Dioxin monitoring method and device based on improved gray model, medium and equipment

Abstract

The invention provides a dioxin monitoring method and device based on an improved gray model, a medium and equipment, and relates to the technical field of environmental protection. Through combination of an improved grey prediction model and a neural network model, useful information in XRF and XRD data and the grey prediction model are fully utilized to fit a complex nonlinear relationship between a precursor and dioxin concentration, the concentration of a related compound is indirectly measured to predict the dioxin concentration, the generalization ability of the neural network model is enhanced, and the accuracy of the dioxin concentration prediction is improved. The high cost and complexity of direct measurement of dioxin in a traditional method are avoided, the measurement cost is greatly reduced, and rapid and large-scale monitoring of dioxin in fly ash is realized.

Description

Technical Field

[0001] The present disclosure relates to the field of environmental protection technology, and in particular to a dioxin monitoring method, device, medium and equipment based on an improved grey model. Background Art

[0002] Fly ash is a type of solid waste, mainly derived from waste incineration, industrial production and coal combustion, and is one of the important sources of dioxins. Dioxins are a class of polycyclic aromatic hydrocarbons containing chlorine. Due to their stable chemical properties, difficulty in decomposition, high bioaccumulation and toxicity, long-term exposure may cause skin lesions, immune system damage, nervous system damage and even cancer. They have been widely regarded as important pollutants for the environment and public health.

[0003] In recent years, with the development of environmental monitoring technology, many improved methods have been proposed to improve measurement efficiency and accuracy. For example, isotope dilution-high resolution gas chromatography-high resolution mass spectrometry provides higher resolution and selectivity; the application of sample pretreatment techniques such as solid phase extraction (SPE) and solid phase microextraction (SPME) helps to improve the sensitivity and selectivity of the analysis.

[0004] However, when rapid, large-scale and continuous monitoring of dioxins in fly ash is required, the content of dioxins in fly ash samples is often very low and the sample matrix is ​​complex. Even with pre-treatment technologies such as SPE and SPME, it is difficult to completely eliminate the matrix effect, which affects the accuracy and sensitivity of the detection. In traditional machine learning, too little data and sample size will also lead to poor model uniformity. In addition, although traditional dioxin analysis methods, such as gas chromatography-mass spectrometry, have high accuracy and sensitivity, the process is cumbersome, costly, and time-consuming, and is not suitable for large-scale real-time monitoring. Summary of the invention

[0005] The present disclosure provides a dioxin monitoring method, device, medium and equipment based on an improved grey model, which can realize large-scale rapid real-time monitoring of dioxins in fly ash and reduce monitoring costs.

[0006] According to a first aspect of the present disclosure, a dioxin monitoring method based on an improved grey model is provided, comprising the following steps: Obtaining X-ray fluorescence spectral data and X-ray diffraction pattern data of multiple fly ash samples; Using an improved grey prediction model to process the X-ray fluorescence spectrum data, identifying the single element with the largest contribution to dioxin concentration in the fly ash sample; Determining the concentration of the related compound corresponding to the single element according to the X-ray diffraction pattern data; The concentration of dioxins is predicted according to the concentration of the related compounds in the new fly ash sample using a neural network model; wherein the neural network model is trained based on training data consisting of the concentrations of the related compounds of the plurality of fly ash samples and the corresponding dioxin concentrations.

[0007] In one possible implementation manner, the improved grey prediction model is constructed in advance through the following steps: According to the concentration of dioxins in the fly ash sample and the element ratio data in the X-ray fluorescence spectrum data, an original data sequence X is generated. (0) ; X (0) Perform an accumulation to generate a new sequence X (1) , ; According to the new sequence X (1) , generate the adjacent mean value sequence z(t) as the background value of the improved grey prediction model, ; According to the new sequence X (1) , establish the first-order linear differential equation, ; This model is recorded as the improved grey prediction model GM(1,1); Among them, a is the development coefficient, b is the grey action; the least squares method is used to estimate the parameters a and b, and the formula is as follows: ; ; Solving the first-order linear differential equation yields X (1) The predicted expression is converted back to X (0) The predicted value of Compare X (0) The original actual value and predicted value are used to evaluate the prediction accuracy of the improved grey prediction model. If the qualified index is not met, the first-order linear differential equation in the improved grey prediction model is changed to an n-order linear differential equation, which is recorded as the improved grey prediction model G(n,1), where n is an integer and the value of n increases from 2 until the prediction accuracy of the improved grey prediction model meets the qualified index.

