Adaptive Multi-Feature Fusion for Vehicle Micro-Motor Noise Recognition Considering Auditory Perception

The deployment of vehicle micro-motors has witnessed an expansion owing to the progression in electrification and intelligent technologies.However,some micro-motors may exhibit design deficiencies,component wear,assembly errors,and other imperfections that may arise during the design or manufacturing phases.Conse-quently,these micro-motors might generate anomalous noises during their operation,consequently exerting a substantial adverse influence on the overall comfort of drivers and passengers.Automobile micro-motors exhibit a diverse array of structural variations,consequently leading to the manifestation of a multitude of distinctive auditory irregularities.To address the identification of diverse forms of abnormal noise,this research presents a novel approach rooted in the utilization of vibro-acoustic fusion-convolutional neural network(VAF-CNN).This method entails the deployment of distinct network branches,each serving to capture disparate features from the multi-sensor data,all the while considering the auditory perception traits inherent in the human auditory sys-tem.The intermediary layer integrates the concept of adaptive weighting of multi-sensor features,thus affording a calibration mechanism for the features hailing from multiple sensors,thereby enabling a further refinement of features within the branch network.For optimal model efficacy,a feature fusion mechanism is implemented in the concluding layer.To substantiate the efficacy of the proposed approach,this paper initially employs an augmented data methodology inspired by modified SpecAugment,applied to the dataset of abnormal noise sam-ples,encompassing scenarios both with and without in-vehicle interior noise.This serves to mitigate the issue of limited sample availability.Subsequent comparative evaluations are executed,contrasting the performance of the model founded upon single-sensor data against other feature fusion models reliant on multi-sensor data.The experimental results substantiate that the suggested methodology yields heightened recognition accuracy and greater resilience against interference.Moreover,it holds notable practical significance in the engineering domain,as it furnishes valuable support for the targeted management of noise emanating from vehicle micro-motors.

Noise and vibration measurements in a Viking military vehicle

Noise and whole-body vibration measurements were made in a Viking military vehicle to determine the variation that should be expected during repeat measures,the effect of speed(up to 60 km/h in 5 km/h increments),and during travel over different types of terrain(comprising concrete road,gravel track and rough cross-country).Measurements were made at various crew positions(including the driver and commander)in both the front and the rear cabs in the vehicles.Three translational axes of vibration were measured in each seat.Two speeds were investigated over road(35 km/h and 55e60 km/h)and gravel(20 km/h and 35 km/h)surfaces.The effect of varying speed of the vehicle on the measured noise and vibration magnitudes was also investigated.The highest sound pressure level(LAeq)of 104 dB(A)was measured at the commander’s standing position during travel over concrete road at 55 km/h.Higher noise levels occurred for a standing commander compared with when sitting on the seat.A maximum single axis frequency-weighted vibration magnitude of 1.0 m/s^(2) r.m.s.was measured on the driver’s seat during travel over track at 35 km/h.Higher vibration magnitudes occurred during travel over track compared with travel over road.Both noise and vibration exposure of crew within the Viking vehicle increased with increasing speed of the vehicle.

Optimization of Vibration and Noise Performance of Permanent Magnet Synchronous Motor for Electric Vehicles

In the design of the motor used for electric vehicles(EVS),vibration and noise problems are often ignored,which reduce the reliability and service life of the motor.In this paper,an interior permanent magnet synchronous motor(IPMSM)with high power density is taken as an example,and its electromagnetic vibration and noise problem is investigated and optimized.Firstly,the factors that generate the electromagnetic force harmonic of IPMSM are analyzed by theoretical derivation.Furthermore,the mode and electromagnetic harmonic distribution of the motor are calculated and analyzed by establishing the electromagnetic-structure-sound coupling simulation model.Then,by combining finite element method(FEM)with modern optimization algorithm,an electromagnetic vibration and noise performance optimization method is proposed in the electromagnetic design stage of the motor.Finally,an IPMSM is optimized by this method for electromagnetic vibration and noise performance.The results of comparison between before and after optimization prove the feasibility of the method.

A Novel Ultra Short-Term Load Forecasting Method for Regional Electric Vehicle Charging Load Using Charging Pile Usage Degree

Electric vehicle(EV)charging load is greatly affected by many traffic factors,such as road congestion.Accurate ultra short-term load forecasting(STLF)results for regional EV charging load are important to the scheduling plan of regional charging load,which can be derived to realize the optimal vehicle to grid benefit.In this paper,a regional-level EV ultra STLF method is proposed and discussed.The usage degree of all charging piles is firstly defined by us based on the usage frequency of charging piles,and then constructed by our collected EV charging transactiondata in thefield.Secondly,these usagedegrees are combinedwithhistorical charging loadvalues toform the inputmatrix for the deep learning based load predictionmodel.Finally,long short-termmemory(LSTM)neural network is used to construct EV charging load forecastingmodel,which is trained by the formed inputmatrix.The comparison experiment proves that the proposed method in this paper has higher prediction accuracy compared with traditionalmethods.In addition,load characteristic index for the fluctuation of adjacent day load and adjacent week load are proposed by us,and these fluctuation factors are used to assess the prediction accuracy of the EV charging load,together with the mean absolute percentage error(MAPE).

