Wavelet Decomposition Algorithm Research Articles

A fast and accurate hybrid method for short-term forecasting significant wave height

This paper proposes a hybrid method for forecasting significant wave height (SWH). The wavelet decomposition algorithm is applied to decompose the original signal into different components, and suitable models are selected to predict wave heights for each component. To validate the performance of the proposed hybrid model in forecasting SWH with varying characteristics, wave data measured from three stations are utilized for both model training and accuracy validation. For each station, a total of 2000 h of continuously observed wave data is selected. The initial 80% of the data is used as training samples, while the remaining 20% is utilized as test samples. Additionally, a data normalization procedure is conducted. In addition to the hybrid model, a set of benchmark models is employed to evaluate the performance of the hybrid model using a series of statistical scoring metrics. The hybrid model, along with benchmark models, is utilized to perform 1-h and multiple-hour lead SWH predictions. The results validate that the proposed hybrid model provides accurate and efficient short-term SWH predictions. The correlation coefficients between the one- and three-step predictions and their corresponding observations are greater than 0.99, while they are greater than 0.96 for the five and eight-step predictions.

Construction and simulation of a joint scale model for power electronic converters based on wavelet decomposition and reconstruction algorithms.

In power electronics systems, system design and operation often involve multiple time and space scales, ranging from nanosecond switching dynamics to hour-level system operation behavior. Due to the complexity of these systems and the rise of wide-gap semiconductor technology, a series of multi-scale phenomena have emerged that are difficult to ignore. The high frequency of switching operations makes multi-scale effects particularly significant, including the fast dynamic response of the power loop, EMI, and heat conduction problems. They are key factors that must be considered in the design to ensure the efficient and reliable operation of power electronic devices. This study proposes the construction and simulation of a joint scale model for power electronic converters based on wavelet decomposition and reconstruction algorithms to address the multi-scale phenomenon and limitations of single-scale power electronic converters. Firstly, a joint scale model for power electronic converters at both macro and micro-scales was established, targeting both single-scale models and simple combinations of multiple scale models for power electronic converters. The traditional single-scale model is sufficient to describe the average behavior of the converter, but it has serious limitations in capturing fast transient processes and high-frequency switching behavior in power electronic systems. These limitations often manifest themselves when there is a need to capture fine timescales of detail. By transforming between the time domain and the frequency domain, wavelet decomposition enables the model to capture both macroscopic average characteristics and microscopic transient dynamics. The wavelet reconstruction algorithm can simulate all kinds of fast changes in the actual working process more accurately and compress irrelevant information while retaining key signal features, so as to optimize the simulation performance of the model. Secondly, this algorithm is used to analyze BC in short time scale. Finally, the short time scale characteristics of power electronic converters are analyzed. Experimental results show that the fusion of wavelet decomposition and reconstruction algorithm enhances the accuracy of the power electronic converter model and improves the performance of the system. The model achieves an error reduction of nearly 3% in the calculation step size of 10-7s, which has a significant impact on the high precision requirements of high-frequency operations. In addition, the optimal calculation step size of 8×10-8s achieves an error reduction of more than 14%, making an important contribution to the transient analysis and fine structure simulation. The wavelet algorithm can improve the accuracy of multi-scale modeling in power electronic system and reduce the simulation time. The reduction of error not only shows the improvement of the accuracy of the model, but also shows its practical significance in the design and test of the actual power electronic system. The reduction in error reveals the ability to more accurately predict and mitigate potential performance problems in matching tests with actual hardware, as well as its ability to adapt to emerging wide bandgap semiconductor materials and structures.

Construction of short-term electricity demand streaming forecasting model in demand-side dynamic response

In the context of demand-side dynamic response, the electricity supply-demand relationship undergoes constant changes, and short-term electricity load exhibits strong randomness and volatility, making load conditions challenging to predict. Hence, this paper proposes a short-term electricity demand streaming forecasting model that combines wavelet decomposition with Random Forest to enhance the accuracy of short-term electricity load forecasting. This model establishes a load feature system, utilizing a three-scale wavelet decomposition algorithm to break down the load sequence into several sub-sequences of different frequency bands. Subsequently, Random Forest load forecasting models are separately established for these sub-sequences. The final load prediction is obtained through reconstruction. This approach enables quasi-real-time short-term forecasting analysis of demand-side electricity demand.

