A method of processing spectral data

Abstract

The present application discloses a kind of spectrum data processing methods, the range of the abscissa value of the spectrum curve studied is divided into two subintervals, the spectrum area of subinterval is compared with ideal isospectral operation, the result of comparison operation has two forms of non-equilibrium output, balanced feedback output, both can be used to finely characterize the environmental sensitivity of spectrum;Because the spectrum data processing process of the method is similar to the form of Wheatstone bridge to realize resistance-voltage conversion, it is named as bridge-type analysis method of spectrum.The present application is generally applicable to the analysis of various shapes of spectrum such as discrete linear spectrum, band spectrum and continuous spectrum;The present application can improve the efficiency of spectrum data processing, reduce the difficulty of reading spectrum, and has higher precision than the conventional analysis method of spectrum peak such as direct reading and fitting.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of chemical metrology, and relates to a method for analyzing a spectrum, named as a bridge-type analysis method for a spectrum, and particularly relates to a spectrum data processing method. BACKGROUND

[0002] Spectrum, microwave spectrum, electromagnetic resonance spectrum, mechanical vibration spectrum and other types of spectrum technology are effective means for studying the composition and structure of a substance, and even the material and structure of an object. Environmental factors can cause changes in the composition or structure of a substance, and thus spectrum technology can also be applied to environmental parameter sensing.

[0003] When spectrum is used for analyzing the composition and structure of a substance, the spectrum used should have fingerprint characteristics, that is, a series of characteristic spectral lines at corresponding positions on the horizontal axis; when the spectrum is greatly affected by the environment or the spectral lines are very dense, the spectral lines are widened into bands; and a larger range of indistinguishable spectral lines form a continuous spectrum, the characteristic of which is generally the position of the cutoff edge. When environmental factors cause changes in the spectral characteristics, the general detection is the peak frequency shift of the spectral line, the movement of the cutoff edge, the change in the spectral profile such as the relative intensity change of the spectrum curve, and the change in the line width or bandwidth. For a same group of spectra continuously changed by a control variable, each spectral characteristic has its own physical meaning, and different spectral characteristics do not necessarily have consistent change rules, and even some spectral characteristics are difficult to extract or have very low reading accuracy. For example, in experiments, spectral data processing often needs to determine spectral profile parameters including bandwidth and spectral peak position through peak fitting, but wideband spectrum is usually difficult to accurately fit due to the complexity of the widening mechanism, and thus the accuracy of directly reading or locating the spectral peak by using the derivative method is limited by the resolution of the spectrum instrument and the signal-to-noise ratio. When the subtle changes in the spectrum are difficult to accurately characterize, the spectrum can no longer be used for environmental parameter sensing.

[0004] The present application proposes a spectrum data processing method for accurately characterizing a spectrum: dividing a range of values of a horizontal axis on a spectrum curve into two subintervals, and comparing the spectrum areas of the subintervals with ideal equipotential spectra. The comparison operation process is similar to the way in which a Wheatstone bridge realizes resistance-voltage conversion, and the results of the comparison operation have two forms of unbalanced output and balanced feedback output, which reflect the same change in the spectral profile. The spectral characteristics given by the method are intuitive and easy to calculate, which improves the efficiency of spectrum data processing and reduces the difficulty of reading the spectrum, and has higher accuracy than the conventional analysis method of directly reading or fitting the spectral peak. Because the spectrum data processing process of the present application is similar to the form of resistance-voltage conversion realized by a Wheatstone bridge, it is named as a bridge-type analysis method for a spectrum. SUMMARY

[0005] The present application aims to provide a spectrum data processing method, which is a spectrum data processing method for transforming the intensity distribution of a spectrum into other spectrum features, and is universally applicable to the analysis of various types of spectrum with shapes such as discrete linear spectrum, band spectrum and continuous spectrum, and can improve the efficiency of spectrum data processing, reduce the difficulty of spectrum reading and improve the accuracy of spectrum reading.

