CN-121978181-A - Seawater sample residual chlorine on-line monitoring method combined with intelligent sensing

CN121978181ACN 121978181 ACN121978181 ACN 121978181ACN-121978181-A

Abstract

The invention relates to the technical field of environment monitoring and intelligent sensing, and discloses an online monitoring method for residual chlorine in a seawater sample by combining intelligent sensing. According to the method, multi-source heterogeneous data are synchronously acquired through a multi-mode electrochemical sensing array and a miniature ultraviolet-visible absorption spectrum module, characteristic decoupling is carried out by utilizing a depth self-encoder and independent component analysis combined model, free residual chlorine dedicated components are extracted, salinity, turbidity and organic matter interference are corrected by combining a dynamic environment compensation model, and finally high-selectivity and high-sensitivity real-time concentration calculation is realized. The invention has the self-adaptive signal decoupling and on-line model fine tuning capability, suppresses cross interference and long-term drift, realizes reagent-free and continuous on-line monitoring, and improves the accuracy and reliability of residual chlorine detection in marine environment.

Inventors

SHEN XUPING
LI PANFENG
GUO ZHIMING
YUAN ZHIGANG
Pan boyuan
ZHANG JIANBIN
LI ZENGGUANG
WEN HONGJIAN

Assignees

新奥(舟山)液化天然气有限公司

Dates

Publication Date: 20260505
Application Date: 20260403

Claims (10)

