High-precision anti-interference cold mirror dew point measurement method based on semiconductor refrigeration

Abstract

The application discloses a high-precision anti-interference dew point measurement method of a cold mirror based on semiconductor refrigeration. The device used in the measurement method mainly comprises an optical measurement module, a temperature measurement module and a control processing unit. The method cools the mirror surface through a semiconductor refrigerator, and the control processing unit integrates a self-adaptive PID algorithm to dynamically adjust the refrigeration power. The light emitted by a collimated light source is reflected by the mirror surface and is received by a reflected photoelectric sensor. Through the change of reflectivity, the reflected light intensity is dynamically compensated based on a reference light signal to eliminate common-mode interference. The measurement method comprises the following steps: initializing and recording the baseline of multiple light signals; controlling the stepwise cooling of the mirror surface and synchronously monitoring; calculating and processing the light signal in real time, determining the dew starting based on a dynamic threshold; switching to PID control to finely balance the temperature near the dew point, and recording the stable value; correcting and outputting the final dew point value by using a compensation model; stopping the refrigeration and evaporating and resetting the mirror surface by using the residual heat, and entering the next cycle.

Description

Technical Field

[0001] This invention relates to the field of atmospheric environment monitoring and precision instrument technology, specifically to a cold mirror dew point meter for measuring gas dew point temperature and its measurement method. Background Technology

[0002] Dew point temperature is a crucial parameter for measuring the absolute humidity of a gas, with wide applications in meteorological observation, industrial process control, laboratory analysis, and aerospace. Cold mirror dew point meters, due to their direct measurement principle, high accuracy, and good long-term stability, are considered one of the benchmark methods for dew point measurement. However, traditional cold mirror dew point meters still face several technical challenges in practical applications: Mirror contamination interference: Long-term exposure of the mirror to the gas being measured makes it highly susceptible to adsorbing dust, oil, and other contaminants from the air, leading to unpredictable drift in the reflectivity baseline and severely affecting the accuracy of condensation detection, even resulting in misjudgments. Frequent manual cleaning and maintenance significantly impact the continuous operation capability of the equipment. Ambient light and light source fluctuation interference: Accidental intrusion of ambient light or aging and fluctuations in the light source itself can be received by the photoelectric sensor, introducing measurement noise. In low dew point (extremely dry) measurements, this interference may prevent the accurate detection of subtle changes in reflectivity. The condensation criterion is singular: traditional methods often use a fixed threshold for reflectivity change as the condensation criterion. However, the dynamic process of condensation formation differs under different humidity ranges and cooling rates, resulting in poor adaptability of fixed thresholds and a tendency to make premature or delayed judgments, affecting accuracy. Temperature measurement and system errors, particularly the accuracy and location of mirror temperature measurement, directly impact the final result. Thermal resistance, self-heating effects, and the thermal inertia of the thermoelectric cooler at the sensor's installation location can introduce systematic temperature measurement errors. Traditional cold mirror dew point meters often suffer from insufficient heat dissipation design, leading to excessively high hot-end temperatures of the thermoelectric cooler, preventing the establishment of an effective temperature difference and limiting their measurement lower limit and reliability under high ambient temperatures. Therefore, there is an urgent need for a highly reliable cold mirror dew point measurement device and method that is resistant to contamination and interference, intelligently judges, and automatically compensates for system errors. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a high-precision, anti-interference, cold mirror dew point measurement device and method. This device, through multiple innovative designs, significantly improves measurement accuracy, long-term stability, and automation in complex environments. To achieve the above objectives, the core innovation of this invention lies in integrating a three-in-one technical solution of passive protection, active anti-interference, and intelligent algorithm compensation. First, at the passive protection level, this invention adds a mirror anti-fouling coating to the core reflective mirror surface. This coating uses a hydrophobic, oleophobic, and optically transparent material (such as silica), which effectively prevents water droplets, oil droplets, and most particulate contaminants from directly adhering to the mirror substrate, making it easier for contaminants to be carried away by flowing gas, greatly extending the mirror's cleaning cycle and maintaining the long-term stability of the reflectivity baseline. Second, at the active anti-interference level, this invention innovatively introduces a reference optical path module. In addition to the reflective photoelectric sensor that measures the intensity of reflected light, a reference photoelectric sensor is added to directly or indirectly monitor the light source output or ambient background light. The control processing unit dynamically cancels out common-mode interference caused by fluctuations in light source intensity and changes in ambient light by calculating the difference between the signals from two sensors. This extracts the reflectivity change signal purely caused by condensation on the mirror surface, significantly improving the signal-to-noise ratio, especially when measuring low dew points. Finally, at the intelligent processing level, the control processing unit is the brain of this invention. It performs the following key tasks: dynamic condensation determination, no longer relying on a fixed threshold, but calculating the rate of change (derivative) of the compensated reflected light intensity signal in real time. When the rate of change first exceeds a dynamically set threshold based on historical data, it is determined as the condensation initiation point. This method is more sensitive and accurate in responding to weak condensation signals. Intelligent temperature control and compensation: A temperature control strategy of "rapid pre-cooling + precise balancing" is adopted. Simultaneously, a temperature drift compensation model for the cold-end temperature sensor is built-in. This model, obtained through experimental calibration, can perform real-time software compensation for the measured mirror temperature, eliminating systematic temperature measurement errors.

[0004] This invention is achieved through the following technical solution: a high-precision anti-interference cold mirror dew point measuring device based on semiconductor cooling, comprising: a measuring cavity 1, having a channel that allows the gas to be measured to flow through; a cavity temperature sensor 2 disposed within the measuring cavity 1; a reflective mirror 3, made of a high reflectivity material, disposed within the measuring cavity 1; a semiconductor cooler 4, the cold end of which is thermally connected to the reflective mirror 3 for adjusting the temperature of the reflective mirror 3; a cold end temperature sensor 5, embedded in the reflective mirror 3 for precisely measuring the temperature of the reflective mirror 3; and an optical measuring module, comprising a collimating light source 6 and a reflective photoelectric sensor 7, wherein the light emitted by the collimating light source 6 is reflected by the reflective mirror 3 and received by the reflective photoelectric sensor 7, for detecting changes in the reflectivity of the reflective mirror 3 to determine the condensation state.

[0005] It also includes: a scattering photoelectric sensor 8, disposed within the measuring cavity 1, with its photosensitive surface facing the reflective mirror 3, but not coplanar with the incident light path of the collimating light source 6 and the receiving light path of the reflecting photoelectric sensor 7, used to receive the scattered light signal generated by dew droplets when condensation forms on the reflective mirror 3. A reference light path module, including a reference photoelectric sensor 9 and a reflector 10, used to monitor the light intensity fluctuations of the collimating light source 6 or interference from ambient background light in real time. A control processing unit 11, electrically connected to the cold end temperature sensor 5, the reflecting photoelectric sensor 7, the scattering photoelectric sensor 8, the reference photoelectric sensor 9, and the semiconductor cooler 4, and configured to perform the following operations: real-time acquisition and fusion processing of the reflected light signal from the reflecting photoelectric sensor 7, the scattered light signal from the scattering photoelectric sensor 8, and the reference light signal from the reference photoelectric sensor 9; dynamic compensation of the reflected light intensity based on the difference between the reflected light signal and the reference light signal to eliminate common-mode interference. Based on the compensated reflected light intensity signal S... r The rate of change and the scattered light intensity signal S s A fusion judgment is performed to adaptively determine the starting point and stable state of condensation on the mirror surface. After condensation is determined, compensation calculations are performed based on the reading of the cold end temperature sensor 5 and the pre-stored historical temperature drift data model to output the final dew point temperature value.

