PMF source analysis method based on PM2.5 / CO characteristic ratio
By constructing a dynamic ratio database and embedding it into the PMF model, combining the β-ray method with non-dispersive infrared technology, the difficulty of the PMF model in distinguishing between coal-fired sources and fireworks sources was solved, and high-precision pollution source analysis and contribution rate determination were achieved.
Patent Information
- Application Number
- CN202510753489.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-06
- Publication Date
- 2025-09-19
AI Technical Summary
The existing PMF model has difficulty distinguishing the cumulative effects and instantaneous bursts of coal-burning sources and fireworks sources, resulting in misjudgment of source contribution rates and a lack of capture of the synergistic release characteristics of PM2.5 and CO.
A PMF source apportionment method based on the PM2.5/CO characteristic ratio is adopted. By constructing a dynamic ratio database and embedding dynamic characteristic thresholds as constraints to optimize the PMF model, the β-ray method and non-dispersive infrared technology are combined to eliminate monitoring errors and enhance the constraints on emission dynamic characteristics.
It significantly improves the accuracy of source analysis results, reduces the overlapping errors of source contribution rates, provides clear pollution source types and contribution rates, and provides a scientific basis for targeted management and control.
Smart Images
Figure CN120673914A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of air particulate matter source analysis, in particular to a method based on PM 2.5 PMF source apportionment method based on the characteristic ratio of CO / CO. Background Art
[0002] Air particulate matter source analysis technology is the core means of precise pollution control. It can identify PM 2.5 The contribution rate of emission sources of pollutants such as pollutants provides a scientific basis for targeted control.
[0003] Currently, the traditional element labeling method is generally used to sample the PMF source apportionment model. This method has difficulty in distinguishing the collinearity between the cumulative effect and instantaneous burst of coal-burning sources and fireworks and firecrackers sources. In the short-term high pollution events in winter, there are the following significant limitations: (1) Both coal-burning and fireworks and firecrackers sources have high sulfur characteristics, which leads to confusion of δ34S isotope signals in the PMF model; (2) Fireworks and firecrackers cause abnormal fluctuations in the background value of potassium; (3) Characteristic elements such as Se and As are not applicable in short-term high-intensity emission scenarios; (4) Existing ratio methods (such as K + / OC) is significantly affected by the secondary aerosol formation process. In addition, the traditional PMF model cannot capture PM2.5 due to the lack of emission dynamics constraints. 2.5 The synergistic release characteristics of CO (such as low-ratio steady-state emissions from coal-fired sources vs. instantaneous high-ratio bursts from fireworks and firecrackers) ultimately lead to misjudgment of source contribution rates. Summary of the Invention
[0004] In order to solve the above technical problems, the present invention provides a PM-based 2.5 The PMF source apportionment method based on the characteristic ratio of CO / CO is particularly suitable for distinguishing the cumulative effects of coal burning sources and fireworks and firecracker sources and solving the collinearity problem of instantaneous bursts.
[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:
[0006] A PM-based 2.5 The PMF source apportionment method for the CO / CO characteristic ratio includes the following steps:
[0007] Collect PM in the target time period and location 2.5 Chemical composition and CO concentration data are standardized to generate a data matrix;
[0008] From PM 2.5 Extract the dynamic characteristic thresholds of coal burning sources and fireworks and firecracker sources from the CO2 / CO2 dynamic ratio database;
[0009] The dynamic feature threshold is embedded as a constraint condition into the objective function of the PME model to optimize the PME model;
[0010] The data matrix is input into the optimized PME model to analyze pollution sources and quantify source class contributions.
[0011] Preferably, the construction of the dynamic ratio database includes the following steps:
[0012] Collect PM from coal burning and fireworks sources 2.5 Data on chemical composition and CO concentration;
[0013] Based on the data, PM 2.5 / CO characteristic ratio interval;
[0014] Fireworks and firecrackers source PM is established based on the data during the Spring Festival 2.5 / CO characteristic ratio interval;
[0015] Calculate PM in different pollution events 2.5 The median and fluctuation range of the / CO ratio.
