Particle Swarm Optimization Support Vector Machine Research Articles

Annual Runoff Forecasting Based on Multi-Model Information Fusion and Residual Error Correction in the Ganjiang River Basin

Accurate forecasting of annual runoff time series is of great significance for water resources planning and management. However, considering that the number of forecasting factors is numerous, a single forecasting model has certain limitations and a runoff time series consists of complex nonlinear and nonstationary characteristics, which make the runoff forecasting difficult. Aimed at improving the prediction accuracy of annual runoff time series, the principal components analysis (PCA) method is adopted to reduce the complexity of forecasting factors, and a modified coupling forecasting model based on multiple linear regression (MLR), back propagation neural network (BPNN), Elman neural network (ENN), and particle swarm optimization-support vector machine for regression (PSO-SVR) is proposed and applied in the Dongbei Hydrological Station in the Ganjiang River Basin. Firstly, from two conventional factors (i.e., rainfall, runoff) and 130 atmospheric circulation indexes (i.e., 88 atmospheric circulation indexes, 26 sea temperature indexes, 16 other indexes), principal components generated by linear mapping are screened as forecasting factors. Then, based on above forecasting factors, four forecasting models including MLR, BPNN, ENN, and PSO-SVR are developed to predict annual runoff time series. Subsequently, a coupling model composed of BPNN, ENN, and PSO-SVR is constructed by means of a multi-model information fusion taking three hydrological years (i.e., wet year, normal year, dry year) into consideration. Finally, according to residual error correction, a modified coupling forecasting model is introduced so as to further improve the accuracy of the predicted annual runoff time series in the verification period.

Blood platelet RNA enables the detection of multiple sclerosis.

BackgroundIn multiple sclerosis (MS), clinical assessment, MRI and cerebrospinal fluid are important in the diagnostic process. However, no blood biomarker has been confirmed as a useful tool in the diagnostic work-up.ObjectivesBlood platelets contain a rich spliced mRNA repertoire that can alter during megakaryocyte development but also during platelet formation and platelet circulation. In this proof of concept study, we evaluate the diagnostic potential of spliced blood platelet RNA for the detection of MS.MethodsWe isolated and sequenced platelet RNA of blood samples obtained from 57 MS patients and 66 age- and gender-matched healthy controls (HCs). 60% was used to develop a particle swarm-optimized (PSO) support vector machine classification algorithm. The remaining 40% served as an independent validation series.ResultsIn total, 1249 RNAs with differential spliced junction expression levels were identified between platelets of MS patients as compared to HCs, including EPSTI1, IFI6, and RPS6KA3, in line with reported inflammatory signatures in the blood of MS patients. The RNAs were subsequently used as input for a MS classifier, capable of detecting MS with 80% accuracy in the independent validation series.ConclusionsSpliced platelet RNA may enable the blood-based diagnosis of MS, warranting large-scale validation.

A grinding force prediction model for SiCp/Al composite based on single-abrasive-grain grinding

Grinding is the main processing method for particle-reinforced composites, and grinding force prediction models are very important for research on removal mechanisms. In this study, single-abrasive-grain grinding experiments on SiCp/Al composites were conducted to determine the grinding forces at different grinding process parameters. In addition, a prediction model for the single-abrasive-grain grinding force was established to study the influence of the grinding process parameters and grinding grain angle on the grinding force of SiCp/Al composite. Moreover, multi-abrasive-grain grinding experiments were conducted at different grinding process parameters, which resulted in different grinding forces. The support vector machine (SVM) prediction method based on particle swarm optimisation (PSO) was used to establish a prediction model for the multi-abrasive-grain grinding force; the single-abrasive-grain grinding forces at different grinding grain angles were the input, and the average experimental grain grinding force was the output. The results show that the error between the predicted and experimental grinding forces is below 12%. Furthermore, the grinding force decreases with increasing wheel speed and increases with increasing feed velocity and grinding depth. The PSO–SVM algorithm–based grinding force prediction model can accurately predict the grinding force of SiCp/Al composite and provides theoretical support for improved surface quality.

