Fly ash treatment monitoring method and system based on internet of things

Abstract

The application relates to the technical field of data processing, and discloses a fly ash treatment monitoring method and system based on the Internet of Things. The method comprises the following steps: obtaining fly ash component concentration and environmental data through a sensor array to generate a concentration gradient distribution data set; based on this, an analysis algorithm is used to establish a component interaction intensity model to obtain a quantitative index; the index is analyzed through time series to predict the dynamic law of the binding performance; the binding characteristics are classified by cluster analysis according to the law to obtain a classification result; a control strategy is determined by correlating historical process parameters; and the process is adjusted to form closed-loop control according to the control strategy to obtain stable binding performance. Through the Internet of Things collection, analysis algorithm, time series and cluster analysis, the application solves the problem of difficult control of fly ash binding performance, and improves the incineration efficiency and environmental safety.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a fly ash treatment monitoring method and system based on the Internet of Things. Background Technology

[0002] In industrial settings such as waste incineration and industrial waste treatment, fly ash, as a combustion byproduct, has a complex and dynamically changing composition. The interaction of chemical components such as chlorides, sulfates, and silicates directly affects its binding properties. Unstable binding properties can lead to problems such as slagging in incineration equipment and fly ash dust generation, which not only reduce incineration efficiency but also threaten environmental safety. Therefore, precise monitoring of the fly ash treatment process is a critical issue that the industry urgently needs to address.

[0003] Existing technologies for monitoring fly ash treatment have significant shortcomings: First, monitoring methods are limited, focusing primarily on the concentration of single chemical components or isolated environmental parameters, failing to capture the synergistic effects and dynamic changes among multiple components. Second, there is a lack of precise analysis of the complex mapping relationship between environmental conditions (such as temperature, pressure, and humidity) and process parameters, making it difficult to address dynamic changes in components caused by fluctuations in raw material sources and incineration conditions. Third, control strategies are lagging behind, as the inability to establish a dynamic correlation model between component interactions and binding properties leads to untimely responses to issues such as equipment slagging and fly ash dust generation, resulting in insufficient control accuracy and difficulty in achieving stable control under complex operating conditions.

[0004] This application integrates sensor arrays to collect multi-dimensional data through Internet of Things (IoT) technology, analyzes the interaction strength of components through algorithm modeling, predicts the dynamic changes in bonding performance, and determines the control strategy and forms a closed-loop control by combining clustering classification with historical process parameters, thereby solving the shortcomings of existing technologies. Summary of the Invention

[0005] This application provides an IoT-based fly ash treatment monitoring method and system. By analyzing the interaction intensity of components through algorithm modeling, it predicts the dynamic changes in bonding performance. Combining clustering classification with historical process parameters, it determines the control strategy and forms a closed-loop control. This solves the problems in the existing technology, such as the single monitoring method, inability to capture the synergistic effect of components, insufficient analysis of the mapping relationship between the environment and process parameters, and the lagging and low accuracy of the control strategy. These problems make it difficult to stably control the bonding performance of fly ash under complex operating conditions, resulting in low incineration efficiency and threats to environmental safety.

[0006] Firstly, this application provides an IoT-based fly ash treatment monitoring method, which includes:

[0007] Step S101: Obtain concentration data of multiple chemical components in fly ash and environmental condition data through a sensor array to obtain a component concentration gradient distribution dataset;

[0008] Step S102: Based on the component concentration gradient distribution dataset, an analysis algorithm is used to establish a calculation model for the interaction intensity between chemical components, and a quantitative index of interaction intensity is obtained.

[0009] Step S103: Analyze the interaction strength quantification index using a time series analysis model to predict the dynamic change law of bonding performance;

[0010] Step S104: Based on the dynamic change law of the bonding performance, a clustering analysis algorithm is used to classify the bonding characteristics and obtain the classification prediction results;

[0011] Step S105: By analyzing the correlation between the classification prediction results and historical process parameters, establish the mapping relationship between component synergy effects and process parameters, and determine the control strategy parameters;

[0012] Step S106: Adjust the process flow according to the control strategy parameters to form a closed-loop feedback control system and obtain stable fly ash adhesion performance.

[0013] Optionally, the step of acquiring concentration data of multiple chemical components and environmental condition data in fly ash through a sensor array includes: synchronously collecting concentration data of chloride, sulfate, and silicate in fly ash through the sensor array at a preset monitoring frequency, and recording environmental parameters such as temperature, pressure, and humidity to form time series change information; correcting the concentration data and environmental parameters through a preset calibration algorithm; and generating a multi-dimensional component concentration gradient distribution dataset based on the corrected concentration data, environmental parameters, and time series change information.

[0014] Optionally, the step of establishing a calculation model for the interaction intensity between chemical components using an analysis algorithm based on the component concentration gradient distribution dataset to obtain a quantitative index of interaction intensity includes: establishing a calculation model for the interaction intensity between each chemical component using a multiple regression analysis algorithm based on the component concentration gradient distribution dataset; monitoring the concentration ratio of chloride to sulfate in real time, and performing real-time calculation of the synergistic effect intensity when the concentration ratio exceeds a preset threshold; and generating a quantitative index of dynamic interaction intensity between components based on the real-time calculation result of the synergistic effect intensity.

[0015] Optionally, the step of establishing a calculation model for the interaction intensity between various chemical components using a multiple regression analysis algorithm includes: dividing temperature zones according to a set temperature threshold, classifying areas with temperatures above a first temperature threshold as high-temperature zones, areas with temperatures below a second temperature threshold as low-temperature zones, and areas with temperatures between the first and second temperature thresholds as normal-temperature zones; dividing concentration zones according to a set concentration threshold, classifying areas with component concentration fluctuations above the concentration threshold as fluctuating zones, and areas with component concentration fluctuations below the concentration threshold as flat zones; and performing piecewise fitting of independent sub-models for the high-temperature zone, low-temperature zone, normal-temperature zone, fluctuating zone, and flat zone to obtain the interaction intensity calculation model.

[0016] Optionally, the step of analyzing the interaction strength quantification index through a time series analysis model to predict the dynamic change law of bonding performance includes: inputting the interaction strength quantification index into a time series analysis model; analyzing the changing trend of component synergy effect at different time nodes through a sliding window mechanism; determining whether the changing trend shows that the interaction strength is increasing; if the interaction strength is increasing, predicting the evolution direction of bonding characteristics in future periods to generate the dynamic change law of bonding performance.