1. The online monitoring method for residual chlorine in a seawater sample combined with intelligent sensing is characterized by comprising the following steps of: collecting seawater samples and flowing through the multichannel sensing reaction cavity; Synchronously acquiring a plurality of electrochemical response signals through a multi-mode electrochemical sensing array; The multi-mode electrochemical sensing array comprises a first sensing unit, a second sensing unit and a third sensing unit; The first sensing unit adopts a platinized gold working electrode and is modified with a nano titanium dioxide catalytic layer to generate broad-spectrum response to chlorine-containing oxidizing substances; the second sensing unit adopts a gold-plated working electrode and is fixed with a molecular imprinting polymer film taking free residual chlorine as a template molecule so as to specifically identify the free residual chlorine; the third sensing unit adopts a glassy carbon working electrode and is coated with a polyaniline conductive polymer layer to respond to bromide and chloramine interference substances; Obtaining absorbance values of the seawater sample at different characteristic wavelengths through a miniature ultraviolet-visible absorption spectrum module; performing space-time alignment on the electrochemical response signal and the absorbance value to form a multi-source heterogeneous sensing data set; performing component separation processing on the multi-source heterogeneous sensing data set based on a preset characteristic decoupling model, extracting characteristic components only related to free residual chlorine, wherein the characteristic decoupling model is a neural network model based on a depth self-encoder and independent component analysis combined architecture, and outputting a plurality of mutually statistically independent potential variables; and calculating the concentration value of free residual chlorine in the seawater sample according to the characteristic component, and outputting a monitoring result.
2. The online monitoring method for residual chlorine of seawater samples combined with intelligent sensing according to claim 1, wherein a microfluidic structure is arranged in the multichannel sensing reaction cavity, and comprises a main sample injection channel, a shunt channel, a mixing chamber and a waste liquid outlet; the main sampling channel is connected with a seawater sampling pump to introduce a seawater sample to be tested; the split flow channel equally divides the seawater sample into three paths which are respectively led into detection chambers corresponding to the first sensing unit, the second sensing unit and the third sensing unit; The bottom of each detection chamber is provided with a constant temperature heating plate to maintain the reaction temperature at 25 ℃, and the top of each detection chamber is provided with a miniature stirring magnet driven by an external rotating magnetic field to ensure that the sample uniformly contacts the sensing interface.
3. The on-line monitoring method for residual chlorine in a seawater sample combined with intelligent sensing according to claim 2, wherein the micro ultraviolet-visible absorption spectrum module comprises a deuterium lamp light source, a quartz cuvette, a grating beam splitter and a linear photodiode array detector; The quartz cuvette is integrated in a bypass channel of the multichannel sensing reaction cavity; The grating beam splitter projects the transmitted light to the linear photodiode array detector after dispersing the transmitted light according to wavelength.
4. The on-line monitoring method for residual chlorine in seawater samples combined with intelligent sensing according to claim 3, wherein the time-space alignment of the multi-source heterogeneous sensing data set is realized through a hardware-level synchronous trigger mechanism; the multimode electrochemical sensing array and the miniature ultraviolet-visible absorption spectrum module share the same clock source; Simultaneously recording current response values of a plurality of electrochemical sensing units and absorbance values of a plurality of characteristic wavelengths at each sampling moment to form a multidimensional vector; Continuously collecting data to form a time window as an input of the characteristic decoupling model.
5. The on-line monitoring method for residual chlorine in seawater samples combined with intelligent sensing according to claim 4, wherein the characteristic decoupling model comprises an encoder, a decoder and an independent component constraint layer; the encoder is composed of three layers of fully-connected neural networks; The independent component constraint layer is embedded between the last layer of the encoder and the first layer of the decoder, and a plurality of potential variables which are forcedly output are mutually independent in statistics; Model training used a standard seawater sample dataset of manual formulation, with the tag data being the true concentrations of free residual chlorine, monochloramine, dichloramine, nitrogen trichloride, bromide and hypobromous acid.
6. The on-line monitoring method for residual chlorine in seawater samples combined with intelligent sensing according to claim 5, wherein in a model reasoning stage, the multi-source heterogeneous sensing data acquired in real time are input into a feature decoupling model with completed training, so as to obtain a plurality of independent components; Determining independent components corresponding to the free residual chlorine through a component-substance mapping relation table; the component-substance mapping relation table is established through an off-line calibration experiment, and the specific method comprises the following steps: and (3) adding single standard substances into the chlorine-free artificial seawater in a progressive manner, recording the change trend of the response intensity of each independent component, and marking the component with the response intensity monotonically increasing along with the concentration of the free residual chlorine and weak response to other substances as the residual chlorine exclusive component.
7. The on-line monitoring method of residual chlorine in a seawater sample combined with intelligent sensing according to claim 6, wherein the monitoring method further comprises a dynamic environment compensation step; The dynamic environment compensation model is used for correcting the interference of the salinity, turbidity and organic matter background of the seawater on the monitoring result; The input variables of the dynamic environment compensation model comprise seawater temperature, conductivity, dissolved oxygen concentration and total organic carbon content, and the output is a correction coefficient for the exclusive component of residual chlorine; the seawater temperature is measured by a platinum resistance temperature sensor; the conductivity is measured by a four-electrode conductivity cell, and the dissolved oxygen concentration is measured by a fluorescence quenching type dissolved oxygen probe; the total organic carbon content is obtained through conversion of an absorbance value empirical formula.
8. The on-line monitoring method of residual chlorine in a seawater sample combined with intelligent sensing according to claim 7, wherein the monitoring method further comprises an adaptive signal decoupling algorithm; When the system detects that the residual chlorine concentration variation amplitude is smaller than a specified value and the environmental parameters are stable in continuous time, starting a model fine adjustment mode; In a model fine tuning mode, the system collects a current seawater sample and injects a trace of free residual chlorine standard solution with known concentration to form a disturbance sample; Inputting multisource heterogeneous sensing data of a disturbance sample into a current model, and calculating the deviation of the predicted concentration and the actual added concentration; and if the absolute value of the deviation is larger than the set value, triggering a back propagation algorithm to update the weight of the characteristic decoupling model in a small step.
9. The online monitoring method for residual chlorine of seawater samples combined with intelligent sensing according to claim 8, wherein the monitoring result is uploaded to a central monitoring platform in real time through an industrial Ethernet interface or an RS485 bus; Triggering an audible and visual alarm and recording an event log when detecting that the residual chlorine concentration is greater than or equal to a preset safety threshold value; The event log includes a time stamp, raw sensing data, a decoupling characteristic value, a compensated concentration value, and an environmental parameter.
10. The on-line monitoring method for residual chlorine in seawater samples combined with intelligent sensing according to claim 9, wherein the nano titanium dioxide catalytic layer is synthesized by a sol-gel method.