[0006] The control processing unit 11 is configured to simultaneously acquire and process the reflected light intensity signal S obtained by the reflective photoelectric sensor 7. r and the scattered light intensity signal S obtained by the scattering photoelectric sensor 8 s Among them, the reflected light intensity signal S r The intensity of the scattered light signal S will decrease during condensation. s It will rise when condensation occurs.

[0007] The reflective mirror 3 is provided with a mirror anti-fouling coating, which covers the reflective surface of the reflective mirror 3. The coating is a hydrophobic, oleophobic and optically transparent thin film; the mirror anti-fouling coating is a hard optical coating made of silicon dioxide or magnesium fluoride.

[0008] It also includes a heat sink 12 and a fan 13 connected to the hot end of the thermoelectric cooler 4, with a hot end temperature sensor 14 embedded in the heat sink 12. The heat sink 12 functions as a highly efficient heat sink, dissipating the Peltier heat generated during the cooling process into the environment in a timely manner, thereby ensuring that the hot end temperature of the thermoelectric cooler 4 is maintained at a low and stable level.

[0009] The control processing unit 11 is configured to execute an adaptive temperature control strategy, using PID proportional-integral-derivative control to adjust the power supply current of the semiconductor cooler 4 and regulate the speed of the fan 13. During the pre-cooling stage, it uses maximum power cooling, and when it approaches the predicted dew point temperature, it switches to PID proportional-integral-derivative control to smoothly pass through the dew point and prevent overshoot.

[0010] The measuring cavity 1 is surrounded by an insulation layer 15 to prevent the temperature outside the measuring cavity 1 from affecting the measurement.

[0011] A high-precision, anti-interference cold mirror dew point measurement method based on semiconductor cooling includes the following steps: S1, initialization and baseline establishment: the gas to be measured flows through the measurement cavity 1 at a constant flow rate; the optical measurement module and the reference optical path module are activated, and the control processing unit 11 reads and records the initial value of the reflective photoelectric sensor 7 as the reflected light intensity baseline S. r 0, and simultaneously read and record the initial value of the reference photoelectric sensor 9 as the background light intensity baseline S. b 0, and read the initial value of the scattering photoelectric sensor 8 as the background scattered light baseline S. s 0.

[0012] S2. Active Cooling and Synchronous Monitoring: The control processing unit 11 controls the semiconductor cooler 4 to be powered on and starts the fan 13. The fan uses a stepped or linear cooling strategy to cool the reflective mirror 3 at a first power, causing its temperature to drop. Simultaneously, the mirror temperature T and the reflection photoelectric sensor value S are collected at a preset frequency. r Reference photoelectric sensor value S b and the value S of the scattering photoelectric sensor s .

[0013] S3. Signal Processing and Condensation Initiation Determination: Real-time calculation of the compensated effective reflected light intensity signal S eff Its calculation formula is: S eff = (S r -S r 0) - k (S b -S b 0), where k is a preset calibration coefficient obtained through experimental calibration. It quantifies the proportional relationship between the response of the main optical path (reflected optical path) and the reference optical path to common-mode interference (mainly light source fluctuations), ensuring the accuracy of compensation, and simultaneously calculating the effective reflected light intensity signal S. eff rate of change over time dS eff / dt. Calculate the increment ΔS of the scattered light signal. s =S s -S s 0, and simultaneously calculate the increment ΔS of the scattered light signal. s The rate of change of d(ΔS) with time t s ) / dt;

[0014] The control processing unit 11 is configured to determine the start of condensation on the mirror surface based on at least one of the following logics, and record the mirror surface temperature at this time as the initial dew point temperature T1:

[0015] a) Logic 1: When dS eff / dt exceeds the preset first dynamic threshold THR1 for the first time, and at the same time ΔS s Exceeding the preset scattered light intensity threshold (THR) s At this time, it is determined that condensation has begun;

[0016] b) Logic 2: When ΔS s rate of change d(ΔS) s When ) / dt first exceeds the preset second dynamic threshold THR2, it is determined that condensation has started.

[0017] S4. Temperature Balance and Stability Measurement: The control processing unit 11 switches the control mode of the semiconductor cooler 4 to PID proportional-integral-derivative control mode, finely adjusting its operating current to make the temperature of the reflective mirror 3 fluctuate within a preset small range near the initial dew point temperature T1 until an effective reflection signal S... eff Once the temperature stabilizes at a new steady-state value, the condensation state becomes stable. The average mirror temperature corresponding to this stable state is recorded as the stable dew point temperature T2.

[0018] S5. Temperature Compensation and Result Output: Substitute the stable dew point temperature T2 into the pre-stored temperature compensation model for calculation to obtain the corrected final dew point temperature value T. corrected (T) d It outputs the compensated dew point temperature value and immediately cuts off the power supply to the semiconductor cooler 4 to stop cooling.

[0019] S6. Condensation Evaporation and System Reset: After cooling stops, fan 13 also stops. The residual heat from radiator 12 is used to heat the hot end of semiconductor cooler 4, causing the temperature of its cold end and reflective mirror 3 to rise, resulting in condensation evaporation on the mirror. When the reflected light intensity signal S... r Reaching ≥ baseline reflected light intensity S r When the temperature difference between the cold end temperature sensor 5 and the cavity temperature sensor 2 is less than or equal to a preset temperature difference threshold, the system is determined to be reset to the ready state and automatically returns to step S2 to start the next measurement cycle.

[0020] In step S3, the effective reflected light intensity signal S eff rate of change over time dS eff / dt, its calculation formula is based on discrete-time signal systems, as follows:

[0021] Assuming the system acquires signals with a fixed sampling period Δt, at any nth sampling time:

[0022] S at the current moment eff The value is S eff (n), S at the previous sampling timeeff The value is S eff (n-1);

[0023] Then the rate of change dS at the current moment eff / dt can be approximated using a first-order backward difference, as shown in the formula:

[0024] ,

[0025] in:

[0026] dS eff / dt represents the effective reflected light intensity signal S eff The rate of change over time, in signal units per second, where the signal unit is Digital Unit (DU).

[0027] S eff (n) represents the effective reflected signal value obtained from the nth sampling calculation;

[0028] S eff (n-1) is the effective reflected signal value calculated in the (n-1)th sampling (i.e., the previous sampling);

[0029] Δt is a fixed sampling time interval, which is a fixed parameter of the system;

[0030] The increment of the scattered light signal ΔS s The rate of change of d(ΔS) with time t s The formula for calculating ) / dt is based on discrete-time signal systems, as follows:

[0031] ,

[0032] in:

[0033] d(ΔS s ) / dt is the rate of change of the scattered light signal increment with time, in signal units per second. The signal unit here is Digital Unit (DU).

