A near-infrared scattering-absorbing combined optical water-in-oil detection system

Abstract

The application discloses a kind of near-infrared scattering-absorption composite optical water-in-oil detection systems, it is related to oil detection technical field, including: optical detection module, signal acquisition module, data processing module and control module, optical detection module emits near-infrared light covering moisture characteristic absorption band and non-characteristic absorption band;The application is by constructing scattering-absorption characteristic fusion double-weight model, make full use of the complementary characteristics of absorption signal and scattering signal, solve the detection limitation caused by the single signal of prior art is handled alone, by dynamically allocating signal weight, the adaptation of different water cut interval is realized, effectively improve the detection precision under low water cut scene, while taking into account the detection stability when high water cut, with the feature extraction and fusion calculation of double data block by SPORT-PLS fusion algorithm, the accuracy and reliability of detection result are greatly improved, realize the accurate detection of water cut in full range of oil.

Description

Technical Field

[0001] This invention relates to the field of oil testing technology, specifically to a near-infrared scattering-absorption composite optics system for detecting water content in oil. Background Technology

[0002] Detection of water content in oil is a crucial step in petroleum extraction, transportation, refining, and industrial applications. Water in oil directly affects combustion efficiency and equipment lubrication, and can also cause corrosion, emulsification, and deterioration. Accurate detection of water content in oil is of significant practical importance for ensuring production safety, reducing energy consumption, and improving product quality. Near-infrared scattering-absorption composite optical technology combines the absorption and scattering characteristics of near-infrared spectroscopy. When near-infrared light irradiates oil, the OH bonds in water molecules exhibit characteristic absorption of specific near-infrared wavelengths. Furthermore, when water exists in the form of emulsion droplets or free water droplets, it scatters near-infrared light. This composite optical technology can simultaneously capture both absorption and scattering signals, providing richer characteristic information for oil water content detection. Compared to single detection principles, it can more comprehensively reflect the water content state of oil, laying the foundation for improved detection accuracy.

[0003] However, existing near-infrared oil water content detection technologies still have certain shortcomings. Most of them process absorption signals or scattering signals separately, without fully integrating the complementary characteristics of the two signals. In low water content scenarios, the characteristic absorption signal of water is weak, and relying solely on the absorption signal can easily lead to insufficient detection accuracy. At high water content, the interference of scattering signals is enhanced, and a single signal processing method is difficult to meet the detection needs of different water content ranges, and cannot achieve accurate detection across the entire range. Therefore, it is of great significance to develop a near-infrared scattering-absorption composite optics oil water content detection system. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a near-infrared scattering-absorption composite optics system for detecting water content in oil. This system can fully utilize the complementary characteristics of absorption and scattering signals by constructing a dual-weighted model that fuses scattering and absorption features, overcoming the detection limitations caused by processing a single signal in existing technologies. By dynamically allocating signal weights, it achieves adaptation to different water content ranges, effectively improving detection accuracy in low water content scenarios. Furthermore, by employing the SPORT-PLS fusion algorithm for feature extraction and fusion calculation of dual data blocks, it improves the accuracy and reliability of detection results, achieving precise detection of water content in oil across the entire measurement range.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a near-infrared scattering-absorption composite optics system for detecting water content in oil, the system comprising: an optical detection module, a signal acquisition module, a data processing module, and a control module; The optical detection module emits near-infrared light covering the characteristic absorption band and non-characteristic absorption band of moisture. After the near-infrared light interacts with the oil to be detected, the optical detection module collects the absorbed light signal and scattered light signal generated by the interaction of the oil and outputs them to the signal acquisition module. The signal acquisition module receives the absorbed light signal and scattered light signal output by the optical detection module, amplifies, filters and performs analog-to-digital conversion on the signal, and outputs the converted digital signal to the data processing module. The data processing module receives the digital signal output by the signal acquisition module. It has a built-in dual-weight model for scattering-absorption feature fusion based on the SPORT-PLS fusion algorithm constructed by the idea of ​​multi-block data analysis. Through this model, the absorption spectrum data and scattering spectrum data corresponding to the digital signal are extracted and fused to generate and output the oil water content result. The control module outputs control signals to the optical detection module, signal acquisition module, and data processing module to control the working sequence of each module, and at the same time receives the moisture content results output by the data processing module.

