Intelligent grinding temperature control method and system for powder coating

Abstract

The invention discloses an intelligent powder coating grinding temperature control method and system, and the method comprises the steps: obtaining a temperature signal and a raw material state parameter of a grinding machine, and forming a temperature control reference strategy through correlation mapping; identifying a temperature abnormal interval, performing multi-point temperature acquisition in the abnormal interval to form a temperature rise characteristic sequence, and constructing a cooling control criterion based on the temperature rise characteristic sequence; according to a cooling control criterion, performing temperature zone division to form a high-temperature zone and a steady-state temperature control zone, monitoring transition from the steady-state temperature control zone to the high-temperature zone to obtain a temperature rise rate, and determining cooling response strength through the temperature rise rate; selecting an optimal cooling opportunity based on the cooling response strength, generating a grinding strength adjusting parameter by combining the correlation between the temperature rise characteristic sequence and the high-temperature region, and cooperatively calibrating and configuring a dynamic cooling threshold value and a grinding linkage coefficient; and determining a caking induction condition by comparing the dynamic cooling threshold value with the temperature rise characteristic sequence, executing cooling air regulation and control and grinding strength adjustment, and outputting a temperature control instruction to complete self-adaptive process control of the grinding temperature.

Description

Technical Field

[0001] This invention relates to the field of powder coating processing technology, and in particular to an intelligent powder coating grinding temperature control method and system. Background Technology

[0002] In the production of powder coatings, the raw materials need to be ground to a specified particle size range through a grinding process. During the grinding process, the friction between the grinding disc and the material continuously generates heat. Powder coating raw materials are sensitive to temperature. When the temperature of the grinding chamber exceeds the glass transition temperature of the raw materials, the powder particles are prone to softening, sticking together, or even premature solidification, leading to product agglomeration and equipment blockage and shutdown. Temperature control is a key link in ensuring the quality of the grinding process.

[0003] However, existing grinding temperature control schemes mostly employ fixed threshold-triggered cooling, resulting in single temperature monitoring points and low sampling frequencies, making it difficult to promptly capture temperature gradient changes and localized hotspot formation within the grinding chamber. Furthermore, the cooling response lacks correlation with raw material characteristics; using the same cooling strategy for raw materials with different glass transition temperatures leads to over-cooling of some materials, impacting production efficiency, while insufficient cooling of others causes agglomeration. Simultaneously, cooling control operates independently of the grinding process; the grinding intensity is not adjusted synchronously upon cooling initiation, making it difficult to establish a dynamic balance between heat generation and heat dissipation, resulting in recurring temperature fluctuations. Summary of the Invention

[0004] This invention discloses an intelligent grinding temperature control method for powder coatings, aiming to solve problems such as lagging temperature monitoring, disconnect between cooling response and raw material characteristics, and lack of linkage between cooling control and grinding process during grinding. By acquiring multiple temperature points and extracting temperature rise characteristics, the dynamic change law of the grinding cavity temperature field is grasped. Combined with the thermally sensitive properties of raw materials, the temperature difference limit and cooling trigger threshold are dynamically adjusted, thereby establishing a linkage regulation mechanism between cooling response and grinding intensity, ultimately achieving precise control of grinding temperature and effective prevention of powder agglomeration.

[0005] The first aspect of this invention proposes a method for intelligent grinding temperature control of powder coatings, comprising the following steps: Acquire the temperature signal and raw material state parameters of the grinding mill, and perform correlation mapping between the temperature signal and the raw material state parameters to form a temperature control reference strategy; The temperature control benchmark strategy is used to identify abnormal temperature ranges, and multiple temperature data are collected in the abnormal temperature ranges to form a temperature rise feature sequence. Cooling control criteria are then constructed based on the temperature rise feature sequence. Based on the cooling control criteria, the temperature zone is divided into a high-temperature zone and a steady-state temperature control domain. The transition from the steady-state temperature control domain to the high-temperature zone is monitored to obtain the temperature rise rate. The cooling response intensity is determined by the temperature rise rate. Based on the cooling response intensity, the optimal cooling time is selected in the high-temperature zone. The grinding intensity adjustment parameter is generated according to the correlation between the temperature rise characteristic sequence and the high-temperature zone. The dynamic cooling threshold and grinding linkage coefficient are configured through the coordinated calibration of the optimal cooling time and the grinding intensity adjustment parameter. The agglomeration induction conditions are determined by comparing the dynamic cooling threshold with the temperature rise characteristic sequence. Cooling air regulation is performed to form a cooling regulation amount based on the agglomeration induction conditions and the cooling control criteria. The grinding mill operation intensity is adjusted according to the grinding linkage coefficient to form a grinding regulation amount. A temperature control command is output based on the cooling regulation amount and the grinding regulation amount to complete the process control of grinding temperature.

[0006] A second aspect of this invention provides an intelligent powder coating grinding temperature control system, comprising: The signal acquisition module is used to acquire the temperature signal and raw material state parameters of the grinding mill, and to correlate and map the temperature signal and the raw material state parameters to form a temperature control reference strategy. Anomaly identification module is used to identify abnormal temperature ranges through the temperature control benchmark strategy, collect multiple temperatures in the abnormal temperature range to form a temperature rise feature sequence, and construct a cooling control criterion based on the temperature rise feature sequence. The temperature zone division module is used to divide the temperature zone into a high-temperature zone and a steady-state temperature control domain according to the cooling control criteria, monitor the transition from the steady-state temperature control domain to the high-temperature zone to obtain the temperature rise rate, and establish the cooling response intensity through the temperature rise rate. The threshold configuration module is used to select the optimal cooling time in the high-temperature zone based on the cooling response intensity, generate grinding intensity adjustment parameters according to the correlation between the temperature rise characteristic sequence and the high-temperature zone, and configure dynamic cooling threshold and grinding linkage coefficient through the coordinated calibration of the optimal cooling time and the grinding intensity adjustment parameters. The instruction output module is used to determine the agglomeration induction conditions by comparing the dynamic cooling threshold with the temperature rise characteristic sequence, execute cooling air regulation to form a cooling regulation amount according to the agglomeration induction conditions and the cooling control criteria, adjust the grinding mill operating intensity to form a grinding regulation amount according to the grinding linkage coefficient, and output a temperature control instruction based on the cooling regulation amount and the grinding regulation amount to complete the process control of grinding temperature.

[0007] The beneficial effects of this invention are reflected in the following points: First, when mapping the temperature signal to the raw material state parameters, the temperature threshold and response sensitivity of each monitoring point are determined based on the glass transition temperature and thermal conductivity of the raw material, and different types of raw materials correspond to different temperature control benchmark strategies; after identifying the abnormal temperature range, multiple points are arranged and synchronously collected in the grinding cavity, and the temperature-time curves of each temperature measuring point are analyzed by gradient to extract features such as temperature rise rate and temperature rise acceleration. The construction of the temperature rise feature sequence enables the quantification and characterization of the spatiotemporal law of temperature change. Second, the temperature zone division divides the grinding cavity temperature range into two major regions: the steady-state temperature control domain and the high-temperature zone. The temperature rise rate reflects the speed at which the temperature migrates from the steady-state temperature control domain to the high-temperature zone. The mapping relationship between the cooling response intensity and the temperature rise rate enables the cooling power to be dynamically adjusted according to the temperature rise conditions; the critical temperature proximity converts the temperature rise rate into an indicator of the urgency of the distance to the critical point, and the intensity correction value increases in segments according to the proximity to ensure stronger cooling intervention during high-risk periods. Finally, the selection of the optimal cooling time takes into account both cooling effect and energy efficiency. The grinding intensity adjustment parameter is generated based on the correlation strength between the temperature rise characteristic sequence and the high temperature zone. The coordinated calibration of the dynamic cooling threshold and the grinding linkage coefficient enables the cooling system and the grinding system to cooperate rather than operate independently. The agglomeration induction conditions are determined by the aggregation detection of over-limit records and the verification of risk attributes. The cooling control amount and the grinding control amount are integrated into a unified temperature control command for execution. The grinding temperature is kept within a safe range under closed-loop regulation. Attached Figure Description

[0008] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0009] Figure 1 This is a schematic flowchart of an intelligent grinding temperature control method for powder coatings according to the present invention.

[0010] Figure 2 This is a schematic diagram of the grinding cavity temperature monitoring layout of the present invention.

[0011] Figure 3 This is a schematic diagram of the hardware connection of the temperature control system of the present invention.

[0012] Figure 4 This is a structural block diagram of an intelligent grinding temperature control system for powder coatings according to the present invention.

[0013] Wherein: 1-Grinding chamber; 2-Grinding disc; 3-Grading wheel; 4-Feed inlet; 5-Discharge outlet; 6-Temperature measuring point at the center of the grinding disc; 7-Temperature measuring point at the edge of the grinding disc; 8-Temperature measuring point on the side wall; 9-Temperature measuring point on the grader wheel; 10-Steady-state temperature control zone; 11-High-temperature zone; 12-Transition monitoring zone; 13-Grinding mill; 14-Cooling fan; 15-Temperature sensor group; 16-Frequency converter; 17-Data acquisition card; 18-Control computer; 19-Temperature signal; 20-Control signal; 21-Cooling airflow. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0015] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0016] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0017] The technical solutions of the embodiments of this application will be described below.

[0018] like Figure 1 As shown, this embodiment of the invention provides a method for intelligent grinding temperature control of powder coatings, including the following steps S110-S150: Step S110: Obtain the temperature signal and raw material state parameters of the grinding mill, and perform correlation mapping between the temperature signal and the raw material state parameters to form a temperature control reference strategy.