[0006] The technical solution of the present application is as follows: a spectrum data processing method, characterized in that the range of the abscissa values of the spectrum curve to be studied is divided into two subintervals, and the spectrum areas of the subintervals are compared with an ideal isocapacity spectrum; the comparison result has two forms: non-equilibrium output and balanced feedback output; the operation process is similar to the resistance-voltage conversion mode of a Wheatstone bridge; and the specific steps are as follows:

[0007] Step 1: the range of the abscissa values of the spectrum curve to be studied is divided into two non-overlapping subintervals, and the spectrum areas of the subintervals are S1 and S2, respectively;

[0008] Step 2: an ideal isocapacity spectrum is inserted in the subinterval range divided in step 1, and the spectrum area of the ideal isocapacity spectrum is S1+S2, which is denoted as 2*S0;

[0009] Step 3: S1, S2, S0 and S0 are four spectrum areas corresponding to the four resistance arms of a Wheatstone bridge, and the spectrum areas are converted into other spectrum features that can represent the environmental sensitivity of the spectrum according to the resistance-voltage conversion relationship of the Wheatstone bridge; the bridge has two forms: non-equilibrium output and balanced feedback output, and the non-equilibrium output corresponds to the spectrum feature (S1-S2) / (S1+S2), and the balanced feedback output corresponds to the balanced condition S1=S2 as the spectrum feature.

[0010] The spectrum curve is a discrete linear spectrum, a band spectrum or a continuous spectrum.

[0011] The two spectrum features (S1-S2) / (S1+S2) and the balanced condition S1=S2 have equivalence, reflect the same spectrum profile change, and can be respectively analogized to the two resistance-voltage conversion modes of the Wheatstone bridge: non-equilibrium output and balanced compensation feedback output.

[0012] The non-equilibrium output spectrum feature (S1-S2) / (S1+S2) can be further simplified as S1 / S2 as a spectrum feature representing the spectrum.

[0013] The essence of the present application is to transform the full profile features of a spectrum curve, which are usually difficult to utilize and have not been utilized, into spectrum feature quantities that are easy to measure and have high measurement accuracy, just as the Wheatstone bridge transforms the resistance value change, which is difficult to accurately measure, into an easy-to-measure voltage signal with high measurement accuracy.

[0014] The present application has universal applicability and can be used to analyze discrete linear spectrum, band spectrum and continuous spectrum. The present application gives the equivalent relationship between two different wave spectrum characteristics, improves the efficiency of wave spectrum data processing and reduces the difficulty of reading spectrum. The present application has higher precision than the conventional processing method of wave spectrum peak data such as direct reading and fitting. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 For the analysis of wave spectrum data characteristic parameters and the analogy of circuit conversion mode, the horizontal axis of the example wave spectrum is wavelength λ, the vertical axis is intensity, the dotted line is the dividing line for dividing the wave spectrum into two subintervals, and the dash-dot line is the ideal equal-energy wave spectrum virtually drawn.

[0016] Figure 2 For a group of results of force-sensitive fluorescence spectrum, the horizontal axis is wavelength, and the vertical axis is intensity normalized by peak value.

[0017] Figure 3 For Figure 2 The results of the bridge analysis method of the spectrum non-equilibrium output, the horizontal axis is compressive stress, and the vertical axis is the wave spectrum characteristic (S1-S2) / (S1+S2) of the non-equilibrium output, which is further simplified to S1 / S2.

[0018] Figure 4 For Figure 2 The results of the bridge analysis method of the spectrum balanced compensation feedback output, the horizontal axis is compressive stress, and the vertical axis is the position of the dividing line calculated under the balanced compensation feedback condition S1=S2. DETAILED DESCRIPTION

[0019] The technical content part of the present application describes a method for analyzing the wave spectrum, which is similar to the way in which the Wheatstone bridge converts the change of resistance value into voltage output signal, and converts the profile characteristics of the wave spectrum into two equivalent wave spectrum characteristic quantities, namely non-equilibrium output and balanced compensation feedback output, which are easy to measure and have high measurement precision. The technical method is described as follows:

[0020] For the measured wave spectrum data, the wave spectrum curve is drawn in the two-dimensional plane coordinate system.

[0021] Determine the horizontal axis value range to be studied on the wave spectrum curve, and divide the range into two subintervals, for example Figure 1 a, Figure 1 b, Figure 1 c shows a division method of continuous spectrum, band spectrum and discrete linear spectrum, and the dotted line in the figure is the dividing line for dividing the subintervals.

[0022] Virtually draw an ideal equal-energy wave spectrum on the wave spectrum graph, as shown in Figure 1 a, Figure 1 b, Figure 1The dotted line in c.

[0023] Wheatstone bridge circuit for realizing resistance-voltage conversion Figure 1 d By analogy, the spectral areas S1, S2 of each sub-interval correspond to R1, R2 in the circuit, and the two spectral areas S0, S0 on the ideal equipotential spectrum correspond to R0, R0 in the circuit. Figure 1 d The working mode of the unbalanced output is that when R1+R2=2R0, V / V0=ΔR / 2R0, where ΔR is the difference between R1, R2 and R0. The corresponding spectral characteristic value is (S1-S2) / (S1+S2)=ΔS / 2S0, where ΔS is the difference between S1, S2 and S0. For the convenience of calculation, (S1-S2) / (S1+S2) is simplified, and S1 / S2 is directly calculated. Obviously, it reflects the contour shape of the spectrum as (S1-S2) / (S1+S2).

