Complete Ensemble Empirical Mode Decomposition Research Articles

A combined model using secondary decomposition for crude oil futures price and volatility forecasting: Analysis based on comparison and ablation experiments

To accurately forecast crude oil futures price and volatility, this article presents a novel deep learning combined model using secondary decomposition with West Texas Intermediate crude oil futures (WTI) and North Sea Brent crude oil futures (Brent) as examples. Firstly, a trend subsequence and several noise subsequences are obtained by decomposing the crude oil futures price or volatility using the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and the secondary decomposition is performed on the highest frequency noise subsequence using the variational mode decomposition (VMD). Secondly, the remaining subsequences obtained from CEEMDAN and the subsequences obtained from VMD are predicted separately using the BiGRU-Attention-CNN model. Finally, the predicted crude oil futures price or volatility is calculated by linearly integrating the predicted values of each subsequence. The empirical analysis shows that the novel combined model using secondary decomposition proposed in this paper has the best forecasting performance among many models, both in the comparison experiments and in the ablation experiments. The model is also shown to have good robustness by predicting the volatility at different maturities, varying the ratio of the training set and the test set for crude oil futures price prediction, and predicting the price of crude oil futures after extreme events. Overall, the novel combined forecasting model using secondary decomposition proposed in this paper can help countries grasp the direction of the crude oil market and improve national economic and political security.

A Hybrid Model Based on CEEMDAN-GRU and Error Compensation for Predicting Sunspot Numbers

To improve the predictive accuracy of sunspot numbers, a hybrid model was built to forecast future sunspot numbers. In this paper, we present a prediction model based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), gated recurrent unit (GRU), and error compensation for predicting sunspot numbers. CEEMAND is applied to decompose the original sunspot number data into several components, which are then used to train and test the GRU for the optimal parameters of the corresponding sub-models. Error compensation is utilized to solve the delay phenomenon between the original sunspot number and the predictive result. We compare our method with the informer, extreme gradient boosting combined with deep learning (XGboost-DL), and empirical mode decomposition combined long short-term memory neutral network and attention mechanism (EMD-LSTM-AM) methods, and evaluation metrics, such as RMSE and MAE, are used to measure their performance. Our method decreases more than 2.2813 and 3.5827 relative to RMSE and MAE, respectively. Thus, the experiment can demonstrate that our method has an obvious advantage compared to others.

A defect contour imaging method of small diameter pipeline based on self-made three-dimensional magnetic sensor

Purpose The purpose of this paper is to use the corresponding magnetic sensor and detection method to detect and image the defects of small diameter pipelines. Urban gas pipeline is an energy transportation tool for urban industrial production and social life, which is closely related to urban safety. Preventing the occurrence of urban gas pipeline transportation accidents and carrying out pipeline defect detection are of great significance for the urban economic and social stability. To perform pipeline defect detection, the magnetic flux leakage internal detection method is generally used in the detection of large-diameter long-distance oil and gas pipelines. However, in terms of the internal detection of small-diameter pipelines, due to the heavy weight, large structure of the detection device and small pipe diameter, the detection is more difficult. Design/methodology/approach In order to solve the above matters, self-made three-dimensional magnetic sensor and three-dimensional magnetic flux leakage imaging direct method are proposed for studying the defect identification. Firstly, for adapting to the diameter range of small-diameter pipelines, and containing the complete information of the defect, a self-made three-dimensional magnetic sensor is made in this paper to improve the accuracy of magnetic flux leakage detection. And on the basis of it, a small diameter pipeline defect detection system is built. Secondly, as detection signal may be affected by background magnetic field interference and the jitter interference, the complete ensemble empirical mode decomposition with adaptive noise method is utilized to screen the detected signal. As a result, the useful signal is reconstructed and the interference signal is removed. Finally, the defect contour inversion imaging of detection is realized based on the direct method of three-dimensional magnetic flux leakage imaging, which includes three-dimensional magnetic flux leakage detection data and data segmentation recognition. Findings The three-dimensional magnetic flux leakage imaging experimental results shown that, compared to the actual defects, the typical defects, irregular defects and crack groove defects can be analyzed by the magnetic flux leakage defect contour imaging method in qualitative and quantitative way respectively, which provides a new idea for the research of defect recognition. Originality/value A three-dimensional magnetic sensor is made to adapt the diameter range of small diameter pipeline, and based on it, a small-diameter pipeline defect detection system is built to collect and display the magnetic flux leakage signal.

