Many domains, including communication, signal processing, and image processing, use the Fourier Transform as a mathematical tool for signal analysis. Although it can analyze signals with steady and transitory properti...Many domains, including communication, signal processing, and image processing, use the Fourier Transform as a mathematical tool for signal analysis. Although it can analyze signals with steady and transitory properties, it has limits. The Wavelet Packet Decomposition (WPD) is a novel technique that we suggest in this study as a way to improve the Fourier Transform and get beyond these drawbacks. In this experiment, we specifically considered the utilization of Daubechies level 4 for the wavelet transformation. The choice of Daubechies level 4 was motivated by several reasons. Daubechies wavelets are known for their compact support, orthogonality, and good time-frequency localization. By choosing Daubechies level 4, we aimed to strike a balance between preserving important transient information and avoiding excessive noise or oversmoothing in the transformed signal. Then we compared the outcomes of our suggested approach to the conventional Fourier Transform using a non-stationary signal. The findings demonstrated that the suggested method offered a more accurate representation of non-stationary and transient signals in the frequency domain. Our method precisely showed a 12% reduction in MSE and a 3% rise in PSNR for the standard Fourier transform, as well as a 35% decrease in MSE and an 8% increase in PSNR for voice signals when compared to the traditional wavelet packet decomposition method.展开更多
One of the important issues in the system identification and the spectrum analysis is the frequency resolution, i.e., the capability of distinguishing between two or more closely spaced frequency components. In the mo...One of the important issues in the system identification and the spectrum analysis is the frequency resolution, i.e., the capability of distinguishing between two or more closely spaced frequency components. In the modal identification by the empirical mode decomposition (EMD) method, because of the separating capability of the method, it is still a challenge to consistently and reliably identify the parameters of structures of which modes are not well separated. A new method is introduced to generate the intrinsic mode functions (IMFs) through the filtering algorithm based on the wavelet packet decomposition (GIFWPD). In this paper, it is demonstrated that the GIFWPD method alone has a good capability of separating close modes, even under the severe condition beyond the critical frequency ratio limit which makes it impossible to separate two closely spaced harmonics by the EMD method. However, the GIFWPD-only based method is impelled to use a very fine sampling frequency with consequent prohibitive computational costs. Therefore, in order to decrease the computational load by reducing the amount of samples and improve the effectiveness of separation by increasing the frequency ratio, the present paper uses a combination of the complex envelope displacement analysis (CEDA) and the GIFWPD method. For the validation, two examples from the previous works are taken to show the results obtained by the GIFWPD-only based method and by combining the CEDA with the GIFWPD method.展开更多
The frequency domain division theory of dyadic wavelet decomposition and wavelet packet decomposition (WPD) with orthogonal wavelet base frame are presented. The WPD coefficients of signals are treated as the outputs ...The frequency domain division theory of dyadic wavelet decomposition and wavelet packet decomposition (WPD) with orthogonal wavelet base frame are presented. The WPD coefficients of signals are treated as the outputs of equivalent bandwidth filters with different center frequency. The corresponding WPD entropy values of coefficients increase sharply when the discrete spectrum interferences (DSIs), frequency spectrum of which is centered at several frequency points existing in some frequency region. Based on WPD, an entropy threshold method (ETM) is put forward, in which entropy is used to determine whether partial discharge (PD) signals are interfered by DSIs. Simulation and real data processing demonstrate that ETM works with good efficiency, without pre-knowing DSI information. ETM extracts the phase of PD pulses accurately and can calibrate the quantity of single type discharge.展开更多
Grinding is known as the most complicated material removal process and the method for monitoring the grinding wheel wear has its own characteristics comparing with the approaches for detecting the wear on regular cutt...Grinding is known as the most complicated material removal process and the method for monitoring the grinding wheel wear has its own characteristics comparing with the approaches for detecting the wear on regular cutting tools.Research efforts were made to develop the wheel wear monitoring system due to its significance in grinding process.This paper presents a novel method for identification of grinding wheel wear signature by combination of wavelet packet decomposition(WPD) based energies.The distinctive feature of the method is that it takes advantage of the combinational information of the decomposed frequency components based on the WPD so the extracted features can be customized according to the specific monitored object to get better diagnosis effects.Experiments are researched on monitoring of grinding wheel wear states under different machining conditions.The results show that the energy ratio extracted from the measured vibration signals is consistent with the grinding wheel wear condition evaluated by experiment and the further extracted feature ratio can be used in prediction of wheel wear condition.展开更多
