Intelligent evaluation and analysis method for heat exchange performance of water-cooled furnace

Abstract

This application provides an intelligent evaluation and analysis method for the heat transfer performance of a water-cooled furnace, comprising: collecting measured values ​​of the thickness of the cast-in-place material layer, the surface temperature of the front and rear arch water-cooled walls, the cumulative thickness of the slag layer, and the heat flux density from the water-cooled furnace wall detection system; quantifying the initial heat transfer resistance distribution of the two material layers, the cast-in-place material layer and the slag layer; evaluating the variation trajectory of the cast-in-place material layer thickness measurement value and the surface temperature of the front and rear arch water-cooled walls at different times based on the total comprehensive thermal resistance of the cast-in-place material layer and the slag layer; determining the triggering conditions for dynamic adjustment of the measurement point positions; and identifying the optimal distribution position and acquisition time interval of the measurement points on the front and rear arch water-cooled walls based on the triggering conditions for dynamic adjustment of the measurement point positions and the real-time feedback of the water-cooled wall detection system, thereby obtaining a measurement point layout scheme.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to an intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace. Background Technology

[0002] Optimizing the heat exchange performance of water-cooled furnaces is a core area in the design and operation of industrial boilers, and is of great significance for improving energy efficiency and reducing operating costs. In coal-fired boilers, water-cooled walls, as the main heat exchange components, directly affect the thermal efficiency of the furnace and the lifespan of the equipment. Traditionally, designers have optimized heat exchange effects by adjusting the surface materials or structures of water-cooled walls. However, this process involves complex interactions between thermal and material properties, requiring the balancing of multiple factors in the actual operating environment to achieve efficient heat exchange. Existing methods for optimizing the heat exchange performance of water-cooled walls often focus on adjusting a single material or structural parameter, such as reducing the thickness of the castable refractory to lower thermal resistance. This method assumes that thermal resistance is mainly determined by the castable refractory itself, ignoring the dynamically changing operating environment factors. In particular, during boiler operation, ash and slag produced by fuel combustion gradually deposit on the surface of the water-cooled walls, forming a slag layer. This dynamic process significantly alters the heat exchange characteristics, leading to deviations between static design parameters and actual operating results. Existing methods typically cannot accurately capture this dynamic change, making it difficult to achieve precise assessment and optimization of heat exchange performance. The thickness of the castable refractory is a key factor affecting the surface temperature of the water-cooled wall, which in turn directly determines the melting and adhesion behavior of the ash. When the castable refractory is thin, the surface temperature of the water-cooled wall rises rapidly. The high-temperature environment accelerates the melting and adhesion of the ash, leading to the rapid formation of a thick slag layer. Because the slag layer is porous and contains a large number of air gaps, its thermal conductivity is usually only about one-tenth that of the castable refractory. Therefore, for the same thickness, the thermal resistance of the slag layer is much greater than that of the castable refractory itself, which significantly reduces the heat exchange effect. This interaction between thickness and slag formation rate leads to a contradiction between the traditional design approach of pursuing thin castable refractory and the actual decrease in heat exchange performance during operation. Therefore, in actual boiler operation, accurately assessing the dynamic impact of changes in castable refractory thickness on the surface temperature of the water-cooled wall and the accumulation of the slag layer becomes a key issue for optimizing heat exchange performance. For example, in a coal-fired boiler, when designers reduce the thickness of the water-cooled wall refractory from 10 mm to 5 mm, the initial heat exchange efficiency improves, but severe slagging is induced on the high-temperature surface within hours, leading to a rapid increase in thermal resistance and a decrease in overall heat exchange capacity. This dynamic contradiction makes finding a suitable balance between refractory thickness and slagging control under different operating conditions a key issue in the intelligent evaluation of heat exchange performance. Summary of the Invention

[0003] This invention provides an intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace, enabling precise monitoring and performance evaluation of slagging on the furnace wall and changes in the state of the cast-in-place material layer. The method mainly includes:

[0004] The initial heat transfer resistance distribution of the two layers of materials, namely the cast material layer and the slag layer, was obtained by collecting the thickness of the cast material layer, the surface temperature of the front and rear arch water-cooled walls, the cumulative thickness of the slag layer, and the heat flux density from the water-cooled furnace wall inspection.

[0005] Based on the initial heat transfer resistance distribution, a sensitivity analysis of the surface temperature of the arch water-cooled wall is performed to evaluate the changes in surface temperature rise and ash melting and adhesion when the thickness of the cast material layer is reduced, and to obtain the total comprehensive thermal resistance of the cast material layer and the slag layer.

[0006] Based on the total comprehensive thermal resistance, evaluate the changes in the thickness of the cast-in-place material layer and the surface temperature of the arch water-cooled wall at different times, and determine the triggering conditions for dynamic adjustment of the measurement point position.

[0007] Based on the triggering conditions for the dynamic adjustment of the measurement point positions and the real-time feedback from the water-cooled wall detection system, the optimal distribution positions and acquisition time intervals of the measurement points on the front and rear arch water-cooled walls are identified, and a measurement point layout scheme is obtained.

[0008] The acquisition frequency and acquisition position of the thickness measuring device and temperature monitor are updated by the measurement point layout scheme. The response relationship between the surface temperature rise rate and the slagging rate of the arch water-cooled wall is evaluated based on the cumulative thickness growth rate of the slagging layer. The total combined thermal resistance of the cast material layer and the slagging layer after adjustment is determined.

[0009] Based on the adjusted total thermal resistance, the heat exchange capacity evolution process corresponding to different castable layer thickness measurements is integrated. The change trajectory of the castable layer thickness measurement and the cumulative thickness of the slag layer is continuously collected to obtain the performance evaluation results of the arch water-cooled wall before and after heat exchange under different castable layer thickness conditions.

[0010] Furthermore, the acquisition of measured values ​​of the cast-in-place material layer thickness, surface temperatures of the front and rear arch water-cooled walls, cumulative thickness of the slag layer, and heat flux density from the water-cooled furnace wall inspection, to obtain the initial heat transfer resistance distribution of the two material layers, the cast-in-place material layer and the slag layer, includes:

[0011] The thickness of the cast lining layer was measured at multiple measuring points on the water-cooled furnace wall using an ultrasonic thickness gauge. The surface temperature data of the arch water-cooled wall before and after heat exchange was obtained using an infrared thermal imager. The cumulative thickness of the slag layer was calculated based on the ash deposition pattern. The heat flux density at each measuring point was measured using a furnace heat flow meter to obtain the basic data of the measuring points, which includes four types of parameters: thickness, temperature, slag formation, and heat flux.

