Design of spectrum data acquisition and analysis based on LabVIEW and CCD

This paper introduces the emerging virtual instrument technology and designs a spectrum analysis and data acquisition system based on LabVIEW. Through software writing, the collected signals are filtered again and the gain adjustment function is added. The least square method is used to realize the wavelength calibration of the system and the peak search function of the spectrum curve. Compared with the traditional linear calibration method, the measurement accuracy is further improved. Finally, the experimental results show that the spectrum analysis system can be used to distinguish the characteristic spectrum lines of the mercury lamp spectrum and achieve the purpose of spectrum analysis.

1 Introduction

With the development of science and technology and the extensive research on spectral analysis systems, people have put forward higher and higher requirements on the main indicators of spectral analysis systems, such as spectral measurement range, resolution, accuracy, etc. The current development direction of spectrometers is miniaturization, automation and high precision. Therefore, this paper introduces the emerging virtual instrument technology and designs a spectral analysis and data acquisition system based on LabVIEW, which improves the overall performance of the spectral analysis system, and is simple to operate and powerful.

2. System Design

2.1 System Structure

According to the workflow of the spectrum analysis acquisition system, the entire system is divided into three parts: optical system design, hardware design, and application design. The spectrum analysis system is a typical photoelectric detection instrument based on photoelectric detectors for measurement. Therefore, combined with the design requirements of the system, in order to meet the requirements of miniaturization and low cost, the use of linear array CCD detectors is considered. The light separated by the grating is collected and received by the TCD1304AP linear array CCD, amplified and filtered by the conditioning circuit and sent to the A/D analog-to-digital converter, which converts the analog signal into a digital signal. Finally, the data collected by PXI-6289 is sent to the host computer, and LabVIEW is selected to receive data, display and analyze.

2.2 Spectral system

The spectrometer system is a key component in the spectrum analysis system, which directly determines the spectroscopic performance of the system. There are many spectroscopic methods for spectrometers. According to the different spectral measurement methods, they are divided into filter spectrometry, prism spectrometry, and grating spectrometry. This article uses the grating spectrometry method. This design has a simple structure, low light loss, high resolution, and high signal-to-noise ratio. The popular CaernyTurner spectrometer system is a very compact optical system that simplifies the optical path as much as possible.

2.3 Data Collection

The data acquisition card PXI-6289 from NI is used. It can be directly inserted into the PXI slot of the industrial computer and is plug-and-play. PXI-6289 is a high-performance multi-function data acquisition card equipped with a 2.0GHz dual-core processor, with a maximum conversion speed of 40kHz and provides users with 32 analog input channels. Since LabVIEW provides a driver for it, the collected data can be directly called by the acquisition function in the software, which is very convenient.

According to the system index requirements, the design chooses to use TCD-1304AP. Since the output analog signal is relatively weak, usually only a few hundred millivolts, in order to obtain high-quality output signals and high system signal-to-noise ratio in practical applications, the output signal of the CCD must be conditioned and amplified first, and then amplified through an emitter follower, and the noise must be suppressed to a certain extent to filter out the interference caused by dark current and low-noise signals to the maximum extent. The analog signal output by TCD1304AP is sent to the differential circuit built with CLC409 for processing.

2.4 Data processing

Due to the existence of factors such as intrinsic noise and optical path noise in the data acquisition process, the signal-to-noise ratio of the system is directly affected. Therefore, the obtained sample spectral information must be filtered out.

According to the signal processing direction, filters can be divided into two types:

Analog filters and digital filters. The input and output of traditional analog filters are continuous, while the input and output of digital filters are discrete time signals. In LabVIEW, what we need to study are digital filters (the signals processed in the computer are all digital signals).

(l) Cumulative average of multiple measurement results

When filtering the raw data, we use the method of multiple measurements, accumulating the measurement results and finally taking the average value. The number of measurements can be set by the user according to his or her own requirements. Assuming the number of measurements set by the user is n times, the result is:

(2) Smoothing filtering of spectral images

Different from the above average filtering method, the smoothing filtering method discussed here is to smooth a certain data in space. For N data collected in a certain measurement, we use the following smoothing filtering formula:

Among them, S is the processing neighborhood of x1, and m is the number of data contained in S. For example, if m is 7, then 3 data are taken before and after each processed data for smoothing filtering.

2.5 Software Design

The overall layout of the front panel is designed based on the user's needs for interface operation.

The main interface of the system is shown in Figure 1:

The left part of the interface is the control panel of the system, which mainly controls certain tasks of the execution system, such as automatic search of wave peaks, retrieval of wavelengths and display of values. The middle part of the system is the light intensity amplitude corresponding to each pixel of the CCD. The right part of the system mainly realizes the display of data, the correspondence between light intensity value and pixel, and the correspondence between light intensity value and calibrated wavelength. Therefore, the display function of spectral image is realized.

The source code to realize the functions in the above control panel is written in a graphical programming language. The block diagram contains various graphical functions, constants, variables, algorithm structures and connections. After receiving the data input by the user from the front panel, it will run the relevant instruction parameters inside the device to perform relevant operations. According to the compiled program, it traverses the program in a data flow manner and finally gives the running results to the user.

3. System testing and analysis

This section uses the characteristic spectrum calibration method to calibrate the spectrum accordingly and find the pixel position corresponding to the specific spectrum line on the CCD. In the test, a low-pressure mercury lamp is used to complete the relatively easy spectrum calibration. The mercury lamp has six characteristic spectrum lines: 366.5nm, 404.66nm, 435.8nm, 546.7nm, 576.96nm, 579.96nm. Find the CCD pixel numbers corresponding to several spectrum curves related to the peak wavelength under the reference spectrum, and complete the fitting of the third-order polynomial by using the least squares method. If these six spectrum lines are y1, y2, y3, y4, y5, y6, the pixel numbers associated with the six spectrum lines collected are x1, x2, x3, x4, x5, x6. Due to the narrow selection of the spectrum width and other reasons such as image curvature, the third-order polynomial fitting using the least squares method has a smaller error than the linear fitting. Suppose the fitting curve equation is y=p(x), then:

According to the least squares principle, the coefficient solution of the least squares equation can be obtained, from which the fitting equation can be determined, and then the calibration and measurement of the characteristic spectral lines can be carried out.

4 Conclusion

This paper selects linear array CCD as the photoelectric detection device to achieve wide spectrum and high resolution measurement requirements. LabVIEW combined with PXI acquisition card is used to realize spectrum data acquisition, analysis and display functions, which has the characteristics of short development cycle, simplicity and practicality, good stability, and can meet various spectrum analysis requirements.