[0008] In one embodiment, the input characteristic further includes a chloride concentration.

[0009] In one possible implementation, the neural network model includes an input layer, a plurality of hidden layers and an output layer; The number of nodes in the input layer is equal to the number of input features, the number of nodes in the multiple hidden layers decreases layer by layer, and the output layer is a single node for outputting the concentration of dioxin; Dividing sample data corresponding to the plurality of fly ash samples into a training set, a validation set, and a test set, wherein the sample data includes the concentration of the relevant compound and the corresponding dioxin concentration; training the neural network model using an Adam optimizer based on the training set, the validation set, and the test set until the loss function converges; The loss function is ;in, is the actual value, is the predicted value.

[0010] According to a second aspect of the present disclosure, a dioxin monitoring device based on an improved grey model is provided, comprising: An acquisition unit, used for acquiring X-ray fluorescence spectrum data and X-ray diffraction spectrum data of a plurality of fly ash samples; A processing unit, used to process the X-ray fluorescence spectrum data using an improved grey prediction model to identify the single element in the fly ash that has the largest contribution to the dioxin concentration; a determination unit, configured to determine the concentration of a related compound corresponding to the single element according to the X-ray diffraction pattern data; The prediction unit is used to predict the concentration of dioxins according to the concentration of the related compounds in the new fly ash sample using a neural network model; wherein the neural network model is trained based on training data consisting of the concentrations of the related compounds of the plurality of fly ash samples and the corresponding dioxin concentrations.

[0011] In one possible implementation manner, the improved grey prediction model is constructed by the following units: The first generating unit is used to generate an original data sequence X according to the concentration of dioxins in the fly ash sample and the element ratio data in the X-ray fluorescence spectrum data. (0) ; The second generation unit is used to generate X (0) Perform an accumulation to generate a new sequence X (1) , ; The third generating unit is used to generate a new sequence X according to the (1) , generate the adjacent mean value sequence z(t) as the background value of the improved grey prediction model, ; Establish a unit for (1) , establish the first-order linear differential equation, ; This model is recorded as the improved grey prediction model GM (1, 1); Among them, a is the development coefficient, b is the grey action; the least squares method is used to estimate the parameters a and b, and the formula is as follows: ; ; A solving and reduction unit is used to solve the first-order linear differential equation to obtain X (1) The predicted expression is converted back to X (0) The predicted value of Evaluation unit for comparison with X (0) The original actual value and predicted value are used to evaluate the prediction accuracy of the improved grey prediction model. If the qualified index is not met, the first-order linear differential equation in the improved grey prediction model is changed to an n-order linear differential equation, which is recorded as the improved grey prediction model G(n, 1), where n is an integer and the value of n increases from 2 until the prediction accuracy of the improved grey prediction model meets the qualified index.

[0012] In one embodiment, the input characteristic further includes a chloride concentration.

[0013] In one possible implementation, the neural network model includes an input layer, an output layer, and a plurality of hidden layers; The number of nodes in the input layer is equal to the number of input features, the number of nodes in the multiple hidden layers gradually decreases, and the output layer is a single node for outputting the concentration of dioxin; Dividing sample data corresponding to the plurality of fly ash samples into a training set, a validation set, and a test set, wherein the sample data includes the concentration of the relevant compound and the corresponding dioxin concentration; training the neural network model using an Adam optimizer based on the training set, the validation set, and the test set until the loss function converges; The loss function is ;in, is the actual value, is the predicted value.

[0014] According to a third aspect of the present disclosure, there is provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed 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 execute the dioxin monitoring method based on the improved grey model described in the present disclosure.

[0015] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the dioxin monitoring method based on the improved grey model described in the present disclosure.

[0016] The dioxin monitoring method, device, medium and equipment based on the improved grey model disclosed in the present invention fully utilize the useful information in XRF and XRD data by combining the improved grey prediction model and the neural network model. The grey prediction model fits the complex nonlinear relationship between the precursor and the dioxin concentration, and indirectly measures the concentration of related compounds to predict the dioxin concentration, thereby enhancing the generalization ability of the neural network model, avoiding the high cost and complexity of directly measuring dioxins in traditional methods, greatly reducing the measurement cost, and realizing rapid and large-scale monitoring of dioxins in fly ash.