As an indispensable part of </span><span style="font-family:Verdana;">global</span><span style="font-family:Verdana;"> satellite navigation system, the frequency band of DME...As an indispensable part of </span><span style="font-family:Verdana;">global</span><span style="font-family:Verdana;"> satellite navigation system, the frequency band of DME will overlap with that of the navigation signal, which will cause the signal from the DME platform to be accepted by the Global Navigation Satellite System receiver and form interference. Therefore, it is of great significance to study an effective algorithm to suppress DME pulse interference. This paper has the following research on this problem. In this paper, wavelet packet transform is used to solve for the suppression of </span><span style="font-family:Verdana;">DME</span><span style="font-family:Verdana;"> pulse interference method, wavelet packet analysis belongs to the linear time-frequency analysis method, it has good time-frequency localization characteristics and the signal adaptive ability, due to the function of wavelet packet and parameter selection of DME will affect the ability of interference suppression, combining with the theory of wavelet </span><span style="font-family:Verdana;">threshold</span><span style="font-family:Verdana;">, function type and decomposition series are discussed to prove the validity of the selected parameters on the pulse interference suppression</span></span></span><span style="font-family:Verdana;"><span style="font-family:Verdana;"><span style="font-family:Verdana;">.展开更多
This paper proposes an efficient learning based approach to detect the faults of an industrial oil pump.The proposed method uses the wavelet transform and genetic algorithm(GA)ensemble for an optimal feature extractio...This paper proposes an efficient learning based approach to detect the faults of an industrial oil pump.The proposed method uses the wavelet transform and genetic algorithm(GA)ensemble for an optimal feature extraction procedure.Optimal features,which are dominated through this method,can remarkably represent the mechanical faults in the damaged machine.For the aim of condition monitoring,we considered five common types of malfunctions such as casing distortion,cavitation,looseness,misalignment,and unbalanced mass that occur during the machine operation.The proposed technique can determine optimal wavelet parameters and suitable statistical functions to exploit excellent features via an appropriate distance criterion function.Moreover,our optimization algorithm chooses the most appropriate feature submatrix to improve the final accuracy in an iterative method.As a case study,the proposed algorithms are applied to experimental data gathered from an industrial heavy-duty oil pump installed in Arak Oil Refinery Company.The experimental results are very promising.展开更多
Accurate wind speed prediction has been becoming an indispensable technology in system security,wind energy utilization,and power grid dispatching in recent years.However,it is an arduous task to predict wind speed du...Accurate wind speed prediction has been becoming an indispensable technology in system security,wind energy utilization,and power grid dispatching in recent years.However,it is an arduous task to predict wind speed due to its variable and random characteristics.For the objective to enhance the performance of forecasting short-term wind speed,this work puts forward a hybrid deep learning model mixing time series decomposition algorithm and gated recurrent unit(GRU).The time series decomposition algorithm combines the following two parts:(1)the complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN),and(2)wavelet packet decomposition(WPD).Firstly,the normalized wind speed time series(WSTS)are handled by CEEMDAN to gain pure fixed-frequency components and a residual signal.The WPD algorithm conducts the second-order decomposition to the first component that contains complex and high frequency signal of raw WSTS.Finally,GRU networks are established for all the relevant components of the signals,and the predicted wind speeds are obtained by superimposing the prediction of each component.Results from two case studies,adopting wind data from laboratory and wind farm,respectively,suggest that the related trend of the WSTS can be separated effectively by the proposed time series decomposition algorithm,and the accuracy of short-time wind speed prediction can be heightened significantly mixing the time series decomposition algorithm and GRU networks.展开更多
