Ice rink ice making system energy consumption optimization control method based on Internet of Things

Abstract

The invention provides an ice rink ice making system energy consumption optimization control method based on the Internet of Things, and the method comprises the steps: obtaining ice surface temperature distribution, friction force distribution and crystal microscopic image data, constructing an ice surface state data set, and analyzing a local ice crystal reconstruction region, caused by a friction heat generation phenomenon, of an ice surface; performing feature extraction on the crystal microscopic image data of the local ice crystal reconstruction area, including crystal form and size features, and determining the position and range of local ice crystal reconstruction according to the crystal form and size features; determining a refrigeration power distribution scheme by analyzing the characteristics of friction distribution in the ice surface state data set in real time according to the temperature gradient anomaly spatial and temporal distribution characteristics of the region boundary; and through the optimized water spraying frequency distribution and refrigeration power distribution scheme, the refrigeration power output and the water spraying frequency of each area are adjusted according to the real-time change of the ice surface performance requirement, a control parameter combination is determined, and a dynamic control instruction is obtained.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an energy consumption optimization control method for an ice-making system in an ice rink based on the Internet of Things. Background Art

[0002] Energy optimization of ice rink ice-making systems is a core area of ​​green sports facility operations, directly impacting event quality, athlete performance, and environmental sustainability. An efficient ice-making system must not only meet the diverse performance requirements of different events but also maintain low energy consumption and stable operation under intense use. Current solutions often rely on single-system cooling power regulation or fixed water spray patterns, making it difficult to address the dynamic performance changes of the ice surface caused by event transitions or intensive use. These methods often lack multidimensional data fusion and real-time control when addressing issues such as localized ice surface frictional heating and crystal structure reconstruction, making it difficult to achieve precise energy allocation, resulting in uneven ice surface quality or energy waste. The main challenge facing research lies in dynamically responding to the complex changes in ice surface performance requirements. The formation of ice surface crystal structure is closely related to the friction coefficient. For example, figure skating requires fine ice crystals with a low friction coefficient, while ice hockey requires a harder ice surface to withstand high impact. Controlling the water spray frequency during the ice-making process directly affects the formation and stability of the ice crystal structure. Excessively high or low water spray frequencies can lead to uneven ice hardness or localized liquid water residue, which in turn affects the friction coefficient and ice surface quality. Changes in events or intensive skateboarding by athletes can generate frictional heat, leading to localized ice crystal reconstruction and, in turn, abnormal temperature gradients. These anomalies can trigger pore freezing, disrupting the uniformity of the ice surface and affecting the skating experience and safety. The interaction between crystal structure and temperature gradients makes it difficult to balance ice surface performance and energy optimization with traditional static control. Therefore, how to capture the dynamic response characteristics of the ice surface crystal structure and friction coefficient in real time through multi-dimensional sensor data fusion, and accordingly optimize the spatiotemporal coordination of cooling power allocation and water spray frequency, has become a key issue in optimizing the energy consumption of ice rink ice-making systems. Summary of the Invention

[0003] The present invention provides an energy consumption optimization control method for an ice rink ice-making system based on the Internet of Things, which mainly includes: Obtain ice surface temperature distribution, friction force distribution, and crystal microscopic image data to construct an ice surface state dataset and analyze the local ice crystal reconstruction areas caused by frictional heating on the ice surface; Extract features from the crystal microscopic image data of the local ice crystal reconstruction area, including crystal morphology and size features, and determine the location and range of the local ice crystal reconstruction based on the crystal morphology and size features; Based on the location and range of local ice crystal reconstruction and the real-time monitored ice surface temperature distribution data, the gradient analysis method is used to analyze the temperature gradient anomaly distribution. Based on the analysis results, the boundaries of the abnormal area are identified and the spatiotemporal distribution characteristics of the temperature gradient anomaly at the regional boundaries are extracted. Based on the spatiotemporal distribution characteristics of temperature gradient anomalies at the regional boundaries, the cooling power allocation scheme is determined by real-time analysis of the friction distribution characteristics in the ice surface state dataset. Based on the cooling power allocation plan, the friction distribution data from the ice surface state dataset was integrated to analyze the impact of frictional heating in local areas of the ice surface on the adjustment of ice crystal structure. The adjustment range of the water spray frequency was determined. Based on the adjustment range of the water spray frequency and the abnormal distribution of temperature gradients, the spatiotemporal coordination parameters of the water spray frequency were adjusted to obtain the optimized water spray frequency distribution. Through the optimized water spray frequency distribution and cooling power allocation scheme, the cooling power output and water spray frequency of each area are adjusted according to the real-time changes in ice surface performance requirements, the control parameter combination is determined, and dynamic control instructions are obtained.