[0017] Optionally, the step of classifying the bonding characteristics using a clustering analysis algorithm based on the dynamic change law of the bonding performance to obtain a classification prediction result includes: classifying and identifying the bonding characteristics under different component ratios using a clustering analysis algorithm based on the dynamic change law of the bonding performance; grouping component combinations with similar bonding behaviors into the same cluster; determining the bonding characteristic category to which the current fly ash state belongs based on the cluster; and generating the classification prediction result of the bonding performance.

[0018] Optionally, the step of establishing a mapping relationship between component synergy and process parameters and determining control strategy parameters through correlation analysis between the classification prediction results and historical process parameters includes: acquiring historical process parameter data; performing correlation analysis between the classification prediction results and the historical process parameter data; establishing a mapping relationship between component synergy and optimal process parameters; determining whether the current component interaction intensity exceeds a preset range; if it exceeds the preset range, automatically matching the corresponding control strategy parameter combination to determine the parameter values ​​of temperature adjustment amplitude and processing time.

[0019] Optionally, adjusting the process flow according to the control strategy parameters to form a closed-loop feedback control system and obtain stable fly ash bonding performance includes: adjusting the incineration temperature and fly ash treatment process in real time according to the control strategy parameters; continuously monitoring the change data of chemical composition content in fly ash after control; analyzing the response trend of bonding characteristics after control based on the change data; and adjusting the process flow parameters through the response trend to form a closed-loop feedback control system and generate stable fly ash bonding performance.

[0020] Secondly, this application provides an IoT-based fly ash treatment monitoring system, the IoT-based fly ash treatment monitoring system comprising:

[0021] Data acquisition module: Collects the concentration of components such as chloride in fly ash, as well as environmental parameters such as temperature, pressure and humidity, according to the preset frequency of the sensor array, forming time series information. After being corrected by the calibration algorithm, a multi-dimensional component concentration gradient distribution dataset is generated.

[0022] Interaction intensity modeling module: Based on the multidimensional component concentration gradient distribution dataset, multiple regression modeling is used to monitor the chlorine-sulfur concentration ratio in real time, calculate the synergistic effect intensity when the threshold is exceeded, and generate a quantitative index of dynamic interaction intensity between components.

[0023] Adhesion performance prediction module: Input the interaction strength index into the time series model, analyze the trend through a sliding window, predict the evolution of adhesion properties when the interaction strength increases, and generate its dynamic change law.

[0024] Adhesion property classification module: Based on the adhesion performance rules, cluster analysis is used to classify the components with similar adhesion behavior into the same cluster, determine the current category, and generate classification prediction results.

[0025] Parameter control module: It associates classification results with historical process parameters, establishes a mapping relationship, and matches control parameters to determine the temperature adjustment range and processing time when the component interaction intensity exceeds the range.

[0026] Closed-loop control module: Adjusts the incineration temperature and process flow according to the control parameters, monitors the changes in composition after control, analyzes the response trend of bonding characteristics, adjusts parameters to form a closed loop, and obtains stable performance.

[0027] Thirdly, this application provides an IoT-based fly ash treatment monitoring device, the device including a memory and a processor, the memory storing a computer program that can run on the processor, and the processor executing the computer program to implement the IoT-based fly ash treatment monitoring method.

[0028] This application proposes an IoT-based fly ash treatment monitoring method and system, applicable to fly ash treatment scenarios such as waste incineration and industrial waste treatment. It can solve problems such as unstable binding performance caused by the interaction of multiple chemical components in fly ash, and the limitations of existing technologies in terms of limited monitoring methods and delayed control. Compared with existing technologies, the beneficial effects of this application's technical solution are at least as follows:

[0029] First, by collecting multi-dimensional data through sensor arrays, the limitations of single monitoring methods can be overcome. It can comprehensively capture the concentration changes of components such as chloride, sulfate, and silicate in fly ash, as well as environmental parameters such as temperature, pressure, and humidity, providing a complete data foundation for subsequent analysis.

[0030] Secondly, by using analytical algorithms to establish a calculation model for the interaction strength between chemical components, and combining it with methods such as piecewise fitting, the synergistic effect between components can be accurately quantified. In particular, the model accuracy is improved when there are sudden changes in temperature and component concentration, thus avoiding distortion in the prediction of bonding performance due to inaccurate calculation of interaction strength.

[0031] Third, by predicting the dynamic changes in bonding performance through time series analysis and classifying bonding characteristics through cluster analysis, we can promptly grasp the evolution trend of fly ash bonding performance, provide advance notice for regulation, and solve the problem of lagging response of existing technologies to dynamic changes.

[0032] Fourth, by analyzing the correlation between the classification prediction results and historical process parameters, a mapping relationship between the synergistic effect of components and process parameters can be established, which can quickly determine the appropriate control strategy parameters, making process adjustments more targeted and improving the accuracy of control.

[0033] Fifth, a closed-loop feedback control system is formed, which continuously optimizes the process flow based on the adjusted data. This can stabilize the fly ash adhesion performance, reduce equipment slagging and fly ash dust problems, and ensure incineration efficiency and environmental safety. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a schematic diagram of one embodiment of the fly ash treatment monitoring method based on the Internet of Things in this application.

[0036] Figure 2 This is a schematic diagram of one embodiment of the fly ash treatment monitoring process based on the Internet of Things in this application.

[0037] Figure 3This is a schematic diagram of one embodiment of the fly ash treatment monitoring system based on the Internet of Things in this application.

[0038] Figure 4 This is a schematic block diagram of the fly ash treatment monitoring device based on the Internet of Things in an embodiment of the present invention. Detailed Implementation

[0039] This application provides an IoT-based fly ash handling monitoring method and system. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0040] This application relates to the field of data processing technology, and in particular to an IoT-based method and system for monitoring fly ash treatment. The method includes: acquiring component concentration and environmental data through a sensor array to obtain a concentration gradient distribution dataset; establishing an interaction intensity model using analytical algorithms to obtain quantitative indicators; predicting the change pattern of bonding performance through time series analysis; classifying bonding characteristics using cluster analysis to obtain classification results; determining control strategies by associating with historical process parameters; and adjusting the process accordingly to form closed-loop control, thereby achieving stable bonding performance. This application solves the problem of difficult control of fly ash bonding performance through IoT data acquisition and multi-algorithm analysis, improving incineration efficiency and environmental safety.

[0041] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the fly ash treatment monitoring method based on the Internet of Things in this application includes:

[0042] Step S101: Obtain concentration data of multiple chemical components in fly ash and environmental condition data through a sensor array to obtain a component concentration gradient distribution dataset;

[0043] In one specific embodiment, the process of performing step S101 may specifically include the following steps:

[0044] The sensor array synchronously collects concentration data of chloride, sulfate and silicate in fly ash at a preset monitoring frequency, and records environmental parameters such as temperature, pressure and humidity to form time series change information;

[0045] The concentration data and environmental parameters are corrected using a preset calibration algorithm;

[0046] Based on the corrected concentration data, environmental parameters, and time series change information, a multi-dimensional component concentration gradient distribution dataset is generated.

