Machine learning for parameters diagnosis of spark discharge by electro-acoustic signal

Discharge plasma parameter measurement is a key focus in low-temperature plasma research.Traditional diagnostics often require costly equipment,whereas electro-acoustic signals provide a rich,non-invasive,and less complex source of discharge information.This study harnesses machine learning to decode these signals.It establishes links between electro-acoustic signals and gas discharge parameters,such as power and distance,thus streamlining the prediction process.By building a spark discharge platform to collect electro-acoustic signals and implementing a series of acoustic signal processing techniques,the Mel-Frequency Cepstral Coefficients(MFCCs)of the acoustic signals are extracted to construct the predictors.Three machine learning models(Linear Regression,k-Nearest Neighbors,and Random Forest)are introduced and applied to the predictors to achieve real-time rapid diagnostic measurement of typical spark discharge power and discharge distance.All models display impressive performance in prediction precision and fitting abilities.Among them,the k-Nearest Neighbors model shows the best performance on discharge power prediction with the lowest mean square error(MSE=0.00571)and the highest R-squared value(R^(2)=0.93877).The experimental results show that the relationship between the electro-acoustic signal and the gas discharge power and distance can be effectively constructed based on the machine learning algorithm,which provides a new idea and basis for the online monitoring and real-time diagnosis of plasma parameters.

The development of a new real-time subsurface mooring

Subsurface mooring allows researchers to measure the ocean properties such as water temperature,salinity,and velocity at several depths of the water column for a long period.Traditional subsurface mooring can release data only after recovered,which constrains the usage of the subsurface and deep layer data in the ocean and climate predictions.Recently,we developed a new real-time subsurface mooring(RTSM).Velocity profiles over upper 1000 m depth and layered data from sensors up to 5000 m depth can be realtime transmitted to the small surface buoy through underwater acoustic communication and then to the office through Beidou or Iridium satellite.To verify and refine their design and data transmission process,we deployed more than 30 sets of RTSMs in the western Pacific to do a 1-year continuous run during 2016–2018.The continuous running period of RTSM in a 1-year cycle can reach more than 260 days on average,and more than 95%of observed data can be successfully transmitted back to the office.Compared to the widely-used inductive coupling communication,wireless acoustic communication has been shown more applicable to the underwater sensor network with large depth intervals and long transmission distance to the surface.

Portable FBAR based E-nose for cold chain real-time bananas shelf time detection

Being cheap,nondestructive,and easy to use,gas sensors play important roles in the food industry.However,most gas sensors are suitable more for laboratory-quality fast testing rather than for cold-chain continuous and cumulative testing.Also,an ideal electronic nose(E-nose)in a cold chain should be stable to its surroundings and remain highly accurate and portable.In this work,a portable film bulk acoustic resonator(FBAR)-based E-nose was built for real-time measurement of banana shelf time.The sensor chamber to contain the portable circuit of the E-nose is as small as a smartphone,and by introducing an air-tight FBAR as a reference,the E-nose can avoid most of the drift caused by surroundings.With the help of porous layer by layer(LBL)coating of the FBAR,the sensitivity of the E-nose is 5 ppm to ethylene and 0.5 ppm to isoamyl acetate and isoamyl butyrate,while the detection range is large enough to cover a relative humidity of 0.8.In this regard,the E-nose can easily discriminate between yellow bananas with green necks and entirely yellow bananas while allowing the bananas to maintain their biological activities in their normal storage state,thereby showing the possibility of real-time shelf time detection.This portable FBAR-based E-nose has a large testing scale,high sensitivity,good humidity tolerance,and low frequency drift to its surroundings,thereby meeting the needs of cold-chain usage.

Real-time damage analysis of 2D C/SiC composite based on spectral characters of acoustic emission signals using pattern recognition

In this study,unsupervised and supervised pattern recognition were implemented in combination to achieve real-time health monitoring.Unsupervised recognition(k-means++)was used to label the spectral characteristics of acoustic emission(AE)signals after completing the tensile tests at ambient temperature.Using in-plane tensile at 800 and 1000°C as implementing examples,supervised recognition(K-nearest neighbor(KNN))was used to identify damage mode in real time.According to the damage identification results,four main tensile damage modes of 2D C/SiC composites were identified:matrix cracking(122.6–201 kHz),interfacial debonding(201–294.4 kHz),interfacial sliding(20.6–122.6 kHz)and fiber breaking(294.4–1000 kHz).Additionally,the damage evolution mechanisms for the 2D C/SiC composites were analyzed based on the characteristics of AE energy accumulation curve during the in-plane tensile loading at ambient and elevated temperature with oxidation.Meanwhile,the energy of various damage modes was accurately calculated by harmonic wavelet packet and the damage degree of modes could be analyzed.The identification results show that compared with previous studies,using the AE analysis method,the method has higher sensitivity and accuracy.