[0016] Preferably, the PM2.5-bearing rate of coal combustion sources is established by combining the β-ray method with the non-dispersive infrared technology. 2.5 / CO characteristic ratio range.
[0017] Preferably, the new function obtained by embedding the dynamic feature threshold as a constraint into the objective function of the model is as follows:
[0018]
[0019] Among them, R k PM of the kth pollution source 2.5 / Dynamic feature threshold in CO feature ratio database, X ij is the mass concentration of the jth component or gas in the i-th sample, G ik is the contribution of the kth source class to the i-th sample, F kj is the content of the jth component in the source spectrum of the kth source class, u ij is the uncertainty of the jth component in the i-th sample, and λ is the constraint weight coefficient.
[0020] Preferably, the parsing results are evaluated using residual analysis and source spectrum similarity testing.
[0021] Preferably, the spatial transmission path of the pollution source is verified through the backward trajectory model and matched with the fireworks sales data to confirm the analysis results.
[0022] The present invention also provides an electronic device, comprising: at least one processor; a memory connected in communication with the at least one processor; and the memory storing instructions executable by the at least one processor, the instructions being executed by the at least one processor so that the at least one processor can execute the PM-based 2.5 PMF source apportionment method based on the characteristic ratio of CO / CO.
[0023] The present invention also provides a computer-readable storage medium, wherein a computer program is stored therein, and when the program is executed by a processor, the PM-based 2.5 PMF source apportionment method based on the characteristic ratio of CO / CO.
[0024] The present invention also provides a computer program product, including a computer program / instruction, which is executed by a processor based on PM. 2.5 PMF source apportionment method based on the characteristic ratio of CO / CO.
[0025] The present invention has the following advantages and positive effects due to the adoption of the above technical solution:
[0026] 1. By constraining the dynamic characteristic threshold, it is integrated into the PMF model as a constraint indicator. From this, the differences in emission dynamics (primary emission and secondary transformation) and spatiotemporal distribution of the two pollutants can be utilized to enhance source specificity, that is, to enhance the model's consideration of PM in the analysis process. 2.5 The difference in emission characteristics of CO2 / CO significantly reduces the overlap error of source contribution rate, thereby greatly improving the accuracy of the analysis results;
[0027] 2. The construction of a standardized data matrix is compatible with multi-source heterogeneous data input and can adapt to different monitoring equipment and pollutant combination scenarios. The output results are directly linked to the pollution source type and contribution rate, providing a clear basis for targeted control.
[0028] 3. The β-ray method is combined with NDIR to eliminate the systematic errors of a single method through technical complementarity; high-precision data acquisition provides a technical basis for the extraction of feature thresholds, avoiding model constraint failures caused by monitoring errors;
[0029] 4. By using constraint terms to force the pollution source ratio calculated by the model to approach the characteristic threshold, the emission dynamics characteristics are integrated into the mathematical optimization process to avoid the black box defects of pure data-driven methods. The constraint weight coefficient allows the constraint strength to be dynamically adjusted according to the type of pollution event, balancing the priority of chemical component fitting and ratio constraints. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 Schematic diagram of the steps of the analytical method of the present invention. DETAILED DESCRIPTION
[0031] The present disclosure is described more fully below with reference to the accompanying drawings, which illustrate exemplary embodiments of the present disclosure. The technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present disclosure.
[0032] like Figure 1 As shown, the present invention provides a PM-based 2.5 The PMF source apportionment method for the CO / CO characteristic ratio includes the following steps:
[0033] S100, collect PM in target time period and location 2.5 Chemical composition and CO concentration data are standardized to generate a data matrix;
[0034] The data are first calibrated and preprocessed to eliminate possible noise and errors, and then standardized to facilitate subsequent model input.
[0035] S200, from PM 2.5 Extract the dynamic characteristic thresholds of coal burning sources and fireworks and firecracker sources from the CO2 / CO2 dynamic ratio database;
[0036] The dynamic ratio database is a specially constructed database that stores PM values from different pollution sources under different conditions. 2.5 By querying this database, we can obtain the representative dynamic characteristic thresholds related to the current study area.