Research on Dual-Grating Spacing Calibration Method Based on Multiple Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise Combined with Hilbert Transform

The paper proposes a method for the calibration of spacing in dual-grating based on Multiple Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) combined with Hilbert Transform (HT), referred to as Multiple ICEEMDAN-HT. This method addresses the potential impact of nonlinear factors on phase extraction accuracy, consequently on ranging precision in the homodyne interference of the dual-grating. Building upon the ICEEMDAN algorithm, the signal undergoes iterative decomposition and reconstruction using the sample entropy criterion. The intrinsic mode functions (IMFs) obtained from multiple iterations are then reconstructed to obtain the complete signal. Through a simulation and comparison with other signal decomposition methods, the repeatability and completeness of signal reconstruction by Multiple ICEEMDAN are verified. Finally, an actual dual-grating ranging system is utilized to calibrate the spacing of the planar grating. Experimental results demonstrate that the calibration relative error of the Multiple ICEEMDAN-HT phase unwrapping method can be reduced to as low as 0.07%, effectively enhancing the signal robustness and spacing calibration precision.

Short-term prediction of PM2.5 concentration by hybrid neural network based on sequence decomposition.

Accurate forecasting of PM2.5 concentrations serves as a critical tool for mitigating air pollution. This study introduces a novel hybrid prediction model, termed MIC-CEEMDAN-CNN-BiGRU, for short-term forecasting of PM2.5 concentrations using a 24-hour historical data window. Utilizing the Maximal Information Coefficient (MIC) for feature selection, the model integrates Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Convolutional Neural Network (CNN), and Bidirectional Recurrent Gated Neural Network (BiGRU) to optimize predictive accuracy. We used 2016 PM2.5 monitoring data from Beijing, China as the empirical basis of this study and compared the model with several deep learning frameworks. RNN, LSTM, GRU, and other hybrid models based on GRU, respectively. The experimental results show that the prediction results of the hybrid model proposed in this question are more accurate than those of other models, and the R2 of the hybrid model proposed in this paper improves the R2 by nearly 5 percentage points compared with that of the single model; reduces the MAE by nearly 5 percentage points; and reduces the RMSE by nearly 11 percentage points. The results show that the hybrid prediction model proposed in this study is more accurate than other models in predicting PM2.5.

Traffic Flow Prediction by an Ensemble Framework with Data Denoising and Functional Principal Components Analysis

Accurate traffic flow data are important for congestion identification and real-time traffic control. Raw traffic flow data may be disturbed by different noises during the acquisition process, which leads to the degradation of model prediction performance. To address this problems, this paper proposes a data denoising method based on the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) combined with wavelet threshold denoising (WTD) to suppress potential outliers in the data, and then introduces functional principal component analysis to accomplish the traffic forecasting task. We use the proposed framework for real traffic flow prediction and validate the prediction performance of the proposed framework using root mean square error and mean absolute error. In the prediction performance comparison, five denoising methods are considered: empirical mode decomposition (EMD), ensemble empirical mode decomposition, complete ensemble empirical mode decomposition with adaptive noise, ICEEMDAN, and WTD. The prediction results show that the model combined with the denoising method outperforms the model without the denoising method. ICEEMDAN combined with WTD’s denoising method improves prediction performance more than other methods. Furthermore, the WTD method with dmey type achieves higher accuracy than other types.

A Hybrid Model for Carbon Price Forecasting Based on Improved Feature Extraction and Non-Linear Integration