[0019] Specifically, the temperature signal and raw material state parameters of the grinding mill are acquired. During the grinding of powder coatings, the friction between the grinding disc and the material continuously generates heat. This accumulated heat causes the grinding chamber temperature to rise. Excessively high temperatures can cause the powder coating to soften, clump, or even solidify prematurely, severely impacting product quality and production efficiency. Temperature sensors are installed at key locations in the grinding mill, including the grinding chamber inlet, grinding disc surface, discharge port, and classifying wheel area. Each sensor collects the temperature value at its location in real time and uploads it to the control system, forming a temperature signal data stream. The temperature signal is sampled at fixed time intervals, with the sampling period set according to the grinding process requirements, typically on the order of seconds or sub-seconds to capture rapid temperature changes. The sampled data includes three fields: timestamp, sensor number, and temperature value. The timestamp is used for the temporal location of subsequent temperature anomaly intervals, and the sensor number is used for spatial location association when collecting temperature data at multiple points. Simultaneously, the physicochemical properties of the current batch of raw materials are acquired as raw material state parameters. These parameters include properties such as the glass transition temperature (Tg), melting point (Tm), thermal conductivity (λ), moisture content, and resin type. The glass transition temperature of epoxy powder coatings is typically between 50 and 60 degrees Celsius, while that of polyester powder coatings can reach 65 to 75 degrees Celsius. Different types of raw materials exhibit varying degrees of temperature sensitivity. These raw material state parameters are obtained before the start of each production batch through online monitoring or formulation database queries, and together with the temperature signal, constitute the input data for temperature control decisions.

[0020] A temperature control baseline strategy is formed by mapping temperature signals to raw material state parameters. Temperature signals reflect the current thermodynamic state of the grinding mill, while raw material state parameters determine the raw material's tolerance to temperature changes; there is an inherent correlation between the two. By analyzing the numerical range and variation characteristics of the temperature signals, and combining this with the glass transition temperature (Tg) in the raw material state parameters, the upper limit of the acceptable temperature (Tmax) for the current raw material during grinding is determined. The calculation formula is Tmax = Tg - ΔTs, where ΔTs is a safety margin, typically 5 to 10 degrees Celsius. For epoxy raw materials, a Tg of 55 degrees Celsius corresponds to a Tmax of approximately 45 to 50 degrees Celsius. The priority of the early warning response is adjusted based on the resin type and moisture content in the raw material state parameters. Raw materials with a moisture content higher than 2% are prone to bubble formation at high temperatures, requiring earlier warning triggers. Thermosetting resins have a narrower temperature tolerance than thermoplastic resins, requiring stricter threshold settings. The mapping rules between each monitoring point of the temperature signal and the raw material state parameters are integrated to form a temperature control baseline strategy for the current production batch. The temperature control baseline strategy includes parameters such as the temperature threshold Tmax, the baseline temperature Tbase, and the safety margin ΔTs for each monitoring point. The temperature threshold of the monitoring point on the grinding disc surface is calculated based on the glass transition temperature. The temperature threshold of the monitoring point at the feed inlet is reduced by 5 degrees Celsius to allow for preheating space. The baseline temperature Tbase is set to 70% of the glass transition temperature as a reference for normal operation.

[0021] Step S120: Identify abnormal temperature ranges through a temperature control benchmark strategy, collect multiple temperatures within the abnormal temperature ranges to form a temperature rise feature sequence, and construct cooling control criteria based on the temperature rise feature sequence.

[0022] In some embodiments, identifying abnormal temperature ranges through the temperature control benchmark strategy includes: determining a temperature deviation index and raw material compatibility parameters based on the temperature control benchmark strategy; adaptively adjusting a preset temperature difference limit according to the raw material compatibility parameters to obtain a dynamic temperature difference limit; defining suspected abnormal segments by comparing the temperature deviation index with the dynamic temperature difference limit; and setting a safety boundary to limit the abnormal temperature range based on the suspected abnormal segments.

[0023] Temperature deviation indicators and raw material adaptation parameters are determined based on a temperature control baseline strategy. The temperature control baseline strategy includes baseline value settings for temperature monitoring points and a parameterized description of the thermal sensitivity characteristics of the raw materials, from which key indicators for anomaly identification are extracted. The temperature deviation indicator characterizes the degree of deviation of the current temperature from the baseline value, calculated using the formula D=(Tcur-Tbase) / Tbase, where Tcur is the current temperature value, Tbase is the baseline temperature value set in the temperature control baseline strategy, and D is the dimensionless relative deviation. The baseline temperature value Tbase for each monitoring point in the temperature control baseline strategy is set according to the raw material characteristics and process requirements. The baseline temperature value for the grinding disc surface is set to 70% of the glass transition temperature; the baseline temperature value for the feed inlet is set to the ambient temperature plus a process temperature rise of 15 degrees Celsius. Thermal properties directly related to dynamic temperature difference adjustment are extracted from the temperature control baseline strategy as raw material adaptation parameters. These parameters focus on two core attributes: glass transition temperature (Tg) and thermal conductivity (λ). Tg determines the softening critical point of the raw material, directly affecting the stringency of the temperature difference limit, while λ affects the rate of heat diffusion in the material, thus influencing the probability of local hotspot formation. The temperature deviation index (D) and the raw material adaptation parameters together constitute the input elements for anomaly identification. The temperature deviation index reflects the actual temperature deviation, while the raw material adaptation parameters determine the raw material's tolerance to deviation; a lower Tg value indicates a lower tolerance. The two are combined for anomaly detection.

[0024] Dynamic temperature difference limits are obtained by adaptively adjusting the preset temperature difference limits based on raw material compatibility parameters. Different types of powder coating raw materials have significantly different sensitivities to temperature changes, making it difficult to adapt to the diversity of raw materials using fixed temperature difference limits. Therefore, dynamic adjustments are made based on raw material compatibility parameters. The lower the glass transition temperature (Tg) in the raw material compatibility parameters, the easier the raw material is to soften upon heating, and the corresponding temperature difference limit should be tightened; the higher the glass transition temperature (Tg), the better the heat resistance of the raw material, and the temperature difference limit can be appropriately relaxed. The baseline temperature difference limit Dbase is set to 0.15 and adjusted according to Tg and λ in the raw material compatibility parameters. The adjustment formula is Ddyn = Dbase × (Tg / Tg0) × (λ / λ0)^0.5, where Tg0 = 60 degrees Celsius is the baseline glass transition temperature, and λ0 = 0.2 W / (m·K) is the baseline thermal conductivity. The Tg of epoxy powder coatings is approximately 55 degrees Celsius. Substituting this into the formula, the calculated dynamic temperature difference limit is approximately 0.14, which is about 7% tighter than the baseline temperature difference limit. The Tg of polyester powder coatings is approximately 70 degrees Celsius, and the calculated dynamic temperature difference limit is approximately 0.18, which is about 20% wider than the baseline temperature difference limit. The thermal conductivity λ in the raw material compatibility parameters affects the rate of heat transfer in the material. Raw materials with high thermal conductivity diffuse heat quickly and are less prone to forming localized hot spots. This effect is reflected in the formula through the square root of λ.

[0025] Suspected abnormal zones are identified by comparing temperature deviation indicators with dynamic temperature difference limits. The time series of the temperature deviation indicator D at each monitoring point is compared with the corresponding dynamic temperature difference limit Ddyn to determine whether the temperature deviation at each moment exceeds the allowable range. A temperature deviation indicator D less than the dynamic temperature difference limit Ddyn is considered normal; a temperature deviation indicator D greater than or equal to the dynamic temperature difference limit Ddyn is considered a suspected abnormal state. The comparison results are scanned along the time axis, and time periods continuously in a suspected abnormal state are marked; these time periods are the suspected abnormal zones. The start point tstart of a suspected abnormal zone is the moment when the temperature deviation indicator D first exceeds the dynamic temperature difference limit Ddyn, and the end point tend is the moment when D falls back below Ddyn. The duration of the zone is the difference between tend and tstart. The duration and deviation magnitude of the suspected abnormal zones are statistically analyzed. The deviation magnitude is the ratio of the maximum value of the temperature deviation indicator D within the zone to the dynamic temperature difference limit Ddyn. A longer duration and a larger deviation magnitude indicate a more severe temperature anomaly. Suspected abnormal segments may briefly deviate and fall back before exceeding the threshold again. If the interval between two abnormalities is less than the set merging threshold (usually 3 to 5 seconds), they are merged into a continuous suspected abnormal segment for processing, thus avoiding excessive fragmentation of abnormal segments that may affect analysis efficiency.

[0026] Safety boundaries are established to define temperature anomaly ranges based on suspected anomaly zones. Suspected anomaly zones identify the time range within which temperatures deviate beyond the limit. To ensure the effectiveness of cooling intervention, the temperature anomaly range is appropriately extended forward and backward from the suspected anomaly zone to form a temperature anomaly range encompassing the safety boundary. The forward-extended safety boundary covers the stage where the temperature begins to rise but has not yet exceeded the limit, allowing cooling intervention to be initiated before the temperature becomes completely out of control. The backward-extended safety boundary covers the stage where the temperature has fallen but may rebound again, ensuring that the cooling effect is fully consolidated. The extent of the safety boundary extension is determined based on the severity of the suspected anomaly zone, which is assessed by a combination of deviation magnitude and duration. Zones with a deviation magnitude greater than 1.5 times the dynamic temperature difference limit or a duration greater than 30 seconds are classified as high severity. For high severity zones, the forward extension is 30% of the zone duration, and the backward extension is 40% of the zone duration; for low severity zones, the forward extension is 20% of the zone duration, and the backward extension is 30% of the zone duration. The time range of suspected abnormal sections is added with the preceding and following safety boundaries to form the temperature abnormality interval. The temperature abnormality interval is recorded in the form of start and end timestamps, and is associated with attribute information such as the location of the monitoring point where the abnormality occurred, the abnormality level, and the triggering cause. The identification results of the temperature abnormality interval are output to the multi-point temperature acquisition stage for refined monitoring.

[0027] In some embodiments, the step of collecting multiple temperatures in the abnormal temperature range to form a temperature rise feature sequence includes: performing multi-point arrangement of the grinding cavity in the abnormal temperature range to obtain a measurement point distribution map; performing synchronous temperature reading on the measurement point distribution map to obtain an instantaneous temperature distribution; extending along the time direction of the instantaneous temperature distribution to collect continuous temperature change data; and constructing a temperature rise feature sequence based on the gradient characterization of the continuous temperature change data.

