Search

CN-121999907-A - Water body sanitation index real-time monitoring system based on biosensor

CN121999907ACN 121999907 ACN121999907 ACN 121999907ACN-121999907-A

Abstract

The invention relates to the technical field of environmental monitoring, and particularly discloses a water body sanitation index real-time monitoring system based on a biosensor. The heterogeneous integrated sensing array is used for synchronously detecting various pollutants, and the deep belief network and the evidence theory are used for data fusion and reliability assessment, so that multi-index concentration inversion, combined pollution risk quantification and intelligent grading early warning are finally realized, and the comprehensiveness, accuracy and decision support capability of monitoring are improved.

Inventors

  • ZHANG KUNFENG
  • YANG YUTING
  • LI JINGHUI

Assignees

  • 佛山市沃特测试技术服务有限公司

Dates

Publication Date
20260508
Application Date
20260115

Claims (10)

  1. 1. Water health index real-time supervision system based on biosensor, its characterized in that includes: the integrated multi-sensor array module is used for synchronously sensing various pollutants in the water body; The microfluidic preprocessing and signal conditioning module is connected with the upstream and the downstream of the integrated multi-sensor array module and is used for preprocessing a water sample and normalizing an original sensing signal; The multi-mode data fusion and intelligent analysis module is used for receiving 3 paths of standardized digital signals from the microfluidic preprocessing and signal conditioning module and carrying out pollutant identification and result reliability evaluation; And the decision output and early warning module is connected to the downstream of the multi-mode data fusion and intelligent analysis module and is used for generating a grading early warning instruction based on the evaluated multi-index concentration data.
  2. 2. The biosensor-based real-time monitoring system for water body health index according to claim 1, wherein the integrated multi-sensor array module adopts a heterogeneous integrated architecture, at least 3 sensing units based on different biorecognition principles are integrated in situ on a common miniaturized substrate, and the sensing units comprise an electrochemical enzyme sensing unit, a surface plasmon resonance immunosensor unit and a whole-cell microorganism sensing unit; the electrochemical enzyme sensing unit forms a double-enzyme cascade reaction system through immobilized acetylcholinesterase and choline oxidase, converts an event of inhibiting enzyme activity of a target object into change of hydrogen peroxide concentration, and detects a current signal by a miniature three-electrode system; The surface of a sensing chip of the surface plasma resonance immunosensor unit is fixed with a monoclonal antibody aiming at a specific heavy metal ion integration, when heavy metal ions in a water sample are pre-combined with the integration on the surface of the chip to form an antigen, the antigen and the antibody are specifically combined to cause the refractive index change on the surface of the chip, and a built-in optical system is used for detecting resonance angular displacement; The whole-cell microorganism sensing unit adopts genetically engineered escherichia coli as a reporting cell, the reporting cell carries a fluorescent protein reporting gene controlled by an environment stress response promoter, when toxic substances or ammonia nitrogen and nitrate with specific concentration exist in a water body, the reporting gene expression is induced or inhibited, a micro photomultiplier detects the change of fluorescence intensity, and 3 sensing units share the same micro-flow path inlet driven by a micro peristaltic pump and are respectively provided with independent signal output ports.
  3. 3. The biosensor-based water body health index real-time monitoring system according to claim 2, wherein the microfluidic pretreatment and signal conditioning module comprises a multistage filtration and enrichment unit, a buffer solution and reagent injection unit and a parallel signal conditioning circuit, wherein the multistage filtration and enrichment unit sequentially carries out fine filtration on water inflow and guides water to selectively enrich through a microcolumn integrated with heavy metal ion chelating resin and organic pollutant adsorption resin, the buffer solution and reagent injection unit is precisely controlled by a plurality of micro electromagnetic valves, buffer solution with specific pH value, enzyme reaction substrate or excitation reagent required by fluorescence detection are injected into a shunted water sample on line, the parallel signal conditioning circuit respectively receives original electric signals or optical signals from 3 types of sensing units, and standardized processing including amplification, filtration and analog-to-digital conversion is carried out on the signals, wherein a phase-locked amplification technology is adopted for electrochemical signals, a correlated double sampling technology is adopted for optical signals, and finally a digital signal sequence with 3 paths of time synchronization and unified dimension is output.