[0012] Based on the basic data of the measuring points, which includes four types of parameters: thickness, temperature, slagging, and heat flow, the thermal resistance of the castable material is calculated according to the thermal conductivity of the castable material and the measured thickness of the castable material layer. The thermal resistance of the slagging layer is calculated according to the thermal conductivity of the slagging layer and the cumulative thickness of the slagging layer. The thermal resistance of the castable material and the thermal resistance of the slagging layer are added together by the principle of thermal resistance superposition to obtain the initial heat transfer resistance distribution at each measuring point.

[0013] Furthermore, the sensitivity analysis of the surface temperature of the arch water-cooled wall based on the initial heat transfer resistance distribution evaluates the changes in surface temperature rise and ash molten adhesion as the thickness of the cast-in-place layer decreases, obtaining the total combined thermal resistance of the cast-in-place layer and the slag layer, including:

[0014] The surface temperature gradient of the water-cooled wall under different pouring material layer thicknesses is calculated based on the initial heat transfer resistance distribution. The temperature difference between the ash softening point and the surface temperature rise rate is obtained to determine the adhesion acceleration time node when the temperature is close to the ash softening point.

[0015] For the adhesion acceleration time node when the temperature is close to the softening point of the ash, based on the difference between the thermal conductivity of the slag layer and the thermal conductivity of the castable, the thermal resistance increment per unit thickness of the slag layer is calculated through the linear relationship between thermal resistance and thickness, and the growth rate function of the total thermal resistance as a function of the cumulative thickness of the slag layer is obtained.

[0016] By using the growth rate function, a dynamic superposition relationship between the thermal resistance of the cast-in-place material layer and the thermal resistance of the slag layer is established. Based on the temperature distribution difference between the front arch area and the rear arch area, a weighting coefficient is set according to the temperature difference ratio to determine the total comprehensive thermal resistance for different operating stages.

[0017] Furthermore, based on the total comprehensive thermal resistance, the evaluation of the measured thickness of the cast-in-place material layer and the surface temperature of the arch water-cooled wall at different times, and the determination of the triggering conditions for dynamic adjustment of the measurement point position, include:

[0018] Based on the time series data of total thermal resistance, the hourly heat flux density numerical sequence and the corresponding slag layer thickness increment sequence are extracted. The Pearson correlation coefficient between the two hourly heat flux density numerical sequences and the corresponding slag layer thickness increment sequence is calculated as the synchronization coefficient. When the synchronization coefficient exceeds the synchronization coefficient threshold, it is determined that the rapid slag accumulation stage has been entered.

[0019] For the rapid accumulation stage of slag, the rate of change of heat flux density is detected hourly to identify the inflection point of heat flux density decrease, and the measured value of the thickness of the cast material layer and the surface temperature data of the arch before and after heat exchange are recorded at the inflection point.

[0020] Using the measured thickness of the poured material layer at the inflection point and the surface temperature data of the arch before and after heat exchange, a temperature change trajectory curve over time is plotted. When the slope of the trajectory curve or the temperature difference between the front and rear arches exceeds a set threshold, the trigger condition for dynamic adjustment of the measurement point position is determined.

[0021] Furthermore, the triggering conditions for the dynamic adjustment of the measurement point positions and the real-time feedback from the water-cooled wall detection system identify the optimal distribution positions and acquisition time intervals of the measurement points on the front and rear arch water-cooled walls, resulting in a measurement point layout scheme, including:

[0022] Based on the trigger condition signal dynamically adjusted according to the location of the measurement point, the real-time temperature gradient and heat flux density decay rate of each measurement point of the arch before and after heat exchange are obtained from the water-cooled wall detection device. The measurement point data are spatially grouped according to the magnitude of the temperature gradient by a clustering algorithm, the regional boundaries are identified, and the preliminary distribution location of the measurement points of the arch water-cooled wall is determined.

[0023] Based on the initial distribution of the measurement points of the arch water-cooled wall, the growth rate is obtained by calculating the difference in the time series data of the slag layer thickness at each measurement point. According to the magnitude of the growth rate, the area is divided into a fast slag zone, a medium-speed slag zone, and a slow slag zone. The key evolution region of heat exchange capacity is determined according to the type of slag zone.

[0024] Based on the spatial distribution of the key evolution regions of the heat exchange capacity, the sampling time intervals for the rapid slagging zone, the medium-speed slagging zone, and the slow slagging zone are set, and a layout matrix containing the correspondence between the measurement point location and the sampling frequency is constructed.

[0025] Based on the acquisition frequency of each measurement point in the layout matrix, priority identifiers are assigned according to regional priority. The measurement point coordinates, acquisition frequency, and priority identifiers are integrated to obtain the measurement point layout scheme that includes acquisition frequency and acquisition location.

[0026] Furthermore, the step of updating the acquisition frequency and acquisition position of the thickness measuring device and temperature monitor through the measurement point layout scheme, evaluating the response relationship between the surface temperature rise rate and the slagging rate of the arch water-cooled wall based on the cumulative thickness growth rate of the slagging layer, and determining the adjusted total combined thermal resistance of the cast-in-place layer and the slagging layer includes:

[0027] Based on the acquisition frequency and acquisition position parameters in the measurement point layout scheme, the ultrasonic probe scanning cycle of the thickness measurement device is adjusted, and the infrared sensor of the temperature monitor is driven to move to the corresponding coordinate point to obtain the real-time thickness data and temperature data of each measurement point after adjustment.

[0028] Using the real-time thickness data and the temperature data, the cumulative thickness growth rate of the slag layer and the surface temperature rise rate of the front and rear arch water-cooled walls are calculated, and a linear regression relationship is established.

[0029] The thermal resistance calculation weights are adjusted according to the slope coefficient of the linear regression relationship. The thermal resistance of the cast-in-place layer and the thermal resistance of the slag layer are calculated, and the total adjusted comprehensive thermal resistance is determined by weighted summation.

[0030] Furthermore, after updating the acquisition frequency and acquisition position of the thickness measuring device and temperature monitor through the measurement point layout scheme, the real-time thickness data and temperature data of each measuring point after the update are obtained. The first derivative of the cumulative thickness of the slag layer with respect to time is calculated as the growth rate, and the first derivative of the surface temperature of the arch water-cooled wall before and after heat exchange with respect to time is calculated as the temperature rise rate. The least squares method is used to fit the two sequences of the surface temperature change rate of the arch water-cooled wall before and after heat exchange to establish a linear regression relationship.

[0031] Furthermore, based on the adjusted total comprehensive thermal resistance, the heat transfer capacity evolution process corresponding to different castable layer thickness measurements is integrated, and the change trajectory of the castable layer thickness measurement and the cumulative thickness of the slag layer is continuously collected to obtain the performance evaluation results of the arch water-cooled wall before and after heat transfer under different castable layer thickness conditions, including:

[0032] Based on the adjusted total thermal resistance, the thermal resistance values ​​corresponding to different pouring material layer thicknesses are calculated. Using the collected pouring material layer thickness measurements and the data on the change of the cumulative thickness of the slag layer over time, a heat exchange capacity evolution curve is constructed.