[0028] A multi-point measurement point distribution map was obtained by implementing a measurement point layout within the grinding cavity for the temperature anomaly range. The identification of the temperature anomaly range determined the key monitoring time range. Within this time range, refined data collection of the temperature distribution within the grinding cavity was performed. First, a plan for the layout of temperature measurement points within the grinding cavity space was planned. For example... Figure 2 As shown, the grinding chamber 1 has a cylindrical geometry. The grinding disc 2 is located at the bottom and performs the grinding function, while the classifying wheel 3 is located at the top and is responsible for particle size separation. The material enters through the feed inlet 4 and flows and grinds between the grinding disc 2 and the classifying wheel 3. The ground fine powder is discharged from the discharge outlet 5. The arrangement of temperature measuring points covers the key heat source areas and heat conduction paths of the grinding chamber 1. Temperature measuring points 6 at the center of the grinding disc 2 and 7 at the edge of the grinding disc are arranged on the surface of the grinding disc 2. The temperature measuring point 6 at the center of the grinding disc is located in the area of ​​most intense friction and is used to capture the temperature peak. The temperature measuring point 7 at the edge of the grinding disc monitors the radial heat conduction. Multiple sidewall temperature measuring points 8 are arranged along the height direction on the sidewall of the grinding chamber 1 to monitor the temperature gradient of heat transfer from the grinding disc 2 upwards. Temperature measuring points 9 of the classifying wheel are arranged in the area of ​​the classifying wheel 3 to monitor the fine powder outlet temperature. The temperature range inside the grinding chamber 1 is divided into two major regions: a steady-state temperature control zone 10 and a high-temperature zone 11. A transition monitoring zone 12 is set between the two regions to monitor the process of temperature migration from the steady-state temperature control zone 10 to the high-temperature zone 11. The spatial coordinates and numbering information of each temperature measuring point are organized to form a measuring point distribution map. The location of each temperature measuring point is marked on the measuring point distribution map with the coordinate system of the grinding cavity 1 column as a reference.

[0029] Synchronous temperature readings are performed on the measurement point distribution map to obtain the instantaneous temperature distribution. Temperature sensors at each measurement point on the distribution map take readings simultaneously to obtain an instantaneous snapshot of the grinding cavity temperature field. Synchronization accuracy is required to be within 10 milliseconds to ensure the time consistency of data from all measurement points. Synchronous reading uses a unified clock trigger mechanism. The control system sends acquisition commands to all temperature sensors simultaneously. Upon receiving the command, each sensor immediately performs temperature measurement and returns a value, including the sensor number, temperature value, and acquisition timestamp. The measurement point located at the center of the grinding disc on the distribution map reads the temperature value at that location at the current moment, which is typically the highest temperature inside the grinding cavity, fluctuating around the glass transition temperature. Measurement points located on the sidewalls of the grinding cavity read the temperature values ​​at their corresponding heights; the temperature on the bottom sidewall is approximately 10 to 15 degrees Celsius higher than that on the top sidewall. The synchronous readings of each temperature measuring point are organized according to the spatial coordinates of the measuring point distribution map to form an instantaneous temperature distribution data structure. The instantaneous temperature distribution can be represented as a discrete sampling matrix of a three-dimensional temperature field, or as a temperature value vector indexed by the temperature measuring point number. This intuitively presents the spatial distribution pattern of heat in the grinding cavity. Areas with higher temperature values ​​correspond to areas with concentrated heat, while areas with lower temperature values ​​correspond to areas with good heat dissipation.

[0030] For example, the step of collecting continuous temperature change data along the time direction of the instantaneous temperature distribution includes: performing time-series correlation analysis on the instantaneous temperature distribution to obtain correlation coefficient records; using the correlation coefficient records to perform peak identification to determine the peak interval duration; performing stability verification on the peak interval duration to form periodic stability; and performing dynamic filtering on the peak interval duration based on the periodic stability to obtain continuous temperature change data.

[0031] A time-series correlation analysis is performed on instantaneous temperature distributions to obtain correlation coefficient records. Instantaneous temperature distributions, continuously collected over time, form a time series of temperature fields. There is a certain correlation between temperature distributions at adjacent moments; analyzing this correlation helps identify the inherent patterns and abnormal fluctuations in temperature changes. Instantaneous temperature distribution data from two consecutive moments are taken, and the temperature values ​​at each measurement point are combined into a vector. The Pearson correlation coefficient between the two vectors is calculated. A correlation coefficient close to 1 indicates that the temperature distributions at two moments are highly similar, and the temperature field changes smoothly; a significantly decreased correlation coefficient indicates a significant change in temperature distribution, possibly involving local temperature abrupt changes or an overall accelerated temperature rise. The distribution correlation coefficient between adjacent instantaneous temperature distributions is calculated by sliding along the time axis, forming a time series of distribution correlation coefficients. Under normal operating conditions, the distribution correlation coefficient between adjacent moments typically remains above 0.95. When the distribution correlation coefficient drops below 0.90, it indicates a significant change in temperature distribution. The distribution correlation coefficients at each moment are organized to form a correlation coefficient record. Moments showing a significant decrease in the distribution correlation coefficient in the correlation coefficient record usually correspond to moments of temperature abrupt changes. These moments are candidate points for peak identification and also support the identification of periodic patterns and the detection of abnormal fluctuations.

[0032] Peak interval durations were determined by using correlation coefficient records for peak identification. The fluctuations in the distributed correlation coefficients within the records reflect the rhythmic characteristics of temperature changes. Identifying the peaks and troughs of the distributed correlation coefficients allows for the capture of periodic patterns in temperature changes. Factors such as grinding disc rotation, feeding intervals, and cooling cycles during mill operation can all cause periodic temperature fluctuations. A grinding disc rotation speed of 300 rpm corresponds to a rotation period of 0.2 seconds. This periodicity helps distinguish between normal fluctuations and abnormal trends. A peak detection algorithm was applied to the time series of distributed correlation coefficients in the correlation coefficient records. The local extremum method was used to identify the moments when the distributed correlation coefficients reached local maxima. The sensitivity threshold for peak detection was set to 95% of the mean of the distributed correlation coefficients. The time difference between adjacent peak moments was calculated to obtain the peak interval duration sequence, which reflects the length of the temperature change cycle. At a grinding disc rotation speed of 300 rpm, the temperature fluctuation cycle is approximately 0.2 seconds, corresponding to a peak interval duration of approximately 200 milliseconds; at a feeding cycle of 10 seconds, the corresponding peak interval duration is approximately 10,000 milliseconds. The time intervals between all adjacent peaks extracted from the correlation coefficient records constitute a peak interval duration list, and the numerical distribution characteristics in the list reflect the stability of the temperature change cycle.

[0033] Stability verification is performed on the peak interval durations to determine periodic stability. The consistency of values ​​in the peak interval duration list reflects the stability of the temperature change cycle. A stable cycle indicates that temperature fluctuations are within a controllable and normal range, while an unstable cycle may indicate abnormal operating conditions or measurement interference. Statistical characteristics of the peak interval duration list are calculated, including the mean, standard deviation, coefficient of variation, and range. A smaller coefficient of variation indicates greater periodic stability; a coefficient of variation less than 0.1 is considered highly stable, between 0.1 and 0.3 is considered moderately stable, and a coefficient of variation greater than or equal to 0.3 is considered unstable. A sliding window analysis is performed on the peak interval duration list, with the window size set to 10 consecutive peak interval duration values. The coefficient of variation for each value within the window is calculated. The coefficient of variation for each window is converted into a periodic stability index; a smaller coefficient of variation indicates periodic stability closer to 1, and a larger coefficient of variation indicates periodic stability closer to 0. The periodic stability series is aligned with the temperature time series. Data periods with high periodic stability values ​​are less affected by disturbances, while data periods with low periodic stability values ​​may contain noise or anomalies.

[0034] Continuous temperature variation data is obtained by dynamically filtering peak interval duration based on periodic stability. Periodic stability indicates the reliability of temperature data in each time period. Periods with high stability have better data quality and are suitable for inclusion in the analysis, while periods with low stability may be disturbed and require careful handling or removal. A periodic stability screening threshold is set. Periods with stability above the threshold are considered valid data segments, while those with stability below the threshold are considered suspicious data segments. The threshold is set according to data quality requirements, typically between 0.7 and 0.8. Temperature data collected for periods with periodic stability above 0.8 are directly included in the valid dataset and marked with high confidence. Data for periods with periodic stability between 0.6 and 0.8 are included in the valid dataset but marked with medium confidence. Data for periods with periodic stability below 0.6 are marked as suspicious data and require further verification. For suspicious data segments, the degree of anomaly in the peak interval duration is considered to determine whether to remove them or perform interpolation repair. If the peak interval duration of a period deviates from the mean by more than 3 times the standard deviation, it is identified as an outlier and removed. After removal, linear interpolation is performed using adjacent valid data points to fill the gap. The valid data retained after periodic stability screening constitutes continuous temperature change data. The continuous temperature change data eliminates interference data from unstable periods and retains high-quality data with clear and identifiable temperature change patterns. The continuous temperature change data stores the temperature values ​​of each temperature measurement point using timestamps as indexes, and also labels the periodic stability of each data point as a confidence level reference. The typical valid data retention rate is between 85% and 95%.

[0035] The gradient representation based on continuous temperature change data constitutes a temperature rise characteristic sequence. The continuous temperature change data includes temperature-time curves for each measurement point. Gradient analysis of these curves extracts the rate characteristics of temperature change. The temperature rise rate curve is obtained by calculating the first derivative of the temperature-time curve for each measurement point. The numerical calculation uses the central difference method. Positive values ​​on the temperature rise rate curve indicate temperature increase, while negative values ​​indicate temperature decrease; the absolute value reflects the speed of temperature change. Further calculation of the second derivative of the temperature rise rate curve yields the temperature rise acceleration curve, which reflects the trend of the temperature rise rate. Positive values ​​indicate that the temperature rise is accelerating, while negative values ​​indicate that the temperature rise is slowing down. A set of characteristic indicators for each measurement point is extracted from the continuous temperature change data, including the peak temperature rise rate, the peak temperature rise acceleration, and the time when the temperature reaches its peak. The peak temperature rise rate at the center measurement point of the grinding disc reaches 2.5 degrees Celsius per minute, the peak temperature rise acceleration reaches 0.5 degrees Celsius per minute squared, and the temperature reaches its peak approximately 180 seconds after the anomaly begins. These characteristic indicators quantitatively describe the temperature rise behavior at that location. The temperature rise characteristic indicators of each temperature measurement point are organized according to spatial coordinates and timestamps to form a temperature rise characteristic sequence. The temperature rise characteristic sequence stores a complete characteristic description of the temperature change of the grinding cavity within the temperature anomaly range in a structured form.