[0006] Furthermore, the optical detection module includes a near-infrared light source, a detection channel, and a photodetector. The near-infrared light source is a broadband light-emitting device that emits near-infrared light covering both the characteristic absorption band and the non-characteristic absorption band of moisture. The detection channel is made of a light-transmitting material and forms an internal channel for the oil to be detected to flow through. The inner wall of the channel is smooth. Two photodetectors are provided, one arranged along the transmission direction of the near-infrared light and the other arranged along the scattering direction of the near-infrared light. Sealed interfaces are provided at both ends of the detection channel, and the sealed interfaces are made of oil-resistant rubber material. An insulation layer is wrapped around the outside of the detection channel, and the insulation layer is fitted into the outer wall of the detection channel.

[0007] Furthermore, the signal acquisition module processes signals by including the following steps: The absorbed light signal and the scattered light signal output by the optical detection module are received, and the two weak signals are linearly amplified by a low-noise operational amplifier respectively. The amplified signal is transmitted to the RC filter circuit to filter out environmental interference noise in the signal. The filtered analog signal is transmitted to the analog-to-digital converter unit, converted into a digital signal, and then output to the data processing module.

[0008] Furthermore, the dual-weight model in the data processing module processes data including the following steps: The digital signal output from the signal acquisition module is received, and the corresponding absorption spectrum data and scattering spectrum data are divided into two independent data blocks. Partial least squares regression was used on the absorption spectrum data blocks to analyze the correlation between absorption spectra and moisture concentration, and characteristic parameters of moisture concentration were extracted. Parallel orthogonalization techniques were used to eliminate interference from irrelevant variables in the scattering spectral data blocks, and the characteristic parameters of water droplet size distribution were extracted. ; Based on the signal intensity of the absorption spectral data block, the water content range of the oil to be tested is initially determined. Based on the preset weighting rules, the absorption signal weighting coefficients are calculated. With the weighting coefficient of the scattered signal The calculation formula is , ,in, To obtain the real-time signal intensity of the absorption spectral data block, The preset maximum absorption signal intensity is determined statistically from the spectral data of oil samples with high water content gradients. The preset minimum scattered signal intensity was determined statistically from the spectral data of oil samples with low water content gradients. The signal sensitivity adjustment coefficient is determined by cross-validation of multiple sets of detection data from samples with different moisture contents, selecting the value that minimizes the detection error across the entire moisture content range. Based on the calculated weight coefficients, the extracted feature parameters are fused to generate the oil water content result. The fusion calculation formula is: ,in, The correlation function between the droplet size distribution characteristic parameters and the water content is established by fitting experimental data of the scattering characteristics of water droplets of different sizes with the corresponding measured values ​​of water content.

[0009] Furthermore, the control module includes a timing control unit, a data receiving unit, and a result output unit. The timing control unit stores a preset control program, outputs a light source start / stop command to the optical detection module, and outputs a signal sampling frequency command to the signal acquisition module. The data receiving unit is connected to the data processing module and receives the moisture content calculation result output by the data processing module. The result output unit is connected to an external display device or storage device and transmits the moisture content result to the external display device or storage device.

[0010] Furthermore, the data processing module also has a built-in data calibration unit, which stores calibration parameters corresponding to different types of oil. When detecting different types of oil, the data processing module calls the calibration parameters corresponding to the oil type in the data calibration unit to correct the parameters in the feature extraction process.

[0011] Furthermore, the control module is also equipped with a manual control interface, which is connected to an external operating terminal. The external operating terminal inputs control commands to the manual control interface, and the control module adjusts the light source intensity and signal acquisition frequency parameters after receiving the commands.

[0012] Furthermore, the photodetector surface of the optical detection module is provided with an anti-oil coating, which covers the signal receiving surface of the photodetector. A light shield is provided outside the near-infrared light source, and the light shield is arranged around the near-infrared light source. The inner wall of the light shield is made of matte material.

