Pollution source analysis method, device and storage medium

Abstract

The present invention discloses a pollution source analysis method, device and storage medium, which belongs to the field of water environment protection and monitoring technology. The method includes collecting polluted water samples; using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to detect water samples to obtain multidimensional spectral data; using an entropy weight method to reduce the dimensionality of the multidimensional spectral data, and respectively calculating the weights of the spectrum matrix allocation; based on the weights of the spectrum matrix, a two-dimensional weight matrix is ​​constructed; using the Euclidean distance judgment analysis method, the similarity between the two-dimensional weight matrix and the weight matrix in the pre-constructed database is calculated to determine the pollution source. The present invention uses electrochemical perturbations to enhance the active ingredients in water quality, providing more information for water pollution source analysis to mine hidden water quality spectral information, and increase the recognition of water quality detection and tracing; at the same time, the entropy weight method-Euclidean distance is used to perform pollution source tracing analysis to improve the accuracy of pollution source identification and analysis.

Description

Technical Field

[0001] The present invention relates to the field of water environment protection and monitoring technology, and in particular to a pollution source analysis method, device and storage medium. Background Art

[0002] The existence of illegal sewage discharge and accidental pollution discharge has caused a serious impact on the quality of the water environment. For example, illegal discharge of excessive chemical oxygen demand (COD), ammonia nitrogen, total phosphorus and other pollutants will have a huge impact on urban sewage treatment plants and have a serious impact on river water quality. How to quickly find the source of sewage is the focus and difficulty of current water pollution warning and management research.

[0003] At present, the commonly used sewage source tracing methods are: on-site sampling source tracing method and numerical simulation method. The on-site sampling source tracing method includes isotope tracing method, water pattern recognition method and spectroscopy method. The source tracing results of this method are highly reliable, but it requires a lot of manpower and material resources to use; numerical simulation method includes deterministic method and uncertain method. Numerical simulation method is a method that uses a model or algorithm to weigh the similarity between the measured value and the simulated value. This method can achieve rapid source tracing, but the accuracy is often not guaranteed.

[0004] In the related art, an invention patent application with publication number CN 112198144 A discloses a method for rapid sewage source tracing, which is implemented in the following steps: S01) Establishment of a pollution source database, by performing water quality data detection and three-dimensional fluorescence spectrum analysis on historical water samples collected from outlet nodes of each enterprise in the monitoring basin, and calculating the three-dimensional fluorescence spectrum using the entropy weight method to obtain A=(anⅹ1) and combining the detected water quality data X1, X2, X3, X4, ...Xn with A into a new matrix Si=(anⅹ1,X1,X2, ...Xn), the new matrix Si is a snapshot of the water quality of the water sample, and then the enterprise , pipe network, river name and geographical location as the enterprise's water quality snapshot label entry into the database, forming the enterprise's water quality unique information; S02) sewage tracing, through the water quality online monitoring equipment and the three-dimensional fluorescence spectrometer, real-time monitoring of the water quality and three-dimensional fluorescence spectrum data of the water body, and then through the same processing method as step S01) to combine the water quality data and the three-dimensional fluorescence spectrum, to obtain the current water quality snapshot of the water sample, and then the current water quality snapshot of the water sample and all the water quality snapshot information in the database are calculated for similarity, and finally the water quality snapshot information with the highest similarity is the pollution source.

[0005] The three-dimensional fluorescence fingerprint spectrum contains a lot of information and can be used to identify and analyze the sources of water pollution. However, the three-dimensional fluorescence spectroscopy technology can only detect fluorescent substances in water. When the concentration of the solution is high, the internal filtration effect will reduce the intensity of the excitation light and affect the linear relationship between the fluorescence intensity and the concentration of the fluorescent substance. In addition, environmental factors, such as solvents, temperature, medium pH and viscosity, heavy atom effects, etc., will affect the linear relationship, resulting in uncertainty in the identification of fluorescent substances in water bodies, causing errors in further source identification and analysis. Summary of the invention

[0006] The technical problem to be solved by the present invention is how to improve the accuracy of pollution source analysis.