[0033] Based on the heat transfer capacity evolution curve, the time variation law of thermal resistance value under different lining thicknesses is compared to obtain the heat transfer performance evaluation results of the arch water-cooled wall under different lining thicknesses.

[0034] Furthermore, the heat transfer performance evaluation results of the arch water-cooled wall under different refractory layer thicknesses are obtained, including:

[0035] Extract the thermal resistance values ​​corresponding to different pouring material layer thicknesses from the adjusted total comprehensive thermal resistance;

[0036] Continuously collect data on the changes in the thickness of the poured material layer and the cumulative thickness of the slag layer over time;

[0037] Construct a heat exchange capacity evolution curve based on the aforementioned change data;

[0038] Based on the heat transfer capacity evolution curve, the thermal resistance variation law under different pouring material layer thicknesses was analyzed;

[0039] Based on the variation law of thermal resistance under different pouring material layer thicknesses, the heat exchange performance evaluation results of the water-cooled wall under different conditions are generated, and a comprehensive performance report is formed.

[0040] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0041] This invention discloses an intelligent evaluation and analysis method for the heat transfer performance of water-cooled furnaces. Addressing the problem of decreased heat transfer capacity due to slagging on the surface of the water-cooled wall and variations in the thickness of the cast-in-place material layer, the method quantifies the initial heat transfer resistance distribution of the cast-in-place material layer and the slagging layer using the principle of thermal resistance superposition. Combined with surface temperature sensitivity analysis and heat flux density variation characteristics, it assesses the response relationship between slagging accumulation and thermal resistance growth, dynamically adjusting the layout of measurement points and the acquisition frequency. By continuously monitoring the trajectory of thickness and temperature changes and optimizing the location and time interval of measurement points, this invention accurately identifies key nodes in the accelerated formation of slagging, ultimately achieving a comprehensive evaluation of the heat transfer performance of the front and rear arch water-cooled walls. Its core technological advantage lies in effectively improving the monitoring accuracy and response speed of the water-cooled wall's operating status through dynamic updating of the total thermal resistance and analysis of the evolution of heat transfer capacity, providing a scientific basis for furnace wall protection and operational optimization.

[0042] Furthermore, by utilizing the principle of thermal resistance superposition, the influence of the cast-in-place material layer and the slag layer is uniformly quantified into the core indicator of "total comprehensive thermal resistance," overcoming the limitations of single-parameter evaluation. Combined with surface temperature change gradient analysis and heat flux density inflection point identification technology, it can keenly capture the critical point of accelerated slag formation, thereby achieving early warning and precise quantification of the heat transfer performance degradation process, significantly improving the real-time performance and accuracy of condition monitoring.

[0043] Based on data change trajectories, such as temperature gradients and heat flux density decay rates, intelligent triggering of measurement point adjustments is achieved. Clustering algorithms are used to automatically classify areas with abnormal temperatures and areas with severe slagging, enabling focused and adaptive tracking of key areas. By dynamically adjusting the spatial distribution of measurement points and the frequency of data acquisition, monitoring resources are prioritized for critical areas of performance degradation, realizing a shift from "uniform distribution" to "on-demand focusing," improving resource utilization efficiency while ensuring monitoring effectiveness.

[0044] By continuously collecting and adjusting the data and cyclically updating the total thermal resistance, it is possible to construct evolution curves of heat transfer capacity under different refractory layer thicknesses and different slagging stages, providing a scientific basis for operation optimization and maintenance decisions. This process reveals the dynamic law of thermal resistance changing with operating conditions. The generated analysis results and comprehensive performance reports can be directly used to guide operating strategies such as soot blowing optimization and load adjustment, as well as the determination of maintenance cycles and scope, providing data-driven decision support for the safe, efficient, and economical operation of the furnace. Attached Figure Description

[0045] Figure 1 This is a flowchart of an intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to the present invention.

[0046] Figure 2 This is a schematic diagram of an intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to the present invention.

[0047] Figure 3 This is another schematic diagram of an intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to the present invention. Detailed Implementation

[0048] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] like Figure 1-3 This embodiment of the intelligent evaluation and analysis method for the heat transfer performance of a water-cooled furnace may specifically include:

[0050] S101. Collect the measured values ​​of the thickness of the cast-in-place material layer, the surface temperature of the front and rear arch water-cooled walls, the cumulative thickness of the slag layer, and the heat flux density from the water-cooled furnace wall detection, and quantify the initial heat transfer resistance distribution of the two layers of materials, the cast-in-place material layer and the slag layer.

[0051] The thickness of the cast-in-place refractory layer was collected from multiple measuring points on the water-cooled furnace wall using an ultrasonic thickness gauge. Infrared thermal imagers were used to acquire surface temperature data of the front and rear arch water-cooled walls. Based on the ash deposition pattern, the cumulative thickness of the slag layer was estimated from the surface temperature. A dedicated in-furnace heat flux meter, equipped with a protective cover and cooling system, was used to measure the heat flux density at each measuring point, resulting in basic measuring point data including four parameters: thickness, temperature, slag formation, and heat flux. Based on this basic measuring point data, the thermal resistance of the cast-in-place refractory layer was calculated using its thermal conductivity and thickness. The thermal resistance of the slag layer was calculated using its thermal conductivity and cumulative thickness. The thermal resistance of the cast-in-place refractory layer and the thermal resistance of the slag layer were then added together using the principle of thermal resistance superposition to obtain the initial heat transfer resistance distribution at each measuring point.

[0052] For example, the measuring points on the water-cooled furnace wall are arranged in a grid pattern, with 8 measuring points in the front arch area and 6 measuring points in the rear arch area, and the spacing between the measuring points is controlled within the range of 500 mm to 800 mm. A non-contact laser thickness gauge is used to calculate the actual thickness of the cast-in-place material layer by emitting laser pulses and calculating the reflection time difference.

[0053] Specifically, the cumulative thickness of the slag layer is determined through a temperature mapping relationship. When the surface temperature of the water-cooled wall exceeds 450 degrees Celsius, the ash begins to soften and adhere; for every 50 degrees Celsius increase, the slag formation rate increases by approximately 30%. By establishing a table corresponding to temperature and slag thickness, the estimated slag layer thickness is obtained directly from the table by looking up the surface temperature values ​​collected by the infrared thermal imager. In the thermal resistance calculation, the thermal conductivity of the castable refractory material ranges from 0.8 to 1.2 W / (m·K), while the thermal conductivity of the slag layer is only 0.08 to 0.15 W / (m·K). Using the principle of thermal resistance superposition, the thermal resistance of the castable refractory material is obtained by dividing its thermal conductivity by the thickness of the castable refractory material layer, and the thermal resistance of the slag layer is obtained by dividing its thermal conductivity by the thickness of the slag layer. The sum of the two is the total heat transfer resistance. The initial heat transfer resistance distribution reflects the difference in heat transfer capacity at different measuring points.