[0017] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, in which: In the drawings, the same or corresponding reference numerals represent the same or corresponding parts.

[0019] Figure 1 A flow chart of a dioxin monitoring method based on an improved grey model provided by an embodiment of the present disclosure is shown; Figure 2 A schematic diagram showing an error result of a neural network model provided by an embodiment of the present disclosure; Figure 3 A schematic diagram showing a correlation coefficient of a trained neural network model provided by an embodiment of the present disclosure; Figure 4 A structural diagram of a dioxin monitoring device based on an improved grey model provided by an embodiment of the present disclosure is shown; Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0020] In order to make the purpose, features, and advantages of the present disclosure more obvious and easy to understand, 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 in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0021] In order to solve the above problems existing in the prior art, the embodiments of the present disclosure provide a dioxin monitoring method, device, medium and equipment based on an improved grey model, which are described in detail below.

[0022] Embodiment 1: like Figure 1 As shown, Figure 1 A flowchart of a dioxin monitoring method based on an improved grey model is provided in an embodiment of the present disclosure. The method comprises the following steps: First, multiple fly ash samples are collected from the end of the solid waste incinerator and the bag. XRF (X-ray Fluorescence) is performed using high-precision instruments to obtain the composition data of the elemental composition in the fly ash; at the same time, XRD (X-Ray Diffraction) is performed to determine the chemical composition and crystal structure of the fly ash. In addition, auxiliary parameters such as temperature, humidity, and pH value of the fly ash can also be recorded. Although these parameters are not directly used for modeling in this method, they can be used for reference or optimization in subsequent studies.

[0023] Import the collected XRF and XRD data into the data processing software, such as using MATLAB's load function to import the data into the workspace, clean and organize it, remove outliers, fill in missing values, and ensure the integrity and accuracy of the data. At the same time, normalize the XRF data so that the proportions of different elements can be compared on the same scale. For XRD data, extract the content information of key compounds to provide basic data for subsequent modeling.

[0024] The concentration of dioxins in the fly ash sample and the element ratio data in the XRF spectrum are used as the original data sequence X (0) Each sample contains a set of dioxin concentration values ​​and corresponding element ratio data.

[0025] For the original data sequence X (0) Perform an accumulation generation to obtain a new sequence X (1) The purpose of cumulative generation is to reduce the randomness of the data and make the data present a more obvious regularity. The cumulative generation formula is: .

[0026] According to the new sequence X (1) , generate the adjacent mean sequence z(t) as the background value of the improved grey prediction model. The formula is: .

[0027] Based on the new sequence X (1) And the adjacent mean sequence z(t), establish a first-order linear differential equation: , as the improved grey prediction model G(1,1).

[0028] Here, a is the development coefficient and b is the gray action. The least squares method is used to estimate the parameters a and b using the following formula: ; .

[0029] And solve the differential equation to get X (1) The prediction expression is: .

[0030] Then, by reducing the formula, X (1) The predicted values ​​are converted back to the original data X (0) Predicted value of: .

[0031] Compare X (0) The original actual value and predicted value are calculated, the residual, the residual sum of squares and the correlation coefficient R are calculated, the prediction accuracy of the improved grey prediction model is evaluated, and the plot function is used to display the prediction results.

[0032] The calculation formula of the residual is: ; The residual sum of squares is an indicator for evaluating the goodness of fit of the improved grey prediction model. The calculation formula is: ; The smaller the SSE, the closer the predicted value of the improved grey prediction model is to the actual observed value, and the better the fit; The formula for calculating the correlation coefficient is: ; The correlation coefficient R is used to evaluate the correlation between the actual observed values ​​and the mean of the predicted values.

[0033] Set a qualified threshold set. If SSE is greater than the corresponding qualified threshold and R is less than the corresponding qualified threshold, it means that the prediction accuracy of the model does not meet the standard. Then change the first-order linear differential equation in the improved grey prediction model to a second-order linear differential equation to obtain G(2,1). Re-execute the above steps and evaluate the prediction accuracy. If it still does not meet the standard, change it to a third-order linear differential equation to obtain G(3,1). This process is repeated until the prediction accuracy of the improved grey prediction model meets the qualified index.

[0034] The XRF spectrum data is input into the above-mentioned improved grey prediction model to determine the contribution of a single element to dioxin, thereby finding the single element with the largest contribution to the dioxin concentration in the fly ash sample. In the embodiment disclosed in the present invention, calcium is used as an example for illustration without limitation.