Measurement of Total Nitrogen Concentration in Surface Water Using Hyperspectral Band Observation Method

Nitrogen overload is one of the main reasons for the deterioration of surface water quality. Hence, monitoring nitrogen loadings is vital in maintaining good surface water quality. Increasingly, the use of spectral reflectance to monitor nitrogen concentration in water has shown potentials, but it poses some problems. Therefore, it is necessary to explore new methods of quantitative monitoring of nitrogen concentration in surface water. In this paper, hyperspectral data from surface water in the Ebinur Lake watershed are used to select sensitive bands using spectral transformation, the spectral index, and a coupling of these two methods. The particle swarm optimization support vector machine (PSO-SVM) model, constructed on the basis of sensitive bands, is used quantitatively to estimate the total nitrogen concentration in surface water and subsequently to verify its accuracy. The results show that the bands near 680, 850, and 940 nm can be used as sensitive bands for estimation of the total nitrogen concentration of surface water in arid regions. Compared with the best estimation models constructed by sensitive bands selected using the spectral transformation or the spectral index alone, the best model based on the coupling of these two measures is more accurate (R2 = 0.604, Root Mean Square Error (RMSE) = 1.61 mg/L, Residual Prediction Deviation (RPD) = 2.002). This coupling method leads to a robust, accurate, and strong predictability model, and can contribute to improved quantitative estimation of water quality indexes of rivers in arid regions.

Characterization of human motion by the use of an accelerometer-based detection system

Human motion mode detection is of great value in evaluating motor ability. In this study, a method of human motion mode detection based on an acceleration sensor was characterized. A human motion mode detection system was designed using the standard deviation, peak, valley and peak-valley interval as characteristic values and a particle swarm optimization – support vector machine (PSO-SVM) algorithm as the detection method. The measurements of 10 subjects in static, walking and running stages were collected and evaluated. The experimental results showed that the static, walking and running detection accuracy values were 96%, 90% and 92% respectively, and the overall value was 94%. Using the same number of samples, the detection accuracy of SVM was 82.57%, while that of PSO-SVM was 93.1%, which had the highest reliability. This study demonstrates the validity of the human motion mode detection method based on an acceleration sensor which can accurately detect static state, walking and running, and the method is worth further promotion and application.

A particle swarm optimization–support vector machine hybrid system with acoustic emission on damage degree judgment of carbon fiber reinforced polymer cables

The feasibility of machine learning in damage degree judgment of carbon fiber reinforced polymer cables was first verified by the improved b-value method and wavelet packet spectrum analysis. Then, a hybrid system with support vector machine classification and particle swarm optimization algorithms was proposed to realize the prediction. The b-value calculated with all acoustic emission events has better performance when noise cannot be avoided. The 1/b-value has almost the same trend with acoustic emission signal cumulative energy, which can meet the preliminarily needs of health monitoring. The particle swarm optimization clustering algorithm works by using nine characteristic parameters of acoustic emission signals. It demonstrates that the characteristic parameters of acoustic emission signals are closely related to the failure mode of the carbon fiber reinforced polymer cable. This indicates their correspondence to the cable’s damage degree and their ability to work as training data for machine learning. With particle swarm optimization, the trained support vector machine can reach at least 77% accuracy of a single acoustic emission signal when predicting the corresponding current damage degree. In addition, using the voting mechanism can promote the performance of support vector machine. This demonstrates the practicability of applying acoustic emission combined with machine learning as a damage degree judgment method for carbon fiber reinforced polymer cables.

Diagnosis of the Severity of Fusarium Head Blight of Wheat Ears on the Basis of Image and Spectral Feature Fusion.