[0054] S102. Based on the sensitivity analysis of the initial heat transfer resistance distribution of the two layers of materials, the casting material layer and the slag layer, to the surface temperature of the front and rear arch water-cooled walls, the change process of the surface temperature rising and the ash melting and adhesion accelerating when the thickness of the casting material layer decreases is evaluated, and the total comprehensive thermal resistance of the casting material layer and the slag layer is obtained.

[0055] Based on the initial heat transfer resistance distribution data, Fourier's law of thermal conductivity is used to calculate the surface temperature gradient of the water-cooled wall under different refractory layer thicknesses. If the refractory layer thickness decreases by more than a preset threshold, the surface temperature rise rate exceeds ten degrees Celsius per hour. The temperature difference between the ash softening point and the surface temperature rise rate is obtained. When the temperature approaches the ash softening point, the adhesion acceleration time node is determined. For the adhesion acceleration time node, the growth curve is extracted from the time-series data of slag layer thickness collected from the water-cooled furnace wall detection. Based on the characteristic that the thermal conductivity of the slag layer is only one-tenth of that of the refractory material, the thermal resistance increment per unit slag layer thickness is calculated through the linear relationship between thermal resistance and thickness, obtaining the growth rate function of the total thermal resistance as a function of the cumulative thickness of the slag layer. Through the growth rate function, a dynamic superposition relationship between the thermal resistance of the refractory layer and the thermal resistance of the slag layer is established. Based on the temperature distribution difference between the front arch area and the rear arch area, a weighting coefficient is set according to the temperature difference ratio to determine the total comprehensive thermal resistance for different operating stages.

[0056] For example, in one embodiment, when Fourier's law of thermal conductivity is applied to the heat transfer calculation of water-cooled walls, the heat flux density is directly proportional to the temperature gradient. When the thickness of the cast-in-place layer is reduced from 12 mm to 8 mm, the thermal resistance decreases by approximately 33%, causing the surface temperature to rise from the original 420 degrees Celsius to 480 degrees Celsius. This temperature gradient is calculated using the thermal conductivity equation q = λΔT / δ, where q is the heat flux density, λ is the thermal conductivity, ΔT is the temperature difference, and δ is the thickness. For every 1 mm reduction in the thickness of the cast-in-place layer, the surface temperature increases at a rate of approximately 15 degrees Celsius per hour. The softening point temperature of ash is a key parameter for determining slagging formation. The aluminosilicate components in coal ash begin to soften at around 550 degrees Celsius. When the surface temperature of the water-cooled wall approaches this temperature, the ash particles begin to melt and become sticky, easily adhering to the wall surface. The adhesion acceleration time point occurs when the temperature difference between the surface temperature and the softening point of the ash is less than 30 degrees Celsius, at which point the ash adhesion rate increases sharply from 0.1 mm per hour to 0.8 mm per hour. By continuously monitoring the change in the temperature difference between the surface and the softening point, the starting moment of rapid slag layer accumulation can be accurately captured.

[0057] Specifically, the thermal resistance growth rate function characterizes the influence of slag layer accumulation on the total thermal resistance. Since the thermal conductivity of the slag layer is only about 0.1 W / (m·K), much smaller than the 1.0 W / (m·K) of the cast-in-place material, the thermal resistance generated by the slag layer at the same thickness is 10 times that of the cast-in-place material. The growth rate function is exponential; initially, for every 1 mm increase in the slag layer thickness, the total thermal resistance increases by 0.01 (m²·K) / W. When the slag layer thickness exceeds 5 mm, the thermal resistance increment reaches 0.015 (m²·K) / W per millimeter. In one embodiment, the thermal resistance growth rate function is expressed as follows:

[0058]

[0059] R total R represents the total thermal resistance. base Let denot , where α represents the thermal resistance, β represents the exponential growth parameter, and δ represents the thickness of the slag layer. This formula describes the exponential growth effect of the slag layer thickness on the total thermal resistance.

[0060] Preferably, the temperature distribution difference between the front and rear arch regions determines the setting of the weighting coefficients. Because the front arch region is closer to the combustion zone, its average temperature is 100 to 150 degrees Celsius higher than that of the rear arch, resulting in a stronger tendency for slagging. Based on the temperature difference ratio, the weighting coefficient for the front arch is set to 0.6, and the weighting coefficient for the rear arch is set to 0.4. Through a weighted average method, the total comprehensive thermal resistance equals the thermal resistance of the front arch multiplied by 0.6 plus the thermal resistance of the rear arch multiplied by 0.4, achieving an accurate assessment of the overall heat transfer performance of the furnace.

[0061] S103. Based on the total combined thermal resistance of the cast material layer and the slag layer, evaluate the measured thickness of the cast material layer and the temperature change trajectory of the front and rear arch water-cooled walls at different times, and determine the triggering conditions for dynamic adjustment of the measurement point position.

[0062] Based on the time-series data of total thermal resistance, the hourly heat flux density numerical sequence and the corresponding slag layer thickness increment are extracted. The Pearson correlation coefficient between the hourly heat flux density numerical sequence and the corresponding slag layer thickness increment is calculated as the synchronization coefficient. When the synchronization coefficient exceeds 0.8, it is determined that the rapid slag accumulation stage has begun. For the rapid slag accumulation stage, the heat flux density change rate is detected hourly with a window length of three consecutive hours. If the change rate is negative for three consecutive windows and the absolute value increases beyond a preset increase threshold, the current moment is identified as the inflection point of heat flux density decrease. The measured value of the slurry layer thickness and the surface temperature data of the front and rear arches are recorded at the inflection point. Using the measured value of the slurry layer thickness and the surface temperature data of the front and rear arches at the inflection point, a temperature change trajectory curve over time is plotted. When the slope of the trajectory curve exceeds a preset slope threshold or the temperature difference between the front and rear arches exceeds a preset temperature difference threshold, the trigger condition for dynamic adjustment of the measurement point position is determined.

[0063] For example, in one implementation, the Pearson correlation coefficient is used to quantify the synchronicity between heat flux density and slag layer thickness growth. The correlation coefficient, ranging from -1 to 1, is obtained by calculating the product of the covariance of two time-series data points and their respective standard deviations. When this coefficient exceeds 0.8, it indicates a strong positive correlation between the decrease in heat flux density and the growth of the slag layer, suggesting that the furnace has entered a rapid accumulation phase of heat transfer deterioration.

