Identification Method With Crankshaft Angle Vibration Of Internal Combustion Engine Working State Based On Response Surface Model

With the rapid development and reform of China's rail transportation,high power internal combustion generator sets based on high speed,complexity and electronic automation are widely used in the suburban with short haul transportation and most of the underdeveloped areas.The internal combustion engine is the key power unit of the generator set,serious consequences may be caused when the fault of internal combustion engine occurs during the actual operation and can't be found in time.Therefore,monitoring and fault diagnosis for each cylinder of internal combustion engine is particularly important.The traditional fault diagnosis scheme has large calculation,long period and heavy detection work.It is difficult to meet the needs of on-line rapid detection and diagnosis of internal combustion engine.Based on the basic theory of the axis angle vibration and the moment of the internal combustion engine,this paper proposes a method to construct the identification and detection model of the work state of the internal combustion engines by using the multi harmonic response surface agent model.This method can effectively overcome the shortcomings in the traditional methods.In general,the fault diagnosis work consists of three aspects: collecting signal data,establishing the related information of fault signal and fault mode,and distinguishing fault through related information.Therefore,this paper takes the typical internal combustion generator as the research object,and studies the characteristics of the method of collecting shaft angle vibration signal signal.The method of modifying the shafting dynamic simulation model is studies by using the torsional vibration test results of shaft system.On the basis of these works,a multi harmonic response surface agent model is proposed to identify of the working state of each cylinder through the angular vibration of internal combustion engines,and the method are checked through simulation and experimental.The acquisition and processing of the angular vibration signals are studied and analyzed.Propose the solving of the problem of the magnetoelectric method at low speed through the photoelectric method.The influence of sampling frequency,joint error,eccentricity of test parts,signal-to-noise ratio and other factors on data acquisition are discussed.The results show that the photoelectric method is suitable for measuring the angular vibration of the shaft system under the medium and low rotational speeds,and the magnetoelectric method is suitable for measuring the angular vibration of the shaft system under the medium and high rotational speeds;the effect of the measuring part eccentricity on the magnetoelectric method is larger than that of the photoelectric method;the joint gap error can be effectively eliminated by the five point three resampling method.These researches can provide convenient to check the accuracy of the model.Construction of dynamic model of shaft system,Proposing the method to check the parameters of the axis equivalent model by the starting and stopping conditions and the continuous reduction or increase of speed conditions.The gas harmonic moment method is corrected by the balance working and single cylinder stop condition.The correctness of the method is verified by the comparison between the revised shafting simulation results and the experimental results.The results show that the revised axle system response and experimental results are more appropriate,which can provide the convenience for the accuracy check of the work condition identification model of each cylinder of the internal combustion engine.This paper,based on the basic relationship between the torsional vibration of the shaft system of the internal combustion engine and the torque received,proposes a multi harmonic response surface agent model to construct the identification and detection model of the work state of the internal combustion engine.The results show that the power loss error of each cylinder identified in the simulation test model is only 0.03%,and the maximum error of the power loss identified in the experiment test is 4.79%,indicating the erroe of power loss is accurate enough.The relevant research results in this paper can provide reference for the acquisition and processing of shafting angle vibration signal.It also has a guiding significance for the check of the working condition of each cylinder in the internal combustion engine.