitable benchmark for numerical simulation models in order to evaluate their performance in the resolution of turbulent resonant shallow flows.

Sensitivity Guided Image Fusion for Electrical Capacitance Tomography

As a fast and nonintrusive measurement and visualization technique, electrical capacitance tomography (ECT) is rapidly expanding its applications in the research on multiphase flow, fluidization, drying, combustion, and so on. However, the marked unevenness of the sensitivity maps sometimes causes unexpected effects in imaging reconstruction, particularly in 3-D cases. To exploit the positive potential of this phenomenon, the authors proposed an image fusion method using the data from two units of ECT sensors in this study. This method is used in image fusion on the reconstructed images for a planar sensor and a cylindrical sensor. In contrast to the conventional fusion models that use fixed weight factors for two sources of data, our model forges weight functions that are set preference the strength of the sensitivity maps. The new algorithm is implemented by first extracting the characteristic information out of the ECT images using the latent low-rank representation and then performing a fusion algorithm with linear weight functions in preference to the significance of the sensitivity maps. The simulation results show that the algorithm effectively retains the advantages of the two units of sensors and mutually compensates the weak points of theirs, and significantly improves the reconstruction quality. The fusion image quality by the new method can response the real situation better in different heights. The results imply that this data fusion method can amend the weakness of ECT cause by the uneven sensitivity maps to a significant extend.

Interferometric analysis of flow around a horizontal tube falling film evaporator for MED systems

In this work, an optical shadow method (non-intrusive) was used to evaluate the falling film thickness around the circumference of a horizontal tube. A Mach-Zehnder Interferometer was used to visualize the isotherm formation. Most of the well established non-intrusive techniques for measuring film thickness around the cylinder fails to measure the film thickness near to entry region (θ<30∘) and towards the exit region (θ>140∘) because of the limitation and complexity of the apparatus. Relavance of this work lies on the simplicity of the measurement and the range of circumferential angles that can be investigated (10∘≤θ≤170∘). This technique involves image visualization, thus, serves as a tool for analyzing the film both quantitatively and qualitatively. A lucid algorithm was developed for image processing along with a set of intensity profile tracking procedures making the whole process of film thickness measurement more obvious. The measured film thickness is showing good agreement with the commonly used empirical formula, and the effectiveness of using those empirical formulas for the small diameter tubes was analyzed. Further, this study was extended to determine the effects of flow rate and feed inlet temperature on film thickness. Film thickness variation around the circumference of the cylinder is showing an increase in trend with increasing Reynolds number, but a change in trend with the varying feed inlet temperature is found to be marginal. This study is further extended for different positions of the tube in the order of impingement of film or feed inlet as top, middle and bottom tubes and the film thickness variation was again found to be marginal. Dynamic characterization at θ=10∘ was studied and the maximum film thickness variation was found to be 11.8%. A standard error of mean analysis was performed on every data set which validates the reproducibility and the dynamic measurement capability of the technique used.

Frequency characteristics of in cylinder pressure of a gasoline engine

This presented work analysis of effects of changes in various injection parameters on development of combustion pressure development and noise emissions from engines. Scope of using non-intrusive diagnosis technique has been analyzed by using changing various locations of microphone around the engine. In particular the noise emissions from engines were found to be dependent on quantity of fuel injected inside cylinder. Various characteristic frequency ranges of contributing sources have been identified. Time-Frequency analysis has shown onset of various events associated with working of engine. Based on the identification of various frequency bands, it is possible to device suitable filters in order to extract more information about combustion and motion based noise which is done in later part of this work. It may be used as an effective tool for condition monitoring of engines in order to gain fuel economy. Suitable structural modifications of engine may be also done in order to improve its performance on various noise and vibration benchmarks.

CREAM, a component level coffeemaker electrical activity measurement dataset

Monitoring the internal conditions of a machine is essential to increase its production efficiency and to reduce energy waste. Non-intrusive condition monitoring techniques, such as analysing electrical signals, provide insights by disaggregating a composite signal of a machine as a whole into the individual components to determine their states. Developing and evaluating new algorithms for condition monitoring and maintenance-related analysis tasks require a fully-labelled dataset for a machine, which comprises standard industrial components that are triggered following a typical manufacturing process to produce goods. For this purpose, we introduce CREAM, a component level electrical measurement dataset for two industrial-grade coffeemakers, simulating industrial processes. The dataset contains continuous voltage and current measurements provided at 6400 samples per second, as well as the product and maintenance-related event labels, such as 370600 expert-labelled component-level electrical events, 1734 product ones and 3646 maintenance ones. CREAM provides fully-labelled ground-truth to establish a benchmark and comparative studies of manufacturing-related analysis in a controlled and transparent environment.

The Application of the Inbound Marketing Strategy on Costa del Sol Planning &amp; Tourism Board. Lessons for Post-COVID-19 Revival

The digital era has radically changed the context in which the tourist service is delivered and experienced, changing the decision processes of consumer and company business models. It is necessary to contact the tourism services customer with non-intrusive techniques, at the beginning of their purchase process and accompany them until the final transaction. The main objective of this case study is to analyze the methodology of inbound marketing and show how The Costa del Sol Planning &amp; Tourism Board could work on its sustainable customer relationship model under the concept of seducing the tourist by being pioneers in the application of this strategy to attract tourism after the pandemic caused by COVID-19. Conclusions of this study include measures to restore travelers’ confidence which will play an important role in attracting tourists after crisis. An inbound marketing strategy will provide a response as it is based on contact with the future tourist through highly specialized content.

Soot maturity studies in methane-air diffusion flames at elevated pressures using laser-induced incandescence

Two-dimensional laser-induced incandescence (LII) measurements were used for quantitative soot volume fraction (fv) measurements in methane-air diffusion flames at pressures ranging from 0.1 to 0.5 MPa. Additionally, laser-induced fluorescence (LIF) was used for visualization of polycyclic aromatic hydrocarbons (PAHs) considered as important soot precursors. A heat and mass transfer-based LII model was used for the analysis of experimental LII fluence curves to evaluate soot absorption functions, E(m), at various spatial positions in the flames. Based on previous work, variations in the evaluated E(m) was assumed to be related to soot maturity. Generally, the most mature soot was found at the spatial position of maximum fv along the central axis and at the flame edges. Also, the soot maturity at the position of maximum fv at each pressure was found to increase from the flame at 0.1 MPa to 0.5 MPa. A critical examination of the assumptions and uncertainties in the E(m) analysis was made. The present study emphasizes the limitation of assigning soot a constant E(m) in the overall flame for temperature and soot concentration evaluation using the optical methods spectral soot emission (SSE) and line-of-sight attenuation (LOSA), which are commonly used non-intrusive optical diagnostic techniques in sooting high-pressure flames. The study also demonstrates the critical choice of fluence in quantitative imaging LII measurements of fv when the E(m) of soot spans over large range of values.