This paper proposes a hybrid deep reinforcement learning framework for locomotive axle temperature by combining the wavelet packet decomposition(WPD),long short-term memory(LSTM),gated recurrent unit(GRU)reinforcement...This paper proposes a hybrid deep reinforcement learning framework for locomotive axle temperature by combining the wavelet packet decomposition(WPD),long short-term memory(LSTM),gated recurrent unit(GRU)reinforcement learning and generalized autoregressive conditional heteroskedasticity(GARCH)algorithms.The WPD is utilized to decompose the raw nonlinear series into subseries.Then the deep learning predictors LSTM and GRU are established to predict the future axle temperatures in each subseries.The Q-learning could generate optimal ensembleweights to integrate the predictors to finish the deterministic forecasting and GARCH is used to conduct the deterministic forecasting based on the deterministic forecasting residual.These parts of the hybrid ensemble structure contributed to optimal modelling accuracy and provided effective support in the real-time monitoring and fault diagnosis of transportation.展开更多
Currently, due to the detrimental effects on surface finish and machining system, chatter has been one crucial factor restricting robotic drilling operations, which improve both quality and efficiency of aviation manu...Currently, due to the detrimental effects on surface finish and machining system, chatter has been one crucial factor restricting robotic drilling operations, which improve both quality and efficiency of aviation manufacturing. Based on the matrix notch filter and fast wavelet packet decomposition, this paper presents a novel pre-generated matrix-based real-time chatter monitoring method for robotic drilling. Taking vibration characteristics of robotic drilling into account, the matrix notch filter is designed to eliminate the interference of spindle-related components on the measured vibration signal. Then, the fast wavelet packet decomposition is presented to decompose the filtered signal into several equidistant frequency bands, and the energy of each sub-band is obtained. Finally, the energy entropy which characterizes inhomogeneity of energy distribution is utilized as the feature to recognize chatter on-line, and the effectiveness of the presented algorithm is validated by extensive experimental data. The results show that the proposed algorithm can effectively detect chatter before it is fully developed. Moreover, since both filtering and decomposition of signal are implemented by the pre-generated matrices, calculation for an energy entropy of vibration signal with 512 samples takes only about 0.690 ms. Consequently, the proposed method achieves real-time chatter monitoring for robotic drilling, which is essential for subsequent chatter suppression.展开更多
An approach to smoothing the fluctuations of largescale wind power is investigated using vehicle-to-grid(V2G)systems.First,an energy management and optimization system is designed and modeled.By using the wavelet pack...An approach to smoothing the fluctuations of largescale wind power is investigated using vehicle-to-grid(V2G)systems.First,an energy management and optimization system is designed and modeled.By using the wavelet packet decomposition method,the target grid-connected wind power,the required electric vehicle(EV)power,and supercapacitor power are determined.The energy management model for EVs is then developed by introducing a knapsack problem that can evaluate the needs of an EV fleet.Furthermore,an optimized dispatch strategy for EVs and wind power is developed by using a dynamic programming method.A case study demonstrates that the energy management and optimization method for V2G systems achieves noticeable performance improvements over benchmark techniques.展开更多
Machine chatter is still an unresolved and challenging issue in the milling process,and developing an online chatter identification and process monitoring system towards smart manufacturing is an urgent requirement.In...Machine chatter is still an unresolved and challenging issue in the milling process,and developing an online chatter identification and process monitoring system towards smart manufacturing is an urgent requirement.In this paper,two indicators of chatter detection are investigated.One is the real-time variance of milling force signals in the time domain,and the other one is the wavelet energy ratio of acceleration signals based on wavelet packet decomposition in the frequency domain.Then,a novel classification concept for vibration condition,called slight chatter,is proposed and integrated successfully into the designed multi-classification support vector machine(SVM)model.Finally,a mapping model between image and chatter indicators is established via a distance threshold on the image.The multi-SVM model is trained by the results of three signals as an input.Experiment data and detection accuracy of the SVM model are verified in actual machining.The identification accuracy of 96.66%has proved that the proposed solution is feasible and effective.The presented method can be used to select optimized milling parameters to improve machining process stability and strengthen manufacturing system monitoring.展开更多