[0035] Then, the compound types and contents corresponding to the calcium element are collected from the XRD spectrum, such as calcium chloride, calcium hydroxide, calcium carbonate, calcium sulfate, sodium chloride, potassium chloride, magnesium chloride, calcium chlorate and other compounds.

[0036] In another embodiment, indicators that have a significant impact on dioxin concentration, such as chloride, etc., can also be introduced as input features of the neural network model together with the concentration of the compound corresponding to the above-mentioned calcium element, such as ammonium chloride, calcium oxychloride, calcium chloride hydroxy calcium and other compounds, to improve the predictive ability of the model.

[0037] Normalize or standardize the above input features to make the data in the same scale range. The specific processing methods include: Normalization: ; standardization: .

[0038] The total samples were divided into training set, validation set and test set, with proportions of 70%, 15% and 15% respectively, and K-fold cross validation was used to evaluate the stability of the model.

[0039] Design a neural network model, including an input layer, multiple hidden layers, and an output layer. The number of input layer nodes is equal to the number of input features; the number of hidden layer nodes decreases layer by layer, and the ReLU activation function is used. The formula is as follows: ; The output layer is a single node that uses a linear activation function to output the predicted value of dioxin concentration. It is changed to use a linear activation function or ReLU.

[0040] The loss function is defined as the mean square error, and its expression is: ; in, is the actual value of dioxin concentration, is the predicted value of dioxin concentration. The Adam optimizer is used for model training, and its update rules include momentum adjustment algorithm and learning rate adaptive adjustment to enhance the training effect. The specific formula is: Momentum calculation: ; Second-order momentum: ; Bias correction: ; Weight update: .

[0041] During the training process, an improved dynamic early stopping method is used to prevent overfitting. Set the loss threshold and loss change rate threshold. When the validation set loss is lower than the loss threshold and the loss change rate is lower than the change rate threshold, the early stopping mechanism is triggered and the training process ends.

[0042] The model is validated using the validation set data and the model parameters are adjusted until convergence to ensure that the model has good generalization ability. The final prediction accuracy of the model is verified using the test set.

[0043] After the training of the neural network model is completed, new fly ash samples are collected and XRF and XRD analysis is performed in the same way to obtain the concentration data of related compounds. Then, these data are input into the trained neural network model, and the concentration of chloride can also be introduced as an input feature to predict the dioxin concentration.

[0044] Since an improved grey prediction model is used for preliminary screening and feature extraction, combined with a neural network model for fitting complex nonlinear relationships, it is possible to achieve higher prediction accuracy with less data, effectively reducing the measurement cost and time cost of dioxins.

[0045] like Figure 2 As shown, Figure 2 A schematic diagram showing the error result of a neural network model provided by an embodiment of the present disclosure is shown. The error is 10 -2 In terms of order of magnitude, it can be found from the figure that when the training step size is set to 1, the loss value obtained by the loss function of the neural network model is the optimal value.

[0046] like Figure 3 As shown, Figure 3 A schematic diagram of the correlation coefficient of a trained neural network model provided in an embodiment of the present disclosure is shown. The correlation coefficient is expressed as R=0.92395 in the example, indicating that the gray prediction model provided by the present disclosure has a better fitting effect, good performance and high prediction accuracy after being combined with the neural network model.

[0047] The dioxin monitoring method based on the improved grey model disclosed in the present invention makes full use of the useful information in XRF and XRD data by combining the improved grey prediction model and the neural network model. The grey prediction model fits the complex nonlinear relationship between the precursor and the dioxin concentration, and indirectly measures the concentration of related compounds to predict the dioxin concentration, thereby enhancing the generalization ability of the neural network model, avoiding the high cost and complexity of directly measuring dioxins in traditional methods, greatly reducing the measurement cost, and realizing rapid and large-scale monitoring of dioxins in fly ash.

[0048] Embodiment 2: like Figure 4 As shown, Figure 4 The structural diagram of a dioxin monitoring device based on an improved grey model provided in an embodiment of the present disclosure. The device includes: An acquisition unit, used for acquiring X-ray fluorescence spectrum data and X-ray diffraction spectrum data of a plurality of fly ash samples; A processing unit, used to process X-ray fluorescence spectrum data using an improved grey prediction model to identify the single element in the fly ash that contributes most to the dioxin concentration; A determination unit, for determining the concentration of a related compound corresponding to a single element based on the X-ray diffraction pattern data; The prediction unit is used to predict the concentration of dioxins according to the concentration of related compounds in the new fly ash sample using a neural network model; wherein the neural network model is trained based on training data consisting of the concentrations of related compounds in multiple fly ash samples and the corresponding dioxin concentrations.