  4. 4. The biosensor-based water body health index real-time monitoring system according to claim 3, wherein the multi-modal data fusion and intelligent analysis module comprises a feature extraction sub-module, a pollutant identification and concentration inversion sub-module based on a deep belief network and a collaborative anti-interference and reliability evaluation sub-module, the feature extraction sub-module is used for extracting at least 5 time domain and frequency domain feature parameters including response amplitude, response dynamics curve slope and reaching steady state time from each path of signals, an input layer of the pollutant identification and concentration inversion sub-module based on the deep belief network is an extracted multi-element feature parameter vector, an implicit layer of the multi-modal data fusion and intelligent analysis module is formed by stacking 3 restricted boltzmann machines, the output layer is used for learning high-order nonlinear mapping relations between different pollutants and concentrations and multi-element features of the pollutants, the collaborative anti-interference and reliability evaluation sub-module is used for constructing a data fusion model based on evidence theory, the detection result of the water body state of the 3 types of sensors is regarded as 3 independent sources, the probability of the existence of specific evidence pairs of the water body of the specific evidence pairs is calculated, and the combination of the specific evidence pairs of the water bodies are calculated, and if the degree of the specific evidence pairs of the water body of the specific evidence pairs are not supported by the threshold value is not equal to 1, if the degree of the combination of the specific evidence is not supported by the threshold value is equal to 0, and if the degree of the combination is not equal to 0, and the degree of the threshold is determined to be equal to the threshold, and the degree of the detection result is not equal to the threshold is equal to the threshold.
  5. 5. The biosensor-based water body sanitation index real-time monitoring system according to claim 4, wherein a multi-level water quality safety threshold database is preset in the decision output and early warning module, the database comprises a national standard limit value, an ecological risk reference value and an early warning baseline value based on local historical data statistics, the decision output and early warning module comprises a composite pollution risk index calculation unit and a dynamic early warning strategy generation unit, the composite pollution risk index calculation unit calculates the composite pollution index by adopting a weighted average model based on the received multi-index concentration, and simultaneously calculates the potential ecological toxicity index by adopting an effect addition model, and the dynamic early warning strategy generation unit compares the index, each single index concentration and reliability evaluation result thereof with the threshold database in real time and generates a grading early warning instruction according to the comparison result.
  6. 6. The biosensor-based water body health index real-time monitoring system according to claim 5, wherein the evidence fusion process in the collaborative anti-interference and reliability evaluation submodule is characterized in that basic probability distribution of specific pollution propositions of the evidence sources to the water body is calculated for each sensing unit evidence source, fusion calculation is carried out on the basic probability distribution of 3 evidences through a Dempster combination rule to obtain joint support degree and uncertainty, and the Dempster combination rule is calculated in a mode that products of intersection non-nulls in the basic probability distribution functions of all the evidence sources are summed up, and then divided by 1 minus the sum of products of intersection nulls in the basic probability distribution functions of all the evidence sources.
  7. 7. The system of claim 6, wherein the pollutant identification and concentration inversion sub-module based on deep belief network is further configured to perform online incremental learning by taking the feature parameter vector corresponding to the detection and the finally confirmed pollutant label and concentration as new training samples when the collaborative anti-interference and reliability evaluation sub-module determines that the detection result is highly reliable.
  8. 8. The biosensor-based real-time monitoring system for water body health index according to claim 7, wherein the multistage filtering and enriching unit in the microfluidic pretreatment and signal conditioning module is characterized in that the heavy metal ion chelating resin is iminodiacetic acid resin, the organic pollutant adsorption resin is hydrophobic reversed phase polymer resin, and the two resins are filled in different sections of the same microcolumn in a serial manner.
  9. 9. The biosensor-based real-time monitoring system of water sanitation index according to claim 8, wherein the genetically engineered escherichia coli in the whole-cell microbial sensing unit has an environmental stress response promoter selected from at least 2 of a heavy metal response promoter, an oxidative stress response promoter or a nitrogen metabolism related promoter.
  10. 10. The biosensor-based real-time monitoring system of water body health index according to claim 9, further comprising a self-cleaning and calibration maintenance module for periodically executing a cleaning process and a calibration process, wherein the cleaning process injects a cleaning solution containing protease and surfactant into the micro-flow path, the calibration process sequentially injects a series of standard solutions with concentration, and automatically updates the response curve of each sensing unit and the relevant parameters in the multi-mode data fusion and intelligent analysis module.