[0037] S300, embedding the dynamic feature threshold as a constraint condition into the objective function of the PDM model, thereby optimizing the PDM model;
[0038] S400, input the data matrix into the optimized PDM model to analyze the pollution sources and quantify the contributions of source classes.
[0039] Using the above method, the model is forced to consider PM in the parsing process through dynamic feature threshold constraints. 2.5 / CO emission characteristics, significantly reducing the overlapping error of source contribution rate; the construction of standardized data matrix is compatible with multi-source heterogeneous data input, and can adapt to different monitoring equipment and pollutant combination scenarios; the output results are directly related to the pollution source type and contribution rate, providing a clear basis for targeted control.
[0040] The data standardization process may adopt any known and feasible technology, which is not the innovation of the present invention and will not be described here in detail.
[0041] In order to solve the existing ratio method (such as K + / OC) relies on a static database and cannot reflect the dynamic characteristics of pollution sources in short-term high-intensity emissions, resulting in the problem that model constraints are out of touch with actual emission scenarios. In this embodiment, an implementation method is provided.
[0042] In one embodiment, the construction of the dynamic ratio database includes the following steps:
[0043] Collect PM at the site to be measured 2.5 Data on chemical composition and CO concentration;
[0044] Based on the data, PM 2.5 / CO characteristic ratio interval;
[0045] Based on the data during the Spring Festival, a special PM of fireworks and firecrackers source was established. 2.5 / CO sign ratio interval;
[0046] Calculate PM in different pollution events 2.5 The median and fluctuation range of the / CO ratio.
[0047] In one embodiment, the data may be derived from measured data of multiple coal-fired power plants.
[0048] In order to solve the problem of traditional single monitoring technology (such as β-ray method) in PM 2.5 In the simultaneous monitoring of CO, there are problems of insufficient sensitivity or cross-interference, which leads to limited accuracy of dynamic ratio data. In this embodiment, a method is provided to achieve this. That is, based on this data, the PM2.5 ratio of coal-fired sources is established by combining the β-ray method with the non-dispersive infrared technology (NDIR). 2.5 The β-ray method is combined with non-dispersive infrared technology to eliminate the systematic errors of each method through technical complementarity. High-precision data acquisition provides a technical foundation for the extraction of characteristic thresholds, avoiding model constraint failure caused by monitoring errors.
[0049] The different pollution events include coal-fired heating and fireworks during the Spring Festival. The PM2.5 levels caused by these two pollution events are: 2.5 / CO ratio is different, specifically, for example, coal burning source: PM 2.5 The CO / m ratio was stable at 0.5-1.2 μg / m 3 / ppm (affected by combustion efficiency); Fireworks and firecrackers source: PM 2.5 The instantaneous CO / CO ratio can reach 3.5-5.0 μg / m 3 / ppm(High PM 2.5 Emissions are accompanied by short-term CO peaks).
[0050] Using the above method, we can complete the construction of a database based on hourly medians and fluctuation ranges, thereby effectively capturing the instantaneous outbreak characteristics of pollution events (such as the setting off of fireworks during the Spring Festival) and avoiding model overfitting caused by static thresholds; by statistically analyzing the ratio distribution of different pollution events, we can filter out abnormal fluctuations caused by meteorological conditions and transmission processes, and improve the representativeness of characteristic thresholds.
[0051] In traditional analysis methods, factor analysis models are generally used to calculate the chemical components in particulate matter and identify source types (such as coal burning sources, motor vehicle sources, dust sources, industrial sources, fireworks and firecracker sources, etc.) and quantify the source type contribution (unit: μg / m 3 ), such as PCA-MLR, Unmix and PMF (Positive matrix factorization); the present invention takes the traditional factor analysis model PMF model as an example.
[0052] The basic principle of the traditional PMF model for particulate matter source apportionment adopts the following objective function:
[0053]
[0054] Among them, X ij is the mass concentration of the jth component or gas in the i-th sample, in μg / m 3 ;g ik is the contribution of the kth source class to the i-th sample, in μg / m 3 ;f kj It is the content of the jth component in the source spectrum of the kth source category, in %.