[0004] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention discloses an energy consumption optimization control method for an ice rink ice-making system based on the Internet of Things. The method obtains ice surface status data through the Internet of Things platform and analyzes the local ice crystal reconstruction phenomenon caused by frictional heat generation on the ice surface. The ice crystal morphological characteristics are extracted using image processing technology, abnormal areas are identified in combination with temperature distribution data, and the relationship between friction force distribution and ice crystal structure adjustment is analyzed. According to the low friction coefficient requirement, the cooling power distribution plan is determined, and the water spray frequency distribution is optimized. The present invention generates dynamic control instructions, adjusts the cooling power and water spray frequency in real time, and iteratively updates the control parameters based on feedback information. This method can accurately control the ice surface state, effectively deal with the local frictional heat generation phenomenon, maintain stable ice surface performance, and improve the intelligence level and energy utilization efficiency of the ice rink refrigeration system. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 This is a flow chart of an energy consumption optimization control method for an ice rink ice-making system based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0006] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0007] like Figure 1 In this embodiment, an energy consumption optimization control method for an ice-making system in an ice rink based on the Internet of Things may specifically include: Step S101, obtain the ice surface temperature distribution, friction force distribution and crystal micrograph data, construct the ice surface state data set, and analyze the local ice crystal reconstruction area caused by the friction heat generation phenomenon of the ice surface.

[0008] From the Internet of Things platform, the real-time data stream of the temperature sensor array in the ice rink is obtained, the temperature value of each sensor node and its corresponding spatial coordinates are read, the temperature data is time stamped according to the preset sampling frequency, and a temperature distribution data set containing space-time information is formed. A multi-point friction force sensor is used to collect the friction coefficients at different positions on the ice surface, and the temperature values of each coordinate point in the temperature distribution data set are combined to calculate the temperature gradient between adjacent monitoring points. When the product of the temperature gradient and the friction coefficient exceeds the preset heat threshold, the region is determined as a friction heat generation region. For the friction heat generation region, a micro-imaging device is started to obtain the ice crystal structure image of the region, an edge detection method is used to identify the crystal boundary, and by comparing the crystal boundary characteristics of the unheated region, a pixel point set with changed crystal structure is extracted. According to the distribution density of the pixel point set in each monitoring region, the ratio of the number of reconstructed crystal pixels to the total number of pixels in the region is calculated. When the ratio exceeds the preset reconstruction ratio threshold, the region is determined as an ice crystal reconstruction region, and an ice surface state data set containing the boundary coordinates of the reconstruction region and the reconstruction degree index is output.

[0009] For example, in an actual ice rink monitoring system, the arrangement of the temperature sensor array needs to fully consider the non-uniformity of the ice surface temperature distribution.

[0010] For example, the sensors can be deployed in a grid manner, with each sensor node spaced 2 to 3 meters apart, forming a monitoring network covering the entire ice rink. This dense sensor arrangement can capture subtle changes in ice surface temperature, especially in areas where athletes frequently move, where temperature changes will be more pronounced. Each sensor not only records temperature values, but also needs to mark its precise spatial coordinates, such as using a Cartesian coordinate system to establish a coordinate system with the center of the ice rink as the origin.

[0011] It should be noted that the importance of time stamping lies in the ability to track the dynamic process of temperature changes. When the ice blade slides on the ice surface, the heat generated by friction will cause the local temperature to rise, and this temperature change has a time dependence. By continuously collecting temperature data with time stamps, a four-dimensional space-time temperature field can be constructed, providing basic data support for subsequent heat analysis.

[0012] In one possible implementation, the friction coefficient is measured using a piezoelectric sensor. When the ice skate passes by, the sensor can sense the normal pressure and tangential friction in real time, and then calculate the dynamic friction coefficient. The temperature gradient is calculated based on the temperature difference between adjacent monitoring points divided by the spatial distance. The advantage of this calculation method is that it can intuitively reflect the direction and intensity of heat conduction. When the product of the temperature gradient and the friction coefficient in a certain area is large, it means that the area is undergoing a violent frictional heating process, which is the key area where the ice crystal structure may change.

[0013] Specifically, the application of microscopic imaging equipment requires fast response and high-resolution imaging capabilities. Crystal boundaries can be identified by calculating the gradient change in the image's grayscale using the Sobel operator or the Canny operator. Normal ice crystals exhibit a regular hexagonal structure, but those subjected to frictional heating may exhibit blurred boundaries and irregular shapes. Comparative analysis can accurately identify areas experiencing structural changes.

[0014] Preferably, the quantitative assessment of the degree of reconstruction adopts a pixel statistics method. In each monitoring area, the number of pixels where structural changes have occurred is counted and the ratio is calculated to the total number of pixels in the area. This quantitative method makes the assessment of the ice surface state more objective and accurate. When the reconstruction ratio exceeds a set threshold, such as 30%, it can be determined that the area needs special attention. The formed ice surface state dataset not only contains the location information of the reconstructed area, but also includes quantitative indicators of the degree of reconstruction, providing a scientific basis for ice rink maintenance.