[0036] Cooling control criteria are constructed based on temperature rise characteristic sequences. These sequences reveal the spatiotemporal patterns of temperature changes within the grinding chamber, providing crucial information for developing effective cooling strategies. Analysis of the temperature rise rate distribution at each measurement point in the sequence identifies concentrated heat sources and heat conduction paths. The temperature rise rate at the contact surface between the grinding disc and the material typically reaches 2 to 3 degrees Celsius per minute, with heat conducted radially towards the edge and dissipated outwards through the sidewalls. Analysis of the peak temperature rise acceleration distribution at each measurement point reveals regions with larger acceleration peaks, indicating accelerating temperature rise; the cooling response should prioritize these regions. Based on the temperature rise rate *v* and the current temperature *T* in the sequence, the time window *tw* required to reach the upper temperature limit *Tmax* is calculated using the formula *tw=(Tmax-T) / v*. A shorter time window indicates a more urgent cooling need and requires stronger cooling intervention. Differentiated cooling response schemes are developed for different temperature rise characteristics. When the temperature rise rate is less than 1 degree Celsius per minute, low-power continuous cooling is used, with the power set to 30% to 50% of the rated power. When the temperature rise rate is greater than 2 degrees Celsius per minute, high-power pulse cooling is used, with the power set to 80% to 100% of the rated power. When the temperature of a local hot spot exceeds the surrounding area by more than 5 degrees Celsius, directional cooling is used, with the cooling airflow concentrated towards the hot spot area. The cooling response schemes are correlated with triggering conditions to form cooling control criteria. The cooling control criteria define the selection of cooling methods, cooling power settings, and cooling duration under different temperature rise characteristics in the form of a rule set. When multiple areas have cooling needs at the same time, they are prioritized according to the time window tw from smallest to largest, with the area with the smallest tw receiving priority in cooling resource allocation.

[0037] Step S130: Divide the temperature zone into a high-temperature zone and a steady-state temperature control zone according to the cooling control criteria, monitor the transition from the steady-state temperature control zone to the high-temperature zone to obtain the temperature rise rate, and establish the cooling response intensity through the temperature rise rate.

[0038] In some embodiments, the step of dividing the temperature zone into a high-temperature zone and a steady-state temperature control domain according to the cooling control criteria includes: performing interval segmentation on the temperature threshold in the cooling control criteria to obtain a temperature level standard; determining a temperature control boundary index based on the temperature level standard; dynamically correcting the temperature level standard based on the temperature control boundary index to obtain a temperature control intensity level; and delineating the steady-state temperature control domain and the high-temperature zone through the temperature control intensity level.

[0039] Temperature tier standards are obtained by dividing the temperature thresholds in the cooling control criteria into intervals. The cooling control criteria include multiple threshold parameters such as the upper temperature limit Tmax, the warning threshold Twarn, and the baseline temperature Tbase. These parameters divide the temperature range into intervals with different risk characteristics. Using the temperature thresholds in the cooling control criteria as dividing points, the complete range from ambient temperature to the upper temperature limit is divided into multiple tiered intervals. The first tier is the safe interval below Tbase, where the temperature is low and no cooling intervention is required. The first tier range for epoxy powder coatings is 25 to 35 degrees Celsius. The second tier is the normal operating interval from Tbase to Twarn, where the temperature is within the expected range and only routine monitoring is required. The second tier range for epoxy powder coatings is 35 to 42 degrees Celsius. The third tier is the warning interval from Twarn to Tmax, where the temperature is high and a cooling response needs to be initiated. The third tier range for epoxy powder coatings is 42 to 50 degrees Celsius. The fourth tier is the danger interval above Tmax, where the temperature has exceeded the limit and emergency cooling is required, and shutdown protection should be considered. The boundary values, level numbers, and risk levels of each temperature grading interval are compiled into a temperature grading standard. This standard records the lower and upper temperature limits and risk levels for each level in tabular form. For polyester powder coatings, the third level ranges from 57 to 65 degrees Celsius, and the fourth level ranges above 65 degrees Celsius. For epoxy powder coatings, the third level ranges from 42 to 50 degrees Celsius, and the fourth level ranges above 50 degrees Celsius. This temperature grading standard reflects the different temperature sensitivities of different raw materials.

[0040] Temperature control boundary indicators are determined based on temperature level standards. These standards define multiple temperature levels, and in actual temperature control processes, it is crucial to distinguish which levels fall under steady-state control and which fall under high-temperature warning. This distinction is defined by the temperature control boundary indicators. The risk levels of each level in the temperature level standards are analyzed to identify the critical levels where risk levels undergo significant changes. Levels one and two share the characteristic of requiring no active cooling or only low-intensity sustained cooling, thus falling under steady-state control. Levels three and four share the characteristic of requiring active cooling intervention, thus falling under high-temperature warning. The temperature control boundary indicator is set as the boundary temperature between levels two and three, i.e., the warning line (Twarn). When the temperature is below the temperature control boundary indicator, the system is in steady-state control mode; when the temperature is above the temperature control boundary indicator, the system switches to high-temperature warning mode. The setting of the temperature control threshold also takes into account the location characteristics of the monitoring point. The temperature control threshold at the center of the grinding disc can be appropriately lowered by 3 to 5 degrees Celsius to provide a greater safety margin, because the central area is the main heat source with the fastest temperature rise rate; the temperature control threshold at the discharge port can be appropriately raised because the material stays there for a short time and the heat accumulation is limited.

[0041] Temperature control intensity levels are obtained by dynamically adjusting the temperature level standard based on the temperature control boundary index. While the temperature level standard provides a static division of temperature ranges, the actual temperature control requirements during grinding fluctuate with changing operating conditions. Therefore, the level standard is dynamically adjusted based on the temperature control boundary index. When the grinding load increases, the heat generation rate accelerates, and the temperature control boundary index should be lowered to trigger the cooling response earlier; the index is lowered by 2 degrees Celsius when the load increases by 20%. When the ambient temperature rises, heat dissipation efficiency decreases, and the temperature control boundary index is also lowered; it is lowered by 1 degree Celsius for every 5 degrees Celsius increase in ambient temperature. When the cooling system is running at full load, the cooling margin is limited, and the temperature control boundary index should also be lowered to allow for response time. After adjusting the temperature control boundary index according to the current operating conditions, the boundary values ​​of each level in the temperature level standard are recalculated to obtain a dynamic level division adapted to the current operating conditions. The dynamically corrected temperature levels are correlated with their corresponding cooling response intensities to form temperature control intensity levels. These levels are identified by level numbers, indicating different temperature control states and response intensities. There are four temperature control intensity levels: E1 corresponds to the safe range where no cooling is required; E2 corresponds to the normal range where cooling is maintained; E3 corresponds to the warning range where active cooling is required; and E4 corresponds to the danger range where emergency cooling is required. The temperature control intensity levels directly link the temperature state with the cooling intensity.

[0042] The steady-state temperature control domain and high-temperature zone are defined by temperature control intensity levels. The four levels of temperature control intensity represent four different temperature control states, which are further consolidated into two major temperature zones to simplify control logic. Levels E1 and E2 are merged into the steady-state temperature control domain. These two levels share the characteristic that the temperature is within a controllable range, with low or no active cooling requirements; the system can operate stably within this region without frequent adjustments. Levels E3 and E4 are merged into the high-temperature zone. These two levels share the characteristic that the temperature has entered a risk zone, requiring active cooling intervention. The system continuously monitors and adjusts the cooling intensity according to temperature changes within this zone. The temperature range of the steady-state temperature control domain covers the intervals corresponding to levels E1 and E2, with ambient temperature as the lower limit and the corrected warning line as the upper limit. The temperature range of the high-temperature zone covers the intervals corresponding to levels E3 and E4, with the corrected warning line as the lower limit and the upper temperature limit Tmax as the upper limit, but without a hard upper limit to cover over-limit situations. The boundary between the steady-state temperature control domain and the high-temperature zone is the corrected temperature control boundary index. When the temperature crosses this boundary, it triggers temperature zone switching and response mode change.

[0043] The temperature rise rate is obtained by monitoring the transition from the steady-state temperature control zone to the high-temperature zone. During the grinding process, the temperature does not remain static within a certain range, but dynamically migrates between the steady-state temperature control zone and the high-temperature zone as the grinding load and heat dissipation conditions change. When the temperature moves from the steady-state temperature control zone to the high-temperature zone, it indicates that the rate of heat accumulation exceeds the rate of heat dissipation. This transition process is closely monitored to determine the timing of cooling intervention. A transition monitoring zone with a width of 5 degrees Celsius is set near the upper boundary of the steady-state temperature control zone. When the temperature enters the transition monitoring zone, the system increases the sampling frequency and activates the transition monitoring mode, increasing the sampling frequency from the conventional 1Hz to 10Hz to capture rapid temperature changes. The time taken for the temperature to rise from the lower boundary of the transition monitoring zone to the lower boundary of the high-temperature zone is recorded, and the average temperature change rate during this process is calculated as the temperature rise rate, expressed in degrees Celsius per minute. The temperature rise rate in the central region of the grinding disc is typically higher than that in the peripheral region. The temperature rise rate accelerates with increased feed rate or reduced cooling airflow. Under normal operating conditions, the temperature rise rate is approximately 0.5 to 1.5 degrees Celsius per minute, while under abnormal operating conditions, it can reach 3 to 5 degrees Celsius per minute. The temperature rise rate reflects the speed at which the temperature approaches the high-temperature zone and is a core indicator for determining the intensity of the cooling response.

[0044] In some embodiments, establishing the cooling response intensity based on the temperature rise rate includes: performing a stability analysis on the temperature rise rate to determine a stable maintenance range; generating a critical temperature proximity based on the rate average of the stable maintenance range; performing incremental adjustment of the response intensity based on the critical temperature proximity to form an intensity correction value; and establishing the cooling response intensity based on the intensity correction value.