[0047] Specifically, the sensor array consists of a chloride ion selective electrode, a sulfate ion sensor, a silicate analyzer, and environmental sensors. These are deployed at key nodes in the fly ash treatment pipeline, such as the incinerator outlet and the collector inlet, and are connected to the central control unit via a wireless network for synchronous data transmission. The system synchronously collects concentration data of chloride (e.g., 50-80 mg / kg), sulfate (e.g., 30-50 mg / kg), and silicate (e.g., 200-250 mg / kg) in fly ash at a preset monitoring frequency (with adjustable acquisition intervals to adapt to changes). Simultaneously, it records environmental parameters such as temperature (e.g., 800-900℃), pressure (e.g., 1.0-1.2 atm), and humidity (e.g., 20%-40%). All data are aligned with timestamps to form a time-series change information. For example, the chloride concentration sequence [50, 55, 60, ...] mg / kg corresponds to the temperature sequence [800, 810, 820, ...]℃ and the humidity sequence [20%, 22%, 25%, ...]. The collected data is corrected using a preset calibration algorithm. This algorithm employs the least squares method, first collecting standard sample data of known concentrations to construct a fitting equation (e.g., ...). y =0.95 x +0.02, where y For correction value, x The original concentration data and environmental parameters are then substituted into the equation to calculate the correction value. For example, if the original chloride sensor reading is 1.1 times the actual value, the corrected value is the original reading multiplied by 0.909, ensuring that the error of the corrected data is less than 1%. Based on the corrected concentration data, environmental parameters, and time series change information, a multi-dimensional component concentration gradient distribution dataset is generated. Specifically, each data point is integrated into a vector along the time axis, and parameters such as temperature, pressure, and humidity are incorporated to form a multi-dimensional matrix (rows represent time points, and columns represent chloride concentration, sulfate concentration, silicate concentration, temperature, pressure, and humidity). The gradient value is obtained by calculating the difference between adjacent time points for each dimension (e.g., chloride gradient 5 mg / kg / min, temperature gradient 10℃ / min). The final dataset contains the concentration of each component, environmental parameters, and their gradient changes over time.

[0048] Step S102: Based on the component concentration gradient distribution dataset, an analysis algorithm is used to establish a calculation model for the interaction intensity between chemical components, and a quantitative index of interaction intensity is obtained.

[0049] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0050] Based on the component concentration gradient distribution dataset, a multivariate regression analysis algorithm is used to establish a calculation model for the interaction intensity between various chemical components.

[0051] The concentration ratio of chloride to sulfate is monitored in real time. When the concentration ratio exceeds a preset threshold, the synergistic effect strength is calculated in real time.

[0052] Based on the real-time calculation results of the synergistic effect intensity, a quantitative index of the dynamic interaction intensity between components is generated.

[0053] Specifically, based on the component concentration gradient distribution dataset, a multivariate regression analysis algorithm is used to establish a calculation model for the interaction strength between various chemical components. This requires extracting the concentration gradient data of chloride, sulfate, and silicate, as well as the corresponding time series temperature parameters from the dataset. In the multivariate regression analysis, the independent variables are chloride concentration (C1), sulfate concentration (C2), silicate concentration (C3), and the interaction terms among them (C1×C2, C1×C3, C2×C3). The dependent variable is the interaction strength (I). The least squares method is used to fit 1000 samples in the dataset, and the regression coefficients are calculated, resulting in the equation I=A×C1+B×C2+C×C3+D×C1×C2-E×C1×C3+F×C2×C3. The coefficient AF is determined through sample training to ensure that the deviation between the interaction strength output by the model and the actual observed adhesion correlation data is controlled within 2%. The concentration ratio of chloride to sulfate is monitored in real time. The C1 and C2 values ​​at the same time point are extracted from the dataset, and the ratio R = C1 / C2 is calculated. The preset threshold is set to 1.5 based on historical data. When C1 is 65 mg / kg and C2 is 40 mg / kg at a certain time point, R = 1.625, exceeding the threshold, triggering a real-time calculation program for the synergistic effect strength. The program calls the interaction strength data for this time point and the previous four time points (calculated through a regression model, for example, 1.2, 1.3, 1.5, 1.4, and 1.6 respectively), forming a sliding window of size 5. The average interaction strength within the window is calculated to be 1.4. A temperature parameter is introduced to adjust the weight; for every 10°C increase in temperature, the weight increases by 0.1. For example, the current temperature of 860°C is 60°C higher than the baseline temperature of 800°C, increasing the weight by 0.6. The synergistic effect strength = 1.4 × (1 + 0.6) = 2.24. Based on the real-time calculation results of the synergistic effect strength, a quantitative index of the dynamic interaction strength between components is generated, which needs to be corrected in conjunction with the silicate concentration during the same period. The quantitative indicator Q = synergistic effect strength × log(C3). When C3 is 220 mg / kg, log(C3) ≈ 2.342; when the synergistic effect strength is 2.24, Q = 2.24 × 2.342 ≈ 5.24. This indicator is correlated with a time series, forming a sequence [4.82, 5.01, 5.24, ...], where each Q value corresponds to C1, C2, C3, and temperature parameters at a specific time point, reflecting the strength of the dynamic interaction between chloride, sulfate, and silicate at that moment.

[0054] In one specific embodiment, the step of establishing a calculation model for the interaction strength among various chemical components using a multiple regression analysis algorithm, wherein the interaction strength calculation model employs a piecewise multiple regression fitting method, including:

[0055] Temperature zones are divided according to the set temperature thresholds. Areas with temperatures above the first temperature threshold are classified as high-temperature zones, areas with temperatures below the second temperature threshold are classified as low-temperature zones, and areas with temperatures between the first and second temperature thresholds are classified as normal-temperature zones.

[0056] Concentration zones are divided according to the set concentration thresholds. Zones where the concentration fluctuation of a component is higher than the concentration threshold are classified as fluctuation zones, and zones where the concentration fluctuation of a component is lower than the concentration threshold are classified as flat zones.

[0057] For the high temperature zone, low temperature zone, normal temperature zone, fluctuating zone, and flat zone, an independent sub-model is used for piecewise fitting to obtain the interaction intensity calculation model.