[0015] Step S102 , extracting features from the crystal microscopic image data of the local ice crystal reconstruction area, including crystal morphology and size features, and determining the position and range of the local ice crystal reconstruction based on the crystal morphology and size features.

[0016] Obtain microscopic image data of the local ice crystal reconstruction area, grayscale the image, identify the crystal edge contour by calculating the grayscale difference of adjacent pixels, use the eight-connected domain search method to number and mark each independent crystal, and obtain a contour data set containing a sequence of crystal edge coordinate points. According to the sequence of edge coordinate points in the contour data set, calculate the crystal perimeter by accumulating the distances between adjacent coordinate points, calculate the crystal enclosed area using Green's formula, obtain the ratio of perimeter to area as the shape factor, and calculate the distance between the two farthest points of the contour as the grain size parameter. If the shape factor exceeds the preset morphological threshold, the crystal is determined to be a reconstructed crystal. For the reconstructed crystals whose shape factors exceed the morphological threshold, calculate the arithmetic mean of their contour coordinate points to obtain the geometric center coordinates. According to the grain size parameters and the geometric center coordinates, use the density clustering algorithm to identify the spatial clustering distribution of the reconstructed crystals, and obtain the boundaries of the clustered regions containing multiple reconstructed crystals. Based on the geometric center coordinates of all reconstructed crystals within the boundary of the aggregation area, the maximum and minimum values ​​of the coordinates are calculated to determine the rectangular boundary, and the length and width of the reconstructed area are obtained by the difference in coordinates of the diagonal vertices of the rectangle. Combined with the ratio of the number of reconstructed crystals in the area to the area, the position coordinates and coverage range data of the local ice crystal reconstruction are output.

[0017] For example, grayscale processing of microscopic images is a basic step in crystal feature extraction.

[0018] In one possible implementation, the original color image is converted to a grayscale image using a weighted averaging method, where the red, green, and blue channels are combined with weights of 0.299, 0.587, and 0.114, respectively. This weighting distribution reflects the human eye's sensitivity to different colors, allowing the converted grayscale image to more accurately reflect the light and dark variations in the crystal structure.

[0019] It should be noted that the eight-connected domain search method offers unique advantages in identifying independent crystals. This method considers not only the four directions of pixel alignment—up, down, left, and right—but also the four diagonal directions. When the grayscale value difference between a pixel and its eight neighboring pixels is less than a set threshold, these pixels are classified as belonging to the same connected domain, thus forming a complete crystal outline. This method effectively avoids outline breaks caused by irregular crystal edges.

[0020] Specifically, Green's formula demonstrates its efficiency in calculating the area of ​​irregular crystals. It does this by sequentially connecting discrete coordinate points on the crystal's outline to form a closed polygon, and then calculating the area using the cross-addition of adjacent vertex coordinates.

[0021] For example, for a crystal outline consisting of 100 edge points, Green's formula can quickly determine its precise area. The shape factor, the ratio of the square of the perimeter to the area, quantitatively describes the regularity of the crystal's morphology. The shape factor of a standard hexagonal ice crystal is approximately 13.86. A measured value exceeding 20 indicates significant changes in the crystal's morphology.

[0022] In one embodiment, the measurement of grain size parameters adopts the distance method of the farthest point pair on the contour. By traversing the Euclidean distance of all pairs of points on the contour, the two points with the largest distance are found, and this maximum distance represents the characteristic size of the grain. Compared with the traditional circumscribed rectangle method, this method can better reflect the actual size characteristics of irregular crystals. The density clustering algorithm plays a key role in identifying the reconstructed crystal aggregation area. The algorithm takes the geometric center of each reconstructed crystal as the core point and sets two parameters: the neighborhood radius and the minimum number of points. When the number of other reconstructed crystals contained in the neighborhood of a core point exceeds the minimum number of points, these crystals are classified as the same aggregation area. In this way, the spatial distribution pattern of the reconstructed crystals can be effectively identified, and the dense reconstruction area and the scattered distribution area can be distinguished.

[0023] Preferably, the minimum enclosing rectangle algorithm is used to determine the boundaries of the rectangle. By traversing the geometric center coordinates of all reconstructed crystals within the clustering area, the maximum and minimum x- and y-coordinates are found. The rectangle formed by these four extreme points is the minimum enclosing rectangle. The length and width of the rectangle directly reflect the spatial extension of the reconstructed area, providing precise positioning information for subsequent ice surface maintenance.

[0024] Step S103: Based on the position and range of the local ice crystal reconstruction and the real-time monitored ice surface temperature distribution data, a gradient analysis method is used to analyze the abnormal distribution of the temperature gradient. Based on the analysis results, the boundary of the abnormal area is identified, and the spatiotemporal distribution characteristics of the temperature gradient anomaly at the area boundary are extracted.