[0013] Compared with existing technologies, this near-infrared scattering-absorption composite optics oil water content detection system has the following advantages: This invention constructs a dual-weight model that fuses scattering and absorption features, fully utilizing the complementary characteristics of absorption and scattering signals. This overcomes the detection limitations caused by processing a single signal in existing technologies. By dynamically allocating signal weights, it achieves adaptation to different water content ranges, effectively improving detection accuracy in low water content scenarios while maintaining detection stability in high water content scenarios. By employing the SPORT-PLS fusion algorithm for feature extraction and fusion calculation of dual data blocks, it significantly improves the accuracy and reliability of detection results, achieving accurate detection of water content in oil across the entire range and providing reliable data support for production and operation in petroleum-related industries.

[0014] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0016] Figure 1 This is a schematic diagram of a near-infrared scattering-absorption composite optics system for detecting water content in oil. Figure 2 This is a flowchart of a near-infrared scattering-absorption composite optics system for detecting water content in oil. Detailed Implementation

[0017] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0018] This invention patent provides a near-infrared scattering-absorption composite optics system for detecting water content in oil, specifying a complete technical solution. (See attached document.) Figure 1 and Figure 2 The core content is as follows: The detection system consists of four core parts: an optical detection module, a signal acquisition module, a data processing module, and a control module. These modules work together to achieve accurate detection of the water content of oil products.

[0019] The optical detection module includes a near-infrared light source, a detection channel, and two photodetectors. The near-infrared light source uses a wide-band light-emitting device that emits near-infrared light covering both the characteristic absorption band and the non-characteristic absorption band of moisture, ensuring that it can simultaneously capture the absorption and scattering optical characteristics of moisture. The detection channel is made of a light-transmitting material, forming an internal passage for the oil to be tested. The inner wall of the channel is smooth to reduce oil flow resistance and signal interference. Both ends are equipped with oil-resistant rubber sealing interfaces to ensure no leakage during oil flow. The outer side is wrapped with an insulation layer that fits against the outer wall of the channel to maintain a stable oil temperature inside the channel and prevent temperature fluctuations from affecting the optical signal characteristics.

[0020] Two photodetectors are arranged along the transmission and scattering directions of near-infrared light, respectively. The transmission direction detector is used to collect the absorbed light signal after being absorbed by the oil, and the scattering direction detector is used to collect the scattered light signal after being scattered by water in the oil. The detector surface is covered with an oil-resistant coating to completely cover the signal receiving surface, preventing oil contamination from affecting the signal acquisition accuracy during the detection process. A light shield is set around the outside of the near-infrared light source. The inner wall of the light shield is made of matte material to reduce the interference of ambient stray light on the near-infrared light emitted by the light source and ensure the stability of the light source output.

[0021] After receiving the two optical signals output by the optical detection module, the signal acquisition module first linearly amplifies the weak signal through a low-noise operational amplifier, then filters out environmental interference noise through an RC filter circuit, and finally converts the analog signal into a digital signal by an analog-to-digital converter and outputs it to the data processing module.

[0022] The data processing module incorporates the SPORT-PLS fusion algorithm, a dual-weighted model for scattering and absorption feature fusion, built upon the concept of multi-block data analysis. Upon receiving the digital signal, it divides it into two independent data blocks: the absorption spectrum and the scattering spectrum. Relevant feature parameters are extracted from each block, and the water content range is determined based on the absorption signal intensity. The corresponding signal weight coefficients are then calculated, and the oil water content result is generated through fusion calculation. Simultaneously, the module includes a data calibration unit that stores calibration parameters for different oil types, allowing for correction of the detection parameters.

[0023] The control module includes a timing control unit, a data receiving unit, and a result output unit, and also features a manual control interface. The timing control unit outputs commands such as light source start / stop and signal sampling frequency to control the operating timing of each module; the data receiving unit receives the moisture content results output by the data processing module; and the result output unit transmits the results to an external display or storage device. The manual control interface can be connected to an external operating terminal to adjust parameters such as light source intensity and signal acquisition frequency.

[0024] The system's working logic is as follows: the optical detection module emits near-infrared light that interacts with the oil, collects the absorbed and scattered light signals, and transmits them to the signal acquisition module for processing. The data processing module calculates the water content result through a dual-weight model fusion. The control module manages the working sequence of each module throughout the process and outputs the final result, achieving accurate detection of water content in oil across the entire range.

