1. Introduction With the rapid development of highways and the continuous increase in vehicle speeds, the impact of aerodynamic drag on fuel consumption is becoming increasingly prominent. 85% of aerodynamic drag is pressure drag, and 91% of this pressure drag originates from the rear of the vehicle (its value varies depending on the vehicle's length). Furthermore, the vehicle's wake structure has a decisive influence on its aerodynamic characteristics. Therefore, measuring vehicle wake velocity, understanding its structure, and improving aerodynamic characteristics are crucial for developing low-drag vehicles and reducing fuel consumption. During wake test research, we found that traditional testing instruments are functionally fixed and packaged, specifically designed for a particular test or task, and are expensive. Developing a testing system places very high demands on designers: requiring mastery of the underlying hardware of the testing instrument, extensive computer programming knowledge, and the ability to write hardware drivers. This results in long development cycles and poor flexibility for traditional testing systems. To shorten the development time of the speed testing system and facilitate future maintenance, expansion, and upgrades by users, we introduced the concept of a virtual instrument based on a hot-wire anemometer, developing a dedicated virtual instrument system for measuring vehicle wake velocity. 2. Hot-Wire Anemometer The advent of the hot-wire anemometer represents a significant breakthrough in fluid mechanics testing techniques. It provides a powerful tool for studying unsteady flow, particularly turbulence, in automotive aerodynamics testing. Although laser velocimetry emerged in the 1960s, experimental research revealed that laser velocimetry often yields unreliable results when measuring turbulence parameters due to the loss of particle signals. Furthermore, the high cost and maintenance expenses of laser velocimeters limit their application in turbulence field research compared to hot-wire technology. Hot-wire technology will remain the primary method for measuring automotive turbulence, especially the velocity of the vehicle's wake field. For forced convective heat transfer in incompressible airflow, the Nu number is only related to the Reynolds number and the physical and operating conditions of the hot wire. For convenience, when measuring wind speed U with a hot-wire anemometer, we typically use the following formula: The value of n varies with Re, with experimentally measured exponents n = 0.45–0.51. In the formula, Rw and Rg are the working resistance of the hot wire and the airflow resistance constant, respectively, and I is the hot-wire current. Coefficients A and B are determined by experiment. When the hot wire is working at a constant temperature, the thermal resistance Rw of the hot wire is constant. Since IRw = e, equation (1) can be rewritten as follows: where e is the output voltage value of the hot wire circuit corresponding to the wind speed U. 3. Concept of Virtual Instrument LabVIEW (Laboratory Virtual Instrument Engineering Workbench) is a graphical programming language that uses function icons to create applications, that is, to create virtual instruments (VI, Virtual Instrument). A virtual instrument is a program software whose operation and appearance are similar to physical instruments such as oscilloscopes or multimeters and achieve the same function. In text-based programming languages, instruction statements determine program execution; in LabVIEW, the data flow determines program execution, and programming is performed according to the data flow. The paper developed a virtual instrument system for measuring the speed of car exhaust flow on the LabVIEW platform. The system includes two subsystems: speed calibration and speed measurement. 4. Hot Wire Speed ​​Calibration Subsystem The external flow field of a car is an incompressible continuous flow. The expression of the static calibration equation for hot wire speed measurement is adopted as equation (2). In the hot-wire speed measurement calibration subsystem, we treat n as a variable, thus eliminating the need for dedicated linearizer hardware to linearize the hot-wire output voltage. We directly implement this through software programming: continuously changing the value of n to perform different exponential fittings, and using the mean square error (MSE) between the fitted curve and the experimental data to judge the quality of the fit. This also demonstrates the advantages of virtual instruments: simple development process, saving time and expenses. 4.1 Subsystem Hardware Structure The hardware structure of the hot-wire speed measurement calibration subsystem is shown in Figure 1. As can be seen from the figure, the hardware structure consists of two parts: one part is the hot-wire anemometer circuit, which generates the hot-wire circuit output voltage e; the other part is the Pitot tube section, which generates the corresponding differential pressure (dynamic pressure) electrical signal Ep. Figure 1 Hardware structure of the hot-wire speed measurement calibration subsystem In this paper, the basic circuit of the constant temperature hot-wire anemometer (CTA) is used for hot-wire speed measurement, and the circuit form is a feedback circuit. When the entire system is connected and the power is turned on, the Wheatstone bridge reaches balance. When the airflow velocity increases, the heat exchange between