The current research of master cylinder pressure estimation mainly relies on hydraulic characteristic or vehicle dynamics.But they are not independently applicable to any environment and have their own scope of applic...The current research of master cylinder pressure estimation mainly relies on hydraulic characteristic or vehicle dynamics.But they are not independently applicable to any environment and have their own scope of application.In addition,about the master cylinder pressure control,there are few studies that can simultaneously balance pressure building accuracy,speed,and prevent pressure overshoot and jitter.In this paper,an adaptative fusion method based on electro-hydraulic characteristic and vehicle mode is proposed to estimate the master cylinder pressure.The fusion strategy is mainly based on the prediction performance of two algorithms under different vehicle speeds,pressures,and ABS states.Apart from this,this article also includes real-time prediction of the friction model based on RLS to improve the accuracy of the electro-hydraulic mode.In order to simultaneously balance pressure control accuracy,response speed,and prevent overshoot and jitter,this article proposes an adaptative LQR controller for MC pressure control which uses fuzzy-logic controller to adjust the weights of LQR controller based on target pressure and difference compared with actual pressure.Through mode-in-loop and hardware-in-loop tests in ramp,step and sinusoidal response,the whole estimation and control system is verified based on real hydraulic system and the performance is satisfactory under these scenes.This research proposes an adaptative pressure estimation and control architecture for integrated electro-hydraulic brake system which could eliminate pressure sensors in typical scenarios and ensure the comprehensive performance of pressure control.展开更多
In transportation cyber-physical-systems （T-CPS）, vehicle-to-vehicle （V2V） communications play an important role in the coordination between individual vehicles as well as between vehicles and the roadside infrast...In transportation cyber-physical-systems （T-CPS）, vehicle-to-vehicle （V2V） communications play an important role in the coordination between individual vehicles as well as between vehicles and the roadside infrastructures, and engine cylinder pressure is significant for engine diagnosis on-line and torque control within the information exchange process under V2V communications. However, the parametric uncertainties caused from measurement noise in T-CPS lead to the dynamic performance deterioration of the engine cylinder pressure estimation. Considering the high accuracy requirement under V2V communications, a high gain observer based on the engine dynamic model is designed to improve the accuracy of pressure estimation. Then, the analyses about convergence, converge speed and stability of the corresponding error model are conducted using the Laplace and Lyapunov method. Finally, results from combination of Simulink with GT- Power based numerical experiments and comparisons demonstrate the effectiveness of the proposed approach with respect to robustness and accuracy.展开更多
Portal hypertension(PH)is a clinical syndrome,characterized by elevated pressure gradient between portal vein and inferior vena cava.These elevated pressures gradient due to increased vascular resistance and/or increa...Portal hypertension(PH)is a clinical syndrome,characterized by elevated pressure gradient between portal vein and inferior vena cava.These elevated pressures gradient due to increased vascular resistance and/or increased volume of blood flowing through the portal vein circulation,results in blood outflow difficulty from portal vein to hepatic veins and inferior vena cava.展开更多
Pressure measurement within the body is of pivotal significance in the diagnosis of vascular and organ-related diseases associated with hydrostatic pressure.At present,the most commonly used clinical method is to inse...Pressure measurement within the body is of pivotal significance in the diagnosis of vascular and organ-related diseases associated with hydrostatic pressure.At present,the most commonly used clinical method is to insert a catheter along with a pressure sensor and then guide it to the area of interest through vessels such as central venous pressure(CVP).However,the presence of sensors within the vessel of interest will inevitably cause alterations to the circulation and thus affect blood pressure.Moreover,the use of invasive methods does not allow monitoring of every area inside the body.展开更多