[0025] Example 1 This embodiment applies to the scenario of detecting the water content of refined oil products in oil refineries. After refining and before bottling and transportation, the water content of the refined oil needs to be accurately detected to avoid excessive water content affecting the combustion efficiency of the refined oil, and to prevent moisture from causing corrosion of subsequent transportation equipment or emulsification and deterioration of the oil during storage. (See also...) Figure 1 and Figure 2 This embodiment employs a near-infrared scattering-absorption composite optics system for detecting water content in oil. It continuously detects the water content of gasoline and diesel, two common refined oil products. Through the coordinated operation of various modules of the system, it achieves accurate measurement of water content across the entire range, providing reliable data support for the quality control of refined oil products.

[0026] During the preparation phase, the connection and debugging of all system modules are completed first. The detection channel of the optical detection module is connected to the refined oil delivery pipeline of the refinery through a sealed interface to ensure smooth and leak-free oil flow within the channel. The insulation layer on the outside of the detection channel is fitted to the outer wall to maintain a stable oil temperature within the channel and prevent temperature fluctuations from interfering with the detection signal. A matte material light shield is installed on the outside of the near-infrared light source to reduce interference from ambient stray light on the near-infrared light emitted by the light source. The signal receiving surfaces of both photodetectors are covered with an anti-oil coating to prevent oil from adhering during the detection process and affecting the accuracy of signal acquisition.

[0027] The control module connects to an external operating terminal via a manual control interface. Based on the type of refined oil to be tested (gasoline or diesel), the user inputs a command on the external operating terminal to call the corresponding calibration parameters of the data calibration unit within the data processing module, thus completing the initial calibration of the system.

[0028] After the detection begins, the timing control unit of the control module outputs a light source start / stop command to the optical detection module. The near-infrared light source is activated and emits near-infrared light covering both the characteristic absorption band and the non-characteristic absorption band of moisture. Simultaneously, the timing control unit outputs a preset signal sampling frequency command to the signal acquisition module, and the signal acquisition module enters the signal receiving state. The oil to be tested in the refined oil pipeline of the refinery enters the detection channel of the optical detection module. The near-infrared light interacts with the oil in the channel. Some of the near-infrared light is absorbed by the moisture in the oil, and some is scattered by the moisture in the oil in the form of emulsion droplets and free water droplets. A photodetector arranged along the transmission direction collects the absorbed light signal after being interacted with by the oil, and a photodetector arranged along the scattering direction collects the scattered light signal. The two detectors synchronously output the collected signals to the signal acquisition module.

[0029] After receiving the absorbed light signal and the scattered light signal, the signal acquisition module first linearly amplifies the two weak signals respectively through a low-noise operational amplifier to ensure that the signal strength meets the requirements of subsequent processing. Then, the amplified signal is transmitted to the RC filter circuit to filter out environmental interference noise in the signal and obtain a clean analog signal. Finally, the filtered analog signal is transmitted to the analog-to-digital conversion unit, converted into a digital signal, and then output to the data processing module.

[0030] After receiving the digital signal from the signal acquisition module, the data processing module divides the corresponding absorption spectrum data and scattering spectrum data into two independent data blocks. Partial least squares regression is then applied to the absorption spectrum data blocks to analyze the correlation between the absorption spectrum and moisture concentration, extracting characteristic parameters of moisture concentration. Parallel orthogonalization techniques were used to eliminate interference from irrelevant variables in the scattering spectral data blocks, and the characteristic parameters of water droplet size distribution were extracted. .

[0031] Subsequently, based on the signal intensity of the absorption spectral data block, the water content range of the oil to be tested was initially determined. Based on the preset weighting rules, the absorption signal weighting coefficients were calculated. With the weighting coefficient of the scattered signal The formula for calculating the absorption signal weighting coefficient is: The formula for calculating the weighting coefficient of the scattered signal is: In the formula To obtain the real-time signal intensity of the absorption spectral data block, The preset maximum value of the absorbed signal intensity. The preset minimum value of the scattered signal intensity. This is the signal sensitivity adjustment coefficient. Based on the calculated weighting coefficients, the extracted feature parameters are fused to generate the oil water content result. The fusion calculation formula is: In the formula This is the correlation function between the droplet size distribution characteristic parameters and the water content. Throughout the data processing process, the data processing module calls the pre-set corresponding oil calibration parameters to correct the parameters in the feature extraction process, ensuring the accuracy of the calculation results.