[0034] ΔS s (n) represents the increment of the scattered light signal calculated at the nth sampling, ΔS s (n)=S s (n)-S s 0;

[0035] ΔS s (n-1) represents the increment of the scattered light signal calculated during the (n-1)th sampling (i.e., the previous sampling), ΔS s (n-1)=S s (n-1) -S s 0;

[0036] S s (n),S s (n-1): The original measurement values ​​of the scattering photoelectric sensor 8 during the nth and (n-1)th samplings;

[0037] Δt is a fixed sampling time interval, which is a fixed parameter of the system.

[0038] In step S3, the setting rule for the dynamic threshold is as follows: a higher first threshold is used during the rapid decrease of the mirror temperature, and a second threshold lower than the first threshold is used when the mirror temperature is close to the predicted dew point temperature.

[0039] The temperature compensation model described in step S5 is a polynomial compensation model:

[0040] ,

[0041] Where T2 is the stable dew point temperature, and a, b, c are coefficients obtained through high-precision experimental calibration. corrected This is the compensated dew point temperature value.

[0042] The beneficial effects of this invention are as follows: Compared with existing technologies, this invention features: high measurement accuracy and good long-term stability. The mirror-like anti-fouling coating physically protects the core optical interface and reduces baseline drift; the reference optical path eliminates major external light interference; and intelligent temperature control and temperature compensation models correct system errors. The combination of these three features ensures that the instrument maintains high accuracy across the entire measurement range, especially in low humidity conditions, and requires no manual calibration over the long term. It also boasts strong environmental adaptability and reliability. The dual-optical-path differential anti-interference design makes the instrument insensitive to changes in ambient light, enabling stable operation in more complex industrial environments. The adaptive condensation detection algorithm adapts to different humidity and cooling conditions, exhibiting stronger robustness. Furthermore, it possesses a high degree of intelligence and automation. From automatic condensation detection and adaptive temperature control to final data compensation output, the entire process requires no manual intervention, achieving true automated measurement and reducing the barrier to entry and maintenance requirements. This invention overcomes the inherent shortcomings of traditional on-off or proportional control, such as large temperature fluctuations, easy overshoot, and static error, by introducing an adaptive temperature control strategy based on a PID algorithm. This significantly improves the control accuracy and stability of the mirror temperature, thus laying a solid foundation for the accurate measurement of dew point temperature and effectively enhancing the resolution and reliability of the instrument. Attached Figure Description

[0043] Figure 1 This is a longitudinal cross-sectional view of the dew point measuring device of the present invention in a dry state;

[0044] Figure 2 This is a longitudinal cross-sectional view of the dew point measuring device of the present invention under condensation conditions.

[0045] Figure 3 Cross-sectional view of a rectangular cross-section embodiment of the dew point measuring device of the present invention;

[0046] Figure 4 Cross-sectional view of a circular cross-section embodiment of the dew point measuring device of the present invention;

[0047] Figure 5 This is a schematic diagram of the electrical principle;

[0048] Figure 6 Flowchart of the measurement method of this invention.

[0049] In the diagram: 1-Measuring cavity, 2-Cavity temperature sensor, 3-Reflective mirror, 4-Semiconductor cooler, 5-Cold end temperature sensor, 6-Collimating light source, 7-Reflective photoelectric sensor, 8-Scattering photoelectric sensor, 9-Reference photoelectric sensor, 10-Reflector, 11-Control processing unit, 12-Heat sink, 13-Fan, 14-Hot end temperature sensor, 15-Insulation layer. Detailed Implementation

[0050] To enable those skilled in the art to better understand the present invention, in conjunction with Figures 1-6 To further explain this application, in this description, the terms "upper", "lower", "left", "right", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly understood by those skilled in the art. They are only used to facilitate the description of the present invention and to simplify the description, and do not indicate or imply that the components or parts referred to must have a specific orientation, or be constructed and operated in a specific orientation. The content mentioned in the embodiments is not a limitation of the present invention.

[0051] This invention relates to a high-precision, anti-interference cold mirror dew point measuring device based on semiconductor refrigeration. A longitudinal cross-sectional view of the dew point measuring device is shown below. Figure 1 , Figure 2 , Figure 1 This is a longitudinal cross-sectional view in its dry state. Figure 2 This is a longitudinal cross-sectional view of the condensation state. The measuring device includes: a measuring chamber 1, which has a channel allowing the gas to be measured to flow through, made of rolled metal (such as aluminum alloy) or non-metallic material (such as plastic), or made of a profile with a suitable cross-section. The cross-sectional shape can be rectangular, see... Figure 3 Or cylindrical, see Figure 4 If rectangular, its height is 20-60mm and its width is 20-40mm. For example, measuring cavity 1 can be made of rectangular aluminum alloy profile with a width of 30mm and a height of 50mm. See [reference needed]. Figure 3 Cross-sectional view. If it's a cylinder, the diameter is 40-60mm. The cavity length is 200-300mm. To facilitate the installation and cleaning (mirror surface) of internal components, the measuring cavity 1 is divided into three parts: the middle part is the equipment compartment, and the two sides are airflow compartments, see... Figure 1 、 Figure 2The dotted lines in the diagram represent the connecting parts. The inner wall of the measuring cavity 1 has a rough black surface to prevent scattered light from reflecting off the inner wall under condensation conditions, which could cause measurement errors. A cavity temperature sensor 2, using a PT100 platinum resistance thermometer, is installed inside the measuring cavity 1 to detect the cavity temperature. An insulation layer 15 is installed on the outside of the measuring cavity 1 to prevent the outside temperature from affecting the measurement. The insulation layer 15 can be wrapped with polyurethane foam material. A reflective mirror 3, made of a high-reflectivity material, is installed in the lower part of the measuring cavity 1. Specifically, the reflective mirror 3 can be made of a high thermal conductivity material (such as copper or aluminum) with a polished surface finish of Ra0.025, serving as the reflective surface of the reflective mirror 3. A mirror anti-fouling coating is provided on the reflective surface of the reflective mirror 3. This coating is a hydrophobic, oleophobic, and optically transparent thin film, and the mirror anti-fouling coating is a hard optical coating made of silicon dioxide or magnesium fluoride. A thermoelectric cooler 4 is located at the lower part of the reflective mirror 3 to regulate the temperature of the reflective mirror 3. Its cold end is thermally connected to the reflective mirror 3, and thermally conductive silicone is applied to the contact surface between the reflective mirror 3 and the thermoelectric cooler 4 to minimize contact thermal resistance. In this invention, the selected thermoelectric cooler 4 must meet the requirements of small size, fast response, and cooling capacity matching the size of the reflective mirror 3. The thermoelectric cooler 4 can be 20×20mm in size, and its specific selection is determined by the heat load of the reflective mirror 3, the target minimum dew point temperature, and the required cooling rate of the system. Preferably, the thermoelectric cooler 4 can be the FPH1-12706 series miniature TEC module manufactured by Guangdong Fuxin Technology Co., Ltd., or the Marlow XM-200-127-10 series manufactured by Marlow Industries, Inc. These models are characterized by small size, low thermal inertia, and fast response speed, making them very suitable for precise temperature control of small mirrors. A groove is milled into the high thermal conductivity material on the reverse side of the reflective surface of the reflective mirror 3. A cold-end temperature sensor 5 is embedded in the groove. The cold-end temperature sensor 5 is a PT100 platinum resistance thermometer used for precise measurement of the temperature of the reflective mirror 3. An optical measurement module is set in the upper part of the measuring cavity 1, including a collimating light source 6 and a reflective photoelectric sensor 7. The light emitted by the collimating light source 6 is reflected by the reflective mirror 3 and received by the reflective photoelectric sensor 7 to detect the change in the reflectivity of the reflective mirror 3 to determine the condensation state. The collimating light source 6 uses an LED light-emitting diode and is equipped with a parabolic reflector cup and a second collimating lens combination to shape and collimate the light, so that the collimating light source 6 emits parallel light. The LED light-emitting diode is a professional-grade LED, preferably the OSRAM OSTAR Projection Family LE xx P2S series, model LUW CNP2. The collimating light source 6 is set on one side of the normal of the reflective mirror 3, and the reflective photoelectric sensor 7 is set on the other side of the normal of the reflective mirror 3. The incident angle of the collimating light source 6 is equal to the reflection angle, and the reflective photoelectric sensor 7 is set in the optical path of the reflected light.