[0064] Specifically, the sliding window detection method uses a fixed-length time window that moves gradually across the time-series data. Each window contains three consecutive hours of heat flux density data. The rate of change, k, is calculated by fitting the linear trend of the data within the window using the least squares method, where k = (∑(ti-tmean)(qi-qmean)) / ∑(ti-tmean)^2, ti is the time point, qi is the heat flux density, and mean is the average value. When the rate of change for three consecutive windows is negative, and the absolute value of the rate of change in the subsequent window is more than 20% greater than that in the previous window, the current moment is determined to be an inflection point for the decrease in heat flux density. This threshold is set based on the analysis of historical slagging experimental data to identify the turning point of rapid deterioration. The inflection point marks the key turning point from slow accumulation to rapid deterioration of slagging, and the recorded thickness and surface temperature data of the cast-in-place layer at this time become important references for subsequent adjustments. The temperature change trajectory curve is plotted based on 24 consecutive hours of temperature monitoring data. The front arch area, due to its direct exposure to the flame, exhibits a rapid temperature rise followed by a gradual flattening; the rear arch area experiences a relatively slower temperature rise but lasts longer. The slope of the trajectory for each time period is calculated by piecewise linear fitting. When the slope exceeds 5 degrees Celsius per hour or the temperature difference between the front and rear arches exceeds 200 degrees Celsius, the dynamic adjustment of the measurement point position is triggered.

[0065] Preferably, the triggering condition for the dynamic adjustment of the measurement point position also includes the determination of the slag layer thickness growth rate. When the slag layer thickness growth rate exceeds 1 mm per hour, the measurement point adjustment procedure is initiated even if the temperature condition has not reached the threshold. This multi-condition triggering mechanism can promptly capture early signals of heat transfer deterioration.

[0066] In one possible implementation, once the triggering condition is met, the measurement points are adjusted from their original uniform distribution to a more densely distributed arrangement in key areas. The spacing between measurement points in the high-temperature region of the front arch is reduced from 800 mm to 400 mm, while the spacing in the low-temperature region of the rear arch remains unchanged, thus optimizing the allocation of measurement resources.

[0067] S104. Based on the triggering conditions for dynamic adjustment of the measurement point position and the real-time feedback of the water-cooled wall detection system, identify the optimal distribution position and acquisition time interval of the measurement points on the front and rear arch water-cooled walls to obtain the measurement point layout scheme.

[0068] Based on the trigger condition signal dynamically adjusted by the measurement point location, the real-time temperature gradient and heat flux density decay rate of each measurement point on the front and rear arches are obtained from the water-cooled wall detection device. The measurement point data are spatially grouped according to the magnitude of the temperature gradient using the K-means clustering algorithm. When the temperature gradient difference between adjacent measurement points exceeds a preset threshold, it is identified as a region boundary, determining the preliminary distribution location of the measurement points on the front and rear arch water-cooled walls. For the preliminary distribution location, the growth rate is obtained by subtracting the time-series data of the slag layer thickness at each measurement point over time. Based on the magnitude of the growth rate, the region is divided into a fast slag formation zone, a medium-speed slag formation zone, and a slow slag formation zone. The corresponding castable lining thickness ranges for these three types of zones are obtained, identifying the key evolution regions of heat transfer capacity. Based on the spatial distribution of the key evolution regions of heat transfer capacity and the castable lining thickness ranges, the sampling time interval is set to once per hour for the fast slag formation zone, once every two hours for the medium-speed slag formation zone, and once every four hours for the slow slag formation zone, constructing a layout matrix that includes the correspondence between measurement point locations and sampling frequencies. Based on the acquisition frequency of each measuring point in the layout matrix, priority identifiers are assigned according to the principle that the fast slagging zone has the highest priority, followed by the medium-speed zone, and the slow zone has the lowest priority. The measuring point coordinates, acquisition frequency, and priority identifiers are integrated to obtain a measuring point layout scheme that includes acquisition frequency and acquisition location.

[0069] For example, in one implementation, when applying the K-means clustering algorithm to group measurement point data of water-cooled walls, all measurement points in the front and rear arch regions are first used as sample points. Each sample point contains a three-dimensional feature vector, temperature gradient value, heat flux density decay rate, and spatial coordinates. During algorithm initialization, three cluster centers are randomly selected, representing the high-temperature rapid change zone, the medium-temperature transition zone, and the low-temperature stable zone, respectively. By iteratively calculating the Euclidean distance from each measurement point to the cluster center, the measurement point is assigned to the nearest cluster center, and then the centroid of each cluster is recalculated as the new cluster center. The clustering process converges when the change in the cluster center position is less than 0.01 or the number of iterations reaches 100. If adjacent measurement points belong to different clusters and the temperature gradient difference exceeds 10 degrees Celsius per meter, then that location is marked as the region boundary line.

[0070] Specifically, the differential calculation process processes the time-series data of slag layer thickness at each measuring point. The sampling time interval is set to 1 hour. For the slag layer thickness h(i,t) at measuring point i at time t, its growth rate v(i,t) is calculated using the formula v(i,t)=[h(i,t)-h(i,t-1)] / Δt, where Δt is the time interval. Growth rate data is continuously calculated for 24 hours to form a growth rate time series. Based on the statistical distribution of the growth rate, areas with a rate greater than 1.5 mm per hour are classified as fast slag-forming zones, those with a rate between 0.5 and 1.5 mm are classified as medium-speed slag-forming zones, and those with a rate less than 0.5 mm are classified as slow slag-forming zones. The corresponding castable material layer thickness ranges for these three types of zones are: 5 to 8 mm for the fast slag-forming zone, 8 to 12 mm for the medium-speed slag-forming zone, and 12 to 15 mm for the slow slag-forming zone. This correspondence reflects the negative correlation between castable material thickness and slag-forming rate. The determination of key evolution zones in heat transfer capacity is based on the correlation between slagging rate and heat transfer efficiency. Although the initial heat transfer efficiency is high in the rapid slagging zone, the heat transfer capacity decreases most rapidly due to the sharp increase in thermal resistance caused by the rapid accumulation of the slagging layer; this is a critical zone requiring close monitoring. The heat transfer capacity changes relatively smoothly in the medium-speed slagging zone, while the heat transfer performance remains stable in the slow slagging zone.

[0071] Preferably, the layout matrix uses a two-dimensional array structure to store the measuring point information. The rows of the matrix represent different spatial location numbers, and the columns contain four attributes: horizontal coordinate, vertical coordinate, sampling frequency value, and area type identifier. Twenty measuring points are set in the front arch area, and 15 measuring points are set in the rear arch area, totaling 35 rows of data. The sampling frequency value is directly stored as hours, with 1 for the fast slagging zone, 2 for the medium-speed slagging zone, and 4 for the slow slagging zone. The area type identifier is represented by numbers: 1 for the fast zone, 2 for the medium-speed zone, and 3 for the slow zone.