[0049] Specifically, the improved grey prediction model is constructed through the following units: The first generating unit is used to generate the original data sequence X according to the concentration of dioxins in the fly ash sample and the element ratio data in the X-ray fluorescence spectrum data. (0) ; The second generation unit is used to generate X (0) Perform an accumulation to generate a new sequence X (1) , ; The third generation unit is used to generate a new sequence X (1) , generate the adjacent mean sequence z(t) as the background value of the improved grey prediction model, ; Create a unit for the new sequence X (1) , establish the first-order linear differential equation, ; This model is recorded as the improved grey prediction model GM (1, 1); Among them, a is the development coefficient, b is the grey action; the least squares method is used to estimate the parameters a and b, and the formula is as follows: ; ; Solve the reduction unit to solve the first-order linear differential equation to obtain X (1) The predicted expression is converted back to X (0) The predicted value of Evaluation unit for comparison with X (0)The original actual value and predicted value are used to evaluate the prediction accuracy of the improved grey prediction model. If the qualified index is not met, the first-order linear differential equation in the improved grey prediction model is changed to an n-order linear differential equation, which is recorded as the improved grey prediction model G(n, 1), where n is an integer and the value of n increases from 2 until the prediction accuracy of the improved grey prediction model meets the qualified index.

[0050] Specifically, the input features also include chloride concentration.

[0051] Specifically, the neural network model includes an input layer, an output layer, and multiple hidden layers; The number of nodes in the input layer is equal to the number of input features, the number of nodes in multiple hidden layers gradually decreases, and the output layer is a single node used to output the concentration of dioxins; The sample data corresponding to the plurality of fly ash samples are divided into a training set, a validation set and a test set, wherein the sample data includes the concentration of the relevant compound and the corresponding dioxin concentration; based on the training set, the validation set and the test set, the neural network model is trained using the Adam optimizer until the loss function converges; The loss function is ;in, is the actual value, is the predicted value.

[0052] The dioxin monitoring device based on the improved grey model provided in the second embodiment of the present disclosure has the same implementation principle and technical effects as those of the aforementioned method embodiment 1. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment 1.

[0053] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device and a readable storage medium.

[0054] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. 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 can also represent various forms of mobile devices, such as personal digital processing, 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 required herein.

[0055] like Figure 3As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 803 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0056] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0057] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as a dioxin monitoring method based on an improved gray model. For example, in some embodiments, the dioxin monitoring method based on the improved gray model may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the dioxin monitoring method based on the improved gray model described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to execute the dioxin monitoring method based on the improved grey model in any other appropriate manner (eg, by means of firmware).

[0058] Various implementations of the systems and techniques described above herein can 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), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can 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.

[0059] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0060] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may 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.

[0061] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0062] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0063] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0064] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0065] 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 technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0066] The above is only a specific embodiment of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present disclosure, which should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A dioxin monitoring method based on an improved grey model, characterized in that: The following steps are involved: Obtaining X-ray fluorescence spectral data and X-ray diffraction pattern data of multiple fly ash samples; The X-ray fluorescence spectrum data is processed using an improved grey prediction model to identify the single element with the largest contribution to the dioxin concentration in the fly ash sample; Determining the concentration of the related compound corresponding to the single element according to the X-ray diffraction pattern data; The concentration of dioxins is predicted according to the concentration of the related compounds in the new fly ash sample using a neural network model; wherein the neural network model is trained based on training data consisting of the concentrations of the related compounds of the plurality of fly ash samples and the corresponding dioxin concentrations.