Accurately predicting the price of carbon is an effective way of ensuring the stability of the carbon trading market and reducing carbon emissions. Aiming at the non-smooth and non-linear characteristics of carbon price, this paper proposes a novel hybrid prediction model based on improved feature extraction and non-linear integration, which is built on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), fuzzy entropy (FuzzyEn), improved random forest using particle swarm optimisation (PSORF), extreme learning machine (ELM), long short-term memory (LSTM), non-linear integration based on multiple linear regression (MLR) and random forest (MLRRF), and error correction with the autoregressive integrated moving average model (ARIMA), named CEEMDAN-FuzzyEn-PSORF-ELM-LSTM-MLRRF-ARIMA. Firstly, CEEMDAN is combined with FuzzyEn in the feature selection process to improve extraction efficiency and reliability. Secondly, at the critical prediction stage, PSORF, ELM, and LSTM are selected to predict high, medium, and low complexity sequences, respectively. Thirdly, the reconstructed sequences are assembled by applying MLRRF, which can effectively improve the prediction accuracy and generalisation ability. Finally, error correction is conducted using ARIMA to obtain the final forecasting results, and the Diebold–Mariano test (DM test) is introduced for a comprehensive evaluation of the models. With respect to carbon prices in the pilot regions of Shenzhen and Hubei, the results indicate that the proposed model has higher prediction accuracy and robustness. The main contributions of this paper are the improved feature extraction and the innovative combination of multiple linear regression and random forests into a non-linear integrated framework for carbon price forecasting. However, further optimisation is still a work in progress.

Non-linear Time Series Prediction using Improved CEEMDAN, SVD and LSTM

This study recommends a new time series forecasting model, namely ICEEMDAN - SVD - LSTM model, which coalesces Improved Complete Ensemble EMD with Adaptive Noise, Singular Value Decomposition and Long Short Term Memory network. It can be applied to analyse Non-linear and non-stationary data. The framework of this model is comprised of three levels, namely ICEEMDAN level, SVD level and LSTM level. The first level utilized ICEEMDAN to break up the series into some IMF components along with a residue. The SVD in the second level accounts for de-noising of every IMF component and residue. LSTM forecasts all the resultant IMF components and residue in third level. To obtain the forecasted values of the original data, the predictions of all IMF components and residue are added. The proposed model is contrasted with other extant ones, namely LSTM model, EMD - LSTM model, EEMD - LSTM model, CEEMDAN - LSTM model, EEMD - SVD - LSTM model, ICEEMDAN - LSTM model and CEEMDAN - SVD - LSTM model. The comparison bears witness to the potential of the recommended model over the traditional models.

ICEEMDAN-Informer-GWO: a hybrid model for accurate wind speed prediction.

Extensive research has been diligently conducted on wind energy technologies in response to pressing global environmental challenges and the growing demand for energy. Accurate wind speed predictions are crucial for the effective integration of large wind power systems. This study presents a novel and hybrid framework called ICEEMDAN-Informer-GWO, which combines three components to enhance the accuracy of wind speed predictions. The improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) component improves the decomposition of wind speed data, the Informer model provides computationally efficient wind speed predictions, and the grey wolf optimisation (GWO) algorithm optimises the parameters of the Informer model to achieve superior performance. Three different sets of wind speed prediction (WSP) models and wind farm data from Block Island, Gulf Coast, and Garden City are used to thoroughly assess the proposed hybrid framework. This evaluation focusses on WSP for three specific time horizons: 5minutes, 30minutes, and 1hour ahead. The results obtained from the three conducted experiments conclusively demonstrate that the proposed hybrid framework exhibits superior performance, leading to statistically significant improvements across all three time horizons.

Method for the P-wave arrival pickup of rock fracture acoustic emission signals under strong noise

This research aimed to investigate the accuracy of picking of P-wave arrival times in rock fracture acoustic emission signals. In order to simulate the mining scenario, Gaussian white noise and pulse noise were added to the data collected in the laboratory. Complete ensemble empirical mode decomposition with adaptive noise + Wavelet (CEEMDAN + Wavelet) was improved in this paper, where the Spearman rank correlation coefficient was adopted to effectively select intrinsic mode functions for denoising which retained the inherent characteristics of the rock fracture signal. The absolute amplitude and energy change rate of the envelope signal, calculated based on the Hilbert transform, were used as the input of the short term average/long term average (STA/LTA) normalization algorithm to pickup the P-wave arrival time. The reliability of this method was tested on 30 groups of recorded rock fracture laboratory data and 60 groups of added noise data. Taking the manual pickup results as the standard, the errors of CEEMDAN + Wavelet + STA/LTA + AIC (Akaike information criterion) method with the absolute amplitude of the signal as the input are all within 10 ms, and 86.67% of the results are within 5 ms. The method proposed in this paper effectively addressing the issue of false pickup caused by the sensitivity of AIC and traditional STA/LTA method for strong noise, and achieving relatively high accuracy and stability in processing low signal-to-noise ratio signals. This work contributes to monitor microscopic changes in rock bodies and is of great significance for the prediction and monitoring of geological disasters.