Low energy impact can induce invisible damage of carbon fiber reinforced polymer(CFRP).The damage can seriously affect the safety of the CFRP structure.Therefore,damage detection is crucial to the CFRP structure.Impac...Low energy impact can induce invisible damage of carbon fiber reinforced polymer(CFRP).The damage can seriously affect the safety of the CFRP structure.Therefore,damage detection is crucial to the CFRP structure.Impact location information is the premise of damage detection.Hence,impact localization is the primary issue.In this paper,an impact localization system,based on the fiber Bragg grating(FBG)sensor network,is proposed for impact detection and localization.For the completed impact signal,the FBG sensor and narrow-band laser demodulation technology are applied.Wavelet packet decomposition is introduced to extract available frequency band signals and attenuate noise.According to the energy of the available frequency band signal,an impact localization model,based on the extreme learning machine(ELM),is established with the faster training speed and less parameters.The above system is verified on the 500 mm×500 mm×2 mm CFRP plate.The maximum localization error and the minimum localization error are 30.4 mm and 6.7 mm,respectively.The average localization error is 14.7 mm,and training time is 0.7 s.Compared with the other machine learning methods,the localization system,proposed in this paper,has higher accuracy and faster training speed.This paper provides a practical system for impact localization of the CFRP structure.展开更多
Wind energy is the burgeoning renewable energy. Accurate wind speed prediction is necessary to ensure the stability and reliability of the power grid for wind energy. This study focuses on developing a novel hybrid fo...Wind energy is the burgeoning renewable energy. Accurate wind speed prediction is necessary to ensure the stability and reliability of the power grid for wind energy. This study focuses on developing a novel hybrid forecasting model to tackle adverse effects caused by strong variability and abrupt changes in wind speed. The hybrid model combines data decomposition and error correction strategy for a wind speed forecasting model applied to wind energy. First, wavelet packet decomposition is applied to wind speed series to obtain stationary subseries. Next, outlier robust extreme learning machine is implemented to predict subseries. Finally, an error correction strategy coupled with data decomposition is designed to repair preliminary prediction results. In addition, four measured datasets from China and USAwind farms with different time intervals are used to evaluate the performance of the proposed approach. Experimental analysis indicates that the proposed model outperforms the compared models. Results show that(1) the prediction accuracy of the proposed model is remarkably improved compared with other conventional models;(2) the proposed model can reduce the influence of the end effect in the decomposition-based forecasting model;(3) the coupling framework is successful for enhancing performance of hybrid forecasting model.展开更多
Aiming at the sensor faults of near-space hypersonic vehicles(NSHV), a fault identification method based on the extended state observer and kernel extreme learning machine(ESO-KELM) is proposed in this paper. The meth...Aiming at the sensor faults of near-space hypersonic vehicles(NSHV), a fault identification method based on the extended state observer and kernel extreme learning machine(ESO-KELM) is proposed in this paper. The method is generated by a combination of the model-based method and the data-driven method. As the source of the fault diagnosis, the residual signals represent the difference between the ESO output and the result measured by the sensor in particular. The energy of the residual signals is distributed in both low frequency bands and high frequency bands. However, the energy of the sensor concentrates on the low-frequency bands. Combined with more different features detected by KELM, the proposed method devotes to improving the accuracy. Meanwhile, it is competent to calculate the magnitude of minor faults based on time-frequency analysis. Finally, the simulation is performed on the longitudinal channel of the Winged-Cone model published by the national aeronautics and space administration(NASA). Results show the validity and the accuracy in calculating the magnitude of the minor faults.展开更多
This paper proposes a data-driven hybrid model for forecasting the battery power voltage of transportation robots by combining a wavelet method and a bidirectional deep-learning technique.In the proposed model,the on-...This paper proposes a data-driven hybrid model for forecasting the battery power voltage of transportation robots by combining a wavelet method and a bidirectional deep-learning technique.In the proposed model,the on-board battery power data is measured and transmitted.A WPD(wavelet packet decomposition)algorithm is employed to decompose the original collected non-stationary series into several relatively more stable subseries.For each subseries,a deep learning–based predictor–bidirectional long short-term memory(BiLSTM)–is constructed to forecast the battery power voltage from one step to three steps ahead.Two experiments verify the effectiveness and generalization ability of the proposed hybrid forecasting model,which shows the highest forecasting accuracy.The obtained forecasting results can be used to decide whether the robot can complete the given task or needs to be recharged,providing effective support for the safe use of transportation robots.展开更多
