CN-121994903-A - Electronic waste dismantling pollutant screening method based on non-targeting technology

CN121994903ACN 121994903 ACN121994903 ACN 121994903ACN-121994903-A

Abstract

The application relates to the technical field of environmental monitoring and pollutant tracing, and discloses a non-targeting technology-based electronic waste dismantling pollutant screening method, which utilizes an internet of things perception network module to monitor environmental and working condition characteristics, calculates a real-time pollution load index, and triggers sampling when the index continuously exceeds a threshold value; the integrated online sampling module is used for intercepting particulate matters by adjusting airflow through an inertial separation structure and capturing volatile organic matters through an adsorption medium, a sample is guided into the mobile high-resolution analysis module through flash evaporation thermal desorption and is used for acquiring mass spectrum data through a soft ionization source, the intelligent recognition and tracing module is used for constructing a deep neural network model based on physical constraint, and the contribution rate and the azimuth probability of a dismantling procedure are analyzed through non-negative projection and sparse regularization data and wind field information. The application realizes the accurate capture and space-time tracing of complex pollutants in the process of disassembling electronic wastes.

Inventors

LUO BINTAO
Qiu Shengqun
Bai Wenhe

Assignees

罗斌韬
邱胜群
白文鹤

Dates

Publication Date: 20260508
Application Date: 20260210

Claims (10)

1. The electronic waste dismantling pollutant screening method based on the non-targeting technology is characterized by comprising the following steps of: Monitoring environmental parameters, meteorological parameters, positioning information, working condition information and material characteristics by using an internet of things sensing network module (200), and sending a starting signal to an integrated online sampling module (100) when the characteristic index value exceeds a preset threshold value; The integrated online sampling module (100) receives the starting signal to start a sampling pump, particles are classified and trapped to a filter membrane by utilizing an inertial separation structure, and volatile organic compounds are captured by utilizing an adsorption medium; after the sampling is finished, the integrated online sampling module (100) starts a thermal desorption program to heat the filter membrane and the adsorption medium so as to desorb and gasify pollutants, and the samples are led into the mobile high-resolution analysis module (300); Acquiring mass spectrum data in a full scanning mode by using the mobile high-resolution analysis module (300) by using a soft ionization source to form an original data stream; The original data stream is processed by an intelligent identification and tracing module (400) to generate a feature vector, and the feature vector is compared with a database to determine the type of the pollutant; The intelligent identification and tracing module (400) combines the determined pollutant types, the pollution source fingerprint library and the working condition information, calculates the contribution rate of the disassembling procedure or the material by using a chemometric algorithm, and generates an analysis report.
2. The method for screening the electronic waste dismantling pollutants based on the non-targeting technology according to claim 1, wherein the step of monitoring the environmental parameters and the material characteristics specifically comprises the steps of collecting monitoring data by using a particle sensor provided with a light scattering principle, a photoionization detector and the internet of things sensing network module (200) of a material fingerprint identification unit; Calculating a real-time pollution load index through an edge calculation gateway; The calculation of the real-time pollution load index is achieved by multiplying the ratio of the particulate matter concentration to a preset particulate matter concentration reference threshold, the ratio of the volatile organic matter concentration to a preset volatile organic matter concentration reference threshold and the matching degree of the material spectrum characteristics by corresponding dimensionless weight coefficients respectively and then carrying out weighted summation.
3. The non-targeted technology-based electronic waste dismantling pollutant screening method according to claim 2, wherein the trigger logic for sending the start signal to the integrated on-line sampling module (100) is: And the edge computing gateway judges that the real-time pollution load index is larger than a trigger threshold value and judges that the state duration time of the real-time pollution load index larger than the trigger threshold value exceeds a set minimum confirmation time.
4. The method for screening the electronic waste dismantling pollutants based on the non-targeting technology according to claim 1, wherein the integrated online sampling module (100) controls the average air flow speed at the outlet of the nozzle by adjusting the air extraction flow of the downstream sampling pump, so that coarse particles with the particle size larger than the cutting particle size cannot be flushed into the receiving tube, and fine particles with the particle size smaller than the cutting particle size deflect along with the main air flow and are guided to pass through the built-in filter membrane; in the step of capturing volatile organic compounds by using the adsorption medium, air flow is enabled to flow through the adsorption medium arranged at the downstream of the filter membrane, and the adsorption medium is filled with a weak-polarity adsorbent and a strong-polarity adsorbent in sequence along the air flow direction.
5. The method for screening the electronic waste disassembled pollutant based on the non-targeting technology according to claim 1, wherein the thermal desorption procedure comprises the steps of driving a valve group to act through a flow path control unit, and utilizing inert gas to establish a purging loop to remove residual moisture and oxygen in the adsorption medium; the flash type heating strategy is executed through the heating unit to control the heating rate so as to raise the temperature; In the step of introducing the sample into the mobile high-resolution analysis module (300), the sample is transmitted through a transmission pipeline with an inner wall made of inert treatment materials and an outer wrapping heat tracing belt and a heat insulation layer, and the temperature of the pipe wall is maintained at a constant high temperature state by a PID controller.
6. The method for screening the electronic waste dismantling pollutants based on the non-targeting technology according to claim 1, wherein in the step of collecting mass spectrum data, a mass spectrometer built in the mobile high-resolution analysis module (300) is subjected to vibration isolation protection through a steel wire rope vibration isolation system; and periodically injecting an internal standard compound with known accurate mass number through an independent reference channel, and reversely calibrating the mass axis coefficient according to the actual measurement flight time of the internal standard compound.
7. The method for screening the electronic waste disassembled pollutants based on the non-targeting technology according to claim 1, wherein the step of processing the raw data stream specifically comprises the steps of constructing a deep neural network model based on physical constraints; The depth neural network model comprises an encoder and a decoder, wherein the encoder compresses a normalized concentration matrix constructed based on the feature vector to a hidden layer, and an activation function of the hidden layer selects a correction linear unit; the decoder reconstructs the hidden layer characteristics back to the original dimension, and the connection weight of the decoder corresponds to the source component spectrum; When the updated weights are back propagated, the updated negative weights are zeroed by the projection operator.
8. The non-targeting technology-based electronic waste dismantling pollutant screening method according to claim 7, wherein the training objective of the deep neural network model is achieved by minimizing a loss function; The loss function consists of a weighted reconstruction error term and a sparse regularization term; The weighted reconstruction error term comprises the square of the difference between the measured concentration and the network reconstruction output value, and weighting is carried out by utilizing measurement uncertainty; The sparse regularization term includes a product of an L1 regularization coefficient and a source contribution term corresponding to the hidden layer feature.
9. The method for screening electronic waste dismantling pollutants based on non-targeting technology according to claim 8, wherein before calculating different dismantling procedures or material contribution rates by using a chemometric algorithm, the method further comprises a data washing step: Taking a time stamp of mass spectrum data as a reference, carrying out up-sampling reconstruction on meteorological and positioning data contained in the environmental parameters, and identifying and eliminating dead wind time period data with wind speed smaller than a set value; And pre-analyzing the concentration matrix through singular value decomposition, and determining the number of potential pollution sources according to the characteristic value accumulation contribution rate.
10. The method for screening the electronic waste disassembled pollutants based on the non-targeting technology according to claim 7, wherein after the time contribution sequence of each pollution source is analyzed, a conditional bivariate probability function is calculated by combining synchronous wind field data contained in the environmental parameters, and the contribution of the time dimension is mapped to the azimuth of the space dimension; When the total sample number in a certain wind direction sector is smaller than the minimum statistical threshold value, the probability value in the wind direction sector is forced to be zero, and finally the azimuth probability radar chart of each pollution source is output.