Description

Seawater sample residual chlorine on-line monitoring method combined with intelligent sensing Technical Field The invention belongs to the technical field of environment monitoring and intelligent sensing, and particularly relates to an online monitoring method for residual chlorine in a seawater sample by combining intelligent sensing. Background Along with the rapid development of the fields of marine environment monitoring, sea water desalination, coastal industrial cooling water treatment and the like, the real-time and accurate online monitoring of the concentration of residual chlorine in sea water is increasingly demanded. The residual chlorine is used as a key index for measuring disinfection effect and ecological risk, and the concentration level of the residual chlorine directly influences the safety of a marine ecological system and human health. The traditional residual chlorine detection method mainly depends on an electrochemical sensor or a colorimetric method, and has a certain application foundation in a fresh water system, but the traditional residual chlorine detection method has a serious challenge in a complex seawater matrix. The seawater is rich in high-concentration salt, bromide, iodide and organic amine substances, and is easy to generate side reaction with residual chlorine to generate structural analogues such as chloramine, hypobromous acid and the like, so that the response signals of the sensor are severely cross-interfered, and the specific identification of the target is difficult to realize. The detection technology based on the single sensing principle generally has the problems of poor selectivity and low stability due to the lack of resolving power of complex components. In recent years, the microfluidic chip technology is tried to be used for pretreatment of water quality analysis due to the advantages of low sample consumption, high separation efficiency, strong concentration Cheng Du and the like, but the existing design focuses on physical filtration or simple laminar flow separation, and cannot effectively simulate a dynamic resolution mechanism of a biological system on mixed odor molecules, so that high-efficiency decoupling of an interfering substance and target residual chlorine is difficult to realize at a molecular level. Meanwhile, although the artificial intelligence algorithm is developed in the field of signal processing, the conventional deep learning model depends on a large amount of annotation data, has high calculation cost, cannot meet the requirements of on-line monitoring on low power consumption and real-time response, and lacks bionic reference on a biological olfactory nerve coding mechanism. In the prior art, no scheme has been provided for deep fusion of dynamic recognition logic of bionic olfaction and microfluidic front-end separation function. The design of the microfluidic channel does not introduce a bionic gradient diffusion or selective adsorption structure, so that the space-time response difference of the biological olfactory epithelium to different volatile molecules cannot be reproduced, and the signal processing link still adopts a static classifier, ignores the impulse response characteristic of the multichannel sensor array in the time dimension, and leads to the limitation of the overall specificity of the system. Especially in the sea water environment with high salt and high interference, the defects are further amplified, so that the residual chlorine detection result is easily influenced by chloramine, bromide and other coexisting substances, obvious false positive or false negative is generated, and the reliability and applicability of the online monitoring system are severely restricted. Therefore, a novel online monitoring method for residual chlorine in sea water integrating a bionic recognition mechanism and a microfluidic separation technology is needed to break through the bottleneck of the prior art in terms of specificity and robustness. Disclosure of Invention The invention provides an on-line monitoring method for residual chlorine in a seawater sample by combining intelligent sensing, and aims to solve the technical problems that residual chlorine in seawater is difficult to distinguish from analogues such as chloramine and bromide and the like and has insufficient specificity. According to the method, a multi-mode electrochemical sensing array and spectrum characteristic fusion recognition mechanism is constructed, and a dynamic environment compensation model and a self-adaptive signal decoupling algorithm are combined, so that high-selectivity, high-sensitivity and real-time online monitoring of free residual chlorine in seawater is realized. The invention provides an online monitoring method for residual chlorine in a seawater sample by combining intelligent sensing, which comprises the following steps: collecting seawater samples and flowing through the multichannel sensing reaction cavity; Synchronously acqu