Blood pressure(BP)is an important indicator of individuals’health conditions for the prevention or treatment of cardiovascular disease.However,conventional measurements require inconvenient cuffbased instruments and ...Blood pressure(BP)is an important indicator of individuals’health conditions for the prevention or treatment of cardiovascular disease.However,conventional measurements require inconvenient cuffbased instruments and are not able to detect continuous blood pressure.Advanced methods utilize machine learning to estimate BP by constructing artificial features in plethysmography(PPG)or using an end-to-end deep learning framework to estimate BP directly.Empirical features are limited by current research on cardiovascular disease and are not sufficient to express BP variability,while data-driven approaches neglect expert knowledge and lack interpretability.To address this issue,in this paper we propose a method for continuous BP estimation that extracts both artificial and data-driven features from PPG to take advantage of expert knowledge and deep learning at the same time.Then a deep residual neural network is designed to reduce information redundancy in the gathered features and refine high-level features for BP estimation.The results show that our proposed methods outperforms the compared methods in three commonly used metrics.展开更多
Background Transthoracic Doppler echocardiography is recommended for screening the presence of pulmonary hypertension(PH).However,the accuracy of pulmonary artery systolic pressure(PASP)estimated by Doppler echocardio...Background Transthoracic Doppler echocardiography is recommended for screening the presence of pulmonary hypertension(PH).However,the accuracy of pulmonary artery systolic pressure(PASP)estimated by Doppler echocardiographic is still unknown.Methods We conducted a retrospective study on 102 patients with idiopathic pulmonary arterial hypertension who underwent Doppler echocar-diography within 72 hours before right heart catheterization.During this time,all patients were stable without any specific drug therapy.Results There was moderate correlation between Doppler echocardiographic and right heart catheteriza-tion measurements of PASP(r=0.642,P【0.001).Using Bland-Altman analytic methods,the bias for the echocardio-graphic estimates of PASP was 6.65 mm Hg with 95%limits of agreement ranging from-47.62 to 34.30 mm Hg.There were 58.8%cases with absolute differences over 10 mm Hg between the two methods.Overestimation and underestimation of PASP by Doppler echocardiography occurred in 15.7%(16/102)and 43.1%(44),respectively.The magnitude of pressure underestimation and overestimation was insignificant(24.52±12.15 vs.25.69±16.09,P=0.765),while the corresponding diagnostic categories of severity that each subject would fall into for each technique are not in good agreement.The diagnostic categories of 16 overestimated patients were in accordance.During 44 underestimated patients,20.5%of patients had their pressure underestimated within one diagnostic category(minor error);4.5%of the underestimates were with two diagnostic categories(major error).Conclusions Transthoracic Doppler echocardiography may frequently be inaccurate in estimating PASP and could not replace the right heart catheterization.展开更多
Insight into average oil pressure in gas reservoirs and changes in production （time）, play a critical role in reservoir and production performance, economic evaluation and reservoir management. In all practicality, ...Insight into average oil pressure in gas reservoirs and changes in production （time）, play a critical role in reservoir and production performance, economic evaluation and reservoir management. In all practicality, average reservoir pressure can be conducted only when producing wells are shut in. This is regarded as a pressure build-up test. During the test, the wellbore pressure is recorded as a function of time. Currently, the only available method with which to obtain average reservoir pressure is to conduct an extended build-up test. It must then be evaluated using Homer or MDH （Miller, Dyes and Huchinson） valuation procedures. During production, average reservoir pressure declines due to fluid withdrawal from the wells and therefore, the average reservoirpressure is updated, periodically. A significant economic loss occurs during the entire pressure build-up test when producing wells are shut in. In this study, a neural network model has been established to map a nonlinear time-varying relationship which controls reservoir production history in order to predict and interpolate average reservoir pressure without closing the producing wells. This technique is suitable for constant and variable flow rates.展开更多
In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucia...In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucial for the vehicle design and safe control.However,precise state estimation of the brake pressure is desired to perform safe driving with a high degree of autonomy.In this paper,a sensorless state estimation technique of the vehicle's brake pressure is developed using a deep-learning approach.A deep neural network(DNN)is structured and trained using deep-learning training techniques,such as,dropout and rectified units.These techniques are utilized to obtain more accurate model for brake pressure state estimation applications.The proposed model is trained using real experimental training data which were collected via conducting real vehicle testing.The vehicle was attached to a chassis dynamometer while the brake pressure data were collected under random driving cycles.Based on these experimental data,the DNN is trained and the performance of the proposed state estimation approach is validated accordingly.The results demonstrate high-accuracy brake pressure state estimation with RMSE of 0.048 MPa.展开更多