[0007] The present invention solves the above technical problems through the following technical means:

[0008] In one aspect, the present invention provides a method for pollution source analysis, the method comprising:

[0009] Collecting polluted water samples;

[0010] The water sample is detected by using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to obtain multi-dimensional spectrum data;

[0011] The multi-dimensional spectral data is subjected to dimensionality reduction processing by using an entropy weight method, and the spectral matrix allocation weights are calculated respectively;

[0012] Assign weights based on the atlas matrix to construct a two-dimensional weight matrix;

[0013] The Euclidean distance judgment analysis method is used to calculate the similarity between the two-dimensional weight matrix and the weight matrix in the pre-constructed database to determine the pollution source.

[0014] The present invention utilizes electrochemical perturbation to enhance the active components in water quality, so that the characteristic peaks of the excitation spectrum and their intensities exhibited by the water sample are enhanced, providing more information for water pollution source analysis to mine hidden water quality spectral information and increase the recognition of water quality detection and traceability; at the same time, the entropy weight method is used to reduce the dimension of multidimensional spectral data, and the weights of the spectrum matrix are calculated respectively, and a two-dimensional weight matrix is ​​constructed. Then, based on the two-dimensional weight matrix, the Euclidean distance judgment analysis method is used to perform pollution source tracing analysis to further improve the accuracy of pollution source identification and analysis.

[0015] Furthermore, the water sample is detected by using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to obtain multi-dimensional spectrum data, including:

[0016] The three-dimensional fluorescence spectrum detection method is used to obtain first spectral response characteristic data of the water sample under normal conditions;

[0017] Performing electrochemical disturbance on the water sample by using an electrochemical method, and acquiring second spectral response characteristic data of the water sample in a disturbed state by using the three-dimensional fluorescence spectrum detection method;

[0018] The multi-dimensional spectral data is constructed based on the first spectral response characteristic data and the second spectral response characteristic data.

[0019] Furthermore, the entropy weight method is used to perform dimensionality reduction processing on the multidimensional spectral data, and the spectrum matrix allocation weights are calculated respectively, including:

[0020] Construct a matrix I of n samples and m evaluation indicators, and normalize the matrix I to obtain a matrix M;

[0021] According to the matrix M, the entropy of the n samples and the m evaluation indicators are calculated;

[0022] Based on the entropy of the n samples and the m evaluation indicators, the entropy weight is calculated as the weight for allocating the graph matrix.

[0023] Furthermore, the weights are assigned based on the atlas matrix, and the entropy weights calculated for the background data and the disturbance data are combined into a new two-dimensional weight matrix.

[0024] Furthermore, the Euclidean distance judgment analysis method is used to calculate the similarity between the two-dimensional weight matrix and the weight matrix in the pre-built database to determine the pollution source, including:

[0025] The Euclidean distance judgment analysis method is used to calculate the Euclidean distance between the two-dimensional weight matrix and each weight matrix in the database;

[0026] Based on each of the Euclidean distances, calculating the similarity between the two-dimensional weight matrix and each weight matrix in the database;

[0027] The pollution source corresponding to the weight data in the database having the highest similarity to the two-dimensional weight matrix is ​​determined as the pollution source of the water sample.

[0028] In addition, the present invention also proposes a pollution source analysis device, which includes:

[0029] A collection module, used for collecting polluted water samples;

[0030] A detection module, used to detect the water sample using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to obtain multi-dimensional spectrum data;

[0031] A dimension reduction module is used to perform dimension reduction processing on the multidimensional spectral data using an entropy weight method, and calculate the spectral matrix allocation weights respectively;

[0032] A weight matrix construction module, used to assign weights based on the atlas matrix and construct a two-dimensional weight matrix;

[0033] The determination module is used to use the Euclidean distance judgment analysis method to calculate the similarity between the two-dimensional weight matrix and the weight matrix in the pre-constructed database to determine the pollution source.

[0034] Furthermore, the detection module includes:

[0035] A first detection unit, used for acquiring first spectral response characteristic data of the water sample under normal conditions by using the three-dimensional fluorescence spectrum detection method;

[0036] A second detection unit is used to electrochemically disturb the water sample by an electrochemical method, and to obtain second spectral response characteristic data of the water sample in a disturbed state by using the three-dimensional fluorescence spectrum detection method;

[0037] The data composition unit is used to form the multi-dimensional spectral data based on the first spectral response characteristic data and the second spectral response characteristic data.

