Artificial Intelligence in Internet of Things System for Predicting Water Quality in Aquaculture Fishponds

Aquaculture has long been a critical economic sector in Taiwan.Since a key factor in aquaculture production efficiency is water quality,an effective means of monitoring the dissolved oxygen content(DOC)of aquaculture water is essential.This study developed an internet of things system for monitoring DOC by collecting essential data related to water quality.Artificial intelligence technology was used to construct a water quality prediction model for use in a complete system for managing water quality.Since aquaculture water quality depends on a continuous interaction among multiple factors,and the current state is correlated with the previous state,a model with time series is required.Therefore,this study used recurrent neural networks(RNNs)with sequential characteristics.Commonly used RNNs such as long short-term memory model and gated recurrent unit(GRU)model have a memory function that appropriately retains previous results for use in processing current results.To construct a suitable RNN model,this study used Taguchi method to optimize hyperparameters(including hidden layer neuron count,iteration count,batch size,learning rate,and dropout ratio).Additionally,optimization performance was also compared between 5-layer and 7-layer network architectures.The experimental results revealed that the 7-layer GRU was more suitable for the application considered in this study.The values obtained in tests of prediction performance were mean absolute percentage error of 3.7134%,root mean square error of 0.0638,and R-value of 0.9984.Therefore,thewater qualitymanagement system developed in this study can quickly provide practitioners with highly accurate data,which is essential for a timely response to water quality issues.This study was performed in collaboration with the Taiwan Industrial Technology Research Institute and a local fishery company.Practical application of the system by the fishery company confirmed that the monitoring system is effective in improving the survival rate of farmed fish by providing data needed to maintain DOC higher than the standard value.

Optical Based Gradient-Weighted Class Activation Mapping and Transfer Learning Integrated Pneumonia Prediction Model

Pneumonia is a common lung disease that is more prone to affect the elderly and those with weaker respiratory systems.However,hospital medical resources are limited,and sometimes the workload of physicians is too high,which can affect their judgment.Therefore,a good medical assistance system is of great significance for improving the quality of medical care.This study proposed an integrated system by combining transfer learning and gradient-weighted class activation mapping(Grad-CAM).Pneumonia is a common lung disease that is generally diagnosed using X-rays.However,in areaswith limited medical resources,a shortage of medical personnel may result in delayed diagnosis and treatment during the critical period.Additionally,overworked physicians may make diagnostic errors.Therefore,having an X-ray pneumonia diagnosis assistance system is a significant tool for improving the quality of medical care.The result indicates that the best results were obtained by a ResNet50 pretrained model combined with a fully connected classification layer.A retraining procedure was designed to improve accuracy by using gradient-weighted class activation mapping(Grad-CAM),which detects the misclassified images and adds weights to them.In the evaluation tests,the final combined model is named Grad-CAM Based Pneumonia Network(GCPNet)out performed its counterparts in terms of accuracy,precision,and F1 score and reached 97.2%accuracy.An integrated system is proposed to increase model performance where Grad-CAM and transfer learning are combined.Grad-CAM is used to generate the heatmap,which shows the region that the model is focusing on.The outcomes of this research can aid in diagnosing pneumonia symptoms,as themodel can accurately classify chest X-ray images,and the heatmap can assist doctors in observing the crucial areas.

Hybrid Multi-Object Optimization Method for Tapping Center Machines

This paper proposes a hybrid multi-object optimization method integrating a uniform design,an adaptive network-based fuzzy inference system(ANFIS),and a multi-objective particle swarm optimizer(MOPSO)to optimize the rigid tapping parameters and minimize the synchronization errors and cycle times of computer numerical control(CNC)machines.First,rigid tapping parameters and uniform(including 41-level and 19-level)layouts were adopted to collect representative data for modeling.Next,ANFIS was used to build the model for the collected 41-level and 19-level uniform layout experiment data.In tapping center machines,the synchronization errors and cycle times are important consid-erations,so these two objects were used to build the ANFIS models.Then,a MOPSO algorithm was used to search for the optimal parameter combinations for the two ANFIS models simultaneously.The experimental results showed that the proposed method obtains suitable parameter values and optimal parameter combinations compared with the nonsystematic method.Additionally,the optimal parameter combination was used to optimize existing CNC tools during the commissioning process.Adjusting the proportional and integral gains of the spindle could improve resistance to deformation during rigid tapping.The posi-tion gain and prefeedback coefficient can reduce the synchronization errors significantly,and the acceleration and deceleration times of the spindle affect both the machining time and synchronization errors.The proposed method can quickly and accurately minimize synchronization errors from 107 to 19.5 pulses as well as the processing time from 3,600 to 3,248 ms;it can also shorten the machining time significantly and reduce simultaneous errors to improve tapping yield,there-by helping factories achieve carbon reduction.

Design and Development of Oblique-Incident Interferometer for Form Measurement of Hand-Scraped Surfaces

Precision flat surfaces finished by hand-scraping are extensively used in the sliding parts of precision positioning systems or the surface plates of precision measuring instruments.These hand-scraped surfaces are employed to improve the flatness of the machined surfaces and the motion accuracy of the instruments.The numerous depressions of micrometer-scale depth on a scraped surface are thought to prevent wringing and improve the lubrication properties by forming oil pockets.Form measurement of the micrometer-scale features of a hand-scraped surface is one of the reasonable methods for investigating its effect on the motion accuracy of the positioning systems or the lubrication properties of the sliding parts.Because a hand-scraped surface is a rough surface composed of micrometric complex features,it is difficult to conduct precision measurement of the three-dimensional surface form using conventional measurement instruments,such as vertical-incident interferometers or stylus-type profilometers.In this study,an oblique-incident interferometer based on Abramson interferometry is introduced to measure the form of a rough surface produced by hand-scraping.Because this technique can detect high-intensity reflected light on a rough surface,it allows the measurement of the form of the hand-scraped surface in noncontact condition.In order to shorten the measurement time,the form calculation based on a five-step phase shifting method was introduced.

Spurious-mode vibrations caused by alternating current poling and their solution process for Pb(Mg_(1/3)Nb_(2/3))O_(3)-PbTiO_(3) single crystals

The effects of alternating current poling(ACP)at 80℃on electrical properties of[001]-oriented 0.72 Pb(Mg_(1/3)Nb_(2/3))O_(3)-0.28PbTiO_(3)(PMN-28PT)single crystals(SCs)have been investigated.The square-wave ACP SCs poled at high voltage(HV,5 kV_(rms)/cm)occasionally showed large fluctuations and low opposite values of piezoelectric coefficient(d_(33)=±1370 pC/N)in one plate.This revealed spuriousmode vibrations(SMV)of impedance spectrum.However,after depolarizing and repolarizing the sample with a sine-wave ACP at low voltage(LV,3.5 kV_(rms)/cm),the d_(33) enhanced to be 1720 pC/N(+26%)and did not exhibit large fluctuation or opposite values in one plate any more.The impedance spectrum became clean and the abnormal SMV disappeared.We proposed four possible mechanisms of the SMV,and speculate that the main cause maybe by macro-scale sub-domain structure and/or phase change in the main domain structure and/or phase in the SC plate due to the specific poling conditions not eternal mechanical damage of PMN-PT SCs.This study will be useful to realize a high d_(33) and improve other properties of PMN-PT ACP SC ultrasonic transducers without any SMV for high-frequency medical imaging equipment.