Research on emotion recognition of bimodal bioelectrical features based on D-S evidence theory

As the pace of life becomes faster and faster, people with weaker mental capacity are prone to a range of negative emotions in certain environments, such as mania and anxiety, and these negative emotions can interfere with normal learning and life, and even affect physical and mental health. Therefore, it is more and more important to accurately analyze people's emotional state.In this paper, based on the bimodal signals of EEG and EMG, we combine the broad learning network and D-S evidence theory to do an emotion recognition analysis of bioelectrical signals evoked by visual stimuli. Firstly, a suitable experimental paradigm is designed to acquire EEG signals and EMG signals through emotion evocation, followed by pre-processing, decomposing the signals using a wavelet decomposition algorithm, extracting signal components significantly affected by emotional states, and calculating the corresponding differential entropy as well as power spectrum features; after that, EMG features in time and frequency domains are extracted, and different features are fused and analyzed; BLS is used for classification, and finally, the BLS classification The results are fused with the introduced D-S evidence theory at the decision level. The method can recognize emotions effectively with an accuracy of 85.2% in triple classification scenarios. The proposed emotion recognition model is expected to provide a new technical tool for emotional cognitive science research and emotional brain-computer interface systems.

Research on Load Spectrum Reconstruction Method of Exhaust System Mounting Bracket of a Hybrid Tractor Based on MOPSO-Wavelet Decomposition Technique

To overcome the limitations of the hybrid tractor bumping tests, which include extended cycle times, high costs, and impracticality for single-part reliability verification, this study focuses on the exhaust system mounting bracket of a hybrid tractor. A novel approach that combines multi-objective particle swarm optimization (MOPSO) and wavelet decomposition algorithms was employed to enhance the reconstruction of shock vibration signals. This approach aims to enable the efficient acquisition of input signals for subsequent shaker table testing. The methodology involves a systematic evaluation of the spectral correlation between the original signal and the reconstructed signal at the stent’s response position, along with signal compression time. These parameters collectively constitute the objective function. The multi-objective particle swarm optimization algorithm is then deployed to explore a range of crucial parameters, including wavelet basic functions, the number of wavelet decomposition layers, and the selection of wavelet components. This exhaustive exploration identifies an optimized signal reconstruction method that accurately represents shock vibration loads. Upon rigorous screening based on our defined objectives, the optimal solution vector was determined, which includes the utilization of the dB10 wavelet basic function, employing a 12-layer wavelet decomposition, and selecting wavelet components a12 and d3~d11. This specific configuration enables the retention of 95% of the damage coefficients while significantly compressing the test time to just 46% of the original signal duration. The implications of our findings are substantial as the reconstructed signal obtained through our optimized approach can be readily applied to shaker excitation. This innovation results in a notable reduction in test cycle time and associated costs, making it particularly valuable for engineering applications, especially in tractor design and testing.

Big data-based method for automatic localization of power quality disturbance signal of PV access distribution network

At present, the conventional automatic localization method of the power quality disturbance signal of the distribution network mainly extracts the feature vector of the disturbance signal. It constructs the target localization function, which leads to poor localization accuracy because the noise part of the disturbance signal is ignored. In this regard, the automatic localization method of PV access power quality disturbance signal based on big data is proposed. By using the wavelet decomposition algorithm, the noise part of the power quality disturbance signal is removed, and the measurement matrix and reconstruction function are combined with compressing and reconstructing the disturbance signal. Finally, the automatic positioning of the disturbance signal is achieved by extracting the characteristics of the disturbance signal data and clustering analysis processing. In the experiment, the designed signal localization method is tested for the localization effect. The results can prove that when the proposed method is used to automatically localize the disturbed signal, the localization results are consistent with the spatial feature distribution of the signal and have a more desirable automatic localization effect.