[0058] Specifically, to address the inaccuracy issue of the interaction intensity calculation model under abrupt changes in temperature and component concentration, a piecewise multiple regression fitting method is adopted. This requires first extracting temperature data and concentration fluctuation data of chloride, sulfate, and silicate from the component concentration gradient distribution dataset. Temperature zones are then defined based on set temperature thresholds: a first threshold of 850℃ and a second threshold of 750℃. Areas with temperatures above 850℃ are classified as high-temperature zones, such as periods when the incinerator outlet temperature reaches 860℃ and 880℃; operating conditions with temperatures below 750℃ are classified as low-temperature zones, such as periods during startup when the temperature is 680℃ and 720℃; and areas with temperatures between 750℃ and 850℃ are classified as normal-temperature zones, such as periods during stable operation when the temperature is 780℃ and 820℃. Concentration zones are defined based on a set concentration threshold, with the threshold being a 10% fluctuation range in component concentration. Zones with fluctuations exceeding 10% are classified as fluctuation zones, such as a period where chloride concentration abruptly changes from 50 mg / kg to 65 mg / kg within 5 minutes (a 30% fluctuation). Zones with fluctuations below 10% are classified as level zones, such as a period where sulfate concentration remains between 30-32 mg / kg (a 2% fluctuation). Independent sub-models are used for piecewise fitting in the high-temperature, low-temperature, normal-temperature, fluctuation, and level zones, requiring the construction of an independent multiple regression analysis model for each zone. The high-temperature model uses the concentration data of chloride (60-80 mg / kg), sulfate (30-40 mg / kg), and silicate (200-250 mg / kg) from the high-temperature dataset, along with interaction terms (C1×T, C2×T, where T is temperature). The regression equation I1 = A1×C1 + B1×C2 + D1×C3 + E1×C1×T + F1×C2×T is obtained through least squares fitting. The low-temperature model uses the low-temperature data, with the equation I2 = A2×C1 + B2×T. The equation for the normal temperature region model is I3 = A3 × C1 + B3 × C2 + D3 × C3 + E3 × C1 × C2 + F3 × C2 × C3; the equation for the fluctuating region model introduces the concentration fluctuation rate parameter, and the equation is I4 = A4 × C1 + B4 × C2 + D4 × C3 + E4 × C1 × V (V is the fluctuation rate); the equation for the flat region model simplifies the interaction term, and the equation is I5 = A5 × C1 + B5 × C2 + D5 × C3 + E5 × C1 × C2. Each sub-model is trained using 1000 sets of sample data from the corresponding region, and the fitting error is controlled within 2%. In practical applications, when the temperature suddenly changes to 870℃ (high temperature zone) and chloride fluctuates by 25% (fluctuation zone), the joint sub-model of the high temperature zone and the fluctuation zone is called, and the current concentration data is input to calculate the interaction strength; when the temperature stabilizes at 800℃ (normal temperature zone) and sulfate fluctuates by 3% (smooth zone), the sub-model of the normal temperature zone and the smooth zone is called.The piecewise multiple regression fitting method divides the temperature zone and concentration zone, allowing each sub-model to adapt to the component interaction law under specific environments, thus solving the problem of low model accuracy in abrupt change scenarios. The independent sub-models are fitted in a targeted manner to ensure more reliable calculation of the interaction intensity in different regions, providing accurate input for subsequent prediction and control of bonding performance, and alleviating the problem of control lag or inaccuracy caused by model errors.

[0059] Step S103: Analyze the interaction strength quantification index using a time series analysis model to predict the dynamic change law of bonding performance;

[0060] In one specific embodiment, the process of performing step S103 may specifically include the following steps:

[0061] Input the interaction intensity quantification index into the time series analysis model;

[0062] The changing trend of component synergistic effect at different time points was analyzed using a sliding window mechanism;

[0063] Determine whether the trend of change indicates an upward trend in interaction intensity;

[0064] If the interaction strength shows an upward trend, the evolution direction of the bonding properties in the future period can be predicted, and the dynamic change law of bonding performance can be generated.

[0065] Specifically, during fly ash treatment, the interactions of chemical components such as chlorides, sulfates, and silicates lead to unstable bonding properties, a problem particularly prominent in waste incineration or industrial waste treatment. To predict the dynamic changes in bonding properties, a time series analysis model is needed to process the quantitative index of interaction strength. This index originates from the dynamic interaction data between components generated by a multiple regression analysis model, reflecting the synergistic effect strength of components such as chlorides and sulfates. When inputting the quantitative index of interaction strength into the time series analysis model, the input sequence is first tested for stationarity. If it is non-stationary, differencing is performed to ensure that the sequence meets the modeling requirements. For example, when the interaction strength sequence shows an upward trend, it is transformed into a stationary sequence through differencing, facilitating subsequent parameter estimation.

[0066] A sliding window mechanism is used to analyze the changing trends of component synergistic effects at different time points. The window size is set to 10 time points, moving once per minute to cover continuous interaction intensity data. Within each window, the mean and standard deviation of the interaction intensity are calculated to form a trend line. If the mean values ​​for three consecutive windows are 0.5, 0.6, and 0.7 respectively, and the standard deviation remains stable, then the interaction intensity is considered to be increasing. This mechanism can capture short-term fluctuations, avoid prediction biases caused by data delays, and track changes in fly ash composition at the incinerator outlet in real time when temperature fluctuations cause rapid changes, providing a reliable basis for prediction.

[0067] To determine whether the interaction intensity is increasing, the slope of the trend line within the window is compared. If the slope is positive and exceeds a preset threshold (e.g., 0.05 units per minute), it is considered to be in an upward trend. At this point, a prediction program is triggered, using the Holt-Winters model (which handles seasonal and trend data) combined with historical interaction intensity sequences and current window data to generate a predicted value for the next 5 minutes. For example, if the current interaction intensity is 0.7, it is predicted to rise to 0.8 in the future, resulting in a 10% increase in adhesion performance. This result directly correlates with the actual adhesion behavior of fly ash, providing a basis for process adjustments.

[0068] When generating dynamic patterns of bonding performance, the model combines predicted values ​​with actual monitoring data to form descriptive patterns. For example, if the interaction strength increases by 0.05 units per minute, the bonding index increases by 10%, reflecting the dynamic characteristics of the synergistic effect of components. In waste incineration, if the chloride concentration suddenly increases due to changes in raw materials, the model can quickly predict the evolution trend of bonding performance and promptly trigger control strategies. By adjusting the incineration temperature or treatment time, the model can suppress the increase in interaction strength and prevent equipment slagging or fly ash dust.