[0052] The dew point measuring device of the present invention also includes a scattering photoelectric sensor 8, which is disposed in the measuring cavity 1. Its photosensitive surface faces the reflective mirror 3, but is not coplanar with the incident light path of the collimating light source 6 and the receiving light path of the reflective photoelectric sensor 7. It is used to receive the scattered light signal generated by dew droplets when condensation forms on the mirror surface. Figure 1 The scattering photoelectric sensor 8 is positioned behind the reflecting photoelectric sensor 7 relative to the reflecting optical path for illustrative purposes only; it can also be positioned in front. To eliminate system errors and achieve real-time compensation, thereby improving the absolute accuracy of the measurement, the dew point measuring device includes a reference optical path module. Its core function is to provide a stable reference standard unaffected by the measured target, used for real-time monitoring and compensation of errors in the measurement system, thus significantly improving the accuracy, stability, and reliability of the measurement. The reference optical path module includes a reference photoelectric sensor 9 and a reflector 10, used for real-time monitoring of light intensity fluctuations of the collimated light source 6 or interference from ambient background light. The reflector 10 is positioned on the incident optical path of the collimated light source 6. The reference photoelectric sensor 9 has an extremely narrow viewing angle, only receiving reflected light from the reflector 10; the incident and reflected light from the collimated light source 6, as well as the scattered light under condensation conditions, cannot enter the reference photoelectric sensor 9.

[0053] The reflective photoelectric sensor 7 and the scattering photoelectric sensor 8 are used to accurately measure light intensity signals. Their core detection elements are preferably silicon photodiodes, such as Nichia's NSC1000 series or Hamamatsu's S1223 series. This choice is based on the excellent linearity, stability, and fast response characteristics of photodiodes, which facilitates high-precision capture of minute changes in specular reflectivity. The reference photoelectric sensor 9 is used to monitor the output fluctuations of the collimated light source 6, providing a reference signal for eliminating common-mode interference. Preferably, its core detection element is a photodiode of the same type as the reflective photoelectric sensor 7, ensuring consistent characteristics under different ambient temperatures, thereby achieving optimal common-mode noise suppression.

[0054] The dew point measuring device includes a control processing unit 11 for controlling the semiconductor cooler 4 and determining the condensation temperature and performing real-time control based on various sensors. The control processing unit 11 is electrically connected to the cold junction temperature sensor 5, the reflection photoelectric sensor 7, the scattering photoelectric sensor 8, the reference photoelectric sensor 9, and the semiconductor cooler 4, and is configured to perform the following operations: real-time acquisition and fusion processing of the reflected light signal from the reflection photoelectric sensor 7, the scattered light signal from the scattering photoelectric sensor 8, and the reference light signal from the reference photoelectric sensor 9; and dynamic compensation of the reflected light intensity based on the difference between the reflected light signal and the reference light signal to eliminate common-mode interference. The control processing unit 11 is configured to simultaneously acquire and process the reflected light intensity signal S obtained from the reflection photoelectric sensor 7. rand the scattered light intensity signal S obtained by the scattering photoelectric sensor 8 s Among them, the reflected light intensity signal S r The intensity of the scattered light signal S will decrease during condensation. s It will rise during condensation. The control processing unit 11 determines the reflected light intensity signal S based on the compensated signal. r The rate of change and the scattered light intensity signal S s A fusion judgment is performed to adaptively determine the starting point and stable state of condensation on the mirror surface. After condensation is determined, compensation calculations are performed based on the reading of the cold end temperature sensor 5 and the pre-stored historical temperature drift data model to output the final dew point temperature value.

[0055] The dew point measuring device also includes a heat sink 12 and a fan 13 connected to the hot end of the thermoelectric cooler 4. The heat sink 12 is milled from aluminum alloy, and a groove is milled into the connection surface between the heat sink 12 and the thermoelectric cooler 4 to embed a hot end temperature sensor 14, which is a PT100 platinum resistance thermometer. The heat sink 12 and the hot end of the thermoelectric cooler 4 are tightly connected using thermally conductive silicone to ensure that heat can be efficiently dissipated. The function of the heat sink 12 is to act as an efficient heat sink, dissipating the Peltier heat generated during the cooling process into the environment in a timely manner, thereby ensuring that the hot end temperature of the thermoelectric cooler 4 is maintained at a low and stable level. This is a prerequisite for the cold end to reach and accurately control the target dew point temperature.

[0056] The control processing unit 11 is configured to execute an adaptive temperature control strategy, employing PID proportional-integral-derivative control to adjust the supply current of the semiconductor cooler 4 and regulate the speed of the fan 13. Maximum power cooling is used during the pre-cooling stage, and the system switches to PID proportional-integral-derivative control as the temperature approaches the predicted dew point to smoothly pass through the dew point and prevent overshoot. PID control is a widely used feedback control algorithm used to accurately and quickly stabilize the controlled physical quantity near a set target value. The control processing unit 11 employs a PID proportional-integral-derivative algorithm to precisely control the temperature of the reflective mirror 3. When the temperature difference between the reflective mirror 3 and the target dew point is large, the system prioritizes cooling speed, using maximum power cooling or only proportional (P) control to achieve rapid cooling. When the mirror temperature approaches the target dew point (e.g., temperature difference less than 0.5~1°C), the system switches to full PID control mode. Integral (I) action eliminates steady-state error to ensure measurement accuracy, while derivative (D) action suppresses temperature overshoot and oscillation. This allows the mirror temperature to pass through the dew point smoothly and without overshoot, maintaining a stable temperature near the target temperature, providing crucial assurance for accurately determining the condensation initiation and stabilization points. This adaptive algorithm, through the rapid response of the proportional term, the elimination of steady-state error in the integral term, and the suppression of overshoot in the derivative term, jointly achieves rapid, stable, and precise full-process control of the mirror temperature.