[0025] Based on the location coordinates and coverage of local ice crystal reconstructions, the temperature values ​​of the corresponding area and its surrounding areas are extracted from the real-time monitored ice surface temperature distribution data. The temperature difference between adjacent monitoring points is calculated and divided by the spatial distance to obtain the temperature gradient value. The gradient direction is also determined based on the coordinate difference between adjacent points, forming temperature gradient field data containing the gradient magnitude and direction. Based on this temperature gradient field data, the average temperature gradient of the undisturbed ice surface area is calculated as a baseline value. Monitoring points with gradient values ​​exceeding a preset multiple of the baseline value are identified as outliers. Adjacent outlier points are connected using an eight-neighborhood search method to form a closed boundary contour of the outlier area. Based on the temperature gradient values ​​of each monitoring point on this closed boundary contour, the difference between the current gradient value and the previous gradient value is calculated according to the preset monitoring period and divided by the time interval to obtain the gradient change rate. The time when the gradient value of each contour point first exceeds the anomaly threshold and the time when it returns to normal is recorded, and the difference between these two times is calculated to obtain the anomaly duration. Based on the gradient change rate and anomaly duration data, combined with the spatial coordinates of the contour points, a feature vector is constructed containing the coordinate position, gradient change rate value, and anomaly duration. The resulting feature dataset reflects the spatiotemporal distribution of the temperature gradient anomaly.

[0026] For example, in an ice surface temperature monitoring scenario, the calculation of the temperature gradient involves space vector operations.

[0027] Specifically, after obtaining the temperature values ​​of two adjacent monitoring points, the temperature difference is divided by the Euclidean distance between the two points to obtain the gradient magnitude. Determining the gradient direction requires considering the spatial distribution of the temperature field. The direction of heat transfer is represented by calculating a unit vector pointing from the low temperature point to the high temperature point. This vectorized representation method enables more accurate subsequent identification of abnormal areas.

[0028] It should be noted that establishing a baseline value is crucial for anomaly detection. In practical applications, a static ice surface away from the athlete's activity area in the ice rink is selected as a reference area. Temperature gradient data in this area at different time periods is continuously collected, and its arithmetic mean is calculated as a baseline. This baseline value reflects the temperature distribution characteristics of the ice surface in its natural state. When the gradient value of a certain area exceeds 1.5 or 2 times the baseline value, it can be identified as an abnormal state. The eight-neighborhood search method shows unique advantages in connecting abnormal points to form a closed contour. Starting from any abnormal point, this method checks the adjacent points in eight directions around it. If there are other abnormal points, they are marked as the same area. Through recursive search, the complete abnormal area boundary is eventually formed. This method can effectively handle abnormal areas with irregular shapes and avoid the errors that may be caused by simple geometric shape fitting.

[0029] In one possible implementation, the gradient change rate is calculated using a time-difference method. Assuming a 30-second monitoring period, and the temperature gradient at a boundary point is 0.8 degrees Celsius per meter at the current moment and 0.5 degrees Celsius per meter at the previous moment, the gradient change rate is 0.01 degrees Celsius per meter per second. This value intuitively reflects the intensity of the dynamic temperature field at that point and provides a quantitative basis for determining the evolving trend of ice surface conditions. The mechanism for recording the duration of anomalies requires precise timestamp management. Each monitoring point maintains a status marker. When the gradient value first exceeds the anomaly threshold, the current moment is recorded as the anomaly start time. The status of the point is continuously monitored until the gradient value returns to the normal range, and the recovery time is recorded. The difference between the two times is the duration of the anomaly at that point, a time parameter that is crucial for assessing the severity of ice surface damage.

[0030] Preferably, the feature vector is constructed using multidimensional data fusion. The feature vector of each boundary point contains five components: x-coordinate, y-coordinate, maximum gradient change rate, average gradient change rate, and anomaly duration. This multidimensional representation method comprehensively characterizes the spatiotemporal evolution of the anomaly region.

[0031] Step S104 , determining a cooling power allocation plan by real-time analysis of the friction distribution characteristics in the ice surface state data set based on the spatiotemporal distribution characteristics of the temperature gradient anomaly at the regional boundary.

[0032] According to the spatiotemporal distribution characteristics of the temperature gradient anomalies at the regional boundaries, the friction coefficient values ​​of the corresponding regions are extracted from the ice surface state data set, and the difference between the friction coefficient of each monitoring point and the preset ideal friction coefficient value is calculated to obtain the friction performance deviation value of each monitoring point and its spatial distribution data. With respect to the friction performance deviation value, the crystal microscopic image data of the corresponding monitoring point position is obtained, and the degree of crystal deformation is determined by calculating the ratio of the current crystal shape factor to the standard hexagonal crystal shape factor. The friction coefficient deviation value is divided into intervals and the corresponding mean value of the crystal deformation degree is statistically calculated to construct the mapping relationship data between the friction coefficient deviation and the crystal deformation degree. Based on the mapping relationship data, if it is detected that the friction coefficient of a certain area exceeds the preset performance threshold, the corresponding crystal deformation degree is queried according to the friction coefficient value, and the required temperature adjustment amount is calculated by the empirical relationship between the crystal deformation degree and the temperature change.