Fusarium head blight (FHB), one of the most prevalent and damaging infection diseases of wheat, affects quality and safety of associated food. In this study, to realize the early accurate monitoring of FHB, a diagnostic model of disease severity was proposed based on the fusion features of image and spectral features. First, the hyperspectral image of FHB infected in the range of the 400–1000 nm spectrum was collected, and the color parameters of wheat ear and spot region were segmented based on image features. Twelve sensitive bands were extracted using the successive projection algorithm, gray-scale co-occurrence matrix, and RGB color model. Four texture features were extracted from each feature band image as texture variables, and nine color feature variables were extracted from R, G, and B component images. Texture features with high correlation and color features were selected to participate in the final model building parameters via correlation analysis. Finally, the particle swarm optimization support vector machine (PSO-SVM) algorithm was used to build the model based on the diagnosis model of disease severity of FHB with different combinations of characteristic variables. The experimental results showed that the PSO-SVM model based on spectral and color feature fusion was optimal. Moreover, the accuracy of the training and prediction set was 95% and 92%, respectively. The method based on fusion features of image and spectral features can accurately and effectively diagnose the severity of FHB, thereby providing a technical basis for the timely and effective control of FHB and precise application of a pesticide.

Distributed Systematic Grid-Connected Inverter Using IGBT Junction Temperature Predictive Control Method: An Optimization Approach

Distributed systematic grid-connected inverter practice needs to improve insulated gate bipolar transistor (IGBT) stability to ensure the safe operation. This study is to ensure the safety and reliability operation of the IGBT module in symmetry to meet the reliable and stable distributed systematic grid-connected inverter practice and the junction temperature is a parameter to assess its operating state. It is difficult to accurately acquire the IGBT junction temperature to be solved by a single method of combining the test and the modeling. The saturation voltage drop or collector current and module junction temperature data under different power cycles are measured by the power cycle test and the single pulse test. The improved chicken swarm optimization increases the chickens diversity and self-learning ability. The prediction model of the improved chicken swarm optimization-support vector machine is proposed to forecast the module junction temperature. The result showed to compare with the particle swarm optimization-support vector machine model and chicken swarm optimization-support vector machine model and showed the coincidence degree between the proposed model prediction value and the true value is higher. The mean absolute error ratio indicates the proposed model has a smaller error and a better prediction performance. The proposed model has a positive impact on improving the distributed systematic grid-connected inverter industrial development and promotes the new energy usage.

Mud pulse signal demodulation based on support vector machines and particle swarm optimization

Mud pulse telemetry is the most commonly used communication technology in measurement-while-drilling or logging-while-drilling systems. During the drilling process, the flowing drilling mud is the communication channel that carries the mud pulse signal to surface. Compared with the relatively benign conditions in the wireline communication method, the mud communication channel encounters high signal attenuation, low bandwidth, and numerous sources of noise. The harsh and volatile conditions of the mud channel make it difficult to demodulate the mud pulse signal. In this study, a novel signal demodulation technique based on support vector machines (SVM) is proposed. Meanwhile, the particle swarm optimization (PSO) algorithm is applied for selecting both the penalty parameter and the kernel parameter for the SVM classifier. To overcome the small training set problem and increase the demodulation accuracy, we introduce a training set expansion scheme which is implemented by enlarging the training set with correctly demodulated data based on the error detection method. Field experiment results prove the effectiveness of the training set expansion scheme and the superiority of the proposed methodology compared with the conventional adaptive fractionally-spaced decision-feedback equalizer (FS-DFE). Moreover, the computational time for the whole SVM demodulation procedure is much less than the signal duration. The proposed PSO–SVM method could be a potential solution for mud pulse signal demodulation in mud pulse telemetry systems.

Raman spectroscopy combined with multiple algorithms for analysis and rapid screening of chronic renal failure.