[0072] In one possible implementation, priority allocation follows a risk level principle. The rapid slagging zone, due to its highest risk of heat exchange deterioration, is assigned a priority value of 3, representing the highest priority; the medium-rapid slagging zone has a moderate risk and a priority value of 2; the slow slagging zone has a low risk and a priority value of 1. When multiple measuring points simultaneously trigger the acquisition conditions, the data acquisition task is executed in descending order of priority value.

[0073] For example, after K-means clustering, the front arch area of ​​a coal-fired boiler identified 5 monitoring points for rapid slagging, 8 for medium-speed slagging, and 7 for slow slagging. The monitoring points for rapid slagging are mainly concentrated near the flame center, with a temperature gradient of 15 degrees Celsius per meter and a slagging layer growth rate of 2 millimeters per hour. According to the layout plan, these 5 monitoring points collect data once per hour to monitor the slagging accumulation process in real time. Furthermore, the implementation process of the monitoring point layout plan includes two stages: initial arrangement and dynamic adjustment. In the initial arrangement stage, the distribution of monitoring points is preset based on the furnace structure and burner position. In the dynamic adjustment stage, the monitoring point positions are optimized based on real-time monitoring data. When a region changes from a medium-speed slagging zone to a rapid slagging zone, the monitoring point density in that region is automatically increased from 2 to 4, while the sampling frequency is also increased.

[0074] Understandably, the proposed measurement point layout scheme optimizes the allocation of measurement resources. By setting differentiated acquisition frequencies, it ensures monitoring accuracy in critical areas while reducing data redundancy in non-critical areas. High-frequency acquisition in rapid slagging zones can promptly detect trends of heat transfer deterioration, providing a basis for operational adjustments.

[0075] For example, in actual operation, when three consecutive data collections from a measuring point in the rapid slagging zone show a slagging layer thickness exceeding 10 mm, a soot blowing operation command is triggered. Data from measuring points in the medium-speed slagging zone is used to assess the overall heat transfer trend, while data from measuring points in the slow slagging zone serves as a reference benchmark to verify the accuracy of data from other areas. This hierarchical monitoring system enables precise evaluation of the heat transfer performance of the water-cooled wall.

[0076] S105. Update the acquisition frequency and acquisition position of the thickness measuring device and temperature monitor by the measurement point layout scheme, evaluate the response relationship between the surface temperature rise rate and slagging rate of the front and rear arch water-cooled walls based on the cumulative thickness growth rate of the slagging layer, and determine the total comprehensive thermal resistance of the adjusted cast-in-place material layer and slagging layer.

[0077] Based on the acquisition frequency and location parameters in the measurement point layout scheme, the acquisition frequency parameters are written into the control register of the thickness measurement device to adjust the ultrasonic probe scanning cycle. The acquisition location coordinates are sent to the temperature monitor to drive the infrared sensor to move to the corresponding coordinate point, and real-time thickness and temperature data of each measurement point are obtained after adjustment. Using the real-time thickness and temperature data, the first derivative of the cumulative thickness of the slag layer with respect to time is calculated as the growth rate, and the first derivative of the surface temperature of the front and rear arch water-cooled walls with respect to time is calculated as the temperature rise rate. The least squares method is used to fit the two rate sequences to establish a linear regression relationship. The thermal resistance calculation weights are adjusted according to the slope coefficient of the linear regression relationship. If the slope a is greater than the preset threshold of 0.5, the thermal resistance weight of the slag layer is increased according to the formula w2=0.3+0.4×(a-0.5), where w2 is the weight of the slag layer. Otherwise, the original weights are kept unchanged. The thermal resistance R1=d1 / λ1 of the cast-in-place layer and the thermal resistance R2=d2 / λ2 of the slag layer are calculated, where d1 and d2 are the thicknesses of the cast-in-place layer and the slag layer, respectively, and λ1 and λ2 are their thermal conductivity, respectively. The total combined thermal resistance of the cast-in-place layer and the slag layer after adjustment is determined by weighted summation.

[0078] For example, in one implementation, the control register is written via a serial communication interface. The thickness measuring device has a 16-bit control register, where the high 8 bits store the scan cycle parameter and the low 8 bits store the trigger mode parameter. Upon receiving the acquisition frequency parameter, the frequency value is converted into the corresponding cycle value and written to the high 8 bits. The ultrasonic probe automatically adjusts the pulse emission interval based on this cycle value. The temperature monitor uses a stepper motor to drive an infrared sensor to move on a two-dimensional guide rail. After receiving the acquisition position coordinates, the controller calculates the offset between the current position and the target position and drives the stepper motor to precisely move to the target coordinate point via a pulse signal.

[0079] Specifically, the least squares fitting process handles two sets of rate time-series data. Let the slag layer growth rate be y, and the temperature rise rate be x, collecting n sets of data points (xi, yi). By calculating statistics such as Σxiyi, Σxi, Σyi, and Σxi², the linear regression equation y = ax + b is obtained, where the slope a = (nΣxiyi - ΣxiΣyi) / (nΣxi² - (Σxi)²). The slope a reflects the degree of influence of temperature change on the slag formation rate; a larger slope indicates a stronger promoting effect of temperature rise on slag formation. When the slope exceeds 0.5, it is considered to have entered a strongly correlated interval, requiring adjustment of the thermal resistance calculation weights. The weight adjustment adopts a dynamic allocation mechanism. Initially, the thermal resistance weight of the cast-in-place layer is 0.7, and the thermal resistance weight of the slag layer is 0.3. When the slope coefficient is greater than the preset threshold of 0.5, the weight of the slag layer increases according to the formula w² = 0.3 + 0.4 × (a - 0.5), while the weight of the cast-in-place layer decreases accordingly to maintain a total weight of 1. If the slope is less than or equal to 0.5, the original weight ratio remains unchanged.

[0080] Preferably, the total comprehensive thermal resistance is calculated through a weighted summation. Let the thermal resistance of the cast-in-place layer be R1, and the thermal resistance of the slag layer be R2. The adjusted weights are w1 and w2, respectively. Then, the total comprehensive thermal resistance R = w1 × R1 + w2 × R2. This dynamic weight adjustment mechanism can reflect the changes in thermal resistance contribution at different operating stages, automatically increasing the influence ratio of the slag layer's thermal resistance when slag formation is severe.

[0081] In one possible implementation, the real-time data acquisition frequency is set differently based on the characteristics of the region. Data is collected every 15 minutes in the rapid slagging zone to ensure that rapid changes are captured; every 30 minutes in the medium-speed zone; and every hour in the slow-speed zone.