An interacting multiple model-based two-stage Kalman filter for vehicle positioning

To address the problem that a general augmented state Kalman filter or a two-stage Kalman filter cannot achieve satisfactory positioning performance when facing uncertain noise of the micro-electro-mechanical system（MEMS） inertial sensors, a novel interacting multiple model-based two-stage Kalman filter（IMM-TSKF） is proposed to adapt to the uncertain inertial sensor noise. Three bias filters are developed based on different noise characteristics to cover a wide range of noise levels. Then, an accurate estimation of biases is calculated by the interacting multiple model algorithm to correct the bias-free filter. Thus, the vehicle positioning system can achieve good performance when suffering from uncertain inertial sensor noise. The experimental results indicate that the average position error of the proposed IMMTSKF is 25% lower than that of the general TSKF.

Noise Control of a Two Stroke Engine Car

Aim To control the noise of two stroke engine Methods On the basis of noise identification,a new muffler and acoustic shield were designed,Results the car's pass-by noise below the national limit Conclusion through proper noise controlling measures,the pass-by noise of two stroke engines could be reduced under national permitting limit.

Strong tracking adaptive Kalman filters for underwater vehicle dead reckoning

To improve underwater vehicle dead reckoning, a developed strong tracking adaptive kalman filter is proposed. The filter is improved with an additional adaptive factor and an estimator of measurement noise covariance. Since the magnitude of fading factor is changed adaptively, the tracking ability of the filter is still enhanced in low velocity condition of underwater vehicles. The results of simulation tests prove the presented filter effective.

Statistical Evaluation and Regression Analysis of Vehicle Sound Quality

A statistical approach to evaluate the subjective perception of the annoyance caused by the vehicle noise was presented in this paper. After recording the noises of Sanfeng, Huali and Xiali at speeds of 30, 40, 50, 60, 70 and 80 km/h respectively, the annoyance of the vehicle noises was evaluated in the testing room using paired comparison method, and the sound quality metrics and subjective annoyance were then distilled. Loudness, sharpness, roughness, periodicity and impulsiveness were selected for each of the vehicle noises. By correlation analysis method, it can be found that loudness has a higher correlation （0.91） with annoyance than other parameters. Meanwhile, sharpness, periodicity, roughness and impulsiveness have correlation with subjective perception with correlation coefficients being 0.84, -0.82, 0.62 and 0.87, respectively. The result of multiple regression analysis shows that calculated annoyance obtained by the regression equation can explain the perceptual annoyance and the regressed evaluation model is feasible to evaluate the sound quality of vehicle.

Experimental research of traction motor system for electric vehicle base on sound tensity

The traction motor of electric vehicle is differing from the general industry traction motor completely. Not only frequently start, parking, accelerate, decelerate and low speed, but also high torque in climbing slope, low torque in high speed and wide range speed are requested. Base on the theory of sound intensi- ty, the experiment of noise are study through the measurement at discrete points. The sizing grid is 10mm × 10mm, The sound intensity map of traction motor are protracted at 1000r/min and the result show that the main noise sources are fan, gear-box and the traction motor in turn.

Experimental Study of a Modified Command Governor Adaptive Controller for Depth Control of an Unmanned Underwater Vehicle

Command governor–based adaptive control(CGAC)is a recent control strategy that has been explored as a possible candidate for the challenging task of precise maneuvering of unmanned underwater vehicles(UUVs)with parameter variations.CGAC is derived from standard model reference adaptive control(MRAC)by adding a command governor that guarantees acceptable transient performance without compromising stability and a command filter that improves the robustness against noise and time delay.Although simulation and experimental studies have shown substantial overall performance improvements of CGAC over MRAC for UUVs,it has also shown that the command filter leads to a marked reduction in initial tracking performance of CGAC.As a solution,this paper proposes the replacement of the command filter by a weight filter to improve the initial tracking performance without compromising robustness and the addition of a closed-loop state predictor to further improve the overall tracking performance.The new modified CGAC(M-CGAC)has been experimentally validated and the results indicate that it successfully mitigates the initial tracking performance reduction,significantly improves the overall tracking performance,uses less control force,and increases the robustness to noise and time delay.Thus,M-CGAC is a viable adaptive control algorithm for current and future UUV applications.