A Novel Motor Fault Diagnosis Method Based on Generative Adversarial Learning with Distribution Fusion of Discrete Working Conditions

Many kinds of electrical equipment are used in civil and building engineering. The motor is one of the main power components of this electrical equipment, which can provide stable power output. During the long-term use of motors, various motor faults may occur, which affects the normal use of electrical equipment and even causes accidents. It is significant to apply fault diagnosis for the motors at the construction site. Aiming at the problem that signal data of faulty motor lack diversity, this research designs a multi-layer perceptron Wasserstein generative adversarial network, which is used to enhance training data through distribution fusion. A discrete wavelet decomposition algorithm is employed to extract the low-frequency wavelet coefficients from the original motor current signals. These are used to train the multi-layer perceptron Wasserstein generative adversarial model. Then, the trained model is applied to generate fake current wavelet coefficients with the fused distribution. A motor fault classification model consisting of a feature extractor and pattern recognizer is built based on perceptron. The data augmentation experiment shows that the fake dataset has a larger distribution than the real dataset. The classification model trained on a real dataset, fake dataset and combined dataset achieves 21.5%, 87.2%, and 90.1% prediction accuracy on the unseen real data, respectively. The results indicate that the proposed data augmentation method can effectively generate fake data with the fused distribution. The motor fault classification model trained on a fake dataset has better generalization performance than that trained on a real dataset.

Fault diagnosis of railway point machines based on wavelet transform and artificial immune algorithm

Abstract Aiming at the current problems of high failure rate and low diagnostic efficiency of railway point machines (RPMs) in the railway industry, a short-time method of fault diagnosis is proposed. Considering the effect of noise on power signals in the data acquisition process of the railway centralized signaling monitoring (CSM) system, this study utilizes wavelet threshold denoising to eliminate interference. The results show that the accuracy of fault diagnosis can be improved by 4.4% after denoising the power signals. Then in order to attain a lighter weight and shorten the running time of the diagnosis model, Mallat wavelet decomposition and artificial immune algorithm are applied to RPM fault diagnosis. Finally, voluminous experiments using veritable power signals collected from CSM are introduced, which show that combining these methods can procure higher precision of RPMs and curtail fault diagnosis time. This substantiates the validity and feasibility of the presented approach.

An effective algorithm for offshore air temperature prediction with LSTM neural network and wavelet decomposition and reconstruction

Offshore air temperature is an important parameter in marine scientific research. The change of offshore air as an indicator of the marine ecological environment is not only related to the growth of offshore organisms but also affects the development of the marine economy. Effective prediction of offshore air temperature is significant. The prediction model of offshore air temperature data is established by using wavelet decomposition and reconstruction algorithm combined with long short-term memory neural network (LSTM). We use the wavelet decomposition to decompose the offshore air temperature data of the ocean station into the overview signal and the detail signal, and the decomposed signal is reconstructed by a single branch to obtain the reconstructed signal. Then, input the reconstructed signals into LSTM model to predict the future offshore temperature. Finally, through the experiments, the proposed model is verified that has more advantages than the prediction effect of the LSTM offshore air temperature prediction model based on seasonal-trend decomposition procedure based on seasonal trend loss (STL) decomposition and the traditional LSTM prediction model. The proposed model has better prediction accuracy for offshore air temperature, and the prediction model can achieve effective prediction of the offshore air temperature.

A Wavelet PM2.5 Prediction System Using Optimized Kernel Extreme Learning with Boruta-XGBoost Feature Selection

The fine particulate matter (PM2.5) concentration has been a vital source of info and an essential indicator for measuring and studying the concentration of other air pollutants. It is crucial to realize more accurate predictions of PM2.5 and establish a high-accuracy PM2.5 prediction model due to their social impacts and cross-field applications in geospatial engineering. To further boost the accuracy of PM2.5 prediction results, this paper proposes a new wavelet PM2.5 prediction system (called WD-OSMSSA-KELM model) based on a new, improved variant of the salp swarm algorithm (OSMSSA), kernel extreme learning machine (KELM), wavelet decomposition, and Boruta-XGBoost (B-XGB) feature selection. First, we applied the B-XGB feature selection to realize the best features for predicting hourly PM2.5 concentrations. Then, we applied the wavelet decomposition (WD) algorithm to reach the multi-scale decomposition results and single-branch reconstruction of PM2.5 concentrations to mitigate the prediction error produced by time series data. In the next stage, we optimized the parameters of the KELM model under each reconstructed component. An improved version of the SSA is proposed to reach higher performance for the basic SSA optimizer and avoid local stagnation problems. In this work, we propose new operators based on oppositional-based learning and simplex-based search to mitigate the core problems of the conventional SSA. In addition, we utilized a time-varying parameter instead of the main parameter of the SSA. To further boost the exploration trends of SSA, we propose using the random leaders to guide the swarm towards new regions of the feature space based on a conditional structure. After optimizing the model, the optimized model was utilized to predict the PM2.5 concentrations, and different error metrics were applied to evaluate the model’s performance and accuracy. The proposed model was evaluated based on an hourly database, six air pollutants, and six meteorological features collected from the Beijing Municipal Environmental Monitoring Center. The experimental results show that the proposed WD-OLMSSA-KELM model can predict the PM2.5 concentration with superior performance (R: 0.995, RMSE: 11.906, MdAE: 2.424, MAPE: 9.768, KGE: 0.963, R2: 0.990) compared to the WD-CatBoost, WD-LightGBM, WD-Xgboost, and WD-Ridge methods.