[0038] Furthermore, the dimension reduction module comprises:

[0039] A normalization unit, used to construct a matrix I of n samples and m evaluation indicators, and normalize the matrix I to obtain a matrix M;

[0040] An entropy calculation unit, used to calculate the entropy of the n samples and the m evaluation indicators according to the matrix M;

[0041] A weight allocation unit is used to calculate the entropy weight as the weight of the graph matrix allocation based on the entropy of the n samples and the m evaluation indicators.

[0042] Furthermore, the determination module includes:

[0043] A distance calculation unit, used to calculate the Euclidean distance between the two-dimensional weight matrix and each weight matrix in the database by using the Euclidean distance judgment analysis method;

[0044] A similarity calculation unit, which calculates the similarity between the two-dimensional weight matrix and each weight matrix in the database based on each Euclidean distance;

[0045] The pollution source determination unit is used to determine the pollution source corresponding to the weight data in the database with the highest similarity to the two-dimensional weight matrix as the pollution source of the water sample.

[0046] In addition, the present invention also proposes a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0047] The advantages of the present invention are:

[0048] (1) The present invention uses an electrochemical method to electrochemically excite substances in water that are electrochemically active but not fluorescent, thereby changing the structure or charge state of the compound, realizing fluorescence detection of non-fluorescent substances, collecting spectral response characteristics under excited and non-excited conditions, fusing three-dimensional fluorescence spectrum data and electrochemical spectrum data to form a multidimensional spectrum array, and using electrochemical perturbations to enhance the active components in the water, so that the characteristic peaks of the excitation spectrum exhibited by the water sample and their intensities are enhanced, providing more information for water pollution source analysis to mine hidden water quality spectrum information and increase the recognition of water quality detection and traceability; at the same time, the entropy weight method is used to reduce the dimensionality of the multidimensional spectrum data, and the weights of the spectrum matrix are calculated respectively, and a two-dimensional weight matrix is ​​constructed. Then, based on the two-dimensional weight matrix, the Euclidean distance judgment analysis method is used to perform pollution source tracing analysis to further improve the accuracy of pollution source identification and analysis.

[0049] (2) The entropy weight method-Euclidean distance method is used to realize multi-dimensional information analysis of the source of water samples and sewage. It has the characteristics of fast speed and high accuracy. It can be used to guide the investigation of illegal discharge of urban enterprises, mixed connection of rainwater and sewage pipes, etc., greatly improving the operational efficiency of urban drainage systems.

[0050] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flow chart of the pollution source analysis method of the present invention;

[0052] Figure 2 It is the overall flow chart of the pollution source analysis method in the present invention;

[0053] Figure 3 It is the Euclidean distance diagram of the P1 outlet in the present invention;

[0054] Figure 4 It is the Euclidean distance diagram of the P2 outlet in the present invention;

[0055] Figure 5 It is a structural diagram of the pollution source analysis device in the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Reference Figure 1 The embodiment of the present invention provides a pollution source analysis method, which includes the following steps:

[0058] S10, collect polluted water samples;

[0059] S20, using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to detect the water sample to obtain multidimensional spectrum data;

[0060] It should be noted that the multidimensional spectral data obtained in this embodiment is four-dimensional spectral data.

[0061] S30, performing dimensionality reduction processing on the multidimensional spectral data using an entropy weight method, and calculating the spectral matrix allocation weights respectively;

[0062] S40, assigning weights based on the atlas matrix to construct a two-dimensional weight matrix;

[0063] S50, using the Euclidean distance judgment analysis method to calculate the similarity between the two-dimensional weight matrix and the weight matrix in the pre-constructed database to determine the pollution source.

[0064] In this embodiment, electrochemical perturbation is used to enhance the active components in the water quality, so that the characteristic peaks of the excitation spectrum and their intensities exhibited by the water sample are enhanced, providing more information for water pollution source analysis to mine hidden water quality spectral information and increase the recognition of water quality detection and traceability; at the same time, the entropy weight method is used to reduce the dimension of the multidimensional spectral data, calculate the weights of the spectrum matrix respectively, and construct a two-dimensional weight matrix, and then based on the two-dimensional weight matrix, the Euclidean distance judgment analysis method is used to perform pollution source tracing analysis to further improve the accuracy of pollution source identification and analysis.