[0069] Step S104: Based on the dynamic change law of the bonding performance, a clustering analysis algorithm is used to classify the bonding characteristics and obtain the classification prediction results;

[0070] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0071] Based on the dynamic change law of the bonding performance, a clustering analysis algorithm is used to classify and identify the bonding characteristics under different component ratios;

[0072] Combinations with similar bonding behaviors are grouped into the same cluster;

[0073] Based on the clusters, determine the current fly ash state to which the bonding characteristic category belongs;

[0074] Generate classification prediction results for the bonding properties.

[0075] Specifically, during fly ash treatment, the dynamic changes in bonding performance manifest as varying effects of different chemical composition combinations on bonding characteristics. These differences stem from the complex interactions of components such as chlorides, sulfates, and silicates. To classify bonding characteristics, a clustering analysis algorithm is employed to process the dynamic change data. The input data includes the bonding performance evolution trend generated by a time-series analysis model and the corresponding component ratio information. For example, when the chloride concentration is 15%, sulfate is 20%, and silicate is 65%, the curve of bonding strength changing over time is extracted as a feature vector. The vector dimensions include initial bonding strength, rate of change, and stable value. These feature vectors constitute data points in a multi-dimensional space for clustering analysis.

[0076] The K-means clustering algorithm is used to process data with different component ratios. As a distance-based unsupervised learning method, this algorithm minimizes the intra-cluster variance by iteratively calculating the Euclidean distance from data points to cluster centers. Chloride, sulfate, and silicate concentrations are used as feature vectors to form a set of points in a multi-dimensional space. During algorithm initialization, K center points are randomly selected, and their positions are iteratively adjusted until the intra-cluster variance is minimized. For example, in a fly ash treatment scenario, K=3 corresponds to high, medium, and low bonding characteristics. The algorithm clusters data points with higher chloride concentrations and faster increases in bonding strength into one cluster, and data points dominated by silicate and with stable bonding strength into another. Calculating the distance between the feature vectors of data points and cluster centers is a crucial step in the clustering process. The distance formula is: ,in x i Represents the feature values ​​of data points. c i The coordinates of the cluster center are represented. For a fly ash mixture with a chloride concentration of 15%, sulfate concentration of 20%, and silicate concentration of 65%, its bonding characteristics, such as bonding strength, are calculated and compared with other mixtures to identify high-bonding and low-bonding groups. At an incineration temperature of 850℃, an increase in chloride concentration tends to enhance bonding characteristics. Clustering identifies this type of mixture as belonging to the easily agglomerated category. For the dominant mixture with a silicate concentration of 70%, the clustering algorithm will classify these data points into the low-bonding cluster, because silicate inhibits the synergistic effect of chloride and sulfate, reducing the probability of agglomeration.

[0077] To group components with similar bonding behaviors into the same cluster, it is necessary to calculate the similarity between feature vectors. The cosine similarity formula is used, which is the dot product of two vectors divided by the product of their moduli. Component combinations with a similarity higher than 0.8 are grouped into the same cluster. For example, components with high chloride concentration and similar sulfate concentration are grouped into one cluster, so that the bonding behavior of components within the same cluster remains consistent.

[0078] Based on the clusters, to determine the current fly ash state's binding characteristic category, it is necessary to compare the current fly ash composition concentration data with the distance to the center point of each cluster, and select the closest cluster as the category. For example, when the current chloride concentration is 18%, the distance to the center of the high-binding cluster is calculated to be 0.5, and the distance to the low-binding cluster is 1.2, thus determining it as a high-binding category. In real-time monitoring with a temperature parameter of 900℃, the current state can be determined as a medium-binding category. For multiple clusters, a majority voting mechanism is used to confirm the category to ensure the accuracy of the judgment.

[0079] To generate the classification prediction results of the bonding performance, the obtained category, such as high bonding, medium bonding or low bonding, needs to be output and the timestamp recorded to form a traceable log for subsequent correlation analysis. For example, the high bonding prediction result may indicate that the temperature needs to be reduced by 5°C to regulate and stabilize the bonding performance.

[0080] Step S105: By analyzing the correlation between the classification prediction results and historical process parameters, establish the mapping relationship between component synergy effects and process parameters, and determine the control strategy parameters;

[0081] In one specific embodiment, the process of executing step S105 may specifically include the following steps:

[0082] Obtain historical process parameter data;

[0083] The classification prediction results are correlated with the historical process parameter data;

[0084] Establish the mapping relationship between component synergy and optimal process parameters;

[0085] Determine whether the current component interaction intensity exceeds the preset range;

[0086] If the value exceeds the preset range, the system will automatically match the corresponding combination of control strategy parameters to determine the parameter values ​​for temperature adjustment range and processing time.

[0087] Specifically, historical process parameter data is extracted from the incinerator system database, including temperature setpoints, processing time records, and corresponding fly ash composition measurement results. The data is stored in order of timestamps to ensure temporal consistency. The correlation analysis uses the Pearson correlation coefficient calculation method, as shown in formula (1), with the adhesion characteristic category in the classification prediction results as the independent variable and the historical process parameters as the dependent variable.

[0088] (1)

[0089] In the formula x i and y i These represent the adhesion category code and temperature value, respectively. The linear correlation strength is obtained by calculating the ratio of covariance to standard deviation. When the correlation coefficient between the adhesion characteristic category and the 800-900℃ temperature range reaches 0.75, it indicates a significant positive correlation between high-temperature environments and high adhesion risk. This correlation stems from the intensified synergistic effect of chlorides and sulfates at high temperatures.

[0090] The mapping relationship between component synergy and process parameters is established using a decision tree algorithm. The algorithm input consists of the interaction strength quantification index generated by association analysis and the process parameter dataset. The root node of the decision tree is selected based on the interaction strength feature, and the splitting criterion is maximizing information gain. The information gain is determined by entropy reduction, and the entropy formula is shown in formula (2).

[0091] (2)

[0092] In the formula p i This represents the probability distribution across different process parameter ranges. During training, when nodes with interaction strengths of 0.6-0.8 split into child nodes, the corresponding leaf nodes outputting a temperature of 850℃ and a processing time of 2 hours form mapping rules. For scenarios with a sudden increase in chloride concentration, samples with an interaction strength of 0.7 are matched in the decision tree to a branch that lowers the temperature by 50℃ and extends the processing time by 0.5 hours. This rule is derived from valid records of process adjustments under the same interaction strength in historical data. The random forest method further integrates the mapping results of multiple decision trees and improves prediction stability through a voting mechanism. For example, for sulfate-dominated fly ash components with an interaction strength of 0.5, the optimized parameters of raising the temperature by 40℃ and shortening the processing time by 0.3 hours are determined by combining the outputs of multiple trees.