Chronic renal failure (CRF) is a symptom of kidney damage in the terminal stages. If a patient is not treated, then CRF will progress to uremia, which greatly reduces the lifespan of the patient. However, current screening strategies, including routine urine tests and medical imaging investigations, have poor sensitivity. Therefore, exploring new and efficient screening methods for CRF such as serum spectroscopy is of great significance. In this study, we first used Raman spectroscopy to classify sera from CRF patients and control subjects. A total of 47 samples from CRF patients and 54 samples from control subjects were acquired. The spectra revealed differences in the phospholipids and proteins between the CRF patients and control subjects. The differences between the CRF patients and control subjects were evaluated by building machine learning models. Subsequent principal component analysis (PCA) was first used for feature extraction. Then, back propagation (BP) neural network, extreme learning machine (ELM), genetic algorithms based on support vector machine (GA-SVM), particle swarm optimization-support vector machine (PSO-SVM), grid search-support vector machine (GS-SVM) and simulated annealing particle swarm optimization based on support vector machine (SAPSO-SVM) algorithms were employed to establish diagnostic models; the diagnostic accuracy of the six classifiers was 70.4 %, 71 %, 83.5 %, 86.9 %, 89.7 % and 82.8 %, respectively, for control subjects and CRF patients. The results show the potential of Raman spectroscopy in differentiating between the control subjects and CRF patients. Based on the limitations of current routine diagnostic methods, serum Raman spectroscopy may be an adjunct/replaceable method for the clinical diagnosis of CRF with the prospective validation of more samples.

Steel corrosion prediction based on support vector machines

• The characteristics of sections of corroded bars are quantified to predict sectional corrosion rate of steel. • Seven digital parameters are innovatively proposed to define corroded bar shapes. • The SVM can be used to predict the sectional corrosion rate of steel. • PSO-SVM is more accurate than GS-SVM for prediction sectional corrosion rate of steel. In this paper, the 3D coordinate data of the corrosion condition of rebar are obtained by a 3D scanning method. Seven numerical parameters, such as the roundness, the section roughness, the inscribed circle radius/circumscribed circle radius and the eccentricity, are obtained by the numerical calculation method. These seven parameters are used to characterize the cross-section morphology of rusted steel bars. The particle swarm optimization support vector machine (PSO-SVM) and the grid search support vector machine (GS-SVM) are used to calculate these seven cross-section digitization parameters to predict the sectional corrosion rate of steel. This work concluded that these two optimization support vector machine (SVM) methods can accurately predict the sectional corrosion rate of steel. Compared with the GS-SVM model, the PSO-SVM steel corrosion prediction model is more accurate.

A Deep-Learning-Based Oil-Well-Testing Stage Interpretation Model Integrating Multi-Feature Extraction Methods

The interpretation of well-testing data is a key means of decision-making support for oil and gas field development. However, conventional processing methods have many problems, such as the stochastic nature of the data, feature redundancies, the randomness of the initial weights or thresholds, and fluctuations in the generalization ability with slight changes in the network parameters. These result in a poor ability to characterize data features and a low generalization ability of the interpretation models. We propose a new integrated well-testing interpretation model based on a multi-feature extraction method and deep mutual information classifiers (MFE-DMIC). This model can avoid the low model classification accuracy caused by the simple training structures, lacking of redundancy elimination, and the non-optimal classifier configuration parameters. First, we obtained the initial features according to four classical feature extraction methods. Then, we eliminated feature redundancies using a deep belief network and united the maximum information coefficient method to achieve feature purification. Finally, we calculated the interpretation results using a hybrid particle swarm optimization–support vector machine classification system. We used 572 well-testing field samples, including five working stages, for model training and testing. The results show that the MFE-DMIC model had the highest total stage classification accuracy of 98.18% as well as the least of features (nine) compared with the classical feature extraction and classification methods and their combinations. The proposed model can reduce the efforts of oil analysts and allow accurate labeling of samples to be predicted.

Fault diagnosis of rope tension in hoisting systems based on vibration signals

Fault diagnosis of rope tension is of great significance for safety in hoisting systems. A novel diagnosis method based on the vibration signals of the head sheaves is proposed. First, the signal is decomposed by the ensemble empirical mode decomposition (EEMD); then the main intrinsic module functions (IMFs) are extracted by correlation analysis. Second, the energy and the permutation entropy (PE) of the main IMFs were calculated to create the feature vector. Third, a particle swarm optimization - support vector machine (PSO-SVM) is applied to classify tension states. The effectiveness and advantage of the proposed method are validated by experiments. Compared with the conventional force-sensor-based method, it has clear advantages in sensor installation, data transmission, safety, and reliability.