[0045] Stability analysis was performed on the temperature rise rate to determine the stable maintenance range. The time series data of the temperature rise rate is not always stable; various factors during the grinding process can cause fluctuations in the temperature rise rate. Intermittent changes in the feed rate can cause periodic fluctuations in the heat load; changes in the contact state between the grinding disc and the material can cause random fluctuations in the rate of frictional heat generation; and disturbances in the cooling airflow can also affect the stability of the temperature acquisition in the short term. Directly using the temperature rise rate data during periods of severe fluctuation to determine the cooling response intensity may lead to frequent adjustments in the control system output, resulting in unstable operation of the cooling fan and the grinding mill. A sliding window analysis was performed on the temperature rise rate time series, with the window size set to 10 to 20 seconds. The standard deviation of the temperature rise rate within each window was calculated. The standard deviation reflects the dispersion of the temperature rise rate within that period. A window with a smaller standard deviation indicates that the temperature rise process is relatively stable during that period, with less external interference, and the acquired temperature rise rate data better reflects the true temperature rise trend under the current operating conditions. A window with a larger standard deviation indicates significant fluctuations during that period, which may be caused by feed switching, sudden changes in grinding disc load, or measurement noise. Windows with a standard deviation less than a set threshold are identified as stable windows. Consecutive stable windows are merged to form a stable maintenance zone. The start and end times of the stable maintenance zone indicate the time range within which the temperature rise rate remains relatively stable. The proportion of the stable maintenance zone is usually high during stable operation of the grinding mill, and decreases during feed switching or operating condition adjustments. Identifying the stable maintenance zone allows subsequent analysis to focus on reliable data intervals.

[0046] The critical temperature approach is generated based on the rate average of the steady-state maintenance range. The arithmetic mean of the temperature rise rate within the steady-state maintenance range is calculated as the rate average. This rate average represents the steady-state rate of temperature rise under current operating conditions, more accurately reflecting the true trend of heat accumulation after eliminating short-term fluctuations. The magnitude of the rate average depends on the balance between grinding intensity and heat dissipation capacity. Higher grinding disc speeds or larger feed rates increase frictional heat generation, leading to a higher rate average. Sufficient cooling airflow or lower ambient temperatures accelerate heat dissipation, causing a lower rate average. Determining the cooling response intensity solely based on the rate average is insufficient. The same rate average has different risk implications at different temperature starting points. When the temperature is close to the upper temperature limit, even a low rate average may quickly exceed the limit. When the temperature is still in the middle of the steady-state temperature control range, a higher rate average still provides sufficient buffer space. Based on the rate average and the difference between the current temperature and the upper temperature limit Tmax, the estimated time required to reach the critical point is calculated. The shorter this time, the closer to the critical state. The critical temperature approach is defined as the ratio of the rate average to the temperature margin, converting the rate of temperature rise into an indicator of the urgency of approaching the critical point. In the mid-stage of epoxy powder coating grinding, when the current temperature still has a significant margin before reaching the upper temperature limit, the critical temperature proximity remains in a low range even if the average rate reaches a moderate level. Conversely, in the later stage of grinding, when the temperature is approaching the upper temperature limit, the critical temperature proximity may still remain in a high range even if the average rate decreases. The critical temperature proximity comprehensively considers both the temperature rise rate and the temperature location, and thus more accurately reflects the urgency of cooling intervention than using the temperature rise rate alone.

[0047] The intensity correction value is formed by incrementally adjusting the response intensity based on the proximity to the critical temperature. The proximity to the critical temperature reflects the urgency of the temperature risk; the higher the urgency, the greater the cooling response intensity should be. However, this enhancement should not be a simple linear relationship. In the low urgency range, the temperature is far from the critical point, and a slight enhancement of cooling can maintain temperature stability. Excessive enhancement will lead to energy waste and decreased production efficiency. In the high urgency range, the temperature is approaching the critical point, and cooling must be significantly enhanced to quickly curb the temperature rise. Insufficient response may lead to temperature overshoot and agglomeration accidents. A piecewise incremental mapping relationship is established between the proximity to the critical temperature and the intensity correction value. The increase in the intensity correction value is gradual in the low urgency range and steep in the high urgency range. The piecewise incremental design simulates the response strategies of experienced operators under different risk scenarios: maintaining normal cooling to avoid interfering with normal production when the temperature risk is low, and decisively enhancing cooling to ensure safety as a priority when the temperature risk increases. Polyester-based powder coatings have a higher glass transition temperature, so the strength correction value can be appropriately reduced when the critical temperatures are close. Epoxy-based powder coatings have a lower glass transition temperature and are more sensitive to temperature, so the strength correction value should be increased when the critical temperatures are close.

[0048] The cooling response intensity is established using an intensity correction value. The final cooling response intensity is obtained by adding the intensity correction value to the base mapping value of the temperature rise rate and the cooling response intensity. The base mapping value reflects the cooling requirement corresponding to the temperature rise rate itself, while the intensity correction value reflects the additional risk correction due to the temperature location. The combined cooling response intensity considers both the rate and location characteristics of the temperature rise. In the initial stage of grinding production, the grinding chamber temperature is low and the temperature rise rate is gradual. Both the base mapping value and the intensity correction value are low, and the cooling response intensity is maintained at a low level to save energy. As the grinding process progresses, frictional heat gradually accumulates, the temperature rise rate increases, leading to an increase in the base mapping value. As the temperature approaches the high-temperature zone, the intensity correction value increases, and the cooling response intensity increases accordingly to cope with the increased temperature risk. When the temperature enters the high-temperature zone and the temperature rise rate remains high, both the base mapping value and the intensity correction value reach high levels, and the cooling response intensity approaches or reaches its upper limit to effectively curb the temperature rise. The upper limit of the cooling response intensity is set to 100%. When the calculated result exceeds 100%, it is taken as 100% to avoid placing demands on the cooling equipment beyond its capacity.

[0049] Step S140: Select the optimal cooling time in the high-temperature zone based on the cooling response intensity, generate grinding intensity adjustment parameters according to the correlation between the temperature rise characteristic sequence and the high-temperature zone, and configure the dynamic cooling threshold and grinding linkage coefficient through the coordinated calibration of the optimal cooling time and the grinding intensity adjustment parameters.

[0050] Specifically, the optimal cooling timing is selected within the high-temperature zone based on the cooling response intensity. While the cooling response intensity determines the power level the cooling system should output, the timing of cooling intervention also affects temperature control. Intervention too early may lead to energy waste, while intervention too late may miss the optimal cooling window. The dynamic characteristics of temperature changes within the high-temperature zone are analyzed to identify the optimal time for cooling intervention. When the temperature first enters the high-temperature zone, the temperature rise trend is not yet fully apparent, and the cooling response intensity is at a moderate level of 40% to 50%. Intervention at this point may be considered an over-response. When the temperature continues to rise within the high-temperature zone and approaches the upper temperature limit, the cooling response intensity has reached over 70%, but the cooling space has been compressed. Intervention at this point is considered a delayed response. The relationship between cooling response intensity and temperature location is analyzed. When the cooling response intensity first exceeds 50% and the temperature is located in the middle of the high-temperature zone, it is determined to be a suitable cooling intervention time. The time point that meets the above conditions is marked as the optimal cooling timing, recorded as a timestamp, and associated with contextual information such as the temperature value and cooling response intensity at that moment. The optimal cooling time for the center area of ​​the grinding disc is usually about 15 to 30 seconds earlier than for the edge area, because the temperature rise rate in the center area is faster and requires earlier intervention.

[0051] Grinding intensity adjustment parameters are generated based on the correlation between the temperature rise characteristic sequence and the high-temperature zone. The temperature rise characteristic sequence records features such as the peak temperature rise rate, peak temperature rise acceleration, and the time when the temperature reaches its peak value at each temperature measurement point. The correlation between these features and the high-temperature zone reflects the degree of influence of the grinding process on temperature. Analyzing the distribution of temperature measurement points entering the high-temperature zone in the temperature rise characteristic sequence, the residence time and peak temperature of each measurement point in the high-temperature zone are statistically analyzed. The longer the residence time or the higher the peak temperature, the stronger the correlation between grinding intensity and temperature rise in that region. The peak temperature in the central region of the grinding disc is usually 5 to 8 degrees Celsius higher than that in the edge region, and the residence time in the high-temperature zone is also longer, approximately 1.5 to 2 times that of the edge region, indicating that the grinding intensity in the central region contributes the most to the overall temperature rise. Based on the correlation strength between the temperature rise characteristic sequence and the high-temperature zone, the heat load contribution of the grinding process is calculated. A higher heat load contribution means that reducing the grinding intensity helps with temperature control. The calculation of the heat load contribution is based on the normalization of the product of the residence time and the peak temperature. The heat load contribution is converted into a grinding intensity adjustment parameter, which indicates the extent to which the grinding intensity should be reduced. The value ranges from 0% to 30%, where 0% indicates no adjustment is needed and 30% indicates a significant reduction in grinding intensity to match temperature control. The grinding intensity adjustment parameter is achieved by adjusting the grinding disc speed or feed rate. When the residence time in the high-temperature zone exceeds 60 seconds, the grinding intensity adjustment parameter is set to 15% to 20%; when the residence time exceeds 120 seconds, the grinding intensity adjustment parameter is set to 25% to 30%.

[0052] In some embodiments, configuring a dynamic cooling threshold and a grinding linkage coefficient through the coordinated calibration of the optimal cooling timing and the grinding intensity adjustment parameter includes: performing time positioning to determine the cooling demand at each time point for the optimal cooling timing; performing correlation analysis between the cooling demand and the grinding intensity adjustment parameter to obtain a coordinated control curve; performing optimization interval screening to identify the optimal coordinated segment for the coordinated control curve; and configuring a dynamic cooling threshold and a grinding linkage coefficient based on the parameter boundaries of the optimal coordinated segment.

[0053] The optimal cooling timing is determined by timing the process to assess the cooling demand at each point in time. The optimal cooling timing identifies the suitable time to initiate cooling, but since the cooling process is continuous, the intensity of cooling demand at points before and after this timing is evaluated. A cooling demand assessment window is formed by extending 60 seconds forward and backward from the optimal cooling timing. The cooling demand at each point within this window is calculated. The cooling demand is determined by three factors: the temperature value at that moment, the rate of temperature rise, and the margin before the upper temperature limit. Higher temperatures, faster rates of temperature rise, and smaller margins result in higher cooling demand. After normalizing these three factors, a weighted sum is calculated to obtain the cooling demand at each moment. The weights are allocated as follows: temperature value 0.3, rate of temperature rise 0.4, and margin 0.3. The cooling demand ranges from 0 to 1, where 0 indicates no cooling is needed and 1 indicates maximum cooling intensity is required. The cooling demand is typically low, around 0.3 to 0.5, in the 30 seconds before the optimal cooling point. The cooling demand at the exact moment of the optimal cooling point is approximately 0.6 to 0.7. The cooling demand in the 30 seconds after the optimal cooling point may continue to rise to above 0.8 if the temperature rise is not effectively contained. The cooling demand at each moment forms a time series, and the shape of the cooling demand curve reflects the evolution of cooling demand over time.