Continuous and accurate blood pressure(BP)monitoring is critical for personalized hypertension management.However,most existing methods focus on absolute BP estimation,with limited attention to BP changes.To address t...Continuous and accurate blood pressure(BP)monitoring is critical for personalized hypertension management.However,most existing methods focus on absolute BP estimation,with limited attention to BP changes.To address this limitation,we propose a novel framework named Multi-Perspective Differential Feature Space(MDFSBP)for cuffless BP estimation using photoplethysmography(PPG)signals.MDFSBP extracts three perspective differential features:time-based and points-of-interest features,frequency-domain features,and physiological statistical features.Then,an adaptive Multi-Perspective Differential Feature Mapping Module(MDFMM)integrates reconstruction regularization,trend-aware classification,and self-weighted contrastive learning to enhance feature representation and strengthen the association between features and BP changes.Finally,an AutoML-based regression pipeline automates model optimization,improving predictive accuracy and deployment efficiency.To better test the model's capability in capturing BP changes,we introduce a novel abnormality-aware classification metric.We demonstrate BP estimation performance over state-of-the-art(SOTA)methods on both the Mindray and MIMIC datasets.On the Mindray dataset,the model achieves a regression error of 0.17±5.17 mmHg for SBP and 0.05±3.29 mmHg for DBP,with classification accuracy and F1-score reaching 85.25%and 87.50%,respectively.On the MIMIC dataset,it achieves−0.09±5.70 mmHg for SBP and 0.12±4.27 mmHg for DBP,with the classification accuracy and F1-score of 72.84%and 70.66%,respectively.These results highlight the effectiveness,robustness,and generalizability of the proposed frame-work for non-invasive,real-time,and continuous BP monitoring in both clinical and wearable healthcare systems.展开更多
The first automatic algorithm was designed to estimate the pulse pressure variation (PPVPPV) from arterial blood pressure (ABP) signals under spontaneous breathing conditions. While currently there are a few publicly ...The first automatic algorithm was designed to estimate the pulse pressure variation (PPVPPV) from arterial blood pressure (ABP) signals under spontaneous breathing conditions. While currently there are a few publicly available algorithms to automatically estimate PPVPPV accurately and reliably in mechani-cally ventilated subjects, at the moment there is no automatic algorithm for estimating PPVPPV on sponta-neously breathing subjects. The algorithm utilizes our recently developed sequential Monte Carlo method (SMCM), which is called a maximum a-posteriori adaptive marginalized particle filter (MAM-PF). The performance assessment results of the proposed algorithm on real ABP signals from spontaneously breath-ing subjects were reported.展开更多
基金Supported by National Natural Science Foundation of China(Grant Nos.52202494,52202495)Chongqing Special Project for Technological Innovation and Application Development(Grant No.CSTB2022TIAD-DEX0014).
文摘The current research of master cylinder pressure estimation mainly relies on hydraulic characteristic or vehicle dynamics.But they are not independently applicable to any environment and have their own scope of application.In addition,about the master cylinder pressure control,there are few studies that can simultaneously balance pressure building accuracy,speed,and prevent pressure overshoot and jitter.In this paper,an adaptative fusion method based on electro-hydraulic characteristic and vehicle mode is proposed to estimate the master cylinder pressure.The fusion strategy is mainly based on the prediction performance of two algorithms under different vehicle speeds,pressures,and ABS states.Apart from this,this article also includes real-time prediction of the friction model based on RLS to improve the accuracy of the electro-hydraulic mode.In order to simultaneously balance pressure control accuracy,response speed,and prevent overshoot and jitter,this article proposes an adaptative LQR controller for MC pressure control which uses fuzzy-logic controller to adjust the weights of LQR controller based on target pressure and difference compared with actual pressure.Through mode-in-loop and hardware-in-loop tests in ramp,step and sinusoidal response,the whole estimation and control system is verified based on real hydraulic system and the performance is satisfactory under these scenes.This research proposes an adaptative pressure estimation and control architecture for integrated electro-hydraulic brake system which could eliminate pressure sensors in typical scenarios and ensure the comprehensive performance of pressure control.