A structural depth network embedding stacking model based on Moran’s index and seasonal trend for short-term solar irradiance prediction

Solar irradiance prediction of multi-station is influenced by complex spatial characteristics and temporal fluctuations, so how to obtain accurate short-term irradiance prediction by using photovoltaic cluster data becomes a challenge. In order to address the challenge, a structural depth network embedding stacking model based on Moran’s I and seasonal trend for short-term solar irradiance prediction is proposed. Firstly, the deep spatial characteristics of photovoltaic clusters are analyzed by using bivariate Moran index to capture strong correlations between stations. Secondly, seasonal trend decomposition is used to smooth the seasonal factors in time series, and Complete Ensemble Empirical Mode Decomposition with Adaptive Noise is used to decompose the residual components to reduce the fluctuation of signals and the complexity of data feature mapping. Finally, based on the Stacking framework, Long Short-term Memory Network, Least Absolute Selection and Shrinkage Operator, Gradient Boosted Decision Tree, and Artificial Neural Network are superimposed. And the aggregated sub-modal components are introduced into the robust integration model to predict the solar irradiance in multiple steps. Through the case study of the National Renewable Energy Laboratory open-source dataset in the United States, the conclusions are as follows: (1) The spatio-temporal method in this paper can effectively extract the deep spatial characteristics of photovoltaic stations. (2) The fusion strategy of embedded structured deep network effectively improves the stacking effect. (3) Comparing with many classical forecasting models, it shows that this method can effectively improve the forecasting effect.

Short-term energy consumption prediction for university dormitory buildings based on energy consumption grouping by considering meteorological factors and teaching schedules

Accurate short-term energy consumption prediction is essential for studying renewable energy utilization in building technologies. University dormitory energy consumption is affected by meteorological factors and teaching schedules, presenting a more complex time series compared to residential buildings, rendering traditional methods unsuitable. Therefore, in the present study, we developed a model called energy consumption grouping-Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise-Bi-directional Long Short-Term Memory (ECG-ICEEMDAN-BILSTM) for predicting short-term energy consumption in university dormitory buildings. Model was tested using electricity consumption data from university dormitory buildings in cold regions of China. Based on the energy consumption characteristics, we identified two consumption patterns comprising high power energy consumption (HPEC) and low power energy consumption (LPEC). Using the ECG method, data with similar daily energy consumption distributions for HPEC and LPEC were grouped. Subsequently, individual prediction models were constructed based on ICEEMDAN-BILSTM for each specific energy consumption group. The experimental results showed the ECG-ICEEMDAN-BILSTM model obtained more accurate predictions compared with other neural network models. In addition, the model was more effective at forecasting the complex distribution of the HPEC pattern, with reductions of at least 16.21% and 15.82% in MAPE and PRMSE, respectively. Thus, study still can obtain valuable insights to address the problem of short-term prediction using complex time series.

A novel lidar signal denoising method based on variational mode decomposition optimized using whale algorithm

Original lidar return signals are covered by high levels of noise that seriously affect the accuracy of subsequent data processing and inversion. Therefore, it is important to separate the effective signal from the returned signal with noise interference. In this paper, an efficient denoising method based on the variational mode decomposition (VMD) algorithm optimized using the global search strategy-based whale algorithm and the total variational stationary wavelet transform (GSWOA-VMD-SWTTV) is proposed, and this method is applied to denoising of lidar return signals. First, the global search strategy-based whale optimization algorithm (GSWOA) is used to acquire the optimal parameters of the VMD algorithm adaptively, and the lidar return signal is then decomposed by global search strategy-based whale optimization algorithm (GSWOA)-VMD. The effective modal components are then determined using the cross-correlation coefficient method from the decomposed modal components, and total variation stationary wavelet denoising is performed on each effective mode. Finally, the effective modes are reconstructed to obtain a clean lidar return signal. Moreover, to provide further verification of the effectiveness of the proposed method, it is compared with the ensemble empirical mode decomposition (EEMD) method, the complete EEMD with adaptive noise (CEEMDAN) method, the singular value decomposition (SVD) method, and the wavelet threshold method under sunny, cloudy, and dusty weather conditions. The experimental results demonstrate the superior noise reduction performance of the proposed algorithm, which can filter out strong noise from the signal while retaining the complete signal details without distortion; additionally, the proposed method has the highest signal-to-noise ratio and lowest mean square error.