[0082] S106. Integrate the heat exchange capacity evolution process corresponding to different cast material layer thickness measurements from the total combined thermal resistance of the adjusted cast material layer and slag layer, continuously collect the change trajectory of the cast material layer thickness measurement and the cumulative thickness of the slag layer, and obtain the heat exchange performance evaluation of the front and rear arch water-cooled walls under different cast material layer thickness conditions.

[0083] The thermal resistance values ​​corresponding to different pouring layer thicknesses are calculated from the total combined thermal resistance of the adjusted pouring layer and slag layer. Using the collected measurements of the pouring layer thickness and the data on the cumulative thickness of the slag layer over time, a heat transfer capacity evolution curve is constructed. Based on this curve, the time-varying patterns of thermal resistance values ​​under different pouring layer thicknesses are compared to obtain an evaluation of the heat transfer performance of the front and rear arch water-cooled walls under different pouring layer thicknesses.

[0084] For example, in one implementation, the heat transfer capacity evolution curve is constructed based on 72 hours of continuous monitoring data. A water-cooled furnace wall monitoring device records the thickness of the cast-in-place refractory layer and the cumulative thickness of the slag layer every hour, forming a time-series data point. The evolution curve is plotted with thermal resistance as the ordinate and time as the abscissa. When the cast-in-place refractory layer thickness is 6 mm, the curve shows a steep upward trend; when the thickness is 10 mm, the curve is relatively flat; and when the thickness is 14 mm, the curve remains essentially horizontal.

[0085] Specifically, heat transfer performance evaluation includes three dimensions: thermal resistance growth rate, heat transfer efficiency decay rate, and effective heat transfer duration. The thermal resistance growth rate reflects the speed of slag accumulation, the heat transfer efficiency decay rate characterizes the decline in heat transfer capacity, and the effective heat transfer duration refers to the duration for which normal heat transfer levels are maintained. By comparing the differences in these three indicators under different castable refractory layer thicknesses, the heat transfer performance level for each thickness range is determined.

[0086] Preferably, independent evaluation systems are established for the front arch region and the rear arch region. Due to the higher temperature of the front arch, its heat exchange capacity decreases more rapidly, so the evaluation period is set to 48 hours; the temperature of the rear arch is relatively lower, so the evaluation period can be extended to 96 hours.

[0087] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for intelligent evaluation and analysis of heat exchange performance of a water-cooled furnace, characterized in that, include: The initial heat transfer resistance distribution of the two layers of materials, namely the cast material layer and the slag layer, was obtained by collecting the thickness of the cast material layer, the surface temperature of the front and rear arch water-cooled walls, the cumulative thickness of the slag layer, and the heat flux density from the water-cooled furnace wall inspection. Based on the initial heat transfer resistance distribution, a sensitivity analysis of the surface temperature of the arch water-cooled wall is performed to evaluate the changes in surface temperature rise and ash melting and adhesion when the thickness of the cast material layer is reduced, and to obtain the total comprehensive thermal resistance of the cast material layer and the slag layer. Based on the total comprehensive thermal resistance, evaluate the changes in the thickness of the cast-in-place material layer and the surface temperature of the arch water-cooled wall at different times, and determine the triggering conditions for dynamic adjustment of the measurement point position. Based on the triggering conditions for the dynamic adjustment of the measurement point positions and the real-time feedback from the water-cooled wall detection system, the optimal distribution positions and acquisition time intervals of the measurement points on the front and rear arch water-cooled walls are identified, and a measurement point layout scheme is obtained. The acquisition frequency and acquisition position of the thickness measuring device and temperature monitor are updated by the measurement point layout scheme. The response relationship between the surface temperature rise rate and the slagging rate of the arch water-cooled wall is evaluated based on the cumulative thickness growth rate of the slagging layer. The total combined thermal resistance of the cast material layer and the slagging layer after adjustment is determined. Based on the adjusted total thermal resistance, the heat exchange capacity evolution process corresponding to different castable layer thickness measurements is integrated. The change trajectory of the castable layer thickness measurement and the cumulative thickness of the slag layer is continuously collected to obtain the performance evaluation results of the arch water-cooled wall before and after heat exchange under different castable layer thickness conditions.

2. The intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to claim 1, characterized in that, The process of collecting measurements of the cast-in-place material layer thickness, the surface temperature of the front and rear arch water-cooled walls, the cumulative thickness of the slag layer, and the heat flux density from the water-cooled furnace wall inspection to obtain the initial heat transfer resistance distribution of the two material layers, the cast-in-place material layer and the slag layer, includes: The thickness of the cast lining layer was measured at multiple measuring points on the water-cooled furnace wall using an ultrasonic thickness gauge. The surface temperature data of the arch water-cooled wall before and after heat exchange was obtained using an infrared thermal imager. The cumulative thickness of the slag layer was calculated based on the ash deposition pattern. The heat flux density at each measuring point was measured using a furnace heat flow meter to obtain the basic data of the measuring points, which includes four types of parameters: thickness, temperature, slag formation, and heat flux. Based on the basic data of the measuring points, which includes four types of parameters: thickness, temperature, slagging, and heat flow, the thermal resistance of the castable material is calculated according to the thermal conductivity of the castable material and the measured thickness of the castable material layer. The thermal resistance of the slagging layer is calculated according to the thermal conductivity of the slagging layer and the cumulative thickness of the slagging layer. The thermal resistance of the castable material and the thermal resistance of the slagging layer are added together by the principle of thermal resistance superposition to obtain the initial heat transfer resistance distribution at each measuring point.

3. The intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to claim 1, characterized in that, The sensitivity analysis of the surface temperature of the arch water-cooled wall based on the initial heat transfer resistance distribution is used to evaluate the changes in surface temperature rise and ash molten adhesion as the thickness of the cast refractory layer decreases, and to obtain the total combined thermal resistance of the cast refractory layer and the slag layer, including: The surface temperature gradient of the water-cooled wall under different pouring material layer thicknesses is calculated based on the initial heat transfer resistance distribution. The temperature difference between the ash softening point and the surface temperature rise rate is obtained to determine the adhesion acceleration time node when the temperature is close to the ash softening point. For the adhesion acceleration time node when the temperature is close to the softening point of the ash, based on the difference between the thermal conductivity of the slag layer and the thermal conductivity of the castable, the thermal resistance increment per unit thickness of the slag layer is calculated through the linear relationship between thermal resistance and thickness, and the growth rate function of the total thermal resistance as a function of the cumulative thickness of the slag layer is obtained. By using the growth rate function, a dynamic superposition relationship between the thermal resistance of the cast-in-place material layer and the thermal resistance of the slag layer is established. Based on the temperature distribution difference between the front arch area and the rear arch area, a weighting coefficient is set according to the temperature difference ratio to determine the total comprehensive thermal resistance for different operating stages.