[0054] For example, the step of performing correlation analysis between the cooling demand and the grinding intensity adjustment parameter to obtain a synergistic control curve includes: obtaining a control baseline curve based on the cooling demand along the time dimension; forming a thermosensitive control coefficient by performing reverse correlation based on the grinding intensity adjustment parameter; generating a control response group by bidirectionally adjusting the control baseline curve and the grinding intensity adjustment parameter through the thermosensitive control coefficient; and forming a synergistic control curve based on the ratio of the cooling component to the grinding component of the control response group.

[0055] The control baseline curve is obtained by unfolding the cooling demand along the time dimension. The time series data of cooling demand records the complete process of cooling demand evolution over time. Visualizing this process and extracting features yields the control baseline. A curve is plotted with time on the horizontal axis and cooling demand on the vertical axis, forming the original form of the control baseline curve. The time range of the horizontal axis covers the entire interval of the cooling demand assessment window, i.e., 120 seconds before and after the optimal cooling time, and the value range of the vertical axis is 0 to 1, corresponding to the complete value range of cooling demand. The control baseline curve typically exhibits a single-peak shape, rising first and then falling. Initially, when the temperature enters the high-temperature zone, the cooling demand gradually increases from around 0.3. As the temperature continues to rise, the cooling demand reaches a peak of approximately 0.7 to 0.8 near the optimal cooling time. After cooling intervention, the temperature begins to decrease, and the cooling demand subsequently drops below 0.4. The control baseline curve is smoothed to eliminate the interference of short-term fluctuations. This is achieved using either a moving average or a low-pass filter. The smoothing window for the moving average is set to 5 to 10 seconds, representing the average of 5 to 10 consecutive sampling points. The cutoff frequency for the low-pass filter is set to 0.1 Hz to filter out rapid fluctuations above this frequency. The peak point of the control baseline curve is extracted. The peak point corresponds to the moment of strongest cooling demand. The height of the peak point reflects the maximum cooling demand intensity, typically between 0.7 and 0.9. The location of the peak point usually falls within ±10 seconds of the optimal cooling time. The peak point is a key reference for coordinated control.

[0056] A thermosensitive adjustment coefficient is formed by inversely correlating the grinding intensity adjustment parameter. The grinding intensity adjustment parameter represents the extent to which the grinding intensity should be reduced. This parameter has an inverse correlation with temperature; reducing the grinding intensity will lead to a decrease in heat generation, thereby suppressing temperature rise. The effect of the grinding intensity adjustment parameter on temperature changes is analyzed, establishing a quantitative relationship between grinding intensity adjustment and temperature response. The decrease in the heat generation rate for every 1 percentage point increase in the grinding intensity adjustment parameter is called the thermosensitive response rate. The thermosensitive response rate varies depending on the raw material type and grinding conditions. The thermosensitive response rate for epoxy powder coatings is approximately 0.3 to 0.5 degrees Celsius per minute, and for polyester powder coatings, it is approximately 0.2 to 0.4 degrees Celsius per minute. Multiplying the grinding intensity adjustment parameter by the thermosensitive response rate yields the thermosensitive adjustment coefficient, which represents the expected decrease in the heat generation rate under the current grinding intensity adjustment parameter setting. When the grinding intensity adjustment parameter is 15% and the thermosensitive response rate is 0.4, the thermosensitive adjustment coefficient is approximately 6%, meaning the expected decrease in the heat generation rate is 6%. The thermistor coefficient reflects the potential contribution of grinding regulation to temperature control; the larger the thermistor coefficient, the more significant the auxiliary effect of grinding regulation on temperature control.

[0057] A bidirectional adjustment response group is generated by using a thermosensitive adjustment coefficient to control the reference curve and the grinding intensity adjustment parameter. The reference curve represents the control demand on the cooling side, while the thermosensitive adjustment coefficient represents the control capability on the grinding side. A bidirectional balance adjustment is performed between the two to achieve synergistic optimization. When the thermosensitive adjustment coefficient is high, grinding adjustment can share some of the temperature control burden, and the cooling demand in the reference curve can be appropriately lowered. When the thermosensitive adjustment coefficient is low, the contribution of grinding adjustment is limited, and the cooling demand in the reference curve remains unchanged or even increases. The reference curve is corrected based on the thermosensitive adjustment coefficient. The corrected cooling demand is equal to the original cooling demand multiplied by the correction coefficient (1 - thermosensitive adjustment coefficient / 100). When the thermosensitive adjustment coefficient is 6%, the correction coefficient is 0.94, and the cooling demand is corrected from 0.7 to 0.66. Simultaneously, the grinding intensity adjustment parameter is adjusted inversely based on the cooling demand level of the reference curve. The higher the cooling demand, the larger the grinding intensity adjustment parameter should be to enhance the coordination on the grinding side. The cooling demand and grinding adjustment range after bidirectional adjustment are paired to form an adjustment response group. Each element of the adjustment response group contains three components: time marker, corrected cooling demand, and corrected grinding adjustment. The adjustment response group covers all times within the control period, and the bidirectional adjustment results at each time point reflect the coordinated scheme of cooling and grinding.

[0058] A coordinated control curve is formed based on the ratio of the cooling component to the grinding component in the control response group. Each moment in the control response group includes both cooling and grinding components, and their ratio reflects the relative weighting of the two control methods. The ratio of the cooling component to the grinding component for each element in the control response group is calculated. A ratio greater than 1 indicates that cooling control is dominant at that moment, a ratio less than 1 indicates that grinding control is dominant, and a ratio equal to 1 indicates a balanced combination of the two control methods. The ratios at each moment are plotted as curves, with the horizontal axis representing the cooling demand level in the control response group and the vertical axis representing the ratio of the cooling component to the grinding component, forming a preliminary coordinated control curve. The preliminary curve is smoothed and normalized to eliminate the influence of outliers and boundary effects, resulting in a regularized coordinated control curve. The shape of the coordinated control curve reflects the optimal ratio of cooling and grinding control under different cooling demand levels. A monotonically rising curve indicates that the higher the cooling demand, the more cooling should dominate; an inflection point indicates a critical point for switching control strategies. The coordinated control curve usually shows an inflection point around 0.5 of the cooling demand. Before the inflection point, the marginal benefit of grinding control is relatively high, and after the inflection point, the marginal benefit of cooling control is even higher.

[0059] The optimal coordinated control curve is selected by screening and optimizing the intervals. While the coordinated control curve covers the entire range from low to high demand, not all intervals are suitable as normal operating control areas. The interval with the highest efficiency is selected as the recommended operating range. The coordinated efficiency index is defined as the temperature control benefit obtained per unit control cost. The control cost includes cooling energy consumption and grinding efficiency loss, while the temperature control benefit is the magnitude of temperature decrease or the degree of temperature rise suppression. The coordinated efficiency at each point on the coordinated control curve is calculated. In the low cooling demand interval, the temperature control benefit is limited but the control cost is also low, while in the high cooling demand interval, the temperature control benefit is significant but the control cost increases sharply. The coordinated efficiency curve typically exhibits a characteristic of first rising and then falling, with an interval of highest efficiency. The interval of the coordinated control curve with a coordinated efficiency higher than the average level is marked as the optimal coordinated segment. The cooling demand range corresponding to the optimal coordinated segment is typically 0.4 to 0.7, and the corresponding grinding intensity adjustment parameter range is typically 10% to 20%. The optimal coordinated segment is the region with the highest efficiency in coordinating cooling and grinding regulation; the system operation should be maintained within this interval.

[0060] The dynamic cooling threshold and grinding linkage coefficient are configured based on the parameter boundaries of the optimal synergy segment. The temperature value corresponding to the lower boundary of the cooling demand of the optimal synergy segment is set as the lower limit of the dynamic cooling threshold; when the temperature reaches this value, the cooling response is triggered. The temperature value corresponding to the upper boundary of the cooling demand of the optimal synergy segment is set as the upper limit of the dynamic cooling threshold; when the temperature reaches this value, the cooling response should reach a higher intensity. The dynamic cooling threshold is expressed in interval form, with the lower limit typically from Tmax-8 to Tmax-6 degrees Celsius and the upper limit typically from Tmax-3 to Tmax-1 degrees Celsius. For epoxy powder coatings, the lower limit of the dynamic cooling threshold is approximately 42 to 44 degrees Celsius, and the upper limit is approximately 47 to 49 degrees Celsius. The ratio of the change in grinding intensity adjustment parameter to cooling demand within the optimal synergy segment is set as the grinding linkage coefficient. The formula for calculating the grinding linkage coefficient is k = Δm / Δc, where Δm is the change in grinding intensity adjustment parameter within the optimal synergy segment, and Δc is the change in the corresponding cooling demand. The typical value of the grinding linkage coefficient is 0.2 to 0.4, which means that for every 0.1 increase in cooling demand, the grinding intensity adjustment parameter should be increased by 2% to 4%. After the dynamic cooling threshold and grinding linkage coefficient are configured, they are written into the control system parameter library. During temperature control, the system triggers cooling based on the relationship between the current temperature and the dynamic cooling threshold, and adjusts the grinding intensity synchronously according to the grinding linkage coefficient.

[0061] Step S150: The agglomeration induction conditions are determined by comparing the dynamic cooling threshold with the temperature rise characteristic sequence. Cooling air regulation is performed to form a cooling regulation amount based on the agglomeration induction conditions and cooling control criteria. The grinding mill operation intensity is adjusted according to the grinding linkage coefficient to form a grinding regulation amount. Temperature control commands are output based on the cooling regulation amount and the grinding regulation amount to complete the process control of grinding temperature.