Attention emotion recognition via ECG signals

BackgroundPhysiological signal‐based research has been a hot topic in affective computing. Previous works mainly focus on some strong, short‐lived emotions ( e.g., joy, anger), while the attention, which is a weak and long‐lasting emotion, receives less attraction. In this paper, we present a study of attention recognition based on electrocardiogram (ECG) signals, which contain a wealth of information related to emotions.MethodsThe ECG dataset is derived from 10 subjects and specialized for attention detection. To relieve the impact of noise of baseline wondering and power‐line interference, we apply wavelet threshold denoising as preprocessing and extract rich features by pan‐tompkins and wavelet decomposition algorithms. To improve the generalized ability, we tested the performance of a variety of combinations of different feature selection algorithms and classifiers.ResultsExperiments show that the combination of generic algorithm and random forest achieve the highest correct classification rate (CCR) of 86.3%.ConclusionThis study indicates the feasibility and bright future of ECG‐based attention research.

Composite wavelet decomposition algorithm combined with correlation analysis in white-light scanning interferometry

White-light scanning interferometry (WLSI) is a mature and efficient technique that is widely used in the field of high-precision three-dimensional measurements. However, coherence and phase information in the WLSI technique are inconsistent, false height jumps, such as batwings artifact and ghost steps, are typical obstacles in applying WLSI, which distort measurement results and reduce the measurement accuracy. In this study, we proposed a composite wavelet decomposition algorithm combined with a correlation analysis to solve the above two obstacles without having any prior knowledge about the tested sample. Two calibration step plates with heights of 459.8 ± 3.0 nm and 90.5 ± 2.8 nm were used to evaluate the accuracy of the proposed algorithm, and the results were compared with those of Centroid and Fourier transform algorithms. Consequently, the proposed algorithm exhibited a higher correction effect, and the accuracy was improved by more than two times, which proved that our approach could effectively suppress the batwings effect and eliminate the ghost steps.

CARC3: Characterizing Audio Traits of Various Systemic to Pulmonary Connections in Ductal Dependent Infants

Background: Infants with congenital heart disease who have insufficient systemic or pulmonary blood flow require a connection between the systemic and pulmonary vasculature. There are four types of shunts: right ventricle to pulmonary artery (Sano), Blalock-TaussigThomas (BTTS), central, and ductal arterial stent (DAS). They have a high incidence of mortality and unplanned interventions. Machine learning algorithms evaluate a shunt murmur recorded on a digital stethoscope to differentiate the audio features of various shunts. Methods: A 3M Littman Electronic Stethoscope acquired phonocardiogram (PCG) data from infants with shunt placement on post-operative days 0, 1, 2, 7, 14, discharge and day of shunt take-down, as well as, when imaging (echo, CT, catheterization) was performed. Patient demographics and clinical outcomes noted. A wavelet decomposition algorithm denoised the PCG data with a bandpass filter ranging from 20 to 2000 Hz and then segmented into S1 and S2. One hundred and fifteen audio features, comprising temporal, spectral, statistical, and bandpower domains were extracted from the waveforms. A statistical algorithm including ANOVA, Kruskal Wallis and pairwise comparison tests detected 33 total features showing a statistically significant difference in means (alpha = 0.05). Results: Thirteen subjects (BTTS (n=3), DAS (n=2), and Sano (n=8)) who had second stage palliation were analyzed using the two most recent recordings prior to shunt takedown. Principal component analysis (Figure 1) and tdistributed stochastic neighbor embedding (Figure 2) visualization techniques indicate a separation among shunt types. Conclusion: Audio signatures could exist based on shunt type characteristics and may lead to a novel non-invasive tool to monitor shunt health. Figure 1. Principal Component Analysis (PCA) indicating a separation in BTTS, Sano, and PDA shunt types from selected recordings. Figure 2. t-distributed stochastic neighbor embedding (tSNE) indicating a separation in BTTS, Sano, and PDA shunt types from selected recordings.