[0065] In one embodiment, referring to Figure 2 , said S20 comprises the following steps:

[0066] S21, using the three-dimensional fluorescence spectrum detection method to obtain first spectral response characteristic data of the water sample under normal conditions;

[0067] S22, using an electrochemical method to electrochemically disturb the water sample, and using the three-dimensional fluorescence spectrum detection method to obtain second spectral response characteristic data of the water sample in a disturbed state;

[0068] S23: construct the multi-dimensional spectral data based on the first spectral response characteristic data and the second spectral response characteristic data.

[0069] It can be understood that, in this embodiment, a three-dimensional fluorescence spectrometer and an ultraviolet-visible absorption spectrometer can be used to test water samples in a normal state and in a disturbed state to obtain spectral response characteristic data.

[0070] This embodiment uses an electrochemical method to electrochemically excite substances in water that are electrochemically active but non-fluorescent, thereby changing the structure or charge state of the compound, realizing fluorescence detection of non-fluorescent substances, collecting spectral response characteristics under excited and non-excited conditions, and fusing three-dimensional fluorescence spectral data and electrochemical spectral data to form a multi-dimensional spectral array, so that the characteristic peaks of the excitation spectrum exhibited by the water sample and their intensities are enhanced, providing more information for water pollution source analysis, mining hidden water quality spectral information, and increasing the recognition of water quality detection and traceability.

[0071] In one embodiment, the step S30 includes:

[0072] S31, constructing a matrix I of n samples and m evaluation indicators, and normalizing the matrix I to obtain a matrix M;

[0073] Specifically, the matrix I of n samples and m evaluation indicators is constructed as:

[0074] I=(a ij m×n)

[0075] Normalize the matrix I to get the matrix M:

[0076] M=(b ij m×n)

[0077] in, a ij is the element of matrix I; a max 、a min Indicates the most satisfactory or least satisfactory element in different samples under the same indicator.

[0078] S32, calculating the entropy of the n samples and m evaluation indicators according to the matrix M;

[0079] Specifically, the entropy of n samples and m evaluation indicators

[0080]

[0081] Among them, R i represents the information entropy of index i, b ij is an element in the matrix M, in order to make lnb ij Makes sense, assuming b ij = 0, b ij lnb ij = 0, but when b ij =1, b ik lnb ij =0, so the formula is used For b ij Make corrections.

[0082] S33. Based on the entropy of the n samples and the m evaluation indicators, the entropy weight is calculated as the weight for allocating the graph matrix.

[0083] Specifically, the calculation formula of entropy weight is:

[0084]

[0085] Among them, A i Represents the entropy weight (weight) of index i; R i Represents the information entropy of index i and satisfies ∑A i =1.

[0086] It should be noted that the invention patent application with publication number CN 112198144 A records a method for rapid sewage tracing, which adopts the entropy weight method to reduce the dimension of three-dimensional fluorescence spectrum data to form one-dimensional data, and combines water quality indicators (such as COD, ammonia nitrogen, etc.) to form new one-dimensional data. The reconstructed matrix is ​​one-dimensional, and then the Pearson similarity is used to calculate the similarity of the one-dimensional data.

[0087] In this embodiment, the entropy weight method is used to reduce the dimensions of the background data and the disturbance data of the water sample to form one-dimensional data, and then reorganize them into two-dimensional matrix data; then the Euclidean distance is used to calculate the similarity of the two-dimensional matrix data. The processing method of the array after dimensionality reduction in this embodiment is different from that in the embodiment.

[0088] In one embodiment, the step S40 includes: combining entropy weights calculated from the background data and the disturbance data into a new two-dimensional weight matrix.

[0089] In one embodiment, the step S50 includes the following steps:

[0090] S51, using the Euclidean distance judgment analysis method to calculate the Euclidean distance between the two-dimensional weight matrix and each weight matrix in the database;

[0091] S52, calculating the similarity between the two-dimensional weight matrix and each weight matrix in the database based on each Euclidean distance;

[0092] S53: Determine the pollution source corresponding to the weight data in the database that has the highest similarity to the two-dimensional weight matrix as the pollution source of the water sample.