[0033]

[0034] , where ΔT represents the required temperature adjustment, K represents the proportionality factor, ΔL represents the crystal deformation, L0 represents the initial length of the crystal, and α represents the crystal's thermal expansion coefficient. By measuring the degree of crystal deformation, the required temperature adjustment can be reversely calculated, thereby achieving precise control. Based on the temperature adjustment amount and the spatial location of the abnormal area, the heat change required to adjust from the current temperature to the target temperature is calculated. Combined with the distance from each location to the refrigeration equipment and the thermal conductivity coefficient, the power output of each refrigeration equipment is determined, forming a refrigeration power allocation plan.

[0035] For example, the calculation of the deviation of the coefficient of friction plays a central role in the evaluation of ice rink performance.

[0036] For example, the ideal ice friction coefficient is typically set between 0.003 and 0.007, a range determined based on the skating requirements of professional speed skaters. When the measured friction coefficient is 0.012, the deviation from the ideal value of 0.005 reaches 0.007, indicating that the ice performance in that area has significantly deviated from the optimal state. The spatial distribution of these deviations exhibits significant non-uniformity, with greater deviations typically occurring along paths frequently crossed by skates.

[0037] It should be noted that the crystal shape factor is a quantitative indicator for evaluating crystal deformation, and its calculation is based on the geometric characteristics of the crystal outline. The shape factor of a standard hexagonal ice crystal is approximately 13.86. This value is obtained by calculating the ratio of the square of the perimeter of a regular hexagon to its area. When ice crystals are reconstructed by frictional heat, the crystal boundaries become irregular and the shape factor increases to 20 or even higher. Through statistical analysis of a large amount of experimental data, it was found that for every 0.001 increase in the friction coefficient, the corresponding crystal shape factor increases by an average of about 1.5.

[0038] In one possible implementation, mapping relationship data is constructed using a segmented statistical approach. The friction coefficient deviation is divided into multiple intervals, such as 0-0.002, 0.002-0.004, and 0.004-0.006. Within each interval, corresponding crystal deformation data is collected and the mean is calculated. This segmented mapping approach effectively addresses data discreteness and noise interference, improving the accuracy of subsequent temperature control calculations.

[0039] Specifically, the empirical relationship between the degree of crystal deformation and temperature change is derived from the theory of ice crystal growth kinetics. When the temperature drops from -5 degrees Celsius to -7 degrees Celsius, the growth rate of ice crystals slows and the crystal structure tends to become more regular. Experiments have shown that for every 1 degree Celsius decrease in temperature, the crystal shape factor decreases by an average of 0.8. Based on this empirical relationship, when the shape factor needs to be reduced from 22 to 14, the required temperature adjustment is approximately 10 degrees Celsius. Calculating the heat change value involves estimating the specific heat capacity and mass of ice. The specific heat capacity of ice is 2.09 kilojoules per kilogram per degree Celsius. For an ice layer 5 centimeters thick and 1 square meter in area, the mass is approximately 46 kilograms. To reduce the temperature by 2 degrees Celsius, the required heat absorption is 192 kilojoules. Taking into account the efficiency loss in the heat conduction process, actual refrigeration equipment needs to provide more cooling capacity.

[0040] Optimally, cooling power allocation should take spatial distance into account. Areas farther from the cooling equipment require higher power allocation to compensate for transmission losses due to the longer heat conduction paths. The typical thermal conductivity coefficient is 2.2 watts per meter per degree Celsius. By calculating the thermal resistance at different locations, the power output of each cooling device can be accurately determined. This power allocation method, based on the physical principles of heat transfer, enables precise control of ice surface temperature and effectively restores the ideal friction properties of the ice surface.

[0041] In step S105, according to the cooling power allocation plan, the friction distribution data in the ice surface state data set is integrated to analyze the impact of the frictional heating phenomenon in the local area of ​​the ice surface on the adjustment of the ice crystal structure, determine the adjustment range of the water spray frequency, and adjust the spatiotemporal coordination parameters of the water spray frequency based on the adjustment range of the water spray frequency and the abnormal temperature gradient distribution to obtain the optimized water spray frequency distribution.