Fault diagnosis of motor rolling bearing based on IMF sample entropy and particle swarm optimization SVM

Aiming at the problem that it is difficult to effectively identify rolling bearing faults, a method for motor rolling bearing faults based on IMF sample entropy and particle swarm optimization SVM (PSO-SVM) is proposed. Firstly, the complementary collective empirical mode decomposition (CEEMD) is applied to the adaptive decomposition of bearing vibration signals to obtain a group of Intrinsic Modal Functions (IMFs). Then the sample entropy of the IMF component containing the main fault feature information was calculated to obtain the sample entropy matrix of the signal component, which was input as the feature vector into the particle swarm optimization SVM for training and testing. Through the analysis of simulation and experimental data, the method has a high identification accuracy for fault types.

Intelligent Sliding Mode Adaptive Controller Design for Wind Turbine Pitch Control System Using PSO-SVM in Presence of Disturbance

Today, the generated electrical energy using renewable energy is rapidly growing all over the world. The speed and direction of the wind are fundamental parameters affecting on power production of wind turbine that is constantly changing. One of the methods for controlling wind turbine power is changing the angle of the blades. In this paper, an adaptive control method has been used to control the wind turbine pitch angle. The proposed method is based on sliding mode control, whose coefficients have been calculated using the particle swarm optimization–support vector machine method. In order to evaluate the effectiveness of the proposed method, it has been compared with the Model Reference Adaptive Controller (MRAC) in the operating under disturbance. The results show that the proposed controller has a better performance than that of the MRAC in the presence of disturbance. Simulations show that the produced power at different wind velocities has reached its optimum value faster.

Comparison of a physical model and phenomenological model to forecast groundwater levels in a rainfall-induced deep-seated landslide

Increases in groundwater levels play an important role in triggering deep-seated landslides. As such, it is important to be able to accurately predict variations in groundwater levels. In order to select a more suitable method for the prediction of groundwater levels in relation to deep-seated landslides, a comparative study was conducted between a physical model and phenomenological model. The physical model is a finite-element seepage code named Slide by Rocscience. The phenomenological model is a Particle Swarm Optimization Support Vector Machine (PSO-SVM). In order to obtain more accurate calculated results from the physical seepage model, the input parameters of the physical seepage model were calibrated by using a trial-error method to compare the computed results with actual monitoring data. The input data of the phenomenological model were also processed in order to obtain more accurate calculated results. The results showed that the physical seepage model was difficult to well calibrate in the condition of less data and hence performed poorly. The validation results showed that the phenomenological model performed better. The Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) of phenomenological model were 0.052 m and 0.043 m, respectively, compared with the values of physical seepage model were 0.92 m and 0.81 m, respectively. It means if constructing a satisfactory physical seepage model was difficult or understanding of physical process could not be considered, the phenomenological model might be sufficient.

Transmission Line Obstacle Detection Based on Structural Constraint and Feature Fusion

Accurate detection and identification of obstacles plays an important role in the navigation and behavior planning of the patrol robot. Aiming at the patrol robot with camera mounted symmetrically, an obstacle detection method based on structural constraint and feature fusion is proposed. Firstly, in order to discover the region of interest, the bounding box algorithm is used to propose the region. The location of the detected ground wire is used to constrain the region, and the image block of interest is clipped. Secondly, in order to accurately represent the multi-view and multi-scale obstacle images, the global shape features and the improved local corner features are fused by different weights. Then, the particle swarm-optimized support vector machine (PSO-SVM) is used for classifying and recognizing obstacles. On block data set B containing multi-view and multi-scale obstacle images, the recognition rate of this method can reach up to 86.2%, which shows the effectiveness of weighted fusion of global and local features. On data set A containing complete images of different distances, the detection success rate of long-distance obstacles can reach 80.2%. The validity of the proposed method based on structural constraints and feature fusion is verified.

