Design of Linear CCD Micron-Level Round Steel Photoelectric Diameter Measuring Instrument

　　introduction

　　There are many traditional methods for measuring the diameter of round steel in the steel industry, such as sampling detection using sizing tools and electromagnetic contact measurement. Measuring with sizing tools is not accurate enough and the speed is very slow; electromagnetic measurement is a contact measurement, which is more troublesome and the accuracy and speed are difficult to guarantee. CCD charge-coupled devices are widely used in photoelectric detection and photoelectric imaging fields such as image scanning, non-contact sizing detection, displacement measurement, barcode reading, etc., and have the advantages of self-scanning, high accuracy, high sensitivity, and wide spectral response [1]. The CCD micron-level non-contact round steel photoelectric diameter gauge is a non-contact photoelectric diameter measuring device for round steel based on CCD photoelectric detection technology. It has the advantages of non-contact, fast speed, high accuracy, miniaturization, and simple structure. It can dynamically and non-destructively monitor the slight changes in the diameter of round steel at any time on the production line under harsh conditions such as dark light, high temperature, and high speed, and has high practical value.

　　1 Overall system design

　　1.1 Measuring principle of CCD micron-level non-contact round steel photoelectric diameter gauge

　　The optical path measurement principle of the CCD micron-level non-contact round steel photoelectric diameter gauge is shown in Figure 1.

　　The parallel light source emits a parallel light beam from the mirror L1. This light beam irradiates the round steel workpiece in the optical path and is imaged on the photosensitive surface of the CCD through the optical lens L2. The CCD device converts the light signal on the photosensitive surface into a charge proportional to the light intensity, which is reflected in the output signal U0 waveform of the CCD under the drive of a clock pulse of a certain frequency. Signal processing is performed on U0, and the actual size of the workpiece is calculated and calibrated based on the width of the workpiece image in the CCD output waveform.

　　1.2 Main signal processing process

　　The relationship between the CCD line synchronization pulse FC and the standard pulse SP and the output signal U0 is shown in Figure 2(a). After the workpiece is placed, the U0 waveform portion in the middle of the line synchronization pulse FC becomes a low level, and the width of the low level changes linearly with the diameter of the workpiece, as shown in Figure 2(b).

　　Figure 2 CCD output signal

　　The signal U0 is processed in each row pulse FC cycle, and the process is shown in Figure 3. The U0 signal is passed through a low-pass filter circuit to filter out high-frequency interference; U0 is subjected to a primary differential boundary separation, and then the signal is uniformly processed through an absolute value circuit to facilitate secondary differential processing; then secondary differential processing is performed to improve the resolution, and then the measurement center is found through a zero-crossing detection circuit, and finally binarization processing is performed to prepare for subsequent pulse counting.

　　Figure 3 The process of signal processing U0 during the FC period of the line pulse

　　1.3 System construction plan

　　The system construction plan is shown in Figure 4.

　　Figure 4 Overall system construction plan

　　The light source box is composed of a high-brightness LED and a lens with a suitable focal length, and its characteristics are used to generate good parallel light, which illuminates the object and then forms an image on the CCD through the optical lens. The output signal of the CCD is transmitted to the integration time adjustment and signal processing circuit module through the 9-pin serial port, and then this module transmits the processed signal to the counting and communication circuit module for counting measurement conversion, and communicates with the display module to send the measured value to the ARM processor. Finally, the intelligent display terminal displays the measured value and realizes functions such as calibration and calibration query. [page]

　　2 System Implementation

　　2.1 Integration time adjustment and signal processing circuit

　　2.1.1 Hardware Design

　　The structural block diagram of the integration time adjustment and signal processing circuit is shown in Figure 5.

　　Figure 5: Block diagram of the integration time adjustment and signal processing circuit

　　Since the output signal U0 of CCD is affected by the light intensity, the stronger the light intensity, the larger the amplitude of U0 waveform, so it is necessary to adjust the integration time of CCD in a closed loop to ensure that the maximum amplitude of signal U0 is within the range of 3 to 4 V. The waveform of U0 is compared with 3 V and 4 V voltages through the dual comparator LM393, and the comparison result is input into the single-chip microcomputer AT89C2051. The single-chip microcomputer sets the integration time setting position M0 to M3 of the CCD driver through four data lines according to the result (where 0000 is the shortest integration time and 1111 is the longest integration time) to ensure that there is a suitable integration time so that the maximum amplitude of U0 is within the required range, which is convenient for accurate measurement. After the integration time is adjusted, the line synchronization pulse FC is output through the AND gate control. U0 is processed by filtering, first differentiation, filtering, absolute value, amplification, second differentiation, filtering, and level adjustment built by four dual op amps LM353, and then compared with 0 V by LM393 comparator for zero crossing detection. The signal is input into the microcontroller AT89C2051 for software binarization, and then the signal is output.