the hot wire and the surrounding fluid increases, the hot wire cools down, and the resistance decreases. According to the working principle of the constant temperature hot wire anemometer discussed above, the output voltage will increase. Conversely, the output voltage will decrease. The waveform of the output voltage change over time is displayed by an oscilloscope, while the specific value is displayed by a digital multimeter. At the same time, the analog signal of this output voltage e is converted into a digital signal by the A/D board and then read into the VI system by the data acquisition card. In addition, the differential pressure signal Ep of the static pressure Pitot tube is also read into the VI system by the acquisition card. The data acquisition is completely controlled by the virtual instrument software. 4.2 Subsystem Virtual Instrument Composition The software composition and user interface of the hot wire anemometer speed measurement calibration subsystem are shown in Figures 2 and 3, respectively. Figure 2 Virtual Instrument Structure Diagram of Hot Wire Anemometer Speed ​​Measurement Calibration System Figure 3 Virtual Instrument User Interface of Hot Wire Anemometer Speed ​​Measurement Calibration System This speed measurement calibration system can perform the following functions: (1) Set data acquisition parameters. The data acquisition system can set the signal gain amplification factor, sampling frequency, signal scan number, and the upper and lower limits of the measured signal, etc. Under the condition that the computer does not overflow, the range of the measurement signal of the virtual instrument can be changed arbitrarily, which is something that traditional testing instruments cannot achieve. (2) Initialize environmental variables, including atmospheric pressure and temperature. With the given initial conditions such as atmospheric pressure and temperature, the system can also calculate the air density and air humidity in the wind tunnel at the time of the test. (3) Display the signal to be measured in real time, including the electrical signal corresponding to the Pitot tube pressure difference and the output voltage signal of the hot-wire anemometer. These tasks are completed by the image indicator in the virtual instrument, which has the same function as the actual oscilloscope. According to the signal display, we can know when the change of the signal to be measured is drastic and when it is stable, and perform sampling control as needed. However, the display range of the oscilloscope is very limited and not as flexible as the image indicator. The display range of the image indicator in the virtual instrument can be set arbitrarily as needed, provided that the computer does not overflow. This also reflects the flexibility of the virtual instrument and the convenience of adjustment and upgrading. (4) Control of data acquisition and storage. (5) Data processing. The data consists of two parts: (1) Preliminary processing: Based on the known relationship between the differential pressure signal and the differential pressure, the signal is first converted into a differential pressure value, and then the differential pressure is converted into a velocity based on the relationship between the differential pressure and the wind speed. (2) Curve fitting: The curve fitting function is completed by a dedicated sub-virtual instrument (SubVI), as shown in Figure 4. The data U from the preliminary processing and the hot wire output voltage are used as the data input for the curve fitting VI. Then, according to King's exponential law, i.e., using formula (2), exponential curve fitting is performed to obtain A, B, and n. A typical fitting curve result is shown in Figure 5. In the curve fitting VI, optimal polynomial fitting is also performed. By comparing the two fitting methods, it can be seen that the curve fitting error of King's exponential law is smaller. The results obtained from the calibration system, i.e., the values ​​of coefficients A, B, and n, can be stored on the hard disk or other storage devices via file I/O. Figure 4. Curve fitting for obtaining A, B, and n using a sub-virtual instrument. Figure 5. Experimental data fitting curve . 5. Hot-wire anemometer wake velocity measurement subsystem This virtual instrument subsystem calls the hot-wire speed calibration subsystem to obtain the calibration formula. The hot-wire anemometer can then measure the electrical signal of the car wake velocity and process it through the virtual instrument (VI) to obtain the wind speed. 5.1 Hardware composition of the subsystem The hardware system structure of the measurement subsystem is shown in Figure 6. Figure 6 Hardware composition of the hot-wire anemometer measurement system Compared with the hot-wire calibration system, this system adds a fixed hot-wire probe transfer frame, a control system actuator—servo system, and the A/D board must be bidirectional, i.e., capable of A/D and D/A conversion. Before measuring the wind speed, the wind tunnel wind speed needs to be measured, which is completed by a static pressure Pitot tube anemometer (as shown in the dashed box on the right side of Figure 6). The Pitot tube should be placed upstream of the test section away from the car model to ensure that the flow field here is not disturbed by the flow field around the test model. The A/D board converts the x, y, z coordinate digital signals emitted by the virtual instrument into analog signals and sends them to the server through channel 3 (ch3), triggering the driver to move the measuring frame to the measurement position. After that, the data measurement can be controlled by the virtual instrument program written in LabVIEW. 