A smart adaptive particle swarm optimization–support vector machine: android botnet detection application

Support vector machine (SVM) is a renowned machine learning technique, which has been successfully applied to solve many practical pattern classification problems. One of the difficulties in successful implementation of SVM is its different parameters (i.e., kernel parameter(s), penalty parameter (C) and the features available in the dataset), which should be well adjusted during the training process. In this paper, a new approach called smart adaptive particle swarm optimization–support vector machine (SAPSO–SVM) is developed to adapt the parameters of optimization algorithm (i.e., inertia weight and acceleration coefficients) to the latest changes in the search space, so that each particle explicitly explores the search space based on the latest changes made to Personal best, Global best and other particle locations. In this algorithm, using the changes in Personal best and Global best at each stage of execution, the new evolution factor values are designated and the interference of the intervals of inertia weight is eradicated. Then, the states of each particle (i.e., convergence, exploitation, exploration, jumping-out) at each stage of administration, based on the interval weights, are specified accurately. By fine tuning the parameters of SAPSO, this algorithm can acquire the best optimal responses for SVM parameters. The results obtained from the SAPSO–SVM method demonstrate the superiority of this method in four different measures (i.e., sensitivity, specificity, precision, accuracy) in comparison with the other three similar ones. Finally, the top 20 features of Android botnets are somehow introduced by the proposed approach and three other approaches; firstly, these features are not encrypted by Android botnets, and secondly, are selected based on the best results.

A Hybrid PSO-SVM Model Based on Safety Risk Prediction for the Design Process in Metro Station Construction.

Incorporating safety risk into the design process is one of the most effective design sciences to enhance the safety of metro station construction. In such a case, the concept of Design for Safety (DFS) has attracted much attention. However, most of the current research overlooks the risk-prediction process in the application of DFS. Therefore, this paper proposes a hybrid risk-prediction framework to enhance the effectiveness of DFS in practice. Firstly, 12 influencing factors related to the safety risk of metro construction are identified by adopting the literature review method and code of construction safety management analysis. Then, a structured interview is used to collect safety risk cases of metro construction projects. Next, a developed support vector machine (SVM) model based on particle swarm optimization (PSO) is presented to predict the safety risk in metro construction, in which the multi-class SVM prediction model with an improved binary tree is designed. The results show that the average accuracy of the test sets is 85.26%, and the PSO–SVM model has a high predictive accuracy for non-linear relationship and small samples. The results show that the average accuracy of the test sets is 85.26%, and the PSO–SVM model has a high predictive accuracy for non-linear relationship and small samples. Finally, the proposed framework is applied to a case study of metro station construction. The prediction results show the PSO–SVM model is applicable and reasonable for safety risk prediction. This research also identifies the most important influencing factors to reduce the safety risk of metro station construction, which provides a guideline for the safety risk prediction of metro construction for design process.

Rope Tension Fault Diagnosis in Hoisting Systems Based on Vibration Signals Using EEMD, Improved Permutation Entropy, and PSO-SVM.

Fault diagnosis of rope tension is significantly important for hoisting safety, especially in mine hoists. Conventional diagnosis methods based on force sensors face some challenges regarding sensor installation, data transmission, safety, and reliability in harsh mine environments. In this paper, a novel fault diagnosis method for rope tension based on the vibration signals of head sheaves is proposed. First, the vibration signal is decomposed into some intrinsic mode functions (IMFs) by the ensemble empirical mode decomposition (EEMD) method. Second, a sensitivity index is proposed to extract the main IMFs, then the de-noised signal is obtained by the sum of the main IMFs. Third, the energy and the proposed improved permutation entropy (IPE) values of the main IMFs and the de-noised signal are calculated to create the feature vectors. The IPE is proposed to improve the PE by adding the amplitude information, and it proved to be more sensitive in simulations of impulse detecting and signal segmentation. Fourth, vibration samples in different tension states are used to train a particle swarm optimization–support vector machine (PSO-SVM) model. Lastly, the trained model is implemented to detect tension faults in practice. Two experimental results validated the effectiveness of the proposed method to detect tension faults, such as overload, underload, and imbalance, in both single-rope and multi-rope hoists. This study provides a new perspective for detecting tension faults in hoisting systems.