Description

Electronic waste dismantling pollutant screening method based on non-targeting technology Technical Field The invention relates to the technical field of environmental monitoring and pollutant tracing, in particular to a non-targeting technology-based electronic waste dismantling pollutant screening method. Background The electronic waste dismantling industry is a typical complex pollution emission source, and various procedures such as crushing, heat treatment, chemical leaching and the like are involved in the operation process, so that mixed pollutants including heavy metals, polybrominated diphenyl ethers, polycyclic aromatic hydrocarbons and a large number of unknown pyrolysis products can be released. The components of the pollutants are complex, the toxicity is obvious, the environmental durability is strong, and the pollutants form potential threats to the surrounding ecological environment and the human health. Therefore, the establishment of the efficient pollutant screening and tracing method has important significance for the environment supervision and risk prevention and control of the park. Current electronic waste dismantling pollutant monitoring technologies rely mainly on fixed-point off-line sampling and laboratory analysis modes. The method generally carries out long-time integral sampling according to a preset time schedule, the electronic waste dismantling operation often has obvious paroxysmal and intermittent characteristics, the instantaneous high-concentration emission event is difficult to capture in a fixed sampling period, the key pollution characteristics are easily diluted by background air, and the pollution load in the operation peak period cannot be truly reflected. In addition, offline analysis has a long time lag, and it is difficult to meet the requirement of environmental supervision on quick response to sudden pollution events. In the aspects of analysis detection and tracing, the prior art mostly adopts a target analysis strategy, namely quantitative detection is carried out aiming at a known specific pollutant list. However, the electronic waste can generate a large amount of secondary pollutants or thermal degradation products with unknown structures in the thermal disassembly process, and only the known target objects are focused to cause a large amount of potential toxic and harmful substances to be omitted, so that the environmental risk cannot be comprehensively estimated. Meanwhile, when the traditional data analysis method processes mass data generated by high-resolution mass spectrum, a positive definite matrix factorization receptor model is often adopted for source analysis. The models often lack strict physical constraint in the mathematical solving process, the phenomenon that source contribution values in calculation results are negative numbers and the like and do not accord with physical objective rules easily occurs, and the sparsity problem of high-dimensional data is difficult to process, so that analyzed pollution source fingerprints are difficult to correspond to an actual dismantling process, and specific generation links and space orientations of pollutants cannot be accurately judged. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a non-targeting technology-based electronic waste dismantling pollutant screening method, which solves the problems that the existing electronic waste dismantling pollutant monitoring technology is difficult to capture instantaneous emission due to fixed-point offline sampling, unknown secondary pollutants are easy to miss in targeted analysis, and the tracing accuracy is insufficient due to the lack of physical constraint of a traditional source analysis model. The electronic waste dismantling pollutant screening method based on the non-targeting technology mainly comprises four stages of environment sensing triggering, integrated sampling, analysis and detection and intelligent tracing. In the environment sensing triggering stage, an environment parameter, a meteorological parameter, positioning information, working condition information and material characteristics are monitored in real time through an Internet of things sensing network module, a characteristic index threshold value is set in the system, and when the monitored characteristic index value exceeds the preset threshold value, a starting signal is sent to an integrated online sampling module, so that the targeted capturing of a high pollution period or a specific working condition is realized. And at the integrated sampling stage, the integrated online sampling module starts a sampling pump after receiving a starting signal, the module utilizes an inertial separation structure to carry out graded interception on the particles so as to deposit the particles on the filter membrane, and simultaneously utilizes an adsorption medium to capture volatile organic matters in the airflow, after the sampling is finished, a t