[0055] Through the above calculations, we can obtain the contribution of coal burning sources, motor vehicle sources, dust sources, industrial sources, fireworks and firecracker sources, secondary particles, etc. to particulate matter, and the unit is μg / m 3 .
[0056] However, the traditional PMF objective function relies solely on chemical component fitting and lacks explicit constraints on the dynamic characteristics of pollution source emissions, leading to the problem of collinearity source apportionment bias. To address this problem, the present invention embeds the dynamic characteristic threshold as a constraint condition into the objective function of the traditional PMF model, thereby obtaining a new function as follows:
[0057]
[0058] Among them, R k PM of the kth pollution source 2.5 / Dynamic feature threshold in CO feature ratio database, X ij is the mass concentration of the jth component or gas in the i-th sample, G ik is the contribution of the kth source class to the i-th sample, Fkj is the content of the jth component in the source spectrum of the kth source class, u ij is the uncertainty of the jth component in the i-th sample, and λ is the constraint weight coefficient.
[0059] Compared with the original objective function of the traditional PMF model, the new function can optimize the factor decomposition process, force the pollution source ratio calculated by the model to approach the characteristic threshold through constraint terms, integrate the emission dynamics characteristics into the mathematical optimization process, and avoid the black box defects of pure data-driven; the constraint weight coefficient allows the constraint strength to be dynamically adjusted according to the type of pollution event, balancing the priority of chemical component fitting and ratio constraints.
[0060] Since the PMF model has the above new function, it is equivalent to an optimized PMF model.
[0061] The following examples are provided for comparison:
[0062] Data collection background:
[0063] Monitoring data in a certain city in January showed that PM 2.5 Peak value reaches 250μg / m 3 , CO concentration rose to 4.2ppm.
[0064] 1-hour resolution PM data for this city in January 2024 2.5 Chemical components (such as SO4 2- , K + 、NO3 - After normalizing the OC) and CO concentration data, an n × m matrix was generated;
[0065] For the above matrix, the objective function of the traditional PMF model is first used for analysis, and then the optimized PMF model of the present invention is used for analysis. The comparison of the analytical results is shown in Table 1 below:
[0066] Table 1 Comparison of traditional PMF and optimized PMF analysis results
[0067]
[0068] As can be seen from Table 1 above, using the traditional PMF model, the contribution rate of fireworks and firecrackers is 25% (39.5 μg / m 3 ), coal-fired sources accounted for 28% (44.2 μg / m 3 ), and the contribution rates of coal-fired sources and fireworks and firecrackers sources overlap, both accounting for 25% to 30% and cannot be distinguished; however, using the optimized PMF model of the present invention for analysis, the contribution rate of fireworks and firecrackers sources is increased to 42%, and the coal-fired source is reduced to 18%, with the remainder being secondary sources and dust sources.
[0069] To determine the accuracy of the results of the analysis using the optimized PMF model of this invention, residual analysis (Q / Qexp) and source spectrum similarity (such as cosine similarity) can be used to evaluate the analysis results. Residual analysis evaluates the overall fit of the model, and source spectrum similarity testing ensures the consistency of the analysis results with the actual source spectrum. This dual verification enhances the credibility of the results. Residual distribution is used to identify abnormal samples and support dynamic iterative optimization of the database.
[0070] The results of the residual analysis (Q / Qexp) are shown in Table 2:
[0071] Table 2 Residual improvement results of the traditional PMF model and the optimized model of the present invention for coal source and fireworks source
[0072]
[0073] From the data in Table 2 above, it can be seen that the results of the analysis using the optimized PMF model of the present invention are more accurate.
[0074] In order to solve the problem that traditional analytical methods only rely on model output and lack cross-validation with external data, resulting in the results being out of touch with actual emission scenarios, this embodiment provides a verification method, namely, verifying the spatial transmission path of the pollution source through a backward trajectory model and matching it with fireworks and firecracker sales data to confirm the analytical results.