[0093] Specifically, based on the Euclidean distance discriminant analysis method, the pollution source analysis is carried out, and the main calculation formula is as follows:

[0094] Two n (n = 2) dimensional vectors a(x 11 ,x 12 ,…,x 1n ) and b(x 21 ,x 22 ,…,x 2n ) is:

[0095]

[0096] The Euclidean distance between two points A(x1,y1) and B(x2,y2) on a two-dimensional plane is calculated as follows:

[0097]

[0098] From the above general formula, we can see that the distance between two points is always non-negative, and the maximum value is positive infinity. However, the similarity value should usually be between [-1, 1], so we need to find the inverse of it to control the result between (0, 1].

[0099] Finally, the similarity between two data can be calculated by Euclidean distance, and the formula is:

[0100]

[0101] The greater the distance between two data or users, the smaller their similarity, and the smaller the distance, the greater the similarity. The calculated similarities are sorted, and the pollution source corresponding to the weighted data with the highest similarity in the database is determined as the pollution source of the water sample.

[0102] This embodiment uses the entropy weight method-Euclidean distance method to achieve multi-dimensional information analysis of the source of water samples and sewage, which has the characteristics of fast speed and high accuracy. It can be used to guide the investigation of illegal emissions by urban enterprises, mixed and wrong connections in rainwater and sewage pipes, and other behaviors, greatly improving the operational efficiency of urban drainage systems.

[0103] It should be noted that Euclidean distance is a common algorithm for calculating data similarity. This scheme adopts this method to calculate the similarity of the reconstructed two-dimensional matrix, which is different from the related technologies such as the invention patent application with publication number CN113267607A, which uses Euclidean distance to calculate the similarity between the training set and the actual monitoring data, sort the inversion results, and filter out useless information. The purpose and method of use are different.

[0104] By taking a secondary tributary basin in a certain basin as the research object, we collected sewage samples from the outlets of key polluting enterprises, and at the same time collected water samples from their outlets into the river and key nodes of the sewage pipe network. The following table shows the relevant water samples collected. This experiment selected two pipe network nodes P1 and P2 in the basin where the water quality exceeded the standard, and collected water samples from the outlets of 11 related enterprises around the pipe network nodes, which were recorded as Q1-Q11. Fingerprint detection was performed using three-dimensional fluorescence detection and electrochemical excitation three-dimensional fluorescence detection to construct four-dimensional data. The weights of the spectrum matrix were calculated based on the entropy weight method, and the weights of the initial state and the excited state were combined to construct a two-dimensional weight matrix. Then, the pollution source analysis was performed based on the Euclidean distance discriminant analysis method, and similarity analysis was performed.

[0105] See Table 1 for details of water sample information collection:

[0106] Table 1 Water sample information

[0107]

[0108] The similarity calculation results are shown in Table 2:

[0109] Table 2 Similarity calculation results

[0110]

[0111] The Euclidean distance and similarity calculation data between the outlet water samples Q1-Q11 and outlets P1 and P2 are shown in Table 2. Figures 3 to 4 The Euclidean distance diagrams between the network nodes P1 and P2 and the enterprises are shown. For the network node P1, the overall distance between Q8 and its electrochemical fluorescence spectrum is small, with a distance value of 0.380, which is converted into a similarity of 0.72. The similarity between the outlets of other enterprises and P1 is generally around 0.6. Therefore, it is preliminarily judged that the excessive water quality at this node is mainly affected by enterprise 8.

[0112] Similarly, for the pipeline node P2, the closest Euclidean distance enterprise is Q5, whose Euclidean distance value is 0.345 and similarity value is 0.74. It is preliminarily judged that the water quality exceeding the standard at this node is mainly affected by enterprise 5.

[0113] The present invention uses an in-situ electrochemical fluorescence spectrum test system to test actual water samples from enterprises and pipe network nodes, and obtains three-dimensional fluorescence spectrum data in the initial state and the excited state. Then, the spectrum matrix allocation weights are calculated based on the entropy weight method, and the allocation weights of the initial state and the excited state are combined to construct a two-dimensional weight matrix. Subsequently, the pollution source analysis is performed based on the Euclidean distance discriminant analysis method, and the suspected pollution source is determined.