文摘Many domains, including communication, signal processing, and image processing, use the Fourier Transform as a mathematical tool for signal analysis. Although it can analyze signals with steady and transitory properties, it has limits. The Wavelet Packet Decomposition (WPD) is a novel technique that we suggest in this study as a way to improve the Fourier Transform and get beyond these drawbacks. In this experiment, we specifically considered the utilization of Daubechies level 4 for the wavelet transformation. The choice of Daubechies level 4 was motivated by several reasons. Daubechies wavelets are known for their compact support, orthogonality, and good time-frequency localization. By choosing Daubechies level 4, we aimed to strike a balance between preserving important transient information and avoiding excessive noise or oversmoothing in the transformed signal. Then we compared the outcomes of our suggested approach to the conventional Fourier Transform using a non-stationary signal. The findings demonstrated that the suggested method offered a more accurate representation of non-stationary and transient signals in the frequency domain. Our method precisely showed a 12% reduction in MSE and a 3% rise in PSNR for the standard Fourier transform, as well as a 35% decrease in MSE and an 8% increase in PSNR for voice signals when compared to the traditional wavelet packet decomposition method.
基金supported by the State Key Program of National Natural Science of China (No. 11232009)the Shanghai Leading Academic Discipline Project (No. S30106)
文摘One of the important issues in the system identification and the spectrum analysis is the frequency resolution, i.e., the capability of distinguishing between two or more closely spaced frequency components. In the modal identification by the empirical mode decomposition (EMD) method, because of the separating capability of the method, it is still a challenge to consistently and reliably identify the parameters of structures of which modes are not well separated. A new method is introduced to generate the intrinsic mode functions (IMFs) through the filtering algorithm based on the wavelet packet decomposition (GIFWPD). In this paper, it is demonstrated that the GIFWPD method alone has a good capability of separating close modes, even under the severe condition beyond the critical frequency ratio limit which makes it impossible to separate two closely spaced harmonics by the EMD method. However, the GIFWPD-only based method is impelled to use a very fine sampling frequency with consequent prohibitive computational costs. Therefore, in order to decrease the computational load by reducing the amount of samples and improve the effectiveness of separation by increasing the frequency ratio, the present paper uses a combination of the complex envelope displacement analysis (CEDA) and the GIFWPD method. For the validation, two examples from the previous works are taken to show the results obtained by the GIFWPD-only based method and by combining the CEDA with the GIFWPD method.
基金Funded by the of the Key Teachers Foundation under the State Ministry Education.
文摘The frequency domain division theory of dyadic wavelet decomposition and wavelet packet decomposition (WPD) with orthogonal wavelet base frame are presented. The WPD coefficients of signals are treated as the outputs of equivalent bandwidth filters with different center frequency. The corresponding WPD entropy values of coefficients increase sharply when the discrete spectrum interferences (DSIs), frequency spectrum of which is centered at several frequency points existing in some frequency region. Based on WPD, an entropy threshold method (ETM) is put forward, in which entropy is used to determine whether partial discharge (PD) signals are interfered by DSIs. Simulation and real data processing demonstrate that ETM works with good efficiency, without pre-knowing DSI information. ETM extracts the phase of PD pulses accurately and can calibrate the quantity of single type discharge.
基金the National Key Laboratory of Mechanical Transmission Foundation of China(No. SKLMT-KFKT-200812)
文摘Grinding is known as the most complicated material removal process and the method for monitoring the grinding wheel wear has its own characteristics comparing with the approaches for detecting the wear on regular cutting tools.Research efforts were made to develop the wheel wear monitoring system due to its significance in grinding process.This paper presents a novel method for identification of grinding wheel wear signature by combination of wavelet packet decomposition(WPD) based energies.The distinctive feature of the method is that it takes advantage of the combinational information of the decomposed frequency components based on the WPD so the extracted features can be customized according to the specific monitored object to get better diagnosis effects.Experiments are researched on monitoring of grinding wheel wear states under different machining conditions.The results show that the energy ratio extracted from the measured vibration signals is consistent with the grinding wheel wear condition evaluated by experiment and the further extracted feature ratio can be used in prediction of wheel wear condition.
