Research on simulation experiments for some conductance operations of a new vehicle-based energy storage tank

The control software is the user interface between the user and the test control system. It provides all the communication functions between the application software and the test controller for defining and controlling the test, and can realize all the control functions of the system. The software can combine physical and virtual prototype tests to provide a complete digital test environment. Including the following functions: preparation and visualization of target data; configuration of test bench hardware; generation of excitation signal files; durability test and monitoring; results and damage analysis.

Time domain waveform reconstruction software is the main software used in this road simulation study. Its main function is to use an electro-hydraulic servo test bench in the lab environment. Through software iteration, the load time history of the vehicle on the measured road is reproduced. The basic flow of time-domain waveform reproduction is the basic flow of time-domain waveform reconstruction. The entire process consists of three steps: target signal acquisition, system identification, and target simulation. These three steps are in the road simulation of automobiles. very important.

The theoretical basis of the road simulation theory The target signal obtained under actual operating conditions is the basic premise for the reconstruction of the time domain waveform, that is, the time horizon of the target that is expected to be reproduced. System identification system model identification is the basis for realizing indoor road simulation test. It should be noted that this system refers to the entire test system including the tested part and the test bench, and uses the system's frequency response function matrix A to describe its mathematics. model. The most important thing in this control algorithm is how to judge whether the model is good or bad, and how to deal with similar models obtained in different recognition processes to get more accurate results.

For the identification and acquisition model, the coherence function method and the prediction method can be used to evaluate the model. Since the coherence function reflects the linear dependence between the system input signal and the output signal, in general, if the function value of the coherence function between the input signal and the output signal in the required frequency band is greater than 0.8, the system is considered to be good. The introduction of the prediction method. The so-called prediction method is to calculate the output of the system based on the input of the system and the obtained model, and compare this result with the actual output to determine the quality of the model. Tests have proved that the evaluation of the obtained model by this method can achieve more satisfactory results. For the similar models obtained in the two consecutive recognition processes, the weighted average method is used to obtain more accurate recognition results. According to the different sources of weighting coefficients, they can be divided into two different weighted average methods: The custom weighted coefficient method is Aup=αAnew(1-α)Apre(1) where Aup is the frequency response function matrix of the updated model; α is the weighting coefficient; Anew is the frequency response function matrix of the latest model; Apre is the frequency response function matrix of the model obtained after the last update.

The weighted coefficient method is obtained by the coherence function as Aup=αAnew(1-α)Apre(2) where: the weighting coefficient α=γnewγnewγpre, γnew is the value of the most recently obtained coherence function, and γpre is the value of the coherence function obtained after the last update. . In the experiment, it can be determined according to the actual situation which model average method is used, or the combination of the two.

Target Simulation In the target simulation phase, the frequency response function model and the desired target signal are used to calculate the drive signal required to reproduce the target signal. Also due to the influence of non-linear factors, this is an iterative process. At each step of the iterative process, the existing error, ie the deviation between the measurement signal and the target signal, is corrected.

In iterative calculations, the entire system is simplified into a linear invariant system, as shown. Due to the multiple input and multiple output characteristics of the test system, the frequency response function matrix is ​​A=Y(f)X-1(f). (3) Simplified model of a multiple-input multiple-output system. White-pink noise can excite the required band signals. Out, and can make the energy evenly distributed in the frequency domain, so in the research process with white-pink noise as input, the sensor to collect the actual response signal is output, so as to obtain the system's frequency response function matrix A.

Due to the influence of nonlinear factors, the frequency response function of a complex structure depends to a large extent on the level and form of the excitation. Therefore, according to the specific conditions of the entire system, the white pink noise parameters are adjusted so that the frequency response function converges to one A representative model, this model is valid within the operating range of the target signal obtained, and can be well applied to the next target simulation process.

The main purpose of the target signal simulation is to use the frequency response function matrix A of the model obtained in the previous step to obtain the driving signal whose response signal converges to the target signal through iteration. The first drive signal X1(f) is calculated using the edited target signal Y(f) and the obtained inverse system function matrix A-1. The system is excited by X1(f), and the sensor recovers the signal Y1(f) and Y1 ( f) Compare with Y(f) to get the error signal.

The error signal and system inverse function matrix A-1 are iterated to obtain a correction signal. The correction signal is added to the driving signal X1(f) to obtain the second driving signal X2(f) so as to repeat the above steps again, and is repeated until the error between the response signal Yn(f) and the desired signal Y(f) is small It is acceptable until the final iterated drive signal is used as the test drive signal to complete the iterative simulation of the target signal.

Freedom degree indoor road simulation test specimen installation Test specimen installation, as far as possible in line with the state of the loading; In addition, the installation of the sensor must also be in line with the state of the data collection. According to the analysis, this test decided to place the test piece on the vibration table.

Determination of measuring points The placement and orientation of the sensors are as consistent as possible. In this test, according to the requirements, six single-way sensors were arranged at the designated positions of the battery. The main purpose of the iterative target signal simulation is to use the frequency response function matrix of the model obtained in the previous step to obtain the drive signal whose response signal converges to the target signal through iteration. The calculation of the first drive signal and the correction of the drive signal are C=B-1{A-1BD}(4)Ci1=CiB-1{A-1GB} (5) where: C is the initial drive signal function matrix; B is Fast Fourier transform matrix; D is the target signal function matrix; Ci is the i-th driving signal function matrix; Di is the i-th target signal function matrix; G is a weighting factor, located between 0 and 1.

Analysis of results The experiment was successful in iteration, with good convergence and iterative accuracy of less than 5%. Therefore, the iterative process can be ended and the excitation spectrum can be used as the loading spectrum of the durability test afterwards.

According to the excitation spectrum obtained by iterating the target signal, the battery is excited as the excitation signal of the hydraulic servo device, and the response signal of the measurement point is collected, and these measurement signals are compared with the target signal. Due to space limitations, only the z-signal of the battery is selected. For a period, the comparison results are as shown.

Cementing Tool

Cementing Tool,Casing Thread Protector,Casing And Line Pipe,Solid Body Casing Centralizers

Puyang Zhongshi Group Co.,Ltd , https://www.pycentralizer.com

Posted on