MEMS Gyroscope Temperature Compensation Based on Improved Complete Ensemble Empirical Mode Decomposition and Optimized Extreme Learning Machine

Herein, we investigate the temperature compensation for a dual-mass MEMS gyroscope. After introducing and simulating the dual-mass MEMS gyroscope’s working modes, we propose a hybrid algorithm for temperature compensation relying on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), sample entropy, time–frequency peak filtering, non-dominated sorting genetic algorithm-II (NSGA II) and extreme learning machine. Firstly, we use ICEEMDAN to decompose the gyroscope’s output signal, and then we use sample entropy to classify the decomposed signals. For noise segments and mixed segments with different levels of noise, we use time–frequency peak filtering with different window lengths to achieve a trade-off between noise removal and signal retention. For the feature segment with temperature drift, we build a compensation model using extreme learning machine. To improve the compensation accuracy, NSGA II is used to optimize extreme learning machine, with the prediction error and the 2-norm of the output-layer connection weight as the optimization objectives. Enormous simulation experiments prove the excellent performance of our proposed scheme, which can achieve trade-offs in signal decomposition, classification, denoising and compensation. The improvement in the compensated gyroscope’s output signal is analyzed based on Allen variance; its angle random walk is decreased from 0.531076°/h/√Hz to 6.65894 × 10−3°/h/√Hz and its bias stability is decreased from 32.7364°/h to 0.259247°/h.

Arctic short-term wind speed forecasting based on CNN-LSTM model with CEEMDAN

Accurate wind speed forecasting is of great significance for the utilization of Arctic wind energy resources. The traditional single model is difficult to fully depict the nonlinearity of wind speed and its wide range of variations. In this paper, a hybrid model is proposed for multi-step wind speed forecasting, which combined complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), convolutional neural network (CNN) and long-short term memory neural network (LSTM). The wind speed series is firstly decomposed into several intrinsic mode functions (IMF) by CEEMDAN to provide more stable data. Secondly, four-step-ahead forecasts are realized using well-tuned CNN-LSTM model for each IMF. Finally, the forecasted wind speed is obtained by reconstruction. The effectiveness and feasibility of the proposed method is validated based on thorough evaluation and step-by-step analysis. The RMSE of the proposed model is 0.4046 m/s, which are reduced by 58 % compared with 7 benchmark models. Furthermore, the average prediction interval of the proposed model is also reduced by 20 %, 16 % and 7 % compared to CEEMDAN-FCNN, CEEMDAN-CNN and CEEMDAN-LSTM respectively. The results prove that all three parts of the proposed model contribute to a better performance in wind speed forecasting.

Refined offshore wind speed prediction: Leveraging a two-layer decomposition technique, gated recurrent unit, and kernel density estimation for precise point and interval forecasts

Precise forecasting of offshore wind speeds is paramount for various applications, including offshore wind energy production, disaster prevention and mitigation, and maritime navigation. This study introduces a novel model for point and interval prediction of offshore wind speed, incorporating an innovative two-layer decomposition technique, gated recurrent unit (GRU), and kernel density estimation (KDE). Initially, the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) is utilized to decompose the raw series into multiple subsequences, thus alleviating the inherent data nonstationarity. Subsequently, we employ the refined composite multiscale fuzzy entropy (RCMFE) to reconstruct the subsequences into multiple components, thereby reducing the computational complexity. Variational mode decomposition (VMD) is utilized to enhance the modeling accuracy to decompose the high-frequency component, thereby generating a set of submodels. The GRU, a specialized recurrent neural network architecture, is leveraged to forecast the remaining components and submodels, producing a series of subresults. The subresults are linearly summed to obtain the final point prediction result. Finally, based on point prediction, employing KDE to gauge the probability density distribution of prediction errors culminates in the derivation of the offshore wind speed prediction interval. The experimental results substantiate the superior predictive performance of the proposed prediction model.

[Snoring noise removal method for bowel sound signal during sleep].