[0057] The control processing unit 11 is the brain of the entire measuring device, and the arithmetic module 110 can use a high-performance ARM Cortex-M series microcontroller (such as STM32H7 or GD32H7). To achieve signal acquisition, real-time control, and data transmission functions, the control processing unit 11 needs to connect to various components through interface circuits to acquire analog signals from photoelectric sensors (reflection, reference, scattering) and temperature sensors (cavity, cold end, hot end). These signals need to be amplified first, then converted into numerical signals by an analog-to-digital converter (ADC), and then connected to the arithmetic module 110. The control output of the control processing unit 11 is connected to the semiconductor cooler 4 and the fan 13 through a controller. See the electrical schematic diagram below. Figure 5 The photoelectric signal amplification module 111 and the temperature signal amplification module 112 use ADA4530-1 operational amplifiers, supporting dual input modes of voltage and charge, with a gain range of 0.01~1000 mV / pC. They incorporate adjustable high-pass (0.3~100Hz) and low-pass (0.3~100kHz) filters to optimize the signal-to-noise ratio and provide low noise (≤5μV), suitable for amplifying weak signals. The amplified signal is then fed into an ADC analog-to-digital converter 113 to convert the analog signal into a numerical signal. The ADC 113 can use an AD7606C conversion chip, which has 8 analog input channels (6 are used in this invention), 16-bit resolution, and a sampling rate of 1MSPS. It exhibits excellent performance, enabling rapid and sequential sampling and conversion of multiple signals, achieving synchronous sampling of multiple channels. The semiconductor cooler 4 is connected to the computing module 110 via a cooling controller 114. The cooling controller 114 uses a dedicated TEC controller (such as MAX1968, ADN8834), integrating PID control algorithms and H-bridge power output, supporting precise bidirectional cooling / heating control. The fan 13 is connected to the computing module 110 via a fan controller 115, which uses an independent fan controller (such as MAX31740). The collimating light source 6 is connected to the computing module 110 via a collimating light controller 116, which uses an optical MOS solid-state relay (such as the HSSR-61A0D series). Measurement results are packaged and output via a communication interface 117. Data can be transmitted to a computer via wired connection, such as RS-232 / RS-485, or wirelessly via Wi-Fi / 4G / 5G to a cloud platform or remote server. All module components are mounted on the PCB board and waterproofed within a component box, with an IP68 protection rating.

[0058] The high-precision, anti-interference cold mirror dew point measurement method for semiconductor refrigeration of the present invention is shown in the flowchart below. Figure 6The process includes the following steps: S1, Initialization and Baseline Establishment: The gas to be tested flows through the measuring chamber 1 at a constant flow rate. The system (especially the reflective mirror 3), the airflow, and the ambient temperature (based on the reading of the chamber temperature sensor 2) reach thermal equilibrium. The control processing unit 11 first reads the value of the chamber temperature sensor 2 as the current ambient temperature T of the gas sample. chamber The baseline is set. The optical measurement module and reference optical path module are activated. Under stable conditions where the mirror surface is clean and dry, the control processing unit 11 reads and records the initial value of the reflection photoelectric sensor 7 as the baseline S of the reflected light intensity. r 0, and simultaneously read and record the initial value of the reference photoelectric sensor 9 as the background light intensity baseline S. b 0, and read the initial value of the scattering photoelectric sensor 8 as the background scattered light baseline S. s 0.

[0059] S2. Active cooling and synchronous monitoring: The control processing unit 11 outputs drive power to the semiconductor cooler 4, and simultaneously reads the value T from the hot end temperature sensor 14. hot And based on T hot With ambient temperature T chamber The difference in temperature (T) dynamically adjusts the speed of fan 13. hot -T chamber When the temperature difference is large, a higher speed is used to enhance heat dissipation; when the temperature difference is small, the speed can be reduced to optimize energy consumption and noise. This ensures that the hot end of the semiconductor cooler 4 always maintains the optimal operating temperature. The semiconductor cooler 4 cools the reflective mirror 3, causing its temperature to drop from the ambient temperature. The system uses a first power to cool the reflective mirror 3 using a stepped or linear cooling strategy, causing its temperature to drop. During the cooling process, the control processing unit 11 synchronously triggers and collects data from all sensors at a fixed high sampling frequency (preferably 20Hz~100Hz). A hardware timer ensures strict synchronization of the acquisition actions, providing a data sequence with a consistent time base for subsequent signal processing. The real-time acquired data includes: mirror temperature T, reflective photoelectric sensor value S. r Reference photoelectric sensor value S b and the value S of the scattering photoelectric sensor s The reflected light intensity signal S in this application r Scattered light intensity signal S s Reference light intensity signal S b The unit is the "Digital Unit" (DU) based on the corresponding photoelectric sensor converting the received light signal into a voltage signal output, amplifying it, and performing analog-to-digital conversion (ADC) to obtain a binary value, which is then read by the control processing unit 11.

[0060] S3. Signal Processing and Condensation Initiation Determination: The control processing unit 11 monitors the reflectivity of the reflective mirror 3 in real time through the optical measurement module. It calculates the compensated effective reflected light intensity signal S in real time. eff Its calculation formula is: S eff = (S r -S r 0)- k (S b -S b 0), where k is a preset calibration coefficient obtained through experimental calibration. It quantifies the proportional relationship between the response of the main optical path (reflected optical path) and the reference optical path to common-mode interference (mainly light source fluctuations), ensuring the accuracy of compensation. Simultaneously, the effective reflected light intensity signal S is calculated. eff rate of change over time dS eff / dt. Calculate the increment ΔS of the scattered light signal. s =S s -S s 0, and simultaneously calculate the increment ΔS of the scattered light signal. s The rate of change of d(ΔS) with time t s ) / dt;

[0061] Effective reflected light intensity signal S eff rate of change over time dS eff / dt, its calculation formula is based on discrete-time signal systems, as follows: Assuming the system acquires signals with a fixed sampling period Δt, at any nth sampling time:

[0062] S at the current moment eff The value is S eff (n), S at the previous sampling time eff The value is S eff (n-1);

[0063] Then the rate of change dS at the current moment eff / dt can be approximated using a first-order backward difference, as shown in the formula:

[0064] ,

[0065] in:

[0066] dS eff / dt represents the effective reflected light intensity signal S eff The rate of change over time, in signal units per second, where the signal unit is Digital Unit (DU).

[0067] S eff (n) represents the effective reflected signal value obtained from the nth sampling calculation;

[0068] S eff(n-1) is the effective reflected signal value calculated in the (n-1)th sampling (i.e., the previous sampling);

[0069] Δt is a fixed sampling time interval, which is a fixed parameter of the system (for example, if the sampling frequency is 100 Hz, then Δt = 0.01 seconds).

[0070] Scattered light signal increment ΔS s The rate of change of d(ΔS) with time t s The formula for calculating ) / dt is based on discrete-time signal systems, as follows:

[0071] ,

[0072] in:

[0073] d(ΔS s ) / dt is the rate of change of the scattered light signal increment with time, in signal units per second. The signal unit here is Digital Unit (DU).