[0062] In some embodiments, determining the clumping induction conditions by comparing the dynamic cooling threshold with the temperature rise feature sequence includes: performing a point-by-point comparison of the temperature rise feature sequence based on the dynamic cooling threshold to obtain out-of-limit records; performing abnormal clustering detection on the out-of-limit records to identify out-of-limit clustering areas; verifying the risk attributes of the out-of-limit clustering areas to form clumping risk segments; and determining the clumping induction conditions based on the clumping risk segments.

[0063] A point-by-point comparison of the temperature rise characteristic sequence based on a dynamic cooling threshold is performed to obtain out-of-limit records. Multiple temperature measuring points arranged inside the grinding cavity continuously collect data within the abnormal temperature range. Each measuring point accumulates a complete temperature change record, which is checked one by one to identify situations exceeding the safe range. The records of each measuring point in the temperature rise characteristic sequence are traversed, and the peak temperature and current temperature value of each measuring point are extracted and compared with the upper limit of the dynamic cooling threshold. The comparison process covers all moments in the temperature rise characteristic sequence. Measuring points whose peak temperature exceeds the upper limit of the dynamic cooling threshold are determined to have peak out-of-limit; measuring points whose current temperature value exceeds the upper limit of the dynamic cooling threshold are determined to have real-time out-of-limit; measuring points that meet both conditions are determined to have double out-of-limit, and measuring points with double out-of-limit have the highest risk level and require priority handling. The comparison results of each measuring point are recorded to form an out-of-limit record, which includes four fields: measuring point number, out-of-limit type, out-of-limit magnitude, and out-of-limit time. The exceedance range is the difference between the actual temperature and the upper limit of the dynamic cooling threshold. When the temperature at the center measuring point of the grinding disc is 58 degrees Celsius and the upper limit of the threshold is 55 degrees Celsius, the exceedance range is 3 degrees Celsius. The exceedance records are arranged in chronological order of the exceedance time, and multiple exceedances at the same measuring point at different times are recorded separately.

[0064] Anomaly cluster detection is implemented to identify over-limit clusters in the over-limit records. Isolated, sporadic over-limit events may be caused by measurement noise or instantaneous operating condition fluctuations and do not necessarily represent a true risk of clustering. Only when over-limit events exhibit clustering characteristics in both time and space does it indicate a systemic temperature runaway problem. Over-limit records are grouped by temperature measurement point number, and the number of over-limit events and total duration of over-limit events at each measurement point are counted. Measurement points with a high number of over-limit events or a long total duration of over-limit events are candidates for spatial clustering, and the areas where these measurement points are located may have continuous heat accumulation. Over-limit records are also grouped by time window, and the number of over-limit measurement points within each time window is counted. Measurement points with more than 30% of the total number of measurement points exceeding the limit within a certain time window are identified as candidates for temporal clustering, and the operating conditions corresponding to these time windows may have overall temperature anomalies. Simultaneously, the distribution characteristics of over-limit types in the over-limit records are analyzed. Areas with a high concentration of dual over-limit events have the highest risk level, areas dominated by peak over-limit events indicate that the temperature has reached its peak but the duration is still short, and areas dominated by real-time over-limit events indicate that the temperature is rising and has not yet peaked. Clustering candidates in both spatial and temporal dimensions are cross-validated. Regions that simultaneously satisfy both spatial and temporal clustering characteristics are marked as overlimit clustering areas. Overlimit clustering areas are represented by a combination of spatial and temporal ranges. A region is defined as one that continuously exceeds the limit within a 100 mm radius of the millstone center during the period from the 120th to the 180th second.

[0065] Based on the risk attributes of the over-limit agglomeration zone, a clumping risk segment was identified. Exceeding the temperature limit does not necessarily lead to powder clumping; clumping also depends on the thermal sensitivity of the raw material and the duration of the over-limit. Epoxy powder coatings may not clump if they remain above the glass transition temperature for a short period, but the risk of clumping increases sharply if the duration is too long. The glass transition temperature (Tg) and melting point (Tm) of the raw material were read, and the temperature peak of the over-limit agglomeration zone was compared with these critical temperatures to assess the actual impact of the over-limit temperature on the raw material's state. Temperature peaks in the over-limit agglomeration zone exceeding Tg indicate a softening risk, manifested as sticky particle surfaces that easily adhere to each other; temperature peaks close to Tm indicate a melting risk, manifested as overall particle softening and potential clumping. The duration of the over-limit agglomeration zone was verified to determine if it was sufficient to cause a change in powder state. Epoxy powder coatings may soften and adhere if the temperature remains above Tg for more than 45 seconds, while polyester powder coatings have a slightly longer tolerance time of approximately 60 seconds. Exceeding limits in clustered areas that pass risk attribute verification are upgraded and marked as clumping risk zones. These risk zones, in addition to the existing risk attribute fields for clustered areas, include two more: risk type and risk level. Risk types are categorized as softening risk and melting risk. Risk levels are assessed based on the extent and duration of the exceedance, categorized as low, medium, and high. Low risk is defined as an exceedance of less than 3 degrees Celsius and a duration of less than 30 seconds, while high risk is defined as an exceedance of more than 5 degrees Celsius or a duration of more than 60 seconds.

[0066] Agglomeration induction conditions are determined based on agglomeration risk zones. These risk zones identify the spatiotemporal areas with agglomeration risk and their risk characteristics, from which key condition parameters are extracted to guide cooling regulation. The spatial location information of each risk zone is summarized to form a spatial distribution of exceedance points, represented as a position list in the grinding cavity coordinate system, indicating the target area that the cooling airflow should direct. The difference between the peak temperature and the upper limit of the dynamic cooling threshold for each risk zone is calculated, forming a temperature peak-to-threshold difference sequence; a larger difference indicates a higher temperature risk requiring stronger cooling power. The duration of exceedances in each risk zone is calculated, forming a duration sequence; a longer duration indicates more severe heat accumulation requiring a longer cooling cycle. The priority of the cooling response is determined based on the risk type and risk level within the risk zones, with melting risk taking precedence over softening risk, and high-risk levels taking precedence over low-risk levels. The spatial distribution of exceedance points, the temperature peak-to-threshold difference, duration, and risk priority are integrated to form agglomeration induction conditions, stored as structured data containing information in four dimensions: location, temperature, time, and priority.

[0067] Cooling air regulation is implemented based on agglomeration induction conditions and cooling control criteria to determine the cooling regulation amount. Cooling air regulation is the main means of grinding temperature control. By introducing low-temperature airflow into the grinding cavity, it removes the heat generated by friction, maintaining the grinding cavity temperature within the range that the raw material can withstand. Based on the spatial distribution of the over-limit points in the agglomeration induction conditions, the direction of the cooling airflow is determined. When the center area of ​​the grinding disc exceeds the limit, the cooling airflow is directed towards the center of the grinding disc for concentrated cooling; when the side wall area exceeds the limit, the cooling airflow is directed towards the corresponding side wall position to inhibit heat conduction outwards. If multiple areas exceed the limit simultaneously, the main direction area is determined by referring to the risk priority in the agglomeration induction conditions. The cooling power is determined based on the difference between the temperature peak and threshold in the agglomeration induction conditions. The larger the difference, the higher the temperature risk, and the higher the cooling power should be. When the difference is within 3 degrees Celsius, medium power (approximately 50% to 70% of the rated power) is used; when the difference exceeds 5 degrees Celsius, high power (approximately 80% to 100% of the rated power) is used. Simultaneously, the cooling power setting rules in the cooling control criteria are used for verification. Based on the duration of the agglomeration induction conditions, the cooling cycle is determined. A longer duration indicates more severe heat accumulation, thus requiring a longer cooling cycle to ensure sufficient temperature drop. When the duration is less than 30 seconds, the cooling cycle is set to 60 seconds; when the duration is greater than 60 seconds, the cooling cycle is set to 120 seconds. The cooling airflow direction, cooling power, and cooling cycle are integrated to form a cooling control quantity, represented as a parameter vector containing three components: airflow direction code, power percentage, and cycle duration.

[0068] The grinding intensity is adjusted based on the grinding linkage coefficient to determine the grinding regulation amount. Relying solely on the cooling system to control temperature suffers from response lag and high energy consumption. If the grinding intensity can be reduced simultaneously with cooling to decrease heat generation, temperature control can be achieved more quickly and effectively. The grinding linkage coefficient and the current cooling response intensity are read to calculate the required reduction in grinding intensity. This reduction equals the increase in cooling response intensity multiplied by the grinding linkage coefficient, ensuring synchronization between cooling and grinding regulation. When the cooling response intensity increases from 40% to 70%, the increment is 30%. With a grinding linkage coefficient of 0.3, the grinding intensity should be reduced by 9 percentage points. The reduction in grinding intensity is achieved by decreasing the grinding disc speed or the feed rate. A 9% reduction corresponds to a 9% decrease in speed, and a 9% reduction corresponds to a 9% decrease in feed rate, both of which effectively reduce frictional heat generation. The grinding intensity reduction range and adjustment method are integrated to form the grinding adjustment amount, which includes a speed adjustment component and a feed adjustment component. The two components can be adjusted simultaneously or one of them can be adjusted selectively. The setting of the grinding adjustment amount follows the principle of minimizing production efficiency loss, prioritizing the adjustment of parameters with less impact on output. When the temperature risk level is high, both components are adjusted simultaneously; when the temperature risk level is medium or low, only the feed rate is adjusted.

[0069] Temperature control commands are output based on cooling and grinding adjustment parameters to achieve process control of grinding temperature. For example... Figure 3As shown, the grinding mill 13 and the cooling fan 14 are physically two independent sets of equipment. The grinding mill 13 is driven by the frequency converter 16 to control the grinding disc speed and feed rate. The cooling fan 14 directly receives the control signal 20 to adjust the cooling power and airflow direction. The cooling airflow 21 is output from the cooling fan 14 and enters the grinding chamber of the grinding mill 13 for cooling. The temperature sensor group 15 arranged in the grinding chamber collects the temperature values ​​of each temperature measuring point in real time. The temperature signal 19 is converted from analog to digital by the data acquisition card 17 and then transmitted to the control computer 18. The control computer 18 runs the temperature control strategy algorithm to analyze and process the temperature data and generate temperature control commands. The temperature control commands are sent out in two paths through the control signal 20. One path is sent to the cooling fan 14 to perform cooling air regulation, and the other path is sent to the frequency converter 16 to adjust the operating intensity of the grinding mill 13. The two control signals 20 act synchronously to form a coordinated temperature control effect. The airflow direction, power percentage, and cycle duration parameters in the cooling control quantity are encoded as cooling sub-commands and sent to the cooling fan 14. The speed adjustment component and feed adjustment component parameters in the grinding control quantity are encoded as grinding sub-commands and sent to the frequency converter 16. The two sub-commands are combined to form a complete temperature control command. After the temperature control command is executed, the system continuously monitors the temperature change. If the temperature drops back to the steady-state temperature control domain 10, the control is considered successful. If the temperature fails to drop effectively, a new temperature control command is generated to strengthen the control. The grinding temperature process is continuously controlled under the closed-loop regulation of the temperature control command until the end of the production batch.