文摘As an indispensable part of </span><span style="font-family:Verdana;">global</span><span style="font-family:Verdana;"> satellite navigation system, the frequency band of DME will overlap with that of the navigation signal, which will cause the signal from the DME platform to be accepted by the Global Navigation Satellite System receiver and form interference. Therefore, it is of great significance to study an effective algorithm to suppress DME pulse interference. This paper has the following research on this problem. In this paper, wavelet packet transform is used to solve for the suppression of </span><span style="font-family:Verdana;">DME</span><span style="font-family:Verdana;"> pulse interference method, wavelet packet analysis belongs to the linear time-frequency analysis method, it has good time-frequency localization characteristics and the signal adaptive ability, due to the function of wavelet packet and parameter selection of DME will affect the ability of interference suppression, combining with the theory of wavelet </span><span style="font-family:Verdana;">threshold</span><span style="font-family:Verdana;">, function type and decomposition series are discussed to prove the validity of the selected parameters on the pulse interference suppression</span></span></span><span style="font-family:Verdana;"><span style="font-family:Verdana;"><span style="font-family:Verdana;">.
文摘This paper proposes an efficient learning based approach to detect the faults of an industrial oil pump.The proposed method uses the wavelet transform and genetic algorithm(GA)ensemble for an optimal feature extraction procedure.Optimal features,which are dominated through this method,can remarkably represent the mechanical faults in the damaged machine.For the aim of condition monitoring,we considered five common types of malfunctions such as casing distortion,cavitation,looseness,misalignment,and unbalanced mass that occur during the machine operation.The proposed technique can determine optimal wavelet parameters and suitable statistical functions to exploit excellent features via an appropriate distance criterion function.Moreover,our optimization algorithm chooses the most appropriate feature submatrix to improve the final accuracy in an iterative method.As a case study,the proposed algorithms are applied to experimental data gathered from an industrial heavy-duty oil pump installed in Arak Oil Refinery Company.The experimental results are very promising.
基金This work was supported in part by the National Key Research and Development Project of China(No.2019YFE0105300)the National Natural Science Foundation of China(No.61972443)the Hunan Provincial Key Research and Development Project of China(No.2022WK2006).
文摘Accurate wind speed prediction has been becoming an indispensable technology in system security,wind energy utilization,and power grid dispatching in recent years.However,it is an arduous task to predict wind speed due to its variable and random characteristics.For the objective to enhance the performance of forecasting short-term wind speed,this work puts forward a hybrid deep learning model mixing time series decomposition algorithm and gated recurrent unit(GRU).The time series decomposition algorithm combines the following two parts:(1)the complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN),and(2)wavelet packet decomposition(WPD).Firstly,the normalized wind speed time series(WSTS)are handled by CEEMDAN to gain pure fixed-frequency components and a residual signal.The WPD algorithm conducts the second-order decomposition to the first component that contains complex and high frequency signal of raw WSTS.Finally,GRU networks are established for all the relevant components of the signals,and the predicted wind speeds are obtained by superimposing the prediction of each component.Results from two case studies,adopting wind data from laboratory and wind farm,respectively,suggest that the related trend of the WSTS can be separated effectively by the proposed time series decomposition algorithm,and the accuracy of short-time wind speed prediction can be heightened significantly mixing the time series decomposition algorithm and GRU networks.
基金This study is fully supported by the National Natural Science Foundation of China(Grant No.61873283)the Changsha Sci-ence&Technology Project(Grant No.KQ1707017)the Hunan Province Science and Technology Talent Support Project(Grant No.2020TJ-Q06).