[0114] This method breaks through the quenching effect of traditional three-dimensional fluorescence spectroscopy, and uses electrochemical perturbations to enhance the active components in water quality, so that the characteristic peaks and intensities of the excitation spectrum of the water sample are enhanced, providing more information for the analysis of water pollution sources. Through the entropy weight method and Euclidean distance, the distance between the characteristic vectors is used to obtain the similarity of water samples. The calculated results match the actual results well.

[0115] Reference Figure 5 The embodiment of the present invention provides a pollution source analysis device, the device comprising:

[0116] A collection module 10, used for collecting polluted water samples;

[0117] A detection module 20, for detecting the water sample using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to obtain multi-dimensional spectrum data;

[0118] A dimension reduction module 30 is used to perform dimension reduction processing on the multi-dimensional spectral data using an entropy weight method, and calculate the weights of the spectrum matrix allocation respectively;

[0119] A weight matrix construction module 40, used to assign weights based on the atlas matrix and construct a two-dimensional weight matrix;

[0120] The determination module 50 is used to use the Euclidean distance judgment analysis method to calculate the similarity between the two-dimensional weight matrix and the weight matrix in the pre-constructed database to determine the pollution source.

[0121] In this embodiment, electrochemical perturbation is used to enhance the active components in the water quality, so that the characteristic peaks of the excitation spectrum and their intensities exhibited by the water sample are enhanced, providing more information for water pollution source analysis to mine hidden water quality spectral information and increase the recognition of water quality detection and traceability; at the same time, the entropy weight method is used to reduce the dimension of the multidimensional spectral data, calculate the weights of the spectrum matrix respectively, and construct a two-dimensional weight matrix, and then based on the two-dimensional weight matrix, the Euclidean distance judgment analysis method is used to perform pollution source tracing analysis to further improve the accuracy of pollution source identification and analysis.

[0122] In one embodiment, the detection module 20 includes:

[0123] A first detection unit, used for acquiring first spectral response characteristic data of the water sample under normal conditions by using the three-dimensional fluorescence spectrum detection method;

[0124] A second detection unit is used to electrochemically disturb the water sample by an electrochemical method, and to obtain second spectral response characteristic data of the water sample in a disturbed state by using the three-dimensional fluorescence spectrum detection method;

[0125] The data composition unit is used to form the multi-dimensional spectral data based on the first spectral response characteristic data and the second spectral response characteristic data.

[0126] In one embodiment, the dimension reduction module 30 includes:

[0127] A normalization unit, used to construct a matrix I of n samples and m evaluation indicators, and normalize the matrix I to obtain a matrix M;

[0128] An entropy calculation unit, used to calculate the entropy of the n samples and the m evaluation indicators according to the matrix M;

[0129] A weight allocation unit is used to calculate the entropy weight as the weight of the graph matrix allocation based on the entropy of the n samples and the m evaluation indicators.

[0130] In one embodiment, the determining module 50 includes:

[0131] A distance calculation unit, used to calculate the Euclidean distance between the two-dimensional weight matrix and each weight matrix in the database by using the Euclidean distance judgment analysis method;

[0132] A similarity calculation unit, which calculates the similarity between the two-dimensional weight matrix and each weight matrix in the database based on each Euclidean distance;

[0133] The pollution source determination unit is used to determine the pollution source corresponding to the weight data in the database with the highest similarity to the two-dimensional weight matrix as the pollution source of the water sample.

[0134] This embodiment uses the entropy weight method-Euclidean distance method to achieve multi-dimensional information analysis of the source of water samples and sewage, which has the characteristics of fast speed and high accuracy. It can be used to guide the investigation of illegal emissions by urban enterprises, mixed and wrong connections in rainwater and sewage pipes, and other behaviors, greatly improving the operational efficiency of urban drainage systems.

[0135] It should be noted that other embodiments of the pollution source analysis device of the present invention or implementation methods can refer to the above-mentioned method embodiments, which will not be repeated here.

[0136] In addition, an embodiment of the present invention discloses a computer-readable medium having computer-readable instructions stored thereon, and the computer-readable instructions can be executed by a processor to implement the pollution source analysis method as described above.