基金supported by the National Natural Science Foundation of China(Grant No.61304197)the Scientific and Technological Talents of Chongqing,China(Grant No.cstc2014kjrc-qnrc30002)+2 种基金the Key Project of Application and Development of Chongqing,China(Grant No.cstc2014yykf B40001)the Natural Science Funds of Chongqing,China(Grant No.cstc2014jcyj A60003)the Doctoral Start-up Funds of Chongqing University of Posts and Telecommunications,China(Grant No.A2012-26)
文摘In transportation cyber-physical-systems （T-CPS）, vehicle-to-vehicle （V2V） communications play an important role in the coordination between individual vehicles as well as between vehicles and the roadside infrastructures, and engine cylinder pressure is significant for engine diagnosis on-line and torque control within the information exchange process under V2V communications. However, the parametric uncertainties caused from measurement noise in T-CPS lead to the dynamic performance deterioration of the engine cylinder pressure estimation. Considering the high accuracy requirement under V2V communications, a high gain observer based on the engine dynamic model is designed to improve the accuracy of pressure estimation. Then, the analyses about convergence, converge speed and stability of the corresponding error model are conducted using the Laplace and Lyapunov method. Finally, results from combination of Simulink with GT- Power based numerical experiments and comparisons demonstrate the effectiveness of the proposed approach with respect to robustness and accuracy.
基金Sino-German Mobility Programme of NSFC and DFG,Grant/Award Number:M-0504National Natural Science Foundation of China,Grant/Award Numbers:82071942,82272013+1 种基金Shanghai Pujiang Program,Grant/Award Number:2020PJD008Clinical Research Plan of SHDC,Grant/Award Numbers:SHDC2020CR1031B,SHDC2020CR4060。
文摘Portal hypertension(PH)is a clinical syndrome,characterized by elevated pressure gradient between portal vein and inferior vena cava.These elevated pressures gradient due to increased vascular resistance and/or increased volume of blood flowing through the portal vein circulation,results in blood outflow difficulty from portal vein to hepatic veins and inferior vena cava.
基金National Natural Science Foundation of China(No.82300663)Natural Science Foundation of Sichuan Province(No.2022NSFSC0843)+1 种基金Fundamental Research Funds for the Central Universities(No.2023SCU12055)China Postdoctoral Science Foundation(No.2022M712262).
文摘Pressure measurement within the body is of pivotal significance in the diagnosis of vascular and organ-related diseases associated with hydrostatic pressure.At present,the most commonly used clinical method is to insert a catheter along with a pressure sensor and then guide it to the area of interest through vessels such as central venous pressure(CVP).However,the presence of sensors within the vessel of interest will inevitably cause alterations to the circulation and thus affect blood pressure.Moreover,the use of invasive methods does not allow monitoring of every area inside the body.
文摘Blood pressure(BP)is an important indicator of individuals’health conditions for the prevention or treatment of cardiovascular disease.However,conventional measurements require inconvenient cuffbased instruments and are not able to detect continuous blood pressure.Advanced methods utilize machine learning to estimate BP by constructing artificial features in plethysmography(PPG)or using an end-to-end deep learning framework to estimate BP directly.Empirical features are limited by current research on cardiovascular disease and are not sufficient to express BP variability,while data-driven approaches neglect expert knowledge and lack interpretability.To address this issue,in this paper we propose a method for continuous BP estimation that extracts both artificial and data-driven features from PPG to take advantage of expert knowledge and deep learning at the same time.Then a deep residual neural network is designed to reduce information redundancy in the gathered features and refine high-level features for BP estimation.The results show that our proposed methods outperforms the compared methods in three commonly used metrics.
文摘Background Transthoracic Doppler echocardiography is recommended for screening the presence of pulmonary hypertension(PH).However,the accuracy of pulmonary artery systolic pressure(PASP)estimated by Doppler echocardiographic is still unknown.Methods We conducted a retrospective study on 102 patients with idiopathic pulmonary arterial hypertension who underwent Doppler echocar-diography within 72 hours before right heart catheterization.During this time,all patients were stable without any specific drug therapy.Results There was moderate correlation between Doppler echocardiographic and right heart catheteriza-tion measurements of PASP(r=0.642,P【0.001).Using Bland-Altman analytic methods,the bias for the echocardio-graphic estimates of PASP was 6.65 mm Hg with 95%limits of agreement ranging from-47.62 to 34.30 mm Hg.There were 58.8%cases with absolute differences over 10 mm Hg between the two methods.Overestimation and underestimation of PASP by Doppler echocardiography occurred in 15.7%(16/102)and 43.1%(44),respectively.The magnitude of pressure underestimation and overestimation was insignificant(24.52±12.15 vs.25.69±16.09,P=0.765),while the corresponding diagnostic categories of severity that each subject would fall into for each technique are not in good agreement.The diagnostic categories of 16 overestimated patients were in accordance.During 44 underestimated patients,20.5%of patients had their pressure underestimated within one diagnostic category(minor error);4.5%of the underestimates were with two diagnostic categories(major error).Conclusions Transthoracic Doppler echocardiography may frequently be inaccurate in estimating PASP and could not replace the right heart catheterization.