文摘This paper proposes a hybrid deep reinforcement learning framework for locomotive axle temperature by combining the wavelet packet decomposition(WPD),long short-term memory(LSTM),gated recurrent unit(GRU)reinforcement learning and generalized autoregressive conditional heteroskedasticity(GARCH)algorithms.The WPD is utilized to decompose the raw nonlinear series into subseries.Then the deep learning predictors LSTM and GRU are established to predict the future axle temperatures in each subseries.The Q-learning could generate optimal ensembleweights to integrate the predictors to finish the deterministic forecasting and GARCH is used to conduct the deterministic forecasting based on the deterministic forecasting residual.These parts of the hybrid ensemble structure contributed to optimal modelling accuracy and provided effective support in the real-time monitoring and fault diagnosis of transportation.
基金supported by the National Key R&D Program of China (No. 2017YFB1302601 and 2018YFB1702503)
文摘Currently, due to the detrimental effects on surface finish and machining system, chatter has been one crucial factor restricting robotic drilling operations, which improve both quality and efficiency of aviation manufacturing. Based on the matrix notch filter and fast wavelet packet decomposition, this paper presents a novel pre-generated matrix-based real-time chatter monitoring method for robotic drilling. Taking vibration characteristics of robotic drilling into account, the matrix notch filter is designed to eliminate the interference of spindle-related components on the measured vibration signal. Then, the fast wavelet packet decomposition is presented to decompose the filtered signal into several equidistant frequency bands, and the energy of each sub-band is obtained. Finally, the energy entropy which characterizes inhomogeneity of energy distribution is utilized as the feature to recognize chatter on-line, and the effectiveness of the presented algorithm is validated by extensive experimental data. The results show that the proposed algorithm can effectively detect chatter before it is fully developed. Moreover, since both filtering and decomposition of signal are implemented by the pre-generated matrices, calculation for an energy entropy of vibration signal with 512 samples takes only about 0.690 ms. Consequently, the proposed method achieves real-time chatter monitoring for robotic drilling, which is essential for subsequent chatter suppression.
文摘An approach to smoothing the fluctuations of largescale wind power is investigated using vehicle-to-grid(V2G)systems.First,an energy management and optimization system is designed and modeled.By using the wavelet packet decomposition method,the target grid-connected wind power,the required electric vehicle(EV)power,and supercapacitor power are determined.The energy management model for EVs is then developed by introducing a knapsack problem that can evaluate the needs of an EV fleet.Furthermore,an optimized dispatch strategy for EVs and wind power is developed by using a dynamic programming method.A case study demonstrates that the energy management and optimization method for V2G systems achieves noticeable performance improvements over benchmark techniques.
基金support from the National Key R&D Program of China(Grant No.2017YFEO101400)also appreciate reviewers for their critical comments.
文摘Machine chatter is still an unresolved and challenging issue in the milling process,and developing an online chatter identification and process monitoring system towards smart manufacturing is an urgent requirement.In this paper,two indicators of chatter detection are investigated.One is the real-time variance of milling force signals in the time domain,and the other one is the wavelet energy ratio of acceleration signals based on wavelet packet decomposition in the frequency domain.Then,a novel classification concept for vibration condition,called slight chatter,is proposed and integrated successfully into the designed multi-classification support vector machine(SVM)model.Finally,a mapping model between image and chatter indicators is established via a distance threshold on the image.The multi-SVM model is trained by the results of three signals as an input.Experiment data and detection accuracy of the SVM model are verified in actual machining.The identification accuracy of 96.66%has proved that the proposed solution is feasible and effective.The presented method can be used to select optimized milling parameters to improve machining process stability and strengthen manufacturing system monitoring.
基金This research is supported by the National Natural Science Foundation of China under Grant Nos.61503218 and 61705098Natural Science Foundation of Shandong Province,China under Grant Nos.ZR2017BF042,ZR2016FM36,and ZR2017BF032Ph.D.Programs Foundation of Ludong University under Grant No.LA2017006.