A Dedicated MAC Protocol Package for Real-Time Ocean Current Field Estimation Using Distributed Underwater Acoustic Sensor Networks

Distributed underwater acoustic sensor networks(UASNs)are envisioned in real-time ocean current velocity estimation.However,UASNs at present are still dominated by post-processing partially due to the complexity of on-line detection for travel times and lack of dedicated medium access control(MAC)protocols.In this study,we propose a dedicated MAC protocol package for real-time ocean current velocity estimation using distributed UASNs.First,we introduce the process and requirements of ocean current velocity estimation.Then,we present a series of spatial reuse time division multiple access(TDMA)protocols for each phase of real-time ocean current field estimation using distributed UASNs,followed by numerical analysis.We divide UASNs into two categories according to their computing ability:feature-complete and feature-incomplete systems.The feature-complete systems that have abundant computing ability carry out the presented MAC protocol package in three phases,whereas the feature-incomplete ones do not have enough computing ability and the presented MAC protocol package is reduced to two phases plus an additional downloading phase.Numerical analysis shows that feature-complete systems using mini-slot TDMA have the best real-time performance,in comparison with feature-incomplete systems and other feature-complete counterparts.Feature-incomplete systems are more energy-saving than feature-complete ones,owing to the absence of in-network data exchange.

Non-invasive diagnosis of hepatitis B virus-related cirrhosis

Chronic hepatitis B(CHB)infection is a major public health problem associated with significant morbidity and mortality worldwide.Twenty-three percent of patients with CHB progress naturally to liver cirrhosis,which was earlier thought to be irreversible.However,it is now known that cirrhosis can in fact be reversed by treatment with oral anti-nucleotide drugs.Thus,early and accurate diagnosis of cirrhosis is important to allow an appropriate treatment strategy to be chosen and to predict the prognosis of patients with CHB.Liver biopsy is the reference standard for assessment of liver fibrosis.However,the method is invasive,and is associated with pain and complications that can be fatal.In addition,intra-and inter-observer variability compromises the accuracy of liver biopsy data.Only small tissue samples are obtained and fibrosis is heterogeneous in such samples.This confounds the two types of observer variability mentioned above.Such limitations have encouraged development of non-invasive methods for assessment of fibrosis.These include measurements of serum biomarkers of fibrosis;and assessment of liver stiffness via transient elastography,acoustic radiation force impulse imaging,real-time elastography,or magnetic resonance elastography.Although significant advances have been made,most work to date has addressed the diagnostic utility of these techniques in the context of cirrhosis caused by chronic hepatitis C infection.In the present review,we examine the advantages afforded by use of non-invasive methods to diagnose cirrhosis in patients with CHB infections and the utility of such methods in clinical practice.

Industrial applications of advanced control techniques

This paper discusses two industrial control applications using advanced control techniques. They are theoptimal-tuning nonlinear PID control of hydraulic systems and the neural predictive control of combustor acoustic ofgas turbines. For hydraulic control systems, an optimal PID controller with inverse of dead zone is introduced toovercome the dead zone and is designed to satisfy desired time-domain performance requirements. Using the adaptivemodel, an optimal-tuning PID control scheme is proposed to provide optimal PID parameters even in the case wherethe system dynamics is time variant. For combustor acoustic control of gas turbines, a neural predictive controlstrategy is presented, which consists of three parts: an output model, output predictor and feedback controller. Theoutput model of the combustor acoustic is established using neural networks to predict the output and overcome thetime delay of the system, which is often very large, compared with the sampling period. The output-feedback con-troller is introduced which uses the output of the predictor to suppress instability in the combustion process. The a-bove control strategies are implemented in the SIMULINK/dSPACE controller development environment. Theirperformance is evaluated on the industrial hydraulic test rig and the industrial combustor test rig.