[0032] The data processing module outputs the generated oil moisture content results to the control module. After receiving the moisture content results, the control module's data receiving unit transmits them to an external display device for real-time viewing by staff. Simultaneously, the results are stored on an external storage device for later data traceability and analysis. When switching to test a different type of refined oil, staff input commands through an external operating terminal to the control module's manual control interface to adjust the light source intensity and signal acquisition frequency parameters. Simultaneously, the data processing module retrieves the corresponding oil's calibration parameters, thus initiating a new round of testing.

[0033] In summary, this embodiment utilizes a near-infrared scattering-absorption composite optics system for detecting water content in oil, achieving continuous and accurate detection of water content in refined oil products from petroleum refineries. The system dynamically allocates the weights of the absorbed and scattered signals using a dual-weight model, effectively adapting to different water content ranges. This improves detection accuracy in low-water-content scenarios while ensuring stability in high-water-content scenarios. By employing the SPORT-PLS fusion algorithm for feature extraction and fusion calculation of the two data blocks, combined with oil product adaptation calibration in the data calibration unit, the accuracy and reliability of the detection results are significantly improved. This provides strong technical support for quality control before refined oil product bottling and transportation, helping to reduce problems such as decreased combustion efficiency and equipment corrosion caused by excessive water content, thus lowering production and operational risks.

[0034] Example 2 This embodiment applies to the scenario of water content detection before crude oil is transported from oilfield wellheads. Crude oil extracted from oilfields often contains water. If the water content exceeds the standard and is directly transported, it will increase transportation energy consumption, exacerbate pipeline corrosion, and may also affect the efficiency of subsequent crude oil processing. (See also...) Figure 1 and Figure 2 This embodiment employs a near-infrared scattering-absorption composite optics system for detecting water content in oil. This system performs real-time online detection of crude oil transported from wellheads. By collaboratively capturing the absorption and scattering optical characteristics of water in crude oil through the coordinated operation of various modules, it achieves accurate detection of crude oil in different water content ranges, providing real-time data support for adjusting oilfield crude oil dehydration processes and controlling the quality of transported crude oil.

[0035] Before testing, the system is deployed and debugged. The detection channel of the optical detection module is connected in series with the oilfield wellhead pipeline through a sealing interface made of oil-resistant rubber material to ensure that the channel and pipeline are tightly connected and there is no crude oil leakage. The insulation layer wrapped around the outside of the detection channel is tightly attached to the outer wall to maintain the stable temperature of the crude oil in the channel and avoid the interference of wellhead crude oil temperature fluctuations on the optical signal.

[0036] A matte-finish light shield is installed on the outside of the near-infrared light source to isolate stray light interference from the wellhead working environment. The signal receiving surfaces of both photodetectors are covered with an oil-resistant coating to prevent heavy components in the crude oil from adhering and affecting signal acquisition. The control module connects to the field operation terminal via a manual control interface. Based on the viscosity characteristics of the crude oil at the wellhead, commands are input into the operation terminal to retrieve calibration parameters adapted to high-viscosity crude oil from the data calibration unit within the data processing module, completing system parameter initialization.

[0037] After detection is initiated, the timing control unit of the control module outputs a light source activation command to the optical detection module. The near-infrared light source emits broad-band near-infrared light covering both the characteristic absorption band and the non-characteristic absorption band of water. Simultaneously, the timing control unit outputs a signal sampling frequency command matching the crude oil flow rate to the signal acquisition module, and the signal acquisition module enters the signal receiving ready state. Crude oil continuously flows into the detection channel from the wellhead pipeline. The near-infrared light penetrates the channel and interacts with the crude oil. Water in the crude oil exhibits characteristic absorption of the near-infrared light in a specific band, while dispersed free water droplets and emulsion droplets in the crude oil scatter the near-infrared light. A photodetector arranged along the transmission direction collects the transmitted light signal (i.e., absorbed light signal) after absorption by the crude oil, and a photodetector arranged along the scattering direction collects the scattered light signal after scattering by water droplets. The two detectors synchronously transmit the collected light signals to the signal acquisition module.