4. The intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to claim 1, characterized in that, The process involves evaluating the measured thickness of the cast-in-place material layer and the surface temperature of the arch water-cooled wall at different times based on the total comprehensive thermal resistance, and determining the triggering conditions for dynamic adjustment of the measurement point position, including: Based on the total comprehensive thermal resistance, determine the hourly heat flux density numerical sequence and the corresponding slag layer thickness increment sequence. Calculate the Pearson correlation coefficient between the two hourly heat flux density numerical sequences and the corresponding slag layer thickness increment sequence as the synchronization coefficient. When the synchronization coefficient exceeds the synchronization coefficient threshold, it is determined that the rapid slag accumulation stage has been entered. For the rapid accumulation stage of slag, the rate of change of heat flux density is detected hourly to identify the inflection point of heat flux density decrease, and the measured value of the thickness of the cast material layer and the surface temperature data of the arch before and after heat exchange are recorded at the inflection point. Using the measured thickness of the poured material layer at the inflection point and the surface temperature data of the arch before and after heat exchange, a temperature change trajectory curve over time is plotted. When the slope of the trajectory curve or the temperature difference between the front and rear arches exceeds a set threshold, the trigger condition for dynamic adjustment of the measurement point position is determined.

5. The intelligent evaluation and analysis method for the heat transfer performance of a water-cooled furnace according to claim 1, characterized in that, The triggering conditions for the dynamic adjustment of the measurement point positions and the real-time feedback from the water-cooled wall detection system identify the optimal distribution positions and acquisition time intervals of the measurement points on the front and rear arch water-cooled walls, resulting in a measurement point layout scheme, including: Based on the trigger condition signal dynamically adjusted according to the location of the measurement point, the real-time temperature gradient and heat flux density decay rate of each measurement point of the arch before and after heat exchange are obtained from the water-cooled wall detection device. The measurement point data are spatially grouped according to the magnitude of the temperature gradient by a clustering algorithm, the regional boundaries are identified, and the preliminary distribution location of the measurement points of the arch water-cooled wall is determined. Based on the initial distribution of the measurement points of the arch water-cooled wall, the growth rate is obtained by calculating the difference in the time series data of the slag layer thickness at each measurement point. According to the magnitude of the growth rate, the area is divided into a fast slag zone, a medium-speed slag zone, and a slow slag zone. The key evolution region of heat exchange capacity is determined according to the type of slag zone. Based on the spatial distribution of the key evolution regions of the heat exchange capacity, the sampling time intervals for the rapid slagging zone, the medium-speed slagging zone, and the slow slagging zone are set, and a layout matrix containing the correspondence between the measurement point location and the sampling frequency is constructed. Based on the acquisition frequency of each measurement point in the layout matrix, priority identifiers are assigned according to regional priority. The measurement point coordinates, acquisition frequency, and priority identifiers are integrated to obtain the measurement point layout scheme that includes acquisition frequency and acquisition location.

6. The intelligent evaluation and analysis method for the heat transfer performance of a water-cooled furnace according to claim 1, characterized in that, The process of updating the acquisition frequency and acquisition position of the thickness measuring device and temperature monitor through the measurement point layout scheme, evaluating the response relationship between the surface temperature rise rate and the slagging rate of the arch water-cooled wall based on the cumulative thickness growth rate of the slagging layer, and determining the adjusted total combined thermal resistance of the cast-in-place layer and the slagging layer includes: Based on the acquisition frequency and acquisition position parameters in the measurement point layout scheme, the ultrasonic probe scanning cycle of the thickness measurement device is adjusted, and the infrared sensor of the temperature monitor is driven to move to the corresponding coordinate point to obtain the real-time thickness data and temperature data of each measurement point after adjustment. Using the real-time thickness data and the temperature data, the cumulative thickness growth rate of the slag layer and the surface temperature rise rate of the front and rear arch water-cooled walls are calculated, and a linear regression relationship is established. The thermal resistance calculation weights are adjusted according to the slope coefficient of the linear regression relationship. The thermal resistance of the cast-in-place layer and the thermal resistance of the slag layer are calculated, and the total adjusted comprehensive thermal resistance is determined by weighted summation.

7. The intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to claim 1, characterized in that, After updating the acquisition frequency and acquisition position of the thickness measuring device and temperature monitor through the aforementioned measurement point layout scheme, the real-time thickness data and temperature data of each measuring point after the update are obtained. The first derivative of the cumulative thickness of the slag layer with respect to time is calculated as the growth rate, and the first derivative of the surface temperature of the arch water-cooled wall before and after heat exchange with respect to time is calculated as the temperature rise rate. The least squares method is used to fit the two sequences of surface temperature change rates of the arch water-cooled wall before and after heat exchange to establish a linear regression relationship.

8. The intelligent evaluation and analysis method for the heat transfer performance of a water-cooled furnace according to claim 1, characterized in that, The process involves integrating the heat transfer capacity evolution process corresponding to different castable lining thickness measurements based on the adjusted total thermal resistance, continuously collecting the change trajectory of the castable lining thickness measurements and the cumulative thickness of the slag layer, and obtaining the performance evaluation results of the arch water-cooled wall before and after heat transfer under different castable lining thickness conditions, including: Based on the adjusted total thermal resistance, the thermal resistance values ​​corresponding to different pouring material layer thicknesses are calculated. Using the collected pouring material layer thickness measurements and the data on the change of the cumulative thickness of the slag layer over time, a heat exchange capacity evolution curve is constructed. Based on the heat transfer capacity evolution curve, the time variation law of thermal resistance value under different lining thicknesses is compared to obtain the heat transfer performance evaluation results of the arch water-cooled wall under different lining thicknesses.

9. The intelligent evaluation and analysis method for the heat exchange performance of a water-cooled furnace according to claim 1, characterized in that, The obtained performance evaluation results of the arch water-cooled wall before and after heat transfer under different refractory layer thicknesses include: Extract the thermal resistance values ​​corresponding to different pouring material layer thicknesses from the adjusted total comprehensive thermal resistance; Continuously collect data on the changes in the thickness of the poured material layer and the cumulative thickness of the slag layer over time; A heat exchange capacity evolution curve is constructed based on the changes in the thickness of the cast material layer and the cumulative thickness of the slag layer over time. Based on the heat transfer capacity evolution curve, the thermal resistance variation law under different pouring material layer thicknesses was analyzed; Based on the variation law of thermal resistance under different pouring material layer thicknesses, the heat exchange performance evaluation results of the water-cooled wall under different conditions are generated, and a comprehensive performance report is formed.

Images