[0093] The process for determining whether the current interaction intensity exceeds the preset range of 0.4-0.6 involves directly reading the real-time calculated interaction intensity index for threshold comparison. When the incinerator outlet sensor detects that the chloride concentration has risen to 18%, causing the interaction intensity to reach 0.7, the system determines that it is out of range and triggers parameter matching. A lookup table method retrieves the control strategy corresponding to an interaction intensity of 0.65 from the mapping relationship, outputting a temperature adjustment range of -30℃ and a processing time increment of 0.5 hours. The parameter combination is generated between discrete mapping data through linear interpolation. The temperature adjustment range is obtained by subtracting the mapping target value from the current temperature, for example, adjusting from 900℃ to 870℃. The processing time directly adopts the mapping suggestion value of 1.5 hours. This dynamic matching mechanism solves the control lag problem caused by composition fluctuations in traditional fly ash treatment. By real-time correlation of component synergy effects and process parameters, it effectively suppresses the tendency of chloride and sulfate to bond stronger at high temperatures.

[0094] Step S106: Adjust the process flow according to the control strategy parameters to form a closed-loop feedback control system and obtain stable fly ash adhesion performance.

[0095] In one specific embodiment, the process of executing step S106 may specifically include the following steps:

[0096] The incineration temperature and fly ash treatment process are adjusted in real time according to the control strategy parameters.

[0097] Continuous monitoring of changes in the chemical composition content of fly ash after regulation;

[0098] Based on the aforementioned change data, analyze the response trend of the bonding properties after regulation.

[0099] By adjusting the process parameters based on the aforementioned response trend, a closed-loop feedback control system is formed, resulting in stable fly ash adhesion performance.

[0100] Specifically, in the fly ash caking performance control method, a closed-loop feedback control system is used to dynamically adjust process parameters to address the unstable caking performance caused by fluctuations in fly ash composition during incineration. After the control strategy parameters are input to the incinerator control system, the temperature regulation module adjusts the heating power in real time according to the preset amplitude value. For example, when the strategy parameter requires the temperature to drop from 900°C to 850°C, the thermocouple sensor and PID controller work together to complete the temperature transition within 30 seconds. At the same time, the processing time parameter is synchronously updated to the timing controller of the fly ash conveying system, extending the processing time by 0.5 hours to ensure sufficient reaction of components. Immediately after the parameter adjustment, the sensor array starts a high-frequency monitoring mode, collecting concentration data of chloride, sulfate, and silicate in the fly ash at a frequency of once per second. The data stream is then transmitted to the central processing unit via IoT nodes to form a time series change curve.

[0101] After preprocessing, the monitoring data is input into the real-time analysis model. This model uses a sliding window mechanism to process time-series data, with a window size of 10 consecutive sampling points, calculating the mean and standard deviation of the concentration of each component within the window. When the chloride concentration shows an upward trend within three consecutive windows and the standard deviation exceeds the threshold of 0.5 mg / kg, the model determines that the interaction strength is increasing. At this point, the adhesion characteristic prediction module initiates ARIMA time series analysis, predicting the direction of change of the adhesion index within the next two minutes based on historical 5-minute data. The prediction results, along with the current process parameters, are input into the feedback control algorithm. The algorithm calculates a new temperature adjustment by comparing the deviation between the predicted adhesion index and the target range. For example, if the prediction shows that the adhesion index will exceed the safety threshold in 2 minutes, the algorithm generates a secondary cooling command, reducing the temperature by 20°C from the original 850°C.

[0102] During the feedback control process, environmental parameter sensors continuously record pressure and humidity data within the incinerator. This data serves as a correction factor in the calculation of control parameters. When the pressure sensor detects a 0.2 atmosphere increase in furnace pressure, the humidity data simultaneously increases by 5%. The control algorithm adjusts the sensitivity coefficient of temperature regulation accordingly to avoid over-adjustment due to sudden environmental changes. After each parameter adjustment, fly ash samples are sent to an online analyzer, where X-ray diffractometers determine their mineral phase composition in real time. The data is fed back to the cluster analysis module to re-verify the bonding category. When the proportion of silicate crystalline phase is detected to increase to 65%, the system automatically shortens the processing time by 0.2 hours to balance the control effect.

[0103] The closed-loop system's data flow exhibits multi-layered linkage characteristics: raw concentration data is processed through a sliding window to generate trend indicators, which drive time series predictions. The prediction results, together with environmental correction parameters, generate control commands, and the effectiveness of these commands is verified by a new round of monitoring data. This structure allows incineration temperature and processing time to adaptively adjust according to the dynamic changes in fly ash composition. For example, when the chloride concentration suddenly rises from 15% to 18%, the system completes the entire "monitoring-prediction-control" process within 45 seconds, keeping the interaction intensity within a safe range of 0.4-0.6. Historical adjustment records of process parameters are stored in an optimization database for incremental learning training of the decision tree model, gradually improving the accuracy of the mapping relationship.

[0104] The above describes the fly ash treatment and monitoring method based on the Internet of Things in the embodiments of this application. See also... Figure 2 The following describes the fly ash treatment monitoring process based on the Internet of Things in the embodiments of this application:

[0105] Figure 2 The diagram illustrates the complete workflow of fly ash treatment and monitoring, with each stage, from data acquisition to closed-loop control, tightly integrated. The system first acquires data in real time through sensor arrays deployed at key nodes of the incinerator. Chloride-selective electrodes, sulfate ion sensors, and silicate analyzers simultaneously measure the concentrations of chloride, sulfate, and silicate in the fly ash at a frequency of once per second, with measurement ranges of 50-80 mg / kg, 30-50 mg / kg, and 200-250 mg / kg, respectively. Simultaneously, thermocouple sensors monitor temperature changes at 800-900℃, pressure sensors record operating pressure data at 1.0-1.2 atmospheres, and humidity sensors collect ambient humidity data at 20%-40%. This raw data is processed using a preset calibration algorithm, employing a correction equation constructed using the least squares method (e.g., ...). y =0.95 xThe sensor readings are corrected by +0.02 to ensure that the data error is controlled within 1%. The corrected data is then aligned by timestamp to form a dataset containing multi-dimensional features such as concentration gradients and temperature gradients, providing a foundation for subsequent analysis.