[0070] To implement the above-described method embodiments, a smart grinding temperature control method for powder coatings is proposed to achieve the corresponding functions and technical effects. See also... Figure 4 , Figure 4 This diagram illustrates a structural block diagram of a powder coating intelligent grinding temperature control system 400 according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The powder coating intelligent grinding temperature control system 400 provided in this embodiment includes: The signal acquisition module 401 is used to acquire the temperature signal and raw material state parameters of the grinding mill, and to perform correlation mapping between the temperature signal and the raw material state parameters to form a temperature control reference strategy. Anomaly identification module 402 is used to identify abnormal temperature ranges through the temperature control benchmark strategy, collect multiple temperatures in the abnormal temperature range to form a temperature rise feature sequence, and construct a cooling control criterion based on the temperature rise feature sequence. Temperature zone division module 403 is used to divide the temperature zone into a high temperature zone and a steady-state temperature control domain according to the cooling control criteria, monitor the transition from the steady-state temperature control domain to the high temperature zone to obtain the temperature rise rate, and establish the cooling response intensity through the temperature rise rate. The threshold configuration module 404 is used to select the optimal cooling time in the high temperature zone based on the cooling response intensity, generate grinding intensity adjustment parameters according to the correlation between the temperature rise characteristic sequence and the high temperature zone, and configure dynamic cooling threshold and grinding linkage coefficient through the coordinated calibration of the optimal cooling time and the grinding intensity adjustment parameters. The instruction output module 405 is used to determine the agglomeration induction conditions by comparing the dynamic cooling threshold with the temperature rise characteristic sequence, execute cooling air regulation to form a cooling regulation amount according to the agglomeration induction conditions and the cooling control criteria, adjust the grinding mill operating intensity to form a grinding regulation amount according to the grinding linkage coefficient, and output a temperature control instruction based on the cooling regulation amount and the grinding regulation amount to complete the process control of grinding temperature.

[0071] The aforementioned intelligent powder coating grinding temperature control system 400 can implement the intelligent powder coating grinding temperature control method of the above-described method embodiments. The options in the above method embodiments are also applicable to this embodiment, and will not be detailed here. The remaining contents of this application embodiment can be referred to the contents of the above method embodiments, and will not be repeated in this embodiment.

[0072] The purpose of the above embodiments is to reproduce and derive the technical solution of the present invention by way of example, and to fully describe the technical solution, purpose and effect of the present invention. The purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosure of the present invention, and not to limit the scope of protection of the present invention.

[0073] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A powder coating smart milling temperature control method, characterized by, The method comprises the following steps: acquiring a temperature signal of a grinding mill and a raw material state parameter, and performing correlation mapping on the temperature signal and the raw material state parameter to form a temperature control reference strategy; identifying a temperature abnormal interval through the temperature control reference strategy, performing multi-point temperature acquisition in the temperature abnormal interval to form a temperature rise characteristic sequence, and constructing a cooling control criterion based on the temperature rise characteristic sequence; dividing a temperature zone according to the cooling control criterion to form a high-temperature zone and a steady-state temperature control domain, monitoring a transition of the steady-state temperature control domain to the high-temperature zone to acquire a temperature rise rate, and determining a cooling response strength through the temperature rise rate; selecting an optimal cooling opportunity in the high-temperature zone based on the cooling response strength, generating a grinding intensity adjustment parameter according to the correlation between the temperature rise characteristic sequence and the high-temperature zone, and configuring a dynamic cooling threshold and a grinding linkage coefficient through the synergistic calibration of the optimal cooling opportunity and the grinding intensity adjustment parameter; determining a caking inducing condition through the comparison of the dynamic cooling threshold and the temperature rise characteristic sequence, performing cooling air regulation according to the caking inducing condition and the cooling control criterion to form a cooling regulation amount, adjusting the grinding mill operation intensity according to the grinding linkage coefficient to form a grinding adjustment amount, outputting a temperature control instruction based on the cooling regulation amount and the grinding adjustment amount, and completing the process control of the grinding temperature.

2. The method of claim 1, wherein, The method of identifying a temperature abnormal interval through the temperature control reference strategy comprises the following steps: determining a temperature deviation index and a raw material adaptation parameter according to the temperature control reference strategy; adaptingly adjusting a preset temperature difference limit to obtain a dynamic temperature difference limit according to the raw material adaptation parameter; defining a suspected abnormal section by comparing the temperature deviation index with the dynamic temperature difference limit; setting a safety boundary to limit the temperature abnormal interval based on the suspected abnormal section.

3. The method of claim 1, wherein, The method of performing multi-point temperature acquisition in the temperature abnormal interval to form a temperature rise characteristic sequence comprises the following steps: performing multi-point arrangement in the grinding cavity of the temperature abnormal interval to obtain a measurement point distribution map; synchronously reading the measurement point distribution map to obtain an instantaneous temperature distribution; collecting continuous temperature change data along the time direction of the instantaneous temperature distribution; constructing a temperature rise characteristic sequence based on the gradient representation of the continuous temperature change data.

4. The method of claim 1, wherein, The method of dividing a temperature zone according to the cooling control criterion to form a high-temperature zone and a steady-state temperature control domain comprises the following steps: performing interval segmentation on the temperature threshold in the cooling control criterion to obtain a temperature level standard; determining a temperature control boundary index according to the temperature level standard; dynamically modifying the temperature level standard based on the temperature control boundary index to obtain a temperature control intensity level; delimiting a steady-state temperature control domain and a high-temperature zone through the temperature control intensity level.

5. The method of claim 1, wherein, The method of determining a cooling response strength through the temperature rise rate comprises the following steps: performing stability analysis on the temperature rise rate to determine a smooth maintenance section; generating a critical temperature proximity degree according to the rate mean value of the smooth maintenance section; forming an intensity correction value by increasing the response strength based on the critical temperature proximity degree; determining a cooling response strength through the intensity correction value.

6. The method of claim 1, wherein, The method of configuring a dynamic cooling threshold and a grinding linkage coefficient through the synergistic calibration of the optimal cooling opportunity and the grinding intensity adjustment parameter comprises the following steps: The optimal cooling opportunity is used to determine the cooling demand degree at each time point; The cooling demand degree and the grinding intensity adjustment parameter are correlated to obtain a synergistic regulation curve; An optimal synergistic segment is identified by performing optimization interval screening on the synergistic regulation curve; A dynamic cooling threshold and a grinding linkage coefficient are configured according to the parameter boundary of the optimal synergistic segment.

7. The method of claim 1, wherein, The comparison of the dynamic cooling threshold and the temperature rise characteristic sequence determines the caking induction condition, including: A point-by-point comparison of the dynamic cooling threshold and the temperature rise characteristic sequence is performed to obtain an over-limit record; An abnormal cluster detection is performed on the over-limit record to identify an over-limit cluster area; A risk attribute verification is performed on the over-limit cluster area to form a caking risk segment; The caking induction condition is determined based on the caking risk segment.

8. The method of claim 3, wherein, The continuous temperature change data is collected along the time direction of the instantaneous temperature distribution, including: A time series correlation analysis is performed on the instantaneous temperature distribution to obtain a correlation coefficient record; A peak value is identified by using the correlation coefficient record to determine a peak interval duration; A periodic stability is formed by performing a stability verification on the peak interval duration; Dynamic screening is performed on the peak interval duration based on the periodic stability to obtain continuous temperature change data.

9. The method of claim 6, wherein, The correlation analysis of the cooling demand degree and the grinding intensity adjustment parameter to obtain a synergistic regulation curve includes: A regulation reference curve is obtained based on the cooling demand degree along the time dimension; A heat-sensitive adjustment coefficient is formed by reverse correlation based on the grinding intensity adjustment parameter; A regulation response group is generated by bidirectional adjustment of the regulation reference curve and the grinding intensity adjustment parameter through the heat-sensitive adjustment coefficient; A synergistic regulation curve is formed based on the ratio of the cooling component to the grinding component of the regulation response group.

10. A powder coating smart milling temperature control system, characterized by, It includes: A signal acquisition module is used to acquire the temperature signal and the raw material state parameter of the grinding machine, and to form a temperature control reference strategy by correlating and mapping the temperature signal and the raw material state parameter; An abnormality identification module is used to identify a temperature abnormality interval by the temperature control reference strategy, to form a temperature rise characteristic sequence by multi-point temperature acquisition in the temperature abnormality interval, and to construct a cooling control criterion based on the temperature rise characteristic sequence; A temperature zone division module is used to divide the temperature zone according to the cooling control criterion to form a high temperature zone and a steady state temperature control domain, to acquire a temperature rise rate by monitoring the transition of the steady state temperature control domain to the high temperature zone, and to determine a cooling response intensity by the temperature rise rate; A threshold configuration module is used to select an optimal cooling opportunity in the high temperature zone based on the cooling response intensity, to generate a grinding intensity adjustment parameter according to the correlation between the temperature rise characteristic sequence and the high temperature zone, and to configure a dynamic cooling threshold and a grinding linkage coefficient by synergistic calibration of the optimal cooling opportunity and the grinding intensity adjustment parameter. The instruction output module is configured to determine a caking inducing condition by comparing the dynamic cooling threshold with the temperature rise characteristic sequence, execute cooling air regulation according to the cooling control criterion to form a cooling regulation amount, adjust the grinding mill operation intensity according to the grinding linkage coefficient to form a grinding adjustment amount, output a temperature control instruction based on the cooling regulation amount and the grinding adjustment amount, and complete the process control of the grinding temperature.

Images