PM2.5 concentration prediction based on WD-SA-LSTM-BP model: a case study of Nanjing city.

PM2.5 concentration is an important indicator to measure the concentration of air pollutants, and it is of important social significance and application value to realize accurate prediction of PM2.5 concentration. To further improve the accuracy of PM2.5 concentration prediction, this paper proposes a hybrid machine learning model (WD-SA-LSTM-BP model) based on simulated annealing (SA) optimization and wavelet decomposition. Firstly, the wavelet decomposition algorithm was used to realize the multiscale decomposition and single-branch reconstruction of PM2.5 concentrations to weaken the prediction error caused by time series data. Secondly, the SA optimization method was used to optimize the super-parameters of each machine learning model under each reconstructed component, so as to solve the problem that it is difficult to determine the parameters of machine learning model. Thirdly, the optimized machine learning model was used to predict the PM2.5 concentration, and the error value was calculated from the actual measured value. Then, the optimized machine learning model was used to predict the error value. Finally, the predicted error value was added to the predicted PM2.5 concentration to obtain the final predicted PM2.5 concentration. The study is experimentally validated based on daily PM2.5 data collected from Nanjing air quality monitoring stations. The experimental results showed that the RMSE and MAE values of the constructed WD-SA-LSTM-BP model were 5.26 and 3.72, respectively, which were superior to those of the WD-LSTM and LSTM models, indicating that the hybrid machine learning model based on SA optimization and wavelet decomposition could better predict the PM2.5 concentration.

Analysis of rock microseismic signal based on blind source wavelet decomposition algorithm

At present, microseismic technology is a widely used method for monitoring the rock burst phenomenon during the construction of deep-buried tunnels. The rock fracture in the tunnel will generate seismic waves. The seismic wave has strong randomness and low energy and is usually mixed with environmental noise, which is called the microseismic signal. The microseismic signal received by the geophone will contain other noises. So, there is an urgent need for an algorithm that can quickly decompose the mixed signal. To solve this problem, this paper proposes a blind source wavelet algorithm to extract the rock fracture signal. First, the signal is preprocessed, and the mixed signal matrix is established. Second, the blind source decomposition algorithm is used to process the signal, and the effective signal is reconstructed. Finally, the wavelet algorithm is used to further remove the noise to enhance the microseismic signal. The proposed method is compared with the wavelet decomposition method and the empirical mode decomposition (EMD) method through the laboratory simulation data and the actual signal of Baihetan Hydropower Station. It is concluded that the proposed algorithm can effectively decompose and remove the noise in the mixed signal, and the decomposition accuracy is better than the wavelet decomposition method and the EMD method. The proposed algorithm has certain practical significance for the identification of microseismic signals of rock fracture and for rock burst early warning.

A novel wavelet decomposition and transformation convolutional neural network with data augmentation for breast cancer detection using digital mammogram

Research in deep learning (DL) has continued to provide significant solutions to the challenges of detecting breast cancer in digital images. Image preprocessing methods and architecture enhancement techniques have been proposed to improve the performance of DL models such as convolutional neural networks (CNNs). For instance, the wavelet decomposition function has been used for image feature extraction in CNNs due to its strong compactness. Additionally, CNN architectures have been optimized to improve the process of feature detection to support the classification process. However, these approaches still lack completeness, as no mechanism exists to discriminate features to be enhanced and features to be eliminated for feature enhancement. More so, no studies have approached the use of wavelet transform to restructure CNN architectures to improve the detection of discriminant features in digital mammography for increased classification accuracy. Therefore, this study addresses these problems through wavelet-CNN-wavelet architecture. The approach presented in this paper combines seam carving and wavelet decomposition algorithms for image preprocessing to find discriminative features. These features are passed as input to a CNN-wavelet structure that uses the new wavelet transformation function proposed in this paper. The CNN-wavelet architecture applied layers of wavelet transform and reduced feature maps to obtain features suggestive of abnormalities that support the classification process. Meanwhile, we synthesized image samples with architectural distortion using a generative adversarial network (GAN) model to argue for their training datasets' insufficiency. Experimentation of the proposed method was carried out using DDSM + CBIS and MIAS datasets. The results obtained showed that the new method improved the classification accuracy and lowered the loss function values. The study's findings demonstrate the usefulness of the wavelet transform function in restructuring CNN architectures for performance enhancement in detecting abnormalities leading to breast cancer in digital mammography.