[0106] After data preprocessing, the system enters the interaction intensity modeling stage. Based on temperature thresholds, the system divides the operating conditions into high-temperature (>850℃), low-temperature (<750℃), and normal-temperature (750-850℃) zones. Simultaneously, it distinguishes between fluctuating and stable zones based on a 10% concentration fluctuation threshold. Independent multivariate regression sub-models are established for different regions. For example, the sub-model for the high-temperature fluctuating zone includes an interaction term between chloride concentration and temperature. The least squares method is used to fit the interaction intensity calculation formula for specific operating conditions. The system monitors the chloride to sulfate concentration ratio in real time. When the ratio exceeds a preset threshold of 1.5, a synergistic effect calculation program is triggered. A sliding window mechanism (window size of 5 time points) is used to calculate the average interaction intensity within the window, and this is weighted and corrected using temperature parameters to ultimately generate a quantified interaction intensity index.

[0107] The time series analysis module receives the interaction intensity index as input. It first performs stationarity checks and necessary differencing on the data, then uses a time series analysis model for analysis. By using a sliding window mechanism with a window size of 10, the system can track the changing trend of interaction intensity in real time. When the average interaction intensity of three consecutive windows shows an upward trend with a slope exceeding 0.05 units / minute, it is determined to be an upward trend, and the prediction program is initiated. The Holt-Winters model combines historical data and current trends to predict the direction of change in adhesion performance within the next 5 minutes, generating a specific description of the dynamic change pattern, providing a basis for subsequent decision-making.

[0108] In the adhesive property classification stage, the system combines time series analysis results with component ratio information to construct a multi-dimensional vector containing features such as initial adhesive strength and rate of change. The K-means clustering algorithm (K=3) is used to analyze these feature vectors, grouping components with similar adhesive behaviors into the same cluster by calculating Euclidean distance. Cosine similarity (threshold 0.8) is used during classification to ensure consistency of component combinations within the same cluster, and a majority voting mechanism is used to handle multi-cluster scenarios, ultimately outputting classification results for high, medium, and low adhesive properties.

[0109] The parameter control module performs correlation analysis between the classification results and historical process parameters, determining the mapping relationship between various bonding properties and process parameters through Pearson correlation coefficient calculation. The decision tree algorithm uses interaction intensity as the node splitting condition to establish specific control strategies; for example, an interaction intensity of 0.7 corresponds to a 50°C temperature reduction and a 0.5-hour extension of the processing time. The system monitors the current interaction intensity in real time, and when a value exceeds the preset range (0.4-0.6), it automatically matches the corresponding control parameters and executes adjustments.

[0110] In the closed-loop control phase, a PID controller precisely adjusts process parameters, completing temperature changes within 30 seconds while simultaneously coordinating with the conveying system to adjust processing time. The adjusted fly ash composition data immediately enters a new round of monitoring, with X-ray diffractometer analysis of mineral phase composition changes in real time. The system continuously tracks the control effect through a sliding window mechanism, initiating secondary adjustments when parameters deviate from expectations, forming a complete monitoring-prediction-control closed loop. All process adjustment records are stored in an optimization database for continuous improvement of the decision tree model's mapping relationship, ensuring the system can adapt to various operating conditions. The entire process achieves real-time monitoring and adaptive adjustment of the fly ash treatment process, effectively maintaining the stability of its bonding performance.

[0111] The above describes the fly ash treatment monitoring process based on the Internet of Things (IoT) in the embodiments of this application. The following describes the fly ash treatment monitoring system based on the IoT in the embodiments of this application. Please refer to [link / reference]. Figure 3 One embodiment of the fly ash treatment monitoring system based on the Internet of Things in this application includes:

[0112] Data acquisition module 301: Collects the concentration of components such as chloride in fly ash, as well as environmental parameters such as temperature, pressure and humidity, according to the preset frequency of the sensor array, to form time series information. After being corrected by the calibration algorithm, it generates a multi-dimensional component concentration gradient distribution dataset.

[0113] Interaction intensity modeling module 302: Based on the multidimensional component concentration gradient distribution dataset, multivariate regression modeling is used to monitor the chlorine-sulfur concentration ratio in real time, and the synergistic effect intensity is calculated when the threshold is exceeded, generating a quantitative index of dynamic interaction intensity between components.

[0114] Adhesion performance prediction module 303: Input the interaction strength index into the time series model, analyze the trend through the sliding window, predict the evolution of adhesion properties when the interaction strength increases, and generate its dynamic change law.

[0115] Adhesion property classification module 304: Based on the adhesion performance rules, cluster analysis is used to classify the components with similar adhesion behavior into the same cluster, determine the current category, and generate classification prediction results.

[0116] Parameter control module 305: Associates classification results with historical process parameters to establish a mapping relationship. When the component interaction intensity exceeds the range, it matches control parameters to determine the temperature adjustment range and processing time.

[0117] Closed-loop control module 306: Adjusts the incineration temperature and process flow according to the control parameters, monitors the changes in composition after control, analyzes the response trend of bonding characteristics, adjusts parameters to form a closed loop, and obtains stable performance.

[0118] above Figure 3 The IoT-based fly ash treatment monitoring system in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The IoT-based fly ash treatment monitoring device in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0119] Reference Figure 4 This invention also provides an IoT-based fly ash processing monitoring device, which can be a server, and its internal structure can be as follows: Figure 4 As shown, the IoT-based fly ash processing monitoring device includes a processor 402, a memory 403, a display screen 404, an input device 405, a network interface 406, and a database 407 connected via a system bus 401. The processor, designed as a computer, provides computing and control capabilities. The memory of the IoT-based fly ash processing monitoring device includes a non-volatile storage medium 4031 and internal memory 4032. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database 407 of the IoT-based fly ash processing monitoring device stores the corresponding data in this embodiment. The network interface 406 of the IoT-based fly ash processing monitoring device is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0120] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the IoT-based fly ash processing monitoring device to which the present invention is applied. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0121] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fly ash treatment and monitoring method based on the Internet of Things, characterized in that, include: Step S101: Obtain concentration data of multiple chemical components in fly ash and environmental condition data through a sensor array to obtain a component concentration gradient distribution dataset; Step S102: Based on the component concentration gradient distribution dataset, a multivariate regression analysis algorithm is used to establish a calculation model for the interaction intensity between chemical components, and a quantitative index of interaction intensity is obtained. The interaction intensity calculation model is fitted with the concentrations of chloride, sulfate, and silicate and the interaction term as independent variables. The quantitative index of interaction intensity is generated based on the real-time monitored chemical component concentration ratio combined with temperature parameter weights and silicate concentration correction. Step S103: Analyze the interaction strength quantification index using a time series analysis model to predict the dynamic change law of bonding performance; Step S104: Based on the dynamic change law of the bonding performance, a clustering analysis algorithm is used to classify the bonding characteristics and obtain the classification prediction results. The clustering analysis algorithm uses the initial bonding strength, the rate of change and the stable value as feature vectors to classify the bonding characteristics into high bonding, medium bonding or low bonding categories. Step S105: Through the correlation analysis between the classification prediction results and historical process parameters, a decision tree algorithm is used to establish the mapping relationship between the component synergy effect and the process parameters, and to determine the control strategy parameters. The historical process parameters include temperature setpoint, processing time records and corresponding fly ash component measurement results. Step S106: Adjust the process flow according to the control strategy parameters to form a closed-loop feedback control system and obtain stable fly ash adhesion performance.