Experimental and Numerical Study of Wind Noise Caused by Cowl Area on Volvo XC60

This paper presents a detailed experimental and numerical study of aerodynamically produced noise which occurs due to turbulent structures created by the cowl cavity and side mirror. Measurements were carried out at Volvo aerodynamical wind tunnel on a Volvo XC60 production model. The configurations considered here are： side mirror On/Off with the cowl cavity open/closed. The results of exterior sound source mapping （with the intensity probe placed in the flow stream） have been compared with the results of the measurements inside the car. The contribution of the cowl area to overall wind noise level is measured in terms of AI% （Articulation Index） inside the compartment. It was shown that increase in AI by 2% could be attributed to the cowl generated wind noise. Transient numerical simulations of the turbulent flow around the car have been performed for all configurations. The results of the simulations show similarity to experimental results and give insight to the flow structures around the car.

Mapping Vehicular Noise Pollution in Port Harcourt Metropolis, Rivers State, Nigeria: Implication for a Sustainable Urbanization

This study aims to investigate geo-referenced vehicular noise pollution in the Port Harcourt metropolis of Rivers State,Nigeria.Three types of data were gathered for this study.Data from vehicular traffic noise were measured in decibels(dB)using Noise Dosimeter(ND);data from vehicular traffic counts were carried out by observing and counting traffic flow at junctions and roundabouts as well as vehicular traffic noise location map was established by using Global Positioning System(GPS)instrument processed in the Geographic Information System(GIS)environment.The findings indicated that in the northern segment,Igwurita(99.5 dB)and New road roundabout(96 dB),generated the highest vehicular noise in the spatial distribution.In the eastern road segments,Eleme Flyover(98.1 dB)and Artillery Junction(95.5 dB)contributed the highest vehicular noise levels.In the northern segment,New Road(2311 vehicles)and Igwuruta(1566 vehicles)at the roundabouts,generated the highest vehicular traffic counts in the spatial distribution.Thus,among the eastern roads,Eleme Flyover(6735 vehicles)and Artillery Junction(5539 vehicles)contributed the highest vehicular counts in the area.The results showed that the northern and eastern segments of the Port Harcourt metropolis had the highest level of vehicular traffic noise and traffic flow.Thus,the vehicular noise level values have exceeded the recommended 75 dB national and international health standards.The study recommended the construction of more road networks in the southern and western parts of the Port Harcourt metropolis to decongest traffic flow and noise pollution in the northern and eastern segments of the city.

采用生理响度感知机制的车内噪声声品质预测

现有车内噪声声品质预测的响度计算中没有考虑真实人耳生理解剖结构的传声特性,因此提出了一种基于生理响度感知模型的车内噪声声品质预估方法。首先,采集两款轿车的车内噪声样本,并通过主观评价试验得到车内噪声的主观评价烦恼度。之后,整合中耳集总参数模型与耳蜗传输线模型,构建生理响度模型。然后,以生理响度模型的响度计算值为主要参数,结合尖锐度、粗糙度与车内噪声的主观评价值,通过TabNet模型构建了车辆声品质预测模型。最终,对比分析了所构建声品质模型与基于现有标准响度模型所构建的声品质模型的预测效果。结果表明,采用生理响度模型的声品质预测平均误差百分比仅有4.73%,优于采用Moore响度(6.13%)与Zwicker响度(6.94%)的声品质预测结果。此外,所构建的TabNet声品质预测模型的平均误差百分比也低于基于BP神经网络模型的平均误差百分比(7.60%)。采用生理响度模型的TabNet声品质预测能够提高车内噪声声品质客观评价的准确率。

基于固-气-液耦合的气垫船水下辐射噪声数值分析

气垫船水下声辐射是复杂的船体结构-气垫-周围水体三类介质声振耦合过程,其声传递特性是气垫船声学分析中的难点。本文以某全垫升气垫船为研究对象,结合声固耦合有限元法与解析法,探究气垫船水下辐射噪声传递特性。通过建立固-气-液声固耦合有限元模型,计算机械激励与垫升风机声激励,并以试验固有频率修正数值模型;通过声透射解析公式计算空气螺旋桨、燃气轮机等设备声激励辐射的水下噪声。采用距离左、右舷1 m处线上各点的总声压级作为评价指标。结果表明:气垫船水下辐射噪声为宽带噪声;有限元法计算的水下辐射噪声在低频时集中分布于气垫-周围水体耦合面附近,随着频率的升高,水下噪声增大并逐渐呈不规则干涉状分布;空气螺旋桨等声激励直接辐射水下噪声主要为近场低频噪声,在设备声激励辐射范围内声压平均占比约为40%。本文揭示的水下辐射噪声传递特性对气垫船减振降噪具有重要指导意义。