文摘Insight into average oil pressure in gas reservoirs and changes in production （time）, play a critical role in reservoir and production performance, economic evaluation and reservoir management. In all practicality, average reservoir pressure can be conducted only when producing wells are shut in. This is regarded as a pressure build-up test. During the test, the wellbore pressure is recorded as a function of time. Currently, the only available method with which to obtain average reservoir pressure is to conduct an extended build-up test. It must then be evaluated using Homer or MDH （Miller, Dyes and Huchinson） valuation procedures. During production, average reservoir pressure declines due to fluid withdrawal from the wells and therefore, the average reservoirpressure is updated, periodically. A significant economic loss occurs during the entire pressure build-up test when producing wells are shut in. In this study, a neural network model has been established to map a nonlinear time-varying relationship which controls reservoir production history in order to predict and interpolate average reservoir pressure without closing the producing wells. This technique is suitable for constant and variable flow rates.
文摘In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucial for the vehicle design and safe control.However,precise state estimation of the brake pressure is desired to perform safe driving with a high degree of autonomy.In this paper,a sensorless state estimation technique of the vehicle's brake pressure is developed using a deep-learning approach.A deep neural network(DNN)is structured and trained using deep-learning training techniques,such as,dropout and rectified units.These techniques are utilized to obtain more accurate model for brake pressure state estimation applications.The proposed model is trained using real experimental training data which were collected via conducting real vehicle testing.The vehicle was attached to a chassis dynamometer while the brake pressure data were collected under random driving cycles.Based on these experimental data,the DNN is trained and the performance of the proposed state estimation approach is validated accordingly.The results demonstrate high-accuracy brake pressure state estimation with RMSE of 0.048 MPa.
基金the Capital Medical Technol-ogy Development Project(2024-IG-4032)National Key Research and Development Program of China(No.2023YFC3603500)+1 种基金the National Natural Science Foundation of China(No.62271023)the Beijing Natural Science Foundation(No.7242269).
文摘Continuous and accurate blood pressure(BP)monitoring is critical for personalized hypertension management.However,most existing methods focus on absolute BP estimation,with limited attention to BP changes.To address this limitation,we propose a novel framework named Multi-Perspective Differential Feature Space(MDFSBP)for cuffless BP estimation using photoplethysmography(PPG)signals.MDFSBP extracts three perspective differential features:time-based and points-of-interest features,frequency-domain features,and physiological statistical features.Then,an adaptive Multi-Perspective Differential Feature Mapping Module(MDFMM)integrates reconstruction regularization,trend-aware classification,and self-weighted contrastive learning to enhance feature representation and strengthen the association between features and BP changes.Finally,an AutoML-based regression pipeline automates model optimization,improving predictive accuracy and deployment efficiency.To better test the model's capability in capturing BP changes,we introduce a novel abnormality-aware classification metric.We demonstrate BP estimation performance over state-of-the-art(SOTA)methods on both the Mindray and MIMIC datasets.On the Mindray dataset,the model achieves a regression error of 0.17±5.17 mmHg for SBP and 0.05±3.29 mmHg for DBP,with classification accuracy and F1-score reaching 85.25%and 87.50%,respectively.On the MIMIC dataset,it achieves−0.09±5.70 mmHg for SBP and 0.12±4.27 mmHg for DBP,with the classification accuracy and F1-score of 72.84%and 70.66%,respectively.These results highlight the effectiveness,robustness,and generalizability of the proposed frame-work for non-invasive,real-time,and continuous BP monitoring in both clinical and wearable healthcare systems.
文摘The first automatic algorithm was designed to estimate the pulse pressure variation (PPVPPV) from arterial blood pressure (ABP) signals under spontaneous breathing conditions. While currently there are a few publicly available algorithms to automatically estimate PPVPPV accurately and reliably in mechani-cally ventilated subjects, at the moment there is no automatic algorithm for estimating PPVPPV on sponta-neously breathing subjects. The algorithm utilizes our recently developed sequential Monte Carlo method (SMCM), which is called a maximum a-posteriori adaptive marginalized particle filter (MAM-PF). The performance assessment results of the proposed algorithm on real ABP signals from spontaneously breath-ing subjects were reported.