[0042] Based on the power values ​​for each zone in the cooling power allocation plan, combined with the friction coefficient and sliding speed data for the corresponding zone in the ice surface status dataset, frictional heating power is calculated by multiplying friction force and speed. The net heat flux value for each monitored zone is then subtracted from the cooling power. Based on this net heat flux value, a negative net heat flux indicates heat accumulation in that zone. The amount of heat required to maintain a stable ice surface temperature is calculated. The water spray volume per unit area is determined by dividing the latent heat of evaporation of water at the ice surface temperature by the heat value, thereby obtaining the initial water spray frequency parameter. To determine this initial water spray frequency parameter, the gradient value for each zone in the temperature gradient anomaly distribution data is obtained. The ratio of this gradient value to the normal gradient baseline value is calculated. If the ratio exceeds a preset threshold, the water spray frequency for that zone is adjusted by multiplying it by the ratio. The corresponding water spray interval is then determined based on the adjusted frequency value. According to the adjusted water spray frequency and the corresponding water spray interval time, combined with the net heat flux density distribution of each area, a parameter matrix including regional coordinates, water spray frequency values, water spray interval parameters, and single water spray volume is constructed, and the optimized water spray frequency distribution scheme after differentiation for each area is output.

[0043] For example, the calculation of frictional heat power plays a key role in the thermal management of ice rinks.

[0044] For example, when a skate skate glides across the ice at 15 meters per second, the product of friction and speed is the instantaneous frictional heating power. Assuming a friction coefficient of 0.01 in a certain area and a positive pressure of 700 Newtons generated by the athlete's weight, the friction force is 7 Newtons, corresponding to a frictional heating power of 105 watts. This power value means that 105 joules of heat are injected into the ice surface every second. If the cooling system power in this area is only 80 watts, the net heat flux density is negative 25 watts per square meter, indicating that the area is continuously warming.

[0045] It's important to note that the latent heat of evaporation of water plays a unique role in managing ice surface temperature. At standard atmospheric pressure, the latent heat of evaporation of water at 0°C is approximately 2,500 kilojoules per kilogram. When the ice surface needs to absorb additional heat to maintain a stable temperature, sprayed water removes this heat through evaporation. Each gram of water evaporates and absorbs 2.5 kilojoules of heat. This efficient heat absorption mechanism makes water spraying an effective means of rapidly regulating ice surface temperature.

[0046] In one possible implementation, the determination of the initial water spray frequency is based on the principle of thermal balance. When the net heat flux density of a certain area is negative 50 watts per square meter, it means that 50 joules of heat needs to be absorbed per square meter per second. Calculation shows that 0.02 grams of water need to be evaporated per second to achieve thermal balance. Considering that the actual evaporation efficiency is about 30%, 0.067 grams of water needs to be sprayed. If the single water spray volume is set to 10 grams per square meter, the water spray interval should be 150 seconds, and the corresponding water spray frequency is 24 times per hour.

[0047] Specifically, the calculation of the temperature gradient ratio provides a basis for dynamic adjustments. The baseline temperature gradient for a normal ice surface is typically within 0.5 degrees Celsius per meter. When the measured gradient reaches 1.5 degrees Celsius per meter, the ratio is 3. This ratio directly reflects the degree of thermal anomaly in that area. Multiplying the initial water spray frequency by this ratio yields an adjusted water spray frequency of 72 times per hour, with the corresponding water spray interval shortened to 50 seconds. This dynamic adjustment mechanism based on real-time data ensures that the water spray strategy can rapidly respond to changes in ice surface conditions. The construction of a parameter matrix enables refined management of water spray control. Each matrix element contains four key parameters: the two-dimensional coordinate location of the area, the calculated water spray frequency value, the corresponding time interval, and the water volume per spray. This matrix-based data organization facilitates rapid system query and execution. By dividing the ice rink into multiple control zones, each with independent parameter settings, a differentiated water spray frequency distribution is achieved. The water spray frequency in abnormally high temperature areas may reach 100 times per hour, while that in normal areas may only be 20 times per hour. This precise zoning control effectively improves the efficiency and accuracy of ice surface temperature control.

[0048] In step S106, the optimized water spray frequency distribution and cooling power allocation scheme is used to adjust the cooling power output and water spray frequency of each area according to the real-time changes in ice surface performance requirements, determine the control parameter combination, and obtain dynamic control instructions.

[0049] Based on the optimized water spray frequency distribution and cooling power allocation scheme, the cooling power values, water spray frequency parameters, and water spray intervals for each zone are encoded according to the device communication protocol. A control data packet containing the zone coordinates, power setpoints, and water spray timing is generated and sent to the corresponding refrigeration equipment execution end via the Internet of Things platform. Real-time monitoring data of the ice surface after the control data packet is executed is obtained, including temperature distribution, friction coefficient, and crystal microscopic images. The measured temperature gradient values ​​at each monitoring point are calculated. The degree of reconstruction is determined by the crystal shape factor. The measured values ​​are then compared with the preset performance standard values ​​to obtain a performance deviation dataset. Based on this performance deviation dataset, when the temperature gradient deviation exceeds a preset threshold, a new cooling power value is calculated according to the linear relationship between the deviation value and the power adjustment amount. A frequency correction value is also determined based on the corresponding relationship between the crystal reconstruction degree deviation and the water spray frequency. The control data for the corresponding zone is updated using these new cooling power values ​​and frequency correction values. A control instruction packet containing the updated parameters is re-encoded and sent for execution via the Internet of Things platform, completing the feedback-based parameter iterative update process.