　　2.1.2 Software Design

　　The program flow of integration time adjustment and signal processing is shown in Figure 6.

　　Figure 6 Integration time adjustment and signal processing program flow

　　When there is external light interference in the system, the integration time needs to be adjusted in real time. The program uses the line synchronization pulse FC as the interrupt source. In the line synchronization pulse FC, it is constantly judged whether the amplitude of U0 is within the range of 3 to 4 V. If not, the values ​​of M0 to M3 are adjusted immediately until the amplitude of U0 is appropriate. At this time, the line synchronization pulse FC is output through the AND gate control, and the zero-crossing detection signal is binary-valued and output.

　　2.2 Counting and Communication Module

　　2.2.1 Hardware Design

　　The structural block diagram of the counting and communication module is shown in Figure 7.

　　Figure 7 Counting and communication module structure diagram

　　The single-chip computer AT89C2051 receives the signal from the integration time adjustment and signal processing module. By counting the standard pulses within the line synchronization pulse FC cycle, the number of standard pulses within the two boundaries of the U0 waveform workpiece imaging can be obtained. The relationship between the standard pulses and the measured standard size of the workpiece can be calibrated to obtain the actual size of the workpiece. The measured value can be temporarily stored in the off-chip RAM expanded by 74LS373 and DS1225, and then sent to the display module through the RS232 serial port. [page]

　　2.2.2 Software Design

　　The block diagram of counting and communication program in MCU is shown in Figure 8.

　　Figure 8 Counting and communication program flow in MCU

　　There are serial port interrupt and external interrupt 0 in the program. Set the serial port interrupt as a high priority interrupt. The receiving interrupt of the serial port receives the control instruction from the ARM microprocessor in the display module to determine whether to start measurement, storage or query; the sending interrupt of the serial port sends the measured value to the ARM microprocessor for display. The falling edge of the line synchronization pulse FC is used as the trigger signal of external interrupt 0. After the falling edge of FC arrives and generates an interrupt, the standard pulse counting begins. When the binary signal Q changes from high level to low level, the number of standard pulses N1 is recorded at this time. When the binary signal Q changes from high level to low level, the counting stops and the number of standard pulses N2 is recorded at this time; N=N2-N1. The measured value is calculated according to the coefficient obtained by calibration and converted into ASCII code and temporarily stored in the external RAM for direct display or query.

　　2.3 Display Module

　　The ARM microprocessor receives the touch key operation information and analyzes the operation to be performed, then sends instructions to the counting communication module through the serial port, and receives the measurement information from the counting communication module through the serial port. The measurement information is displayed through the LCD touch screen, and the measurement information and operation interface can also be stored through Flash/RAM.

　　The display module software design process is shown in Figure 9.

　　Figure 9 Display module program flow

　　The program continuously scans the touch keyboard and waits for the serial port interrupt. If a key is pressed, the corresponding instruction is sent to the serial port to control the counting module to measure; if there is a serial port interrupt to determine the corresponding word head, the LCD display is controlled or the system settings are updated.

　　3 System implementation effect

　　The system implementation effect is shown in Figure 10.

　　Figure 10 System implementation effect

　　The overall effect of the system is good, the volume is small and it is non-contact measurement. The measurement accuracy and speed basically meet the requirements of micron-level measurement, and the error is between ±0.005 mm, which is higher than the national production measurement accuracy standard; the human-machine interface is friendly, which can well meet the requirements of static or dynamic measurement of round steel in the production process.

　　Conclusion

　　This paper proposes a design scheme for a linear array CCD micron-level non-contact round steel photoelectric diameter measuring instrument. With ARM microprocessor and single-chip microcomputer as the core, it solves the limitation of contact measurement of the system in the traditional round steel diameter measurement method, and has the characteristics of simple structure, miniaturization, non-contact and high precision. The experimental results show that the system realizes CCD non-contact round steel photoelectric diameter measurement, the measurement results are accurate, high precision, good stability, and the measurement results can be displayed directly and conveniently. The system has been applied to high-standard detection of round steel production in steel plants and has high practical application value.