[0024] As Figure 2 shown in the case, the conventional spectral data processing method cannot distinguish the spectral shift or spectral contour change caused by stress. Using the unbalanced output (S1-S2) / (S1+S2) of the bridge analysis method to simplify the spectral characteristic S1 / S2 as the observation, the compressive stress response curve shown in Figure 3 is obtained.

[0025] The principle circuit of the Wheatstone bridge with balanced compensation feedback output is as follows Figure 1 e To achieve balance, the contact point of the galvanometer G and the resistance R1+R2 needs to be located at the position of the balance condition R1=R2. If any of R1 and R2 changes, the contact point P needs to move to compensate for the resistance change and maintain the balance condition. By analogy, it corresponds to the spectrum, Figure 1 a、 Figure 1 b、 Figure 1 c The spectral areas S1, S2 of each sub-interval still correspond to R1, R2 in the circuit; taking S1=S2 as the balance condition, the position of the sub-interval dividing line dividing the total area S1+S2 is taken as the characteristic quantity of the spectrum. This characteristic quantity moves with the change of the spectral contour, and the compressive stress response curve shown in Figure 2 is obtained by data processing. Figure 4 As a comparison, Figure 4 in which the peak positions corresponding to different compressive stresses obtained by the conventional spectral peak search analysis are also plotted. Figure 2 It can be seen that the peak position data is discrete and irregular.

[0026] A spectrum data processing method, which divides the range of the abscissa value of the studied spectrum curve into two subintervals, and compares the spectrum area of the subintervals with an ideal isometric spectrum, the comparison result has two forms of non-equilibrium output and balanced feedback output, both of which can be used to finely characterize the environmental sensitivity of the spectrum; because the spectrum data processing process of the method is similar to the form of Wheatstone bridge realizing resistance-voltage conversion, it is named as bridge-type analysis method of spectrum. The present application is universally applicable to the analysis of various types of spectrum such as discrete linear spectrum, band spectrum and continuous spectrum; the present application can improve the efficiency of spectrum data processing and reduce the difficulty of reading spectrum, and has higher precision than the conventional analysis method of spectrum peak such as direct reading and fitting.

[0027] Compared with the conventional spectrum peak shift analysis, the two spectrum characteristics of non-equilibrium output and balanced compensation feedback output of the bridge-type analysis method of the present application can significantly improve the experimental precision and reduce the error level, and the superiority of the bridge-type spectrum analysis can be analogized to the precision of the resistance value detected by the Wheatstone bridge being superior to the direct measurement of resistance by the multimeter.

Claims

1. A method of processing spectral data, characterized by, The range of the horizontal axis of the spectrum curve is divided into two subintervals, and the energy spectrum of the subintervals is compared, and the comparison result has two forms of non-equilibrium output and balanced feedback output, and the specific steps are: Step 1, the range of the horizontal axis of the spectrum curve is divided into two non-overlapping subintervals, and the spectrum areas of the subintervals are S1 and S2 respectively; Step 2, a virtual energy spectrum is inserted in the subinterval range divided in step 1, and the spectrum area of the energy spectrum is S1+S2, denoted as 2*S0; Step 3, S1, S2, S0 and S0 are four resistance arms corresponding to the Wheatstone bridge, and the spectrum area is converted into other spectrum characteristics representing the environmental sensitivity of the spectrum according to the resistance-voltage conversion relationship of the Wheatstone bridge, and the bridge has two forms of non-equilibrium output and balanced feedback output, and the non-equilibrium output spectrum characteristic (S1-S2) / (S1+S2) and the balanced feedback output balanced condition S1=S2 are obtained as the spectrum characteristics.

2. A method of processing spectral data according to claim 1, wherein, The spectrum curve is a discrete line spectrum, a band spectrum and a continuous spectrum.

3. The method of claim 1, wherein The non-equilibrium output (S1-S2) / (S1+S2) and the balanced feedback output balanced condition S1=S2 have equivalence, reflect the same spectrum profile change, and can be compared with the non-equilibrium output and the balanced compensation feedback output of the Wheatstone bridge respectively.

4. The method of claim 1, wherein The non-equilibrium output spectrum characteristic (S1-S2) / (S1+S2) can be further simplified to S1 / S2 as a spectrum characteristic representing the spectrum.

Images