[0050] For example, the encoding process of the control data packet plays a bridging role in IoT communication.

[0051] In one possible implementation, each data packet uses a fixed-format byte sequence consisting of four main components: a header identifier, zone coordinates, power value, and timing parameters. Zone coordinates are numbered on a two-dimensional grid; for example, the zone code for row 3, column 5 is 0305. The cooling power value is expressed as a percentage, ranging from 0 to 100, representing the proportion of the device's maximum power. The water spray timing parameters, including the start time and interval duration, are encoded in seconds. This standardized encoding ensures that refrigeration equipment from different manufacturers can correctly interpret and execute control commands.

[0052] It's important to note that the choice of device communication protocol directly impacts control efficiency. The commonly used MQTT protocol, with its lightweight and low latency, is particularly well-suited for scenarios like ice rinks, which require real-time responses. When the control center generates a power adjustment command, the refrigeration equipment in the relevant area receives it within milliseconds through MQTT's publish-subscribe mechanism. Each device has a unique topic identifier, ensuring that the command is accurately delivered to the target device.

[0053] Specifically, the calculation of performance deviations requires the establishment of a clear benchmark. The standard value of the temperature gradient is usually set at 0.5 degrees Celsius per meter, which is determined based on the thermal conductivity characteristics of an ideal ice surface. When the measured temperature gradient reaches 1.2 degrees Celsius per meter, the deviation value is 0.7. The standard value of the crystal shape factor is about 14. When the measured value reaches 18, it indicates that the crystal structure has changed significantly. These deviation data form the quantitative basis for subsequent parameter adjustments. The linear relationship between the deviation value and the power adjustment amount is derived from the principles of thermodynamics. Experimental data show that for every 0.1 degree Celsius per meter increase in temperature gradient, the cooling power needs to be increased by about 5% to compensate. This linear relationship remains stable within a certain range, allowing the control algorithm to quickly calculate the required power adjustment amount.

[0054] For example, when the temperature gradient deviation is 0.7, the power needs to be increased by 35%. This adjustment method based on physical laws avoids complex nonlinear calculations and improves the system's response speed.

[0055] In one embodiment, the correspondence between the degree of crystal reconstruction and the water spray frequency is obtained through experimental calibration. When the crystal shape factor deviation reaches 2, the water spray frequency needs to be increased by 20%; when the deviation reaches 4, the frequency is increased by 40%. This step-by-step correspondence ensures the accuracy of the adjustment while avoiding too frequent parameter changes. The calculation of the frequency correction value also needs to take into account the current base frequency to avoid the adjusted frequency exceeding the operating range of the device. The parameter iterative update process forms a complete closed-loop control. After each parameter adjustment, the system will continuously monitor the changing trend of the ice surface state. If the effect after adjustment is still not up to standard, the system will recalculate the adjustment amount based on the new deviation data to form a continuous optimization process. This iterative mechanism ensures that the ice surface performance can gradually approach the target state and achieve precise control in a dynamic environment.

[0056] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for optimizing energy consumption of an ice-making system in an ice rink based on the Internet of Things, characterized in that: include: Obtain ice surface temperature distribution, friction force distribution, and crystal microscopic image data to construct an ice surface state dataset and identify local ice crystal reconstruction areas caused by frictional heating; performing feature extraction on crystal microscopic image data of the local ice crystal reconstruction region, and determining the position and range of the local ice crystal reconstruction region according to the extracted features; Calculating the temperature gradient distribution based on the position and range of the local ice crystal reconstruction area and combining it with the ice surface temperature distribution data, analyzing the abnormal area boundary of the temperature gradient distribution, and extracting the spatiotemporal distribution characteristics of the temperature gradient at the abnormal area boundary; Determine the correlation between low friction coefficient and crystal structure adjustment based on the temporal and spatial distribution characteristics of the temperature gradient and the friction distribution in the ice surface state data set, and generate a cooling power allocation plan based on the correlation; According to the cooling power distribution scheme and in combination with the friction force distribution, the influence of frictional heat generation on the crystal structure adjustment is analyzed, the water spray frequency adjustment parameters are determined, and the water spray frequency distribution is generated; A refrigeration system control instruction is generated according to the water spray frequency distribution and the refrigeration power allocation scheme.

2. The method for optimizing energy consumption of an ice-making system in an ice rink based on the Internet of Things according to claim 1, characterized in that: The method of obtaining ice surface temperature distribution, friction force distribution, and crystal microscopic image data, constructing an ice surface state data set, and identifying local ice crystal reconstruction areas caused by frictional heat generation includes: Obtain temperature data from the ice rink's temperature sensor array from the IoT platform, and combine the spatial coordinates and timestamps to generate a temperature distribution dataset containing spatiotemporal information. Collecting the friction coefficient of the ice surface, combining it with the temperature data of the temperature distribution data set, calculating the product of the temperature gradient and the friction coefficient at adjacent points, and identifying frictional heating areas where the product exceeds a preset threshold; For the frictional heat generating area, obtaining a crystal microscopic image, identifying the crystal boundary, and extracting a set of pixel points where the crystal boundary changes; The proportion of reconstructed crystal pixels is calculated according to the distribution density of the pixel set, and an area where the proportion exceeds a preset threshold is identified as a local ice crystal reconstruction area.