Monitoring of bowel sounds is an important method to assess bowel motility during sleep, but it is seriously affected by snoring noise. In this paper, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) method was applied to remove snoring noise from bowel sounds during sleep. Specifically, the noisy bowel sounds were first band-pass filtered, then decomposed by the CEEMDAN method, and finally the appropriate components were selected to reconstruct the pure bowel sounds. The results of semi-simulated and real data showed that the CEEMDAN method was better than empirical mode decomposition and wavelet denoising method. The CEEMDAN method is used to remove snoring noise from bowel sounds during sleep, which lays an important foundation for using bowel sounds to assess the intestinal motility during sleep.

A hybrid multi-scale fusion paradigm for AQI prediction based on the secondary decomposition.

With rapid industrialization and urbanization, air pollution has become an increasingly severe problem. As a key indicator of air quality, accurate prediction of the air quality index (AQI) is essential for policymakers to establish effective early warning management mechanisms and adjust living plans. In this work, a hybrid multi-scale fusion prediction paradigm is proposed for the complex AQI time series prediction. First, an initial decomposition and integration of the original data is performed by combining the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and sample entropy (SE). Then, the subsequences, divided into high-frequency and low-frequency groups, are applied to different processing methods. Among them, the variational mode decomposition (VMD) is chosen to perform a secondary decomposition of the high-frequency sequence groups and integrated by using K-means clustering with sample entropy. Finally, multi-scale fusion training of sequence prediction results with different frequencies by using long short-term memory (LSTM) yields more accurate results with R2 of 0.9715, RMSE of 2.0327, MAE of 0.0154, and MAPE of 0.0488. Furthermore, validation of the AQI datasets acquired from four different cities demonstrates that the new paradigm is more robust and generalizable as compared to other baseline methods. Therefore, this model not only holds potential value in developing AQI prediction models but also serves as a valuable reference for future research on AQI control strategies.

A Lithium-Ion Battery Remaining Useful Life Prediction Model Based on CEEMDAN Data Preprocessing and HSSA-LSTM-TCN

Accurate prediction of the Remaining Useful Life (RUL) of lithium-ion batteries is crucial for reducing battery usage risks and ensuring the safe operation of systems. Addressing the impact of noise and capacity regeneration-induced nonlinear features on RUL prediction accuracy, this paper proposes a predictive model based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) data preprocessing and IHSSA-LSTM-TCN. Firstly, CEEMDAN is used to decompose lithium-ion battery capacity data into high-frequency and low-frequency components. Subsequently, for the high-frequency component, a Temporal Convolutional Network (TCN) prediction model is employed. For the low-frequency component, an Improved Sparrow Search Algorithm (IHSSA) is utilized, which incorporates iterative chaotic mapping and a variable spiral coefficient to optimize the hyperparameters of Long Short-Term Memory (LSTM). The IHSSA-LSTM prediction model is obtained and used for prediction. Finally, the predicted values of the sub-models are combined to obtain the final RUL result. The proposed model is validated using the publicly available NASA dataset and CALCE dataset. The results demonstrate that this model outperforms other models, indicating good predictive performance and robustness.

Functional near-infrared spectroscopy-based diagnosis support system for distinguishing between mild and severe depression using machine learning approaches.

Early diagnosis of depression is crucial for effective treatment. Our study utilizes functional near-infrared spectroscopy (fNIRS) and machine learning to accurately classify mild and severe depression, providing an objective auxiliary diagnostic tool for mental health workers. Develop prediction models to distinguish between severe and mild depression using fNIRS data. We collected the fNIRS data from 140 subjects and applied a complete ensemble empirical mode decomposition with an adaptive noise-wavelet threshold combined denoising method (CEEMDAN-WPT) to remove noise during the verbal fluency task. The temporal features (TF) and correlation features (CF) from 18 prefrontal lobe channels of subjects were extracted as predictors. Using recursive feature elimination with cross-validation, we identified optimal TF or CF and examined their role in distinguishing between severe and mild depression. Machine learning algorithms were used for classification. The combination of TF and CF as inputs for the prediction model yielded higher classification accuracy than using either TF or CF alone. Among the prediction models, the SVM-based model demonstrates excellent performance in nested cross-validation, achieving an accuracy rate of 92.8%. The proposed model can effectively distinguish mild depression from severe depression.