文摘Low energy impact can induce invisible damage of carbon fiber reinforced polymer(CFRP).The damage can seriously affect the safety of the CFRP structure.Therefore,damage detection is crucial to the CFRP structure.Impact location information is the premise of damage detection.Hence,impact localization is the primary issue.In this paper,an impact localization system,based on the fiber Bragg grating(FBG)sensor network,is proposed for impact detection and localization.For the completed impact signal,the FBG sensor and narrow-band laser demodulation technology are applied.Wavelet packet decomposition is introduced to extract available frequency band signals and attenuate noise.According to the energy of the available frequency band signal,an impact localization model,based on the extreme learning machine(ELM),is established with the faster training speed and less parameters.The above system is verified on the 500 mm×500 mm×2 mm CFRP plate.The maximum localization error and the minimum localization error are 30.4 mm and 6.7 mm,respectively.The average localization error is 14.7 mm,and training time is 0.7 s.Compared with the other machine learning methods,the localization system,proposed in this paper,has higher accuracy and faster training speed.This paper provides a practical system for impact localization of the CFRP structure.
基金supported by the National Natural Science Foundation of China (Grant Nos. 11732010, 12072185, 11972220, 11825204, and 92052201)
文摘Wind energy is the burgeoning renewable energy. Accurate wind speed prediction is necessary to ensure the stability and reliability of the power grid for wind energy. This study focuses on developing a novel hybrid forecasting model to tackle adverse effects caused by strong variability and abrupt changes in wind speed. The hybrid model combines data decomposition and error correction strategy for a wind speed forecasting model applied to wind energy. First, wavelet packet decomposition is applied to wind speed series to obtain stationary subseries. Next, outlier robust extreme learning machine is implemented to predict subseries. Finally, an error correction strategy coupled with data decomposition is designed to repair preliminary prediction results. In addition, four measured datasets from China and USAwind farms with different time intervals are used to evaluate the performance of the proposed approach. Experimental analysis indicates that the proposed model outperforms the compared models. Results show that(1) the prediction accuracy of the proposed model is remarkably improved compared with other conventional models;(2) the proposed model can reduce the influence of the end effect in the decomposition-based forecasting model;(3) the coupling framework is successful for enhancing performance of hybrid forecasting model.
基金supported by the National Natural Science Foundation of China(62073020)。
文摘Aiming at the sensor faults of near-space hypersonic vehicles(NSHV), a fault identification method based on the extended state observer and kernel extreme learning machine(ESO-KELM) is proposed in this paper. The method is generated by a combination of the model-based method and the data-driven method. As the source of the fault diagnosis, the residual signals represent the difference between the ESO output and the result measured by the sensor in particular. The energy of the residual signals is distributed in both low frequency bands and high frequency bands. However, the energy of the sensor concentrates on the low-frequency bands. Combined with more different features detected by KELM, the proposed method devotes to improving the accuracy. Meanwhile, it is competent to calculate the magnitude of minor faults based on time-frequency analysis. Finally, the simulation is performed on the longitudinal channel of the Winged-Cone model published by the national aeronautics and space administration(NASA). Results show the validity and the accuracy in calculating the magnitude of the minor faults.
基金funded by the German Federal Ministry of Education and Research Germany(FKZ:03Z1KN11,03Z1KI1)supported by the National Natural Science Foundation of China(Grant No.61873283)+3 种基金the Changsha Science&Technology Project(Grant No.KQ1707017)the Shenghua Yu-ying Talents Programme of Central South Universitythe Innovation-Driven Project of Central South Universitythe Wasion Group Limited.
文摘This paper proposes a data-driven hybrid model for forecasting the battery power voltage of transportation robots by combining a wavelet method and a bidirectional deep-learning technique.In the proposed model,the on-board battery power data is measured and transmitted.A WPD(wavelet packet decomposition)algorithm is employed to decompose the original collected non-stationary series into several relatively more stable subseries.For each subseries,a deep learning–based predictor–bidirectional long short-term memory(BiLSTM)–is constructed to forecast the battery power voltage from one step to three steps ahead.Two experiments verify the effectiveness and generalization ability of the proposed hybrid forecasting model,which shows the highest forecasting accuracy.The obtained forecasting results can be used to decide whether the robot can complete the given task or needs to be recharged,providing effective support for the safe use of transportation robots.