3. The method for optimizing energy consumption of an ice-making system in an ice rink based on the Internet of Things according to claim 1, characterized in that: The extracting features from the crystal microscopic image data of the local ice crystal reconstruction area and determining the position and range of the local ice crystal reconstruction area according to the extracted features includes: Gray-scaling the crystal microscopic image of the local ice crystal reconstruction area, calculating pixel grayscale differences to identify crystal edges, and generating a crystal edge coordinate sequence; Calculating the crystal perimeter and area based on the crystal edge coordinate sequence, generating a shape factor, extracting grain size parameters, and identifying reconstructed crystals whose shape factors exceed a preset threshold; Calculating the geometric center according to the coordinates of the reconstructed crystal, identifying the spatial aggregation region of the reconstructed crystal, and generating the aggregation region boundary; The boundary range is calculated according to the coordinates of the boundary of the aggregation area, and the position and coverage range data of the local ice crystal reconstruction area are output.

4. The method for optimizing energy consumption of an ice-making system in an ice rink based on the Internet of Things according to claim 1, wherein: The method of calculating the temperature gradient distribution based on the position and range of the local ice crystal reconstruction area and combining the ice surface temperature distribution data, analyzing the abnormal area boundary of the temperature gradient distribution, and extracting the spatiotemporal distribution characteristics of the temperature gradient at the abnormal area boundary includes: According to the position and range of the local ice crystal reconstruction area, the temperature data of the corresponding area is extracted, the ratio of the temperature difference and the distance between adjacent points is calculated, and the temperature gradient field data is generated; According to the temperature gradient field data, identifying abnormal points where the gradient value exceeds the reference value, and connecting the abnormal points to generate the abnormal area boundary; According to the gradient value of the boundary of the abnormal area, the gradient change rate and the abnormal duration are calculated to generate a feature data set including coordinates, gradient change rate and duration.

5. The method for optimizing energy consumption of ice-making systems in ice rinks based on the Internet of Things according to claim 1, characterized in that: The determining of the correlation between the low friction coefficient and the crystal structure adjustment based on the temporal and spatial distribution characteristics of the temperature gradient and the friction distribution in the ice surface state data set, and generating a cooling power allocation plan based on the correlation, includes: Extracting the friction coefficient of the corresponding area based on the temporal and spatial distribution characteristics of the temperature gradient, calculating the deviation of the friction coefficient from a preset value, and generating friction performance deviation data; Extracting a crystal microscopic image of a corresponding region based on the friction performance deviation data, calculating a ratio of a crystal shape factor to a standard value, generating crystal deformation degree data, and constructing a mapping relationship between the friction coefficient deviation and the crystal deformation degree; According to the mapping relationship, identifying the area where the friction coefficient exceeds a preset threshold, and calculating the temperature adjustment amount corresponding to the degree of crystal deformation; According to the temperature adjustment amount and the regional location, the heat change value is calculated and a power allocation plan for each refrigeration device is generated.

6. The method for optimizing energy consumption of ice-making systems in ice rinks based on the Internet of Things according to claim 1, characterized in that: The method of analyzing the influence of frictional heat generation on crystal structure adjustment based on the cooling power distribution scheme and the friction force distribution, determining the water spray frequency adjustment parameters, and generating the water spray frequency distribution includes: According to the cooling power distribution scheme, combined with the friction coefficient of the friction force distribution, the friction heat power is calculated to generate net heat flux density data; Calculating the water spraying amount and the initial water spraying frequency according to the net heat flux density data; Adjusting the initial water spraying frequency according to the temperature gradient distribution to generate adjusted water spraying frequency and interval parameters; A water spraying frequency distribution scheme including coordinates and water spraying parameters is generated according to the adjusted water spraying frequency and interval parameters.

7. The method for optimizing energy consumption of an ice-making system in an ice rink based on the Internet of Things according to claim 1, characterized in that: Generating a refrigeration system control instruction according to the water spray frequency distribution and the refrigeration power allocation scheme includes: Encoding and generating a control data packet including power and water spraying timing according to the water spraying frequency distribution and the refrigeration power allocation scheme; Based on the ice surface state data after execution, the deviation of the temperature gradient and the crystal reconstruction degree from the preset values ​​is calculated to generate performance deviation data; Calculating new cooling power and water spray frequency correction values ​​based on the performance deviation data; The control data packet is updated according to the new refrigeration power and water spray frequency correction values.

Images