[0075] Using this method, the backward trajectory model verifies the transmission paths of pollution sources, preventing misallocation of source contributions due to cross-regional transmission. By matching the model with fireworks sales data, the model's analytical results are linked to actual emissions, providing direct evidence for law enforcement. Matching the backward trajectory model with fireworks sales data further validates the analytical accuracy of the optimized model.
[0076] Based on the embodiments of the present disclosure, the present disclosure also provides an electronic device, a computer-readable storage medium, and a computer program product.
[0077] An electronic device includes at least one processor; a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the PM-based 2.5 PMF source apportionment method based on the characteristic ratio of CO / CO.
[0078] The electronic device may be implemented in whole or in part by software, hardware, firmware, or any combination thereof, and the electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.
[0079] A computer-readable storage medium stores a computer program, which, when executed, implements the PM-based 2.5 PMF source apportionment method based on the characteristic ratio of CO / CO.
[0080] Various embodiments of the present disclosure may be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system comprising at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0081] A computer program product includes a computer program / instruction, wherein the computer program / instruction is executed by a processor based on PM provided by the present disclosure. 2.5 PMF source apportionment method based on the characteristic ratio of CO / CO.
[0082] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.
[0083] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0084] The embodiments of the present invention are described in detail above, but the contents described are only preferred embodiments of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.
Claims
1. Based on PM 2.5 The PMF source apportionment method based on the characteristic ratio of CO2 / CO is characterized by: The following steps are involved: Collect PM in the target time period and location 2.5 Chemical composition and CO concentration data are standardized to generate a data matrix; From PM 2.5 Extract the dynamic characteristic thresholds of coal burning sources and fireworks and firecracker sources from the CO2 / CO2 dynamic ratio database; The dynamic feature threshold is embedded as a constraint condition in the objective function of the PMF model to optimize the PMF model; The data matrix is input into the optimized PMF model to analyze pollution sources and quantify source class contributions.
2. The PM-based method according to claim 1 2.5 The PMF source apportionment method based on the characteristic ratio of CO2 / CO is characterized by: The construction of the dynamic ratio database comprises the following steps: Collect PM from coal burning sources and during concentrated fireworks and firecrackers emission periods 2.5 Data on chemical composition and CO concentration; Based on the data, PM 2.5 / CO characteristic ratio interval; Based on the data during the Spring Festival, a special PM of fireworks and firecrackers source is established. 2.5 / CO sign ratio interval; Calculate PM in different pollution events 2.5 The median and fluctuation range of the / CO ratio.
3. The PM-based method according to claim 2 2.5 The PMF source apportionment method based on the characteristic ratio of CO2 / CO is characterized by: The PM2.5 monitoring system for coal-fired sources was established by combining β-ray method with non-dispersive infrared technology. 2.5 / CO characteristic ratio range.
4. The PM-based method according to any one of claims 1 to 3. 2.5 The PMF source apportionment method based on the characteristic ratio of CO2 / CO is characterized by: The new function obtained by embedding the dynamic feature threshold as a constraint into the objective function of the PDM model is as follows: Among them, R k PM of the kth pollution source 2.5 / Dynamic feature threshold in CO feature ratio database, X ij is the mass concentration of the jth component or gas in the i-th sample, G ik is the contribution of the kth source class to the i-th sample, F kj is the content of the jth component in the source spectrum of the kth source class, u ij is the uncertainty of the jth component in the i-th sample, and λ is the constraint weight coefficient.
5. The PM-based method according to any one of claims 1 to 4. 2.5 The PMF source apportionment method based on the characteristic ratio of CO2 / CO is characterized by: The parsing results are evaluated using residual analysis and source spectrum similarity test.
6. The PM-based method according to any one of claims 1 to 5. 2.5 The PMF source apportionment method based on the characteristic ratio of CO2 / CO is characterized by: The spatial transmission path of the pollution source was verified through the backward trajectory model and matched with the fireworks sales data to confirm the analysis results.
7. An electronic device, characterized in that: include: at least one processor; A memory communicatively connected to at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1 to 6.
8. A computer-readable storage medium, characterized in that: A computer program is stored therein, and when the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.
9. A computer program product, characterized in that: The method comprises a computer program / instruction which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 6.