2. The dioxin monitoring method based on the improved grey model according to claim 1, characterized in that: The improved grey prediction model is constructed in advance through the following steps: According to the concentration of dioxins in the fly ash sample and the element ratio data in the X-ray fluorescence spectrum data, an original data sequence X is generated. (0) ; X (0) Perform an accumulation to generate a new sequence X (1) , the formula is: ; According to the new sequence X (1) , generate the adjacent mean value sequence z(t) as the background value of the improved grey prediction model, the formula is: ; According to the new sequence X (1) , establish the first-order linear differential equation, ; This model is recorded as the improved grey prediction model GM(1,1); Among them, a is the development coefficient, b is the grey action; the least squares method is used to estimate the parameters a and b, and the formula is as follows: ； ； Solving the first-order linear differential equation yields X (1) The predicted expression is converted back to X (0) The predicted value of Compare X (0) The original actual value and predicted value are used to evaluate the prediction accuracy of the improved grey prediction model. If the qualified index is not met, the first-order linear differential equation in the improved grey prediction model is changed to an n-order linear differential equation, which is recorded as the improved grey prediction model G(n,1), where n is an integer and the value of n increases from 2 until the prediction accuracy of the improved grey prediction model meets the qualified index.

3. The dioxin monitoring method based on the improved grey model according to claim 1, characterized in that: The input characteristics also include chloride concentration.

4. The dioxin monitoring method based on the improved grey model according to claim 1, characterized in that: The neural network model includes an input layer, a plurality of hidden layers and an output layer; The number of nodes in the input layer is equal to the number of input features, the number of nodes in the multiple hidden layers decreases layer by layer, and the output layer is a single node for outputting the concentration of dioxin; Dividing sample data corresponding to the plurality of fly ash samples into a training set, a validation set, and a test set, wherein the sample data includes the concentration of the relevant compound and the corresponding dioxin concentration; training the neural network model using an Adam optimizer based on the training set, the validation set, and the test set until the loss function converges; The loss function is ;in, is the actual value, is the predicted value.

5. A dioxin monitoring device based on an improved grey model, characterized in that: include: An acquisition unit, used for acquiring X-ray fluorescence spectrum data and X-ray diffraction spectrum data of a plurality of fly ash samples; A processing unit, for processing the X-ray fluorescence spectrum data using an improved grey prediction model to identify a single element in the fly ash that has the largest contribution to dioxin concentration; a determination unit, configured to determine the concentration of a related compound corresponding to the single element according to the X-ray diffraction pattern data; The prediction unit is used to predict the concentration of dioxins according to the concentration of the related compounds in the new fly ash sample using a neural network model; wherein the neural network model is trained based on training data consisting of the concentrations of the related compounds of the plurality of fly ash samples and the corresponding dioxin concentrations.

6. The dioxin monitoring device based on the improved grey model according to claim 5, characterized in that: The improved grey prediction model is constructed by the following units: The first generating unit is used to generate an original data sequence X according to the concentration of dioxins in the fly ash sample and the element ratio data in the X-ray fluorescence spectrum data. (0) ; The second generation unit is used to generate X (0) Perform an accumulation to generate a new sequence X (1) , ; The third generating unit is used to generate a new sequence X according to the (1) , generate the adjacent mean value sequence z(t) as the background value of the improved grey prediction model, ; Establish a unit for (1) , establish the first-order linear differential equation, ; This model is recorded as the improved grey prediction model GM (1, 1); Among them, a is the development coefficient, b is the grey action; the least squares method is used to estimate the parameters a and b, and the formula is as follows: ； ； A solving and reduction unit is used to solve the first-order linear differential equation to obtain X (1) The predicted expression is converted back to X (0) The predicted value of Evaluation unit for comparison with X (0) The original actual value and predicted value are used to evaluate the prediction accuracy of the improved grey prediction model. If the qualified index is not met, the first-order linear differential equation in the improved grey prediction model is changed to an n-order linear differential equation, which is recorded as the improved grey prediction model G(n, 1), where n is an integer and the value of n increases from 2 until the prediction accuracy of the improved grey prediction model meets the qualified index.

7. The dioxin monitoring device based on the improved grey model according to claim 5, characterized in that: The input characteristics also include chloride concentration.

8. The dioxin monitoring device based on the improved grey model according to claim 5, characterized in that: The neural network model includes an input layer, an output layer and multiple hidden layers; The number of nodes in the input layer is equal to the number of input features, the number of nodes in the multiple hidden layers gradually decreases, and the output layer is a single node for outputting the concentration of dioxin; Dividing sample data corresponding to the plurality of fly ash samples into a training set, a validation set, and a test set, wherein the sample data includes the concentration of the relevant compound and the corresponding dioxin concentration; training the neural network model using an Adam optimizer based on the training set, the validation set, and the test set until the loss function converges; The loss function is ;in, is the actual value, is the predicted value.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed 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 method according to any one of claims 1 to 4.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-4.

Images