[0074] ΔS s (n) represents the increment of the scattered light signal calculated at the nth sampling, ΔS s (n)=S s (n)-S s 0;

[0075] ΔS s (n-1) represents the increment of the scattered light signal calculated during the (n-1)th sampling (i.e., the previous sampling), ΔS s (n-1)=S s (n-1) -S s 0;

[0076] S s (n),S s (n-1): The original measurement values ​​of the scattering photoelectric sensor 8 during the nth and (n-1)th samplings;

[0077] Δt is a fixed sampling time interval, which is a fixed parameter of the system (for example, if the sampling frequency is 100Hz, then Δt = 0.01 seconds).

[0078] The control processing unit 11 is configured to determine the start of condensation on the mirror surface based on at least one of the following logics, and record the mirror surface temperature at this time as the initial dew point temperature T1:

[0079] a) Logic 1: When dS eff / dt exceeds the preset first dynamic threshold THR1 for the first time, and at the same time ΔS s Exceeding the preset scattered light intensity threshold (THR) sAt this time, it is determined that condensation has begun;

[0080] b) Logic 2: When ΔS s rate of change d(ΔS) s When ) / dt first exceeds the preset second dynamic threshold THR2, it is determined that condensation has started.

[0081] The control processing unit 11 presets a first dynamic threshold THR1, a second dynamic threshold THR2, and a scattered light intensity threshold THR. s The dynamic threshold is not a fixed value, but is set and adaptively adjusted according to one or more of the following strategies to ensure reliable and sensitive condensation initiation determination under different humidity ranges and environmental noise levels. When the control processing unit 11 uses PID proportional-integral-derivative control, the setting rule for the dynamic threshold is as follows: a higher first dynamic threshold THR1 is used during the rapid decrease of the mirror temperature, and a second dynamic threshold THR2, lower than the first dynamic threshold THR1, is used when the mirror temperature approaches the predicted dew point temperature. The method for setting the dynamic threshold is as follows:

[0082] Scattered light intensity threshold (THR) s Setting: Scattered Light Intensity Threshold (THR) s (For logic 1) Primarily used to distinguish between weak background scattering and true condensation scattering signals. Its setting method is as follows:

[0083] Baseline statistics: During the initialization phase, after the system stabilizes, the control processing unit 11 collects the background scattered light baseline S over a period of time (e.g., 5-10 seconds). s 0, and calculate its average value μ s and standard deviation σ s ;

[0084] Threshold calculation, the scattered light intensity threshold THR s Set as:

[0085] ,

[0086] Among them, K s This is a preset confidence coefficient, typically ranging from 3 to 5. This setting means that when the diffuse light signal increment ΔS... s K exceeding the normal range of baseline fluctuations (defined by standard deviation) s Only when the event is multiplied by a factor of 1 is it considered a significant event, thus effectively suppressing false triggering caused by random noise;

[0087] Setting the first dynamic threshold THR1 (reflection rate of change threshold): The first dynamic threshold THR1 (used for logic one) is used to determine whether the reflected light intensity has begun to change drastically due to condensation;

[0088] Noise level assessment: In the initial stage of cooling, when no condensation forms on the reflective mirror 3, the control processing unit 11 calculates the effective reflected light signal S in real time. eff rate of change dS eff / dt, and monitor its fluctuation level;

[0089] The dynamic correlation setting, the first dynamic threshold THR1, is adaptively set based on the current signal-to-noise ratio and cooling rate, and its calculation formula is as follows:

[0090] ,

[0091] Where: σdS / dt is dS eff The standard deviation of / dt represents the noise level of the rate of change; |dT / dt| is the absolute value of the current cooling rate of the mirror; K1 and C are weighting coefficients calibrated experimentally (typically K1 is 0.5~5, C is 0.1~1). Through experiments, a set of (K1, C) was found such that in the "noisy / rapid cooling non-dew point state," THR1 is higher than the signal change, avoiding false alarms; while in the "stable signal, close to the dew point state," THR1 drops low enough to be surpassed by a small, real dew point signal, ensuring accurate triggering. This design allows the threshold to automatically increase when there is high noise or rapid cooling to prevent false alarms; and to decrease the threshold when the signal is stable and close to the dew point, improving sensitivity.

[0092] Setting the second dynamic threshold THR2 (scattered light rate of change threshold): The second dynamic threshold (used for logic 2) is used to determine whether the scattered light has started to increase rapidly;

[0093] The setting principle is similar to that of the first dynamic threshold, but its evaluation object is the increment ΔS of the scattered light signal. s rate of change d(ΔS) s The formula for calculating ) / dt is:

[0094] ,

[0095] Where σdΔS / dt is d(ΔS) s The standard deviation of the fluctuation in the baseline phase is ) / dt, and K2 is a preset coefficient (usually 4~6). This threshold is focused on capturing the abrupt start point of the scattered light signal.

[0096] The aforementioned preset coefficient K s K1 and K2 were calibrated experimentally. The calibration process involved collecting a set of data (≥10) from a controlled experiment, while simultaneously recording the actual state of the system (e.g., whether condensation occurred). The optimal values ​​of the coefficients were determined through step-by-step optimization by adjusting the coefficients and after multiple rounds of testing and verification.

[0097] S4. Temperature Balance and Stability Measurement: After the control processing unit 11 determines that condensation has started and records the initial dew point temperature T1, it immediately switches the control mode of the semiconductor cooler 4 to PID proportional-integral-derivative control mode to finely adjust its operating current, thereby achieving precise fine-tuning of the mirror temperature. The control processing unit 11 presets a small temperature control range centered on T1, for example, [T1-δ, T1+δ], where δ is a minimum value within 0.2°C. The set point of the PID controller can be set to T1 itself, causing the temperature of the reflective mirror 3 to fluctuate within a preset small range near the initial dew point temperature T1. The control objective is to make the mirror temperature perform smooth, damped oscillations without overshoot within this range, and eventually stabilize near T1. Until the effective reflection signal S... eff The value is at a new level, significantly lower than the baseline of reflected light intensity S. r The system remains stable at a level of 0. The condensation state stabilizes, and the average mirror temperature measured under this stable condition is recorded as the stable dew point temperature T2. Once the system determines that the condensation state has stabilized, the control processing unit 11 records the mirror temperature reading sequence during this stable phase. The stable dew point temperature T2 is not an instantaneous value, but is obtained by calculating the arithmetic mean of all temperature sampling points within this time period. Simultaneously, the stable S value during this phase is recorded. eff With S s The signal value serves as confirmation of the presence of condensation.

[0098] S5. Temperature Compensation and Result Output: The control processing unit 11 substitutes the stable dew point temperature T2 obtained in step S4 into the pre-stored temperature compensation model for calculation, and obtains the corrected final dew point temperature value T. corrected (T) d Then, a system reset operation is performed. The specific steps are as follows:

[0099] Temperature compensation: The pre-stored temperature compensation model is a functional relationship obtained by calibrating the system using a high-precision standard dew point meter under different ambient temperatures. Its preferred form is a polynomial model:

[0100] ,

[0101] Where T2 is the stable dew point temperature, and a, b, c are coefficients obtained through high-precision experimental calibration. corrected This is the compensated dew point temperature value;

[0102] The results are packaged and the control processing unit 11 outputs the final dew point temperature value T. corrected In addition to the final dew point temperature, the output data packet may optionally include auxiliary data from this measurement, such as the stable dew point temperature T2 and the compensation amount (T). corrected- T2), the standard deviation of the fluctuation during the mirror temperature equilibrium stage, the reflected light intensity signal value during the condensation stabilization stage, etc., are used by the upper-level system for data quality assessment and traceability.