An Algorithm with the Even-odd Splitting of the Wavelet Transform of Non-Hermitian Splines of the Seventh Degree

The article investigates an implicit method of decomposition of the 7th degree non-Hermitian splines into a series of wavelets with two zero moments. The system of linear algebraic equations connecting the coefficients of the spline expansion on the initial scale with the spline coefficients and wavelet coefficients on the embedded scale is obtained. The even-odd splitting of the wavelet decomposition algorithm into a solution of the half-size five-diagonal system of linear equations and some local averaging formulas are substantiated. The results of numerical experiments on accuracy on polynomials and compression of spline-wavelet decomposition are presented.

Wavelet Decomposition Impacts on Traditional Forecasting Time Series Models

This investigative study is focused on the impact of wavelet on traditional forecasting time-series models, which significantly shows the usage of wavelet algorithms. Wavelet Decomposition (WD) algorithm has been combined with various traditional forecasting time-series models, such as Least Square Support Vector Machine (LSSVM), Artificial Neural Network (ANN) and Multivariate Adaptive Regression Splines (MARS) and their effects are examined in terms of the statistical estimations. The WD has been used as a mathematical application in traditional forecast modelling to collect periodically measured parameters, which has yielded tremendous constructive outcomes. Further, it is observed that the wavelet combined models are classy compared to the various time series models in terms of performance basis. Therefore, combining wavelet forecasting models has yielded much better results.

Hyperspectral Image Classification Using CNN-Enhanced Multi-Level Haar Wavelet Features Fusion Network

Convolutional neural networks (CNNs) are widely utilized in hyperspectral image (HSI) classification due to their powerful capability to automatically learn features. However, ordinary CNN mainly captures the spatial characteristics of HSI and ignores the spectral information. To alleviate the issue, this work proposes a CNN-enhanced multi-level Haar wavelet features fusion network (CNN-MHWF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> N), which combines the spatial features obtained through 2-D-CNN with the Haar wavelet decomposition features to obtain sufficient spectral–spatial features. Specifically, factor analysis is first used to reduce the HSI dimension. Then, four-level decomposition features are obtained through the Haar wavelet decomposition algorithm, which of them are, respectively, concatenated with four-layer convolution features for combining spatial with spectral information. In this way, spectral–spatial features achieve better information interaction. Besides, a double filtrating feature fusion module is designed, which is operated following each level spectral–spatial features to obtain finer characteristics. Finally, those recognizable features are merged via a fusion operator. The whole designed model is conducive to enhancing the final HSI classification performance. In addition, experiments also reveal that the designed model is superior on three benchmark databases compared with the state-of-the-art approaches.

Study on the Law of Harmful Gas Release from Limnoperna fortunei (Dunker 1857) during Maintenance Period of Water Tunnel Based on K-Means Outlier Treatment

It is of great significance for air pollution control and personnel safety guarantee to master the release characteristics of harmful gases in the process of Limnoperna fortunei corruption. In view of the lack of research on the environmental pollution caused by the corruption of Limnoperna fortunei, a model experiment was designed to study the three harmful gases of NH3, H2S, and CH4 in the putrid process of Limnoperna fortunei by considering the density of Limnoperna fortunei and the time of leaving water. The results show that: (1) The recognition and processing of outliers based on wavelet decomposition and K-means algorithm can effectively reduce the standard deviation and coefficient of variation of the data set and improve the accuracy of the data set. (2) The variation of NH3 and H2S gas concentrations with the time of water separation satisfies polynomial linear regression (R2 &gt; 99%). (3) At a density of 0.5–7.0 × 104 mussels/m2, the highest concentration of NH3 reached 47.9777–307.9454 mg/m3 with the increase in the density of Limnoperna fortunei and the extension of the time away from water, far exceeding the occupational exposure limit of NH3 of 30 mg/m3, potentially threatening human health and safety. The highest detection value of H2S concentration is 0.1909–5.0946 mg/m3, and the highest detection concentration of CH4 is 0.02%, both of which can be ignored.