5.2 Subsystem Virtual Instrument Composition The composition of the hot wire anemometer system VI and its user interface are shown in Figure 7 and Figure 8. Figure 7 Hot wire anemometer system VI structure diagram Figure 8 Hot wire anemometer system user interface This speed measurement system can perform the following functions: (1) Setting the parameters for data acquisition, which is the same as the parameter settings for the calibration system. (2) Initializing the test environment variables. In addition to initializing the environment variables of the calibration system, the position (x, y, z) to be measured by the hot wire probe of the hot wire anemometer must also be given. (3) Measuring the wind speed in the wind tunnel, that is, measuring the wind speed of the uniform flow field in the test section of the wind tunnel. As shown in the dashed box on the right in Figure 6, the pressure difference of the Pitot tube is first sensed by the differential pressure sensor and converted into an electrical signal. After being converted by the A/D board, it is processed by the data processing system and displayed as wind speed in the indicator on the panel (V0total, m/s). It is also temporarily stored in the buffer and stored in the memory after the test. The principle of data processing is the same as before. (4) Assign values ​​to A, B, and n. Assign the values ​​of A, B, and n obtained by the calibration system to the control matrix. The system will input these three coefficients into the fitting function to determine the correspondence between wind speed and hot wire output voltage signal. During the speed test, the voltage signal output by the hot wire can be directly converted into wind speed value through the fitting function. (5) Display of the time change of the signal to be measured. (6) Control of data acquisition and storage. 6. Test and results of virtual instrument For a 1:10 model of a domestic car, this virtual instrument system was used to conduct a test study on the speed distribution of the car wake and compared with the results of our previous study to verify the effectiveness of the virtual instrument. 6.1 Test Content Measurement Cross Sections: The velocity distribution of the wake velocity was measured at four cross sections. The model height of the vehicle was h=126mm. The four surfaces were located at distances of 1h, 2h, 3h, and 4h from the rear of the vehicle, corresponding to x/h=1, 2, 3, and 4, respectively. The positions of the four cross sections are shown in Figure 9. Figure 9: Position diagram of the four measurement cross sections Measurement point arrangement: The measurement cross section was divided into six heights. The first height was 5mm from the floor surface, and the first to fifth heights were evenly spaced at 35mm intervals. There were 7 measurement points in the cross section width, all spaced at 35mm intervals, resulting in a total of 42 measurement points per cross section. The measurement point arrangement considered both the orthographic projection area of ​​the vehicle model and the adjustment range of the three-dimensional coordinate measuring frame. Test wind speed: 20m/s. Blockage ratio: 6%. 6.2 Test Results Table 1: Velocity results measured by the virtual instrument system The airflow velocity in the wake was measured and processed using the LabVIEW virtual instrument system according to the correction formula. The experimental velocity values ​​of each point in the wake measurement cross section at a distance x=h from the rear of the car, i.e., x/h=1, are shown in Table 1. After further data processing, the velocity distribution contour map of the wake cross section is obtained, as shown in Figure 10a. Figure 10 Velocity distribution map in the four wake cross sections The experimental results also show that when the yaw angle of the car is zero degrees, the velocity distribution on both sides of the wake relative to the center has good symmetry. As can be seen from the figure, the stepback car has a similar velocity distribution in its wake cross section, forming two laterally symmetrical drag vortices at the rear edge of the trunk lid, which is consistent with the numerical calculation results and the experimental results obtained in previous studies. 7. Conclusion (1) In view of the various characteristics and drawbacks of traditional instruments, the concept of virtual instruments is introduced into the automotive wind tunnel, and a virtual instrument system based on hot-wire anemometer is developed specifically for automotive wake speed measurement. This system can be easily maintained, expanded and upgraded. (2) A wake speed measurement test was conducted on a domestically produced sedan using this virtual instrument system. The test results were consistent with our previous numerical calculation results and experimental results, proving that using this virtual instrument for car wake speed measurement is effective and feasible. (3) Applying this virtual instrument system to measure car wake speed and then studying measures to improve the wake structure is of great significance for improving the aerodynamic characteristics of automobiles, developing low-drag domestically produced automobiles, and reducing automobile fuel consumption.