Description

Water body sanitation index real-time monitoring system based on biosensor Technical Field The invention belongs to the technical field of environmental monitoring, and particularly relates to a water body sanitation index real-time monitoring system based on a biosensor. Background The water body sanitation monitoring is one of core technologies in the fields of environmental science and public sanitation, and the core aims at continuously and accurately acquiring concentration data of various pollutants and sanitation indexes in the water body so as to evaluate water quality safety, early warn pollution events and guide treatment decisions. With the acceleration of the urban process and the increasing frequency of industrial activities, the requirement for real-time and online monitoring of water body health conditions becomes particularly urgent, wherein the biosensor technology has become an important development direction in the field due to high sensitivity, quick response and specific recognition capability. The biosensor-based real-time monitoring system aims at utilizing the specific response of a biological recognition element (such as enzyme, antibody, microorganism or nucleic acid) to a specific target object and converting the specific response into a measurable electrical signal or optical signal, thereby realizing continuous and automatic detection of specific health indexes in a water body. The rationale for such systems is the specific binding or reaction between the biological recognition element and the target contaminant, which in turn triggers a change in the sensor output signal. A single type of biosensor or detection principle is employed to detect a specific class of contaminants, for example, organophosphorus pesticides only with an enzyme sensor, or heavy metal ions only with an immunosensor. The single detection mode has the remarkable limitations that firstly, the types of pollutants in natural water bodies are complex and various, including heavy metals, organic pollutants, pathogenic microorganisms, nutritive salts and the like, and a single sensor cannot realize synchronous coverage and comprehensive evaluation of multiple pollutants, so that monitoring data are on one side, and the overall sanitary condition of the water bodies is difficult to comprehensively reflect. Secondly, the physicochemical properties and interference factors of different pollutants are different, the anti-interference capability of a single sensor is limited, and the single sensor is easily influenced by coexisting substances in a complex water body matrix, so that false positive or false negative appears in a detection result, and the reliability is insufficient. In addition, the existing system focuses on independent monitoring of a single index, lacks a mechanism for effectively fusing and cooperatively analyzing various sensor data, and cannot evaluate the composite pollution risk and the ecological toxicity of the water body from the whole level. Disclosure of Invention The invention aims to provide a water body sanitation index real-time monitoring system based on a biosensor, which is used for solving the technical contradiction that a single type of biosensor cannot synchronously cover multiple types of pollutants, the anti-interference capability is limited, and a multi-source data fusion and collaborative analysis mechanism is lacked, so that the water body sanitation condition is monitored on one side, the reliability is insufficient, and the risk of composite pollution cannot be estimated in the prior art. The technical scheme of the invention is that the water body sanitation index real-time monitoring system based on the biosensor comprises an integrated multi-sensor array module, a microfluidic preprocessing and signal conditioning module, a multi-mode data fusion and intelligent analysis module and a decision output and early warning module. The system realizes full-flow automation from sample collection, multi-index synchronous detection, intelligent data processing and risk decision output through cooperation of physical integration and logic. The integrated multi-sensor array module is a front-end sensing core of the system, and is not used for simply juxtaposing a plurality of sensors, but adopts a heterogeneous integrated architecture. The mass integration architecture is arranged on a common miniaturized substrate, and at least 3 sensing units based on different biological recognition principles are integrated in situ through a micro-nano processing technology. Specifically, the sensing unit comprises an electrochemical enzyme sensing unit, a surface plasmon resonance immunosensor unit and a whole-cell microorganism sensing unit. The electrochemical enzyme sensing unit is used for specifically detecting organophosphorus and carbamate pesticides in water, and a double-enzyme cascade reaction system is formed by immobilized acetylcholinesterase and choline oxidase, so that an even