[0038] After receiving the absorbed light signal and the scattered light signal, the signal acquisition module first linearly amplifies the two weak light signals respectively through a low-noise operational amplifier to ensure that the signal strength meets the requirements of subsequent processing. Then, the amplified signal is transmitted to the RC filter circuit to filter out interference noise generated by the operation of the wellhead motor, environmental vibration, etc., to obtain a clean analog signal. Finally, the filtered analog signal is sent to the analog-to-digital conversion unit, converted into a digital signal, and output to the data processing module in real time.

[0039] After receiving the digital signal, the data processing module first divides the corresponding absorption spectrum data and scattering spectrum data into two independent data blocks. For the absorption spectrum data block, a partial least squares regression algorithm is used to analyze the correlation between the absorption spectrum and the water concentration in crude oil, extracting characteristic parameters of the water concentration. Parallel orthogonalization techniques were applied to the scattering spectral data blocks to eliminate signal interference from irrelevant components such as asphaltene and resins in crude oil, and to extract the characteristic parameters of water droplet size distribution. .

[0040] Subsequently, based on the real-time signal intensity of the absorption spectral data block, the current water content range of the crude oil is preliminarily determined, and the absorption signal weighting coefficient is calculated based on the preset weighting allocation rule. With the weighting coefficient of the scattered signal ,in , Based on the calculated weighting coefficients, the extracted feature parameters are fused and calculated to generate the crude oil water content result. The fusion calculation formula is: Throughout the data processing process, the data processing module calls the pre-set high-viscosity crude oil calibration parameters to correct the parameters in real time during the feature extraction process, ensuring that the calculation results are adapted to the characteristics of the wellhead crude oil.

[0041] The data processing module outputs the generated crude oil water cut results to the control module in real time. After receiving the results, the control module's data receiving unit transmits the water cut data to the wellhead field display terminal, allowing operators to monitor changes in crude oil water cut in real time. Simultaneously, the data is stored on the field server for subsequent dehydration process optimization analysis. When changes in the wellhead crude oil extraction layer lead to alterations in crude oil composition, operators can input commands to the control module via the operating terminal to adjust the near-infrared light source intensity and signal acquisition frequency. The data processing module then synchronously calls the corresponding calibration parameters for the crude oil components, ensuring continuous and accurate system detection.

[0042] In summary, this embodiment utilizes a near-infrared scattering-absorption composite optics system for detecting water content in oil at wellheads, achieving real-time, accurate online detection of crude oil water content. The system dynamically allocates the weights of absorption and scattering signals using a dual-weight model, effectively adapting to fluctuating crude oil water content scenarios. This solves the problem of insufficient detection accuracy caused by weak water absorption signals in low-water-content crude oil, while avoiding the impact of scattering signal interference on detection stability at high water content. Combined with the SPORT-PLS fusion algorithm and calibration parameters adapted to crude oil characteristics, the accuracy and reliability of the detection results are further improved. This provides real-time data support for quality control of crude oil transportation and adjustment of dehydration processes, helping to reduce energy consumption in oil transportation and pipeline corrosion risks, and improving the efficiency of oilfield production and operation.

[0043] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A near-infrared scattering-absorption composite optics system for detecting water content in oil, characterized in that, The system includes: an optical detection module, a signal acquisition module, a data processing module, and a control module; The optical detection module emits near-infrared light covering the characteristic absorption band and non-characteristic absorption band of moisture. After the near-infrared light interacts with the oil to be detected, the optical detection module collects the absorbed light signal and scattered light signal generated by the interaction of the oil and outputs them to the signal acquisition module. The signal acquisition module receives the absorbed light signal and scattered light signal output by the optical detection module, amplifies, filters and performs analog-to-digital conversion on the signal, and outputs the converted digital signal to the data processing module. The data processing module receives the digital signal output by the signal acquisition module. It has a built-in dual-weight model for scattering-absorption feature fusion based on the SPORT-PLS fusion algorithm constructed by the idea of ​​multi-block data analysis. Through this model, the absorption spectrum data and scattering spectrum data corresponding to the digital signal are extracted and fused to generate and output the oil water content result. The control module outputs control signals to the optical detection module, signal acquisition module, and data processing module to control the working sequence of each module, and at the same time receives the moisture content results output by the data processing module.