电动汽车座舱内部噪声的非线性心理声学烦恼度建模

拟研究电动汽车座舱内部噪声特性对声学舒适性的影响。通过道路试验获取4种典型路面工况下某型电动汽车在不同车速下的舱内噪声,计算分析其心理声学参数,开展烦恼度主观评价实验,并基于非线性Zwicker心理声学烦恼度及其修正模型进行建模研究。结果显示,随着车速和路面粗糙程度变大,响度逐渐提高,尖锐度也随车速增大而变大;Zwicker心理声学烦恼度模型更适用于电动汽车座舱内部噪声。

CNN-LSTM车辆运动状态识别的AUKF组合导航方法

针对固定的噪声协方差难以适应车辆不同运动行为下噪声统计特性差异大的问题,提出了一种基于卷积神经网络与长短期记忆网络(CNN-LSTM)的车辆运动状态识别自适应无迹卡尔曼滤波(AUKF)组合导航方法。首先,应用CNN-LSTM网络模型进行车辆运动状态识别,解决车辆自我运动不确定性的问题;其次,将特定运动状态约束下的噪声协方差应用于UKF的时间更新与量测更新;最后,将所提方法在采集的数据集上进行验证。实验结果表明,与经典的UKF算法相比,所提方法的位置均方根误差与速度均方根误差分别下降了22.67%与2.63%,验证了所提方法的有效性。

CRTSⅢ型板和双块式无砟轨道振动噪声特性研究

【目的】探究高速铁路CRTSⅢ型板和双块式无砟轨道的结构振动与噪声特性。【方法】以有限元、多体动力学与边界元相结合的方法为基础,通过建立高速列车-无砟轨道-箱梁耦合动力学模型和边界元模型,研究列车高速通过时的列车-轨道-箱梁耦合系统的结构动力响应,以及噪声辐射特性。【结果】结果表明,列车通过桥上CRTSⅢ型板与双块式轨道结构段时,列车的最大轮重减载率分别为0.0536与0.1657,下部箱梁挠度分别为1/9673与1/8457。与双块式无砟轨道相比,CRTSⅢ型板的轨道板及底座位移响应更小,并且轨道-箱梁系统的竖向振动衰减更快,减振性能更优。另外,比较两种轨道结构条件下的高架箱梁结构噪声,CRTSⅢ型板段的箱梁噪声总声压级更小。但是,桥下远场点噪声峰值差异较小,峰值频率接近,均为31.5Hz左右;而桥下峰值频率差异较大,Ⅲ型板的峰值频率为63Hz,双块式则为20Hz。【结论】CRTSⅢ型板式无砟轨道更适合在高速铁路桥梁段使用。

车辆运行参数自适应软测量的防抱死控制研究

建立了一种将Sage-Husa噪声估计器与无迹卡尔曼滤波器有机的结合的软测量算法,通过自适应软测量算法对车速等车辆运行参数进行有效的测量,结合采用轮速传感器得到的易测量轮速信号,实时计算车辆防抱死系统的实际的滑移率,通过与理想路面滑移率比较,从而实现各轮缸压力的有效控制,避免了车轮的抱死。通过CARSIM/Simulink联合仿真实验表明:建立的自适应软测量算法可以实现车速的准确测量,并且在对开路面上可以有效抑制了制动过程中的车轮抱死,从而保证了车辆的制动稳定性。

基于卷积神经网络的红外弱小车辆目标检测方法

传统方法无法获得理想的红外弱小车辆目标检测结果,导致检测误差大,无法满足实际应用要求,为了解决传统红外弱小车辆目标检测方法存在的局限性,及时检测红外图像中的弱小车辆,提高车辆检测精度,设计了基于卷积神经网络的红外弱小车辆目标检测方法。首先对弱小车辆目标检测需要的红外图像进行采集,并对红外图像噪声进行处理,消除噪声对弱小车辆目标检测的干扰,然后采用卷积神经网络建立弱小车辆目标检测模型,最后通过具体仿真实验测试弱小车辆目标检测方法的性能。结果表明,该方法的弱小车辆目标检测精度超过了90%,大幅度减少了弱小车辆目标的误检率,同时弱小车辆目标检测时间控制在5 s内,可以满足弱小车辆目标检测的实时性要求,具有较高的实际应用价值。

基于φ-OTDR的车辆振动信号检测方法

针对相位敏感光时域反射计(Phase Sensitive Optical Time Domain Reflectometer,φ-OTDR)系统在车辆振动信号检测中信号信噪比低且检测准确率受频率影响的问题,提出了一种引导边缘滤波检测(Guided Edge Filter,GEF)方法。GEF方法在减少噪声影响的同时保留并提取出有效的振动位置信息。使用PZT分别产生不同频率的振动信号,将GEF方法与直接振幅差分法进行实验对比。结果表明,使用GEF方法得到的平均信噪比相较于直接振幅差分法提高了3.26 dB。使用公路上行驶车辆的振动信号进行测试,结果表明,使用GEF方法得到的信噪比最高为3.98 dB,比直接振幅差分法提高了2.65 dB。