[0103] The intelligent cooling system stops cooling immediately after successfully outputting the final dew point temperature value. The control processing unit 11 then generates a control command to cut off the power supply to the thermoelectric cooler 4, thus stopping cooling. This operation aims to prevent condensation on the mirror surface from adsorbing more pollutants from the environment due to prolonged low temperatures, while also reducing system energy consumption. Simultaneously with or after a short delay, the control processing unit 11 controls the fan 13 to stop operating, entering a fully low-power state.

[0104] S6. Condensation Evaporation and System Reset: After completing the recording and output of the dew point temperature, the system enters the reset phase. Its core objective is to utilize residual heat to quickly and controllably dry the mirror surface and accurately determine the reset completion time to initiate the next round of measurements. The control processing unit 11 first cuts off the drive power to the thermoelectric cooler 4, stopping active cooling, and the fan 13 stops accordingly. Then, using the residual heat stored in the heat sink 12, it guides the heat flow to the hot end of the thermoelectric cooler 4 to provide a stable heat source for the mirror surface to recover. The cold end of the thermoelectric cooler 4 and the mirror surface temperature subsequently rise, causing condensation to form and evaporate on the mirror surface. During the natural recovery process, the control processing unit 11 continuously monitors three key temperature parameters: the reading T of the hot end temperature sensor 14... hot The reading T of cold junction temperature sensor 5 cold and the reading T of cavity temperature sensor 2 chamber To comprehensively assess the system status, and to achieve controlled and efficient reset while preventing system temperature runaway, the following safety measures are adopted: tracking the temperature recovery trajectory and monitoring T in real time. cold The rate of recovery and its relationship with T chamber Temperature difference (|T) chamber - T cold |) shrinkage trend; assess heat source status, and simultaneously monitor T hot The changes were used to confirm the stability of waste heat release.

[0105] During the system reset process, the control processing unit 11 synchronously and continuously monitors the following key parameters: Optical reset signal: reflected light intensity signal S r and the increase in the scattered light signal ΔS s Thermal reset signal: reading T of cold junction temperature sensor 5. cold and the reading T of cavity temperature sensor 2 chamber To comprehensively assess the system status, the system employs a combined primary and secondary decision-making strategy to confirm reset completion: Primary criterion (optical criterion): When the reflected light intensity signal S... r Restored to ≥ baseline reflected light intensity S rWhen the condensation on the mirror reaches a set ratio of 0 (dry baseline) (e.g., within the range of 95% to 98%), it is preliminarily determined that the condensation has essentially evaporated, and the optical properties meet the initial conditions for the next measurement. Auxiliary criterion (thermal criterion): Simultaneously, monitor the cold junction temperature T. cold With ambient temperature T chamber The temperature difference. Experiments show that it is not necessary to wait for the mirror temperature to fully recover to the ambient temperature, as long as the optical criteria are met and the cold junction temperature T... cold With the ambient temperature T of the cavity chamber Temperature difference | T chamber - T cold Once the temperature has narrowed to a safe range (e.g., ≤ 10°C~15°C), the system reset can be considered complete. Actual test data shows that after the optical signal is restored, even if the cold junction temperature T... cold With ambient temperature T chamber Even with a certain temperature difference (e.g., 10°C at the cold end and 25°C in the environment), a new round of measurements will be initiated immediately. This is because maximizing measurement efficiency requires more time to reach complete thermal equilibrium. Given that the optical signal has confirmed the mirror is dry, the remaining temperature difference has minimal impact on the initial cooling phase of the next measurement cycle. This thermal criterion ensures that when starting the next measurement, the reflective mirror 3 is in a "cold start" state that does not significantly affect the initial cooling phase, rather than requiring complete thermal equilibrium. However, this significantly shortens the measurement cycle, increases the number of measurements per unit time, avoids unnecessary waiting time, and reduces the average power consumption of the system. The criterion for "system readiness" is: the optical signal confirms the mirror is dry + the temperature field is in a safe state that allows for rapid initiation of the next measurement. This criterion ensures that thermal equilibrium has been reached between the mirror and the gas being measured, thermally eliminating the influence of residual temperature on the initial stage of the next measurement. The thermal criterion serves as an auxiliary criterion, and the control processing unit 11 ultimately determines that the system has been fully reset to the "dry and ready" state. Once the system is determined to be ready, the control processing unit 11 immediately and automatically returns to step S2 to start a new round of measurement cycle, thereby achieving true unattended continuous monitoring.

[0106] To enhance the system's adaptability to different environments (such as gases with different cleanliness levels and light sources with different aging degrees), all the above thresholds can be adaptively updated. The control processing unit 11 can periodically (e.g., after 24 hours of continuous operation), and after the reset phase of each measurement cycle and before the next active cooling begins, recalculate the above standard deviations σs and σs based on the latest acquired baseline noise data. F The thresholds σdS / dt and σdΔS / dt are dynamically updated to achieve long-term stable and reliable detection.

[0107] The specification and drawings of this application are merely one specific embodiment and are not restrictive. Those skilled in the art can make many other modifications based on the teachings of this application without departing from the spirit and scope of the claims, all of which are within the scope of protection of this application.