2. The near-infrared scattering-absorption composite optics oil water content detection system according to claim 1, characterized in that, The optical detection module includes a near-infrared light source, a detection channel, and a photodetector. The near-infrared light source is a broadband light-emitting device that emits near-infrared light covering both the characteristic absorption band and the non-characteristic absorption band of moisture. The detection channel forms a passage for the oil to be tested to flow through. Two photodetectors are provided, one arranged along the transmission direction of the near-infrared light and the other along the scattering direction of the near-infrared light. Sealed interfaces are provided at both ends of the detection channel, and an insulation layer is wrapped around the outside of the detection channel, with the insulation layer attached to the outer wall of the detection channel.

3. The near-infrared scattering-absorption composite optics oil water content detection system according to claim 1, characterized in that, The signal acquisition module processes signals by including the following steps: The absorbed light signal and the scattered light signal output by the optical detection module are received, and the two weak signals are linearly amplified by a low-noise operational amplifier respectively. The amplified signal is transmitted to the RC filter circuit to filter out environmental interference noise in the signal. The filtered analog signal is transmitted to the analog-to-digital converter unit, converted into a digital signal, and then output to the data processing module.

4. The near-infrared scattering-absorption composite optics oil water content detection system according to claim 1, characterized in that, The dual-weight model in the data processing module processes data including the following steps: The digital signal output from the signal acquisition module is received, and the corresponding absorption spectrum data and scattering spectrum data are divided into two independent data blocks. Partial least squares regression was used on the absorption spectrum data blocks to analyze the correlation between absorption spectra and moisture concentration, and characteristic parameters of moisture concentration were extracted. Parallel orthogonalization techniques were used to eliminate interference from irrelevant variables in the scattering spectral data blocks, and the characteristic parameters of water droplet size distribution were extracted. ; Based on the signal intensity of the absorption spectral data block, the water content range of the oil to be tested is initially determined. Based on the preset weighting rules, the absorption signal weighting coefficients are calculated. With the weighting coefficient of the scattered signal The calculation formula is , ,in, To obtain the real-time signal intensity of the absorption spectral data block, The preset maximum value of the absorbed signal intensity. The preset minimum value of the scattered signal intensity. This is the signal sensitivity adjustment coefficient; Based on the calculated weight coefficients, the extracted feature parameters are fused to generate the oil water content result. The fusion calculation formula is: ,in, This is the correlation function between the droplet size distribution characteristic parameters and the water content.

5. The near-infrared scattering-absorption composite optics oil water content detection system according to claim 1, characterized in that, The control module includes a timing control unit, a data receiving unit, and a result output unit. The timing control unit stores a preset control program, outputs a light source start / stop command to the optical detection module, and outputs a signal sampling frequency command to the signal acquisition module. The data receiving unit is connected to the data processing module and receives the moisture content calculation result output by the data processing module. The result output unit is connected to an external display device or storage device and transmits the moisture content result to the external display device or storage device.

6. The near-infrared scattering-absorption composite optics oil water content detection system according to claim 1, characterized in that, The data processing module also has a built-in data calibration unit, which stores calibration parameters corresponding to different types of oil. When detecting different types of oil, the data processing module calls the calibration parameters corresponding to the oil type in the data calibration unit to correct the parameters in the feature extraction process.

7. The near-infrared scattering-absorption composite optics oil water content detection system according to claim 5, characterized in that, The control module is also equipped with a manual control interface, which is connected to an external operating terminal. The external operating terminal inputs control commands to the manual control interface, and the control module adjusts the light source intensity and signal acquisition frequency parameters after receiving the commands.

8. The near-infrared scattering-absorption composite optics oil water content detection system according to claim 1, characterized in that, The photodetector surface of the optical detection module is provided with an anti-oil coating, which covers the signal receiving surface of the photodetector. A light shield is provided outside the near-infrared light source, and the light shield is arranged around the near-infrared light source.

Images