[0137] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0138] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0139] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0140] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0141] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A pollution source analysis method, characterized in that: The method comprises: Collecting polluted water samples; The water sample is detected by using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to obtain multi-dimensional spectrum data; The multidimensional spectral data is subjected to dimensionality reduction processing by using an entropy weight method, and the weights of the spectrum matrix allocation are calculated respectively, including: constructing a matrix I of n samples and m evaluation indicators, and normalizing the matrix I to obtain a matrix M; according to the matrix M, the entropy of the n samples and m evaluation indicators is calculated; based on the entropy of the n samples and m evaluation indicators, the entropy weight is calculated as the weight of the spectrum matrix allocation; Allocating weights based on the atlas matrix to construct a two-dimensional weight matrix includes: combining entropy weights calculated from background data and disturbance data into a new two-dimensional weight matrix; The method adopts the Euclidean distance judgment and analysis method to calculate the similarity between the two-dimensional weight matrix and the weight matrix in a pre-constructed database to determine the pollution source, including: using the Euclidean distance judgment and analysis method to calculate the Euclidean distance between the two-dimensional weight matrix and each weight matrix in the database; based on each Euclidean distance, calculating the similarity between the two-dimensional weight matrix and each weight matrix in the database; determining the pollution source corresponding to the weight data in the database with the highest similarity to the two-dimensional weight matrix as the pollution source of the water sample.

2. The pollution source analysis method according to claim 1, characterized in that: The three-dimensional fluorescence spectrum detection method and the electrochemical spectrum detection method are used to detect the water sample to obtain multi-dimensional spectrum data, including: The three-dimensional fluorescence spectrum detection method is used to obtain first spectral response characteristic data of the water sample under normal conditions; Performing electrochemical disturbance on the water sample by using an electrochemical method, and acquiring second spectral response characteristic data of the water sample in a disturbed state by using the three-dimensional fluorescence spectrum detection method; The multi-dimensional spectral data is constructed based on the first spectral response characteristic data and the second spectral response characteristic data.

3. A pollution source analysis device, characterized in that: The device comprises: A collection module, used for collecting polluted water samples; A detection module, used to detect the water sample using a three-dimensional fluorescence spectrum detection method and an electrochemical spectrum detection method to obtain multi-dimensional spectrum data; A dimension reduction module is used to perform dimension reduction processing on the multidimensional spectral data using an entropy weight method, and calculate the spectral matrix allocation weights respectively; A weight matrix construction module, used to assign weights based on the atlas matrix and construct a two-dimensional weight matrix; A determination module, used to calculate the similarity between the two-dimensional weight matrix and the weight matrix in the pre-built database by using the Euclidean distance judgment analysis method, so as to determine the pollution source; The dimension reduction module comprises: A normalization unit, used to construct a matrix I of n samples and m evaluation indicators, and normalize the matrix I to obtain a matrix M; An entropy calculation unit, used to calculate the entropy of the n samples and the m evaluation indicators according to the matrix M; A weight allocation unit, used to calculate entropy weights as weights for allocating the graph matrix based on the entropies of the n samples and the m evaluation indicators; The determining module comprises: A distance calculation unit, used to calculate the Euclidean distance between the two-dimensional weight matrix and each weight matrix in the database by using the Euclidean distance judgment analysis method; A similarity calculation unit, which calculates the similarity between the two-dimensional weight matrix and each weight matrix in the database based on each Euclidean distance; The pollution source determination unit is used to determine the pollution source corresponding to the weight data in the database with the highest similarity to the two-dimensional weight matrix as the pollution source of the water sample.

4. The pollution source analysis device according to claim 3, characterized in that: The detection module comprises: A first detection unit, used for acquiring first spectral response characteristic data of the water sample under normal conditions by using the three-dimensional fluorescence spectrum detection method; A second detection unit is used to electrochemically disturb the water sample by an electrochemical method, and to obtain second spectral response characteristic data of the water sample in a disturbed state by using the three-dimensional fluorescence spectrum detection method; The data composition unit is used to form the multi-dimensional spectral data based on the first spectral response characteristic data and the second spectral response characteristic data.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 2 is implemented.

Images