2. The method as described in claim 1, characterized in that, The acquisition of concentration data of multiple chemical components in fly ash and environmental condition data through a sensor array includes: The sensor array synchronously collects concentration data of chloride, sulfate and silicate in fly ash at a preset monitoring frequency, and records environmental parameters such as temperature, pressure and humidity to form time series change information; The concentration data and environmental parameters are corrected using a preset calibration algorithm; Based on the corrected concentration data, environmental parameters, and their time-series changes, a multi-dimensional component concentration gradient distribution dataset is generated.

3. The method as described in claim 1, characterized in that, The step involves establishing a calculation model for the interaction intensity between chemical components using a multiple regression analysis algorithm based on the component concentration gradient distribution dataset, to obtain quantitative indicators of interaction intensity, including: Based on the component concentration gradient distribution dataset, a multivariate regression analysis algorithm is used to establish a calculation model for the interaction intensity between various chemical components. The concentration ratio of chloride to sulfate is monitored in real time. When the concentration ratio exceeds a preset threshold, the synergistic effect strength is calculated in real time. Based on the real-time calculation results of the synergistic effect intensity, a quantitative index of the dynamic interaction intensity between components is generated.

4. The method as described in claim 3, characterized in that, The method of establishing a calculation model for the interaction strength between various chemical components using a multiple regression analysis algorithm includes: Temperature zones are divided according to the set temperature thresholds. Areas with temperatures above the first temperature threshold are classified as high-temperature zones, areas with temperatures below the second temperature threshold are classified as low-temperature zones, and areas with temperatures between the first and second temperature thresholds are classified as normal-temperature zones. Concentration zones are divided according to the set concentration thresholds. Zones where the concentration fluctuation of a component is higher than the concentration threshold are classified as fluctuation zones, and zones where the concentration fluctuation of a component is lower than the concentration threshold are classified as flat zones. For the high temperature zone, low temperature zone, normal temperature zone, fluctuating zone, and flat zone, an independent sub-model is used for piecewise fitting to obtain the interaction intensity calculation model.

5. The method as described in claim 1, characterized in that, The step of analyzing the interaction strength quantification index using a time series analysis model to predict the dynamic change law of bonding performance includes: Input the interaction intensity quantification index into the time series analysis model; The changing trend of component synergistic effect at different time points was analyzed using a sliding window mechanism; Determine whether the trend of change indicates an upward trend in interaction intensity; If the interaction strength shows an upward trend, the evolution direction of the bonding properties in the future period can be predicted, and the dynamic change law of bonding performance can be generated.

6. The method as described in claim 1, characterized in that, The step of classifying the bonding characteristics based on the dynamic change law of the bonding performance using a clustering analysis algorithm to obtain the classification prediction results includes: Based on the dynamic change law of the bonding performance, a clustering analysis algorithm is used to classify and identify the bonding characteristics under different component ratios; Combinations with similar bonding behaviors are grouped into the same cluster; Based on the clusters, determine the current fly ash state to which the bonding characteristic category belongs; Generate classification prediction results for the bonding properties.

7. The method as described in claim 1, characterized in that, The process involves analyzing the correlation between the classification prediction results and historical process parameters, using a decision tree algorithm to establish a mapping relationship between component synergistic effects and process parameters, and determining control strategy parameters, including: Obtain historical process parameter data; The classification prediction results are correlated with the historical process parameter data; Establish the mapping relationship between component synergy and optimal process parameters; Determine whether the current component interaction intensity exceeds the preset range; If the value exceeds the preset range, the system will automatically match the corresponding combination of control strategy parameters to determine the parameter values ​​for temperature adjustment range and processing time.

8. The method as described in claim 1, characterized in that, The process flow is adjusted according to the control strategy parameters to form a closed-loop feedback control system, thereby obtaining stable fly ash adhesion performance, including: The incineration temperature and fly ash treatment process are adjusted in real time according to the control strategy parameters. Continuous monitoring of changes in the chemical composition content of fly ash after regulation; Based on the aforementioned change data, analyze the response trend of the bonding properties after regulation. By adjusting the process parameters based on the aforementioned response trend, a closed-loop feedback control system is formed, resulting in stable fly ash adhesion performance.

9. A fly ash treatment monitoring system based on the Internet of Things, characterized in that, For implementing the IoT-based fly ash treatment monitoring method as described in any one of claims 1-8, the IoT-based fly ash treatment monitoring system comprises: The data acquisition module is used to acquire concentration data of various chemical components in fly ash and environmental condition data through a sensor array, and obtain a dataset of component concentration gradient distribution. The interaction intensity modeling module is used to establish a calculation model of the interaction intensity between chemical components based on the component concentration gradient distribution dataset and a multivariate regression analysis algorithm to obtain a quantitative index of interaction intensity. The interaction intensity calculation model is fitted with the concentrations of chloride, sulfate, and silicate and the interaction term as independent variables. The quantitative index of interaction intensity is generated based on the real-time monitored chemical component concentration ratio combined with temperature parameter weights and silicate concentration correction. The bonding performance prediction module is used to analyze the interaction strength quantification index through a time series analysis model to predict the dynamic change law of bonding performance. The bonding property classification module is used to classify the bonding properties according to the dynamic change law of the bonding performance using a clustering analysis algorithm to obtain the classification prediction result. The clustering analysis algorithm uses the initial bonding strength, the rate of change and the stable value as feature vectors to classify the bonding properties into high bonding, medium bonding or low bonding categories. The parameter control module is used to establish a mapping relationship between the component synergy effect and the process parameters by using a decision tree algorithm through the correlation analysis between the classification prediction results and historical process parameters, and to determine the control strategy parameters. The historical process parameters include temperature setpoints, processing time records, and corresponding fly ash component measurement results. The closed-loop control module is used to adjust the process flow according to the control strategy parameters to form a closed-loop feedback control system and obtain stable fly ash adhesion performance.

10. A fly ash treatment monitoring device based on the Internet of Things, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the fly ash processing monitoring method based on any one of claims 1 to 8.

Images