Claims

1. A high-precision, anti-interference cold mirror dew point measurement method based on semiconductor refrigeration, characterized in that, A high-precision, interference-resistant cold mirror dew point measurement device using semiconductor cooling includes: The measuring chamber (1) is provided with a channel through which the gas to be measured can flow. The measuring chamber (1) is provided with a chamber temperature sensor (2). A reflective mirror (3) is disposed inside the measuring cavity (1); A semiconductor cooler (4) has its cold end thermally connected to the reflective mirror (3) for adjusting the temperature of the reflective mirror (3); A cold-end temperature sensor (5) is embedded in a reflective mirror (3) for precise measurement of the temperature of the reflective mirror (3); The optical measurement module includes a collimated light source (6) and a reflective photoelectric sensor (7). The light emitted by the collimated light source (6) is reflected by the reflective mirror (3) and received by the reflective photoelectric sensor (7) to detect the change in the reflectivity of the reflective mirror (3) to determine the condensation state. Also includes: A scattering photoelectric sensor (8) is set inside the measuring cavity (1), with its photosensitive surface facing the reflective mirror (3), but it is not coplanar with the incident light path of the collimating light source (6) and the receiving light path of the reflective photoelectric sensor (7). It is used to receive the scattered light signal generated by dew droplets when condensation forms on the reflective mirror (3). The reference optical path module includes a reference photoelectric sensor (9) and a reflector (10) for real-time monitoring of light intensity fluctuations of the collimated light source (6) or interference from ambient background light; The control processing unit (11) is electrically connected to the cold end temperature sensor (5), the reflection photoelectric sensor (7), the scattering photoelectric sensor (8), the reference photoelectric sensor (9) and the semiconductor cooler (4), respectively. The hot end of the semiconductor cooler (4) is connected to the heat sink (12) and the fan (13), and the hot end temperature sensor (14) is embedded on the heat sink (12). The measurement method includes the following steps: S1. Initialization and baseline establishment: The gas to be measured flows through the measurement cavity (1) at a constant flow rate; the optical measurement module and the reference optical path module are started; the control processing unit (11) reads and records the initial value of the reflective photoelectric sensor (7) as the baseline of the reflected light intensity. S r 0, and simultaneously read and record the initial value of the reference photoelectric sensor (9) as the background light intensity baseline. S b 0, and read the initial value of the scattering photoelectric sensor (8) as the background scattering baseline. S s 0; S2. Active cooling and synchronous monitoring: The control processing unit (11) controls the semiconductor cooler (4) to be powered on and the fan (13) to cool the reflective mirror (3) with a first power using a stepped or linear cooling strategy, so that its temperature drops. The mirror temperature T and reflected light intensity signals are collected synchronously at a preset frequency. S r Reference light intensity signal S b and scattered light intensity signal S s ; S3. Signal Processing and Condensation Initiation Determination: Real-time calculation of the compensated effective reflected light intensity signal. S eff The calculation formula is as follows: S eff = ( S r - S r 0) - k ( S b - S b 0), where k is a preset calibration coefficient, and the effective reflected light intensity signal is calculated simultaneously. S eff rate of change over time dS eff / dt; Calculate the increment Δ of the scattered light signal. S s = S s - S s 0, and simultaneously calculate the increment Δ of the scattered light signal. S s rate of change with time t d (Δ S s ) / dt ; The control processing unit (11) is configured to determine the start of condensation on the mirror surface based on at least one of the following logics, and record the mirror surface temperature at this time as the initial dew point temperature T1: a) Logic 1: When dS eff / dt The first time it exceeds the preset first dynamic threshold THR1, and at the same time ΔS s Exceeding the preset scattered light intensity threshold (THR) s At this time, it is determined that condensation has begun; b) Logic 2: When Δ S s rate of change d (Δ S s ) / dt When the preset second dynamic threshold THR2 is exceeded for the first time, it is determined that condensation has started; S4. Temperature Balance and Stability Measurement: The control processing unit (11) switches the control mode of the semiconductor cooler (4) to PID proportional-integral-derivative control mode, finely adjusts its operating current, and makes the temperature of the reflective mirror (3) fluctuate within a preset small range near the initial dew point temperature T1 until an effective reflection signal is achieved. S eff The condensation point temperature stabilizes at a new steady-state value. The average mirror temperature corresponding to this steady-state condition is recorded as the stable dew point temperature T2. S5. Temperature Compensation and Result Output: Substitute the stable dew point temperature T2 into the pre-stored temperature compensation model for calculation to obtain the corrected final dew point temperature value. T corrected Output the compensated dew point temperature value; and immediately cut off the power supply to the semiconductor cooler (4) to stop cooling; S6, Condensation Evaporation and System Reset: After cooling stops, the fan (13) stops; the residual heat of the radiator (12) is used to heat the hot end of the semiconductor cooler (4), and the temperature of its cold end and the reflective mirror (3) rises accordingly, causing condensation to form on the mirror; when the reflected light intensity signal... S r Reaching ≥ baseline of reflected light intensity S r When the temperature difference between the cold end temperature sensor (5) and the cavity temperature sensor (2) is less than or equal to a preset temperature difference threshold, the system is determined to be reset to the ready state and automatically returns to step S2 to start the next measurement cycle.

2. The high-precision anti-interference cold mirror dew point measurement method based on semiconductor refrigeration according to claim 1, characterized in that, In step S3, the effective reflected light intensity signal S eff rate of change over time dS eff / dt, its calculation formula is based on discrete-time signal systems, as follows: Assume the system uses a fixed sampling period Δ t The signal is acquired, at any given time... n Each sampling time: The current moment S eff Value S eff ( n ), the previous sampling time S eff Value S eff ( n -1); The rate of change at the current moment dS eff / dt It can be approximated by a first-order backward difference, as shown in the formula: ， in: dS eff / dt To effectively reflect light intensity signals S eff The rate of change over time, expressed in signal units per second; S eff ( n ) is the first n The effective reflected signal value obtained from the second sampling calculation; S eff ( n -1) is the first n -1 is the effective reflected signal value obtained from the previous sampling calculation; Δ t The fixed sampling time interval is a fixed parameter of the system. The increment Δ of the scattered light signal S s Over time t rate of change d (ΔS s ) / dt Its calculation formula is based on discrete-time signal systems, as follows: ， in: d (Δ S s ) / d t is the rate of change of the scattered light signal increment over time, in units of signal units per second; Δ S s ( n ) is the first n The increment of the scattered light signal calculated at the next sampling, Δ S s ( n )= S s ( n )- S s 0; Δ S s ( n -1) is the first n The increment of the scattered light signal calculated at the -1st sampling, Δ S s ( n -1)= S s ( n -1) - S s 0; S s ( n ), S s ( n -1): No. n Second and third n -1 sampling, the original measurement value of the scattering photoelectric sensor (8); Δ t The fixed sampling time interval is a fixed parameter of the system.

3. The high-precision anti-interference cold mirror dew point measurement method based on semiconductor refrigeration according to claim 1, characterized in that, In step S3, the setting rule for the dynamic threshold is as follows: a higher first threshold is used during the rapid decrease of the mirror temperature, and a second threshold lower than the first threshold is used when the mirror temperature is close to the predicted dew point temperature.

4. The high-precision anti-interference cold mirror dew point measurement method based on semiconductor refrigeration according to claim 1, characterized in that, The temperature compensation model described in step S5 is a polynomial compensation model: ， in, T 2. To stabilize the dew point temperature, a , b , c These are the coefficients obtained through high-precision experimental calibration. T corrected This is the compensated dew point temperature value.

5. The high-precision anti-interference cold mirror dew point measurement method based on semiconductor refrigeration according to claim 1, characterized in that, The control processing unit (11) is configured to simultaneously acquire and process the reflected light intensity signal obtained by the reflected photoelectric sensor (7). S r and the scattered light intensity signal obtained by the scattered photoelectric sensor (8). S s ; wherein, the reflected light intensity signal S r The intensity of the scattered light signal will decrease during condensation. S s It will rise when condensation occurs.

6. The high-precision anti-interference cold mirror dew point measurement method based on semiconductor refrigeration according to claim 1, characterized in that, The reflective mirror (3) is provided with a mirror anti-fouling coating, which covers the reflective surface of the reflective mirror (3). The coating is a hydrophobic, oleophobic and optically transparent thin film. The mirror anti-fouling coating is a hard optical coating made of silicon dioxide or magnesium fluoride.

7. The high-precision anti-interference cold mirror dew point measurement method based on semiconductor refrigeration according to claim 1, characterized in that, The measuring cavity (1) is surrounded by an insulation layer (15).

Images