Air cooling unit temperature control adjusting method and system based on self-adaptive model prediction

Abstract

The invention discloses an air cooling unit temperature control adjusting method and system based on self-adaptive model prediction, and relates to the technical field of industrial cooling control, and the method comprises the steps: collecting air temperatures and water temperatures of a plurality of spatial positions of an air cooling unit, enabling each spatial position to correspond to an independent fan and a cooling water valve, and obtaining a preset temperature threshold value and a space temperature difference threshold value; the air temperature and the water temperature are input into a prediction model, the predicted temperature of each spatial position is obtained, and the predicted temperature difference between the adjacent spatial positions is calculated; identifying a spatial position where the predicted temperature exceeds a temperature threshold or the predicted temperature difference exceeds a spatial temperature difference threshold as a to-be-adjusted position; determining the rotating speed adjusting quantity of a fan corresponding to the to-be-adjusted position and the opening adjusting quantity of a cooling water valve based on the exceeding amplitude; and the rotating speed adjusting quantity and the opening adjusting quantity are output to a fan and a cooling water valve at corresponding spatial positions. According to the invention, excessive adjustment of all actuators is avoided, the space temperature distribution uniformity is improved, the total cooling energy consumption is reduced, and the thermal fatigue of equipment is relieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial cooling control, in particular to an air-cooled unit temperature control adjustment method and system based on adaptive model prediction. BACKGROUND

[0002] The air-cooled unit drives air to flow through the surface of the heat exchanger by the fan, and realizes heat exchange between the air and the cooling water to reduce the temperature of the cooling water. In actual operation, the temperature control method of the air-cooled unit mainly adopts feedback control based on the measured temperature deviation, and the deviation between the outlet water temperature and the set temperature is detected to uniformly adjust the rotation speed of all fans and the opening degree of the cooling water valve, so that the outlet water temperature is maintained within the target range.

[0003] However, due to the large space scale of large air-cooled units, there are differences in heat exchange conditions, cooling load distribution and air flow state at different spatial positions, resulting in uneven spatial temperature distribution. Although the unified control method has simple control logic, in order to correct the local overheating area, the overall cooling capacity needs to be excessively enhanced, which leads to overcooling in other normal areas, thereby increasing the total cooling energy consumption. In addition, the traditional feedback control relies on the current measured temperature signal to generate control action, and when the temperature anomaly in a certain area is detected, the heat has already affected the adjacent area through space conduction, and the controller can only respond passively after the problem occurs, and cannot prevent the deterioration trend of uneven spatial temperature. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides an air-cooled unit temperature control adjustment method based on adaptive model prediction to solve the problems of uneven temperature distribution at different spatial positions in large air-cooled units, the same control instruction is used for all actuators in the traditional overall control method, the overall cooling is excessively cooled to correct the local overheating, which leads to overcooling in other positions, thereby increasing the total energy consumption, and the feedback control based on the current temperature cannot prevent the trend of expanding spatial temperature difference.

[0006] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides an air-cooled unit temperature control adjustment method based on adaptive model prediction, which comprises collecting air temperature and water temperature at multiple spatial positions of the air-cooled unit, wherein each spatial position corresponds to an independent fan and cooling water valve, and obtaining a preset temperature threshold and a spatial temperature difference threshold; The air temperature and water temperature at multiple spatial positions are input into a prediction model to obtain the predicted temperature of each spatial position at a future time, and the predicted temperature difference between adjacent spatial positions is calculated; comparing the predicted temperature of each space position with a temperature threshold, comparing the predicted temperature difference between adjacent space positions with a space temperature difference threshold, and identifying a space position whose predicted temperature exceeds the temperature threshold or whose predicted temperature difference exceeds the space temperature difference threshold as a position to be adjusted; For the position to be adjusted, determining a speed adjustment amount of a corresponding fan and an opening adjustment amount of a corresponding cooling water valve based on a magnitude by which the predicted temperature exceeds the temperature threshold or a magnitude by which the predicted temperature difference exceeds the space temperature difference threshold. outputting the speed adjustment amount to the fan of the corresponding space position and outputting the opening adjustment amount to the cooling water valve of the corresponding space position.

[0007] As a preferred scheme of the air-cooled unit temperature control adjustment method based on adaptive model prediction, wherein: the air temperature and water temperature of the plurality of space positions are input into a prediction model to obtain a predicted temperature of each space position at a future time, and a predicted temperature difference between adjacent space positions is calculated, comprising: The air temperature and water temperature of the plurality of space positions are input into a prediction model to obtain a predicted temperature of each space position at a future time, and a predicted temperature difference between adjacent space positions is calculated, comprising: selecting a predicted temperature corresponding to a target time from the predicted temperatures corresponding to the plurality of future times as the predicted temperature of each space position at the future time; For each space position, a plurality of adjacent space positions adjacent to the space position are determined, and a temperature deviation between the predicted temperature of the space position at the future time and the predicted temperature of each adjacent space position at the future time is calculated, and the temperature deviation is taken as the predicted temperature difference between the adjacent space positions.

[0008] As a preferred scheme of the air-cooled unit temperature control adjustment method based on adaptive model prediction, wherein: the space position whose predicted temperature exceeds the temperature threshold or whose predicted temperature difference exceeds the space temperature difference threshold is identified as the position to be adjusted, comprising: For each space position, if the predicted temperature exceeds the temperature threshold, the space position is marked as a first type of position to be adjusted; For each space position, if the predicted temperature difference exceeds the space temperature difference threshold, the space position is marked as a second type of position to be adjusted; The first type of position to be adjusted and the second type of position to be adjusted are taken as the position to be adjusted.

[0009] As a preferred scheme of the air-cooled unit temperature control adjustment method based on adaptive model prediction, wherein: for the position to be adjusted, a speed adjustment amount of a corresponding fan and an opening adjustment amount of a corresponding cooling water valve are determined based on a magnitude by which the predicted temperature exceeds the temperature threshold or a magnitude by which the predicted temperature difference exceeds the space temperature difference threshold, comprising: For the first type of position to be adjusted, calculate the first deviation between the predicted temperature and the temperature threshold, and determine the speed adjustment of the fan and the opening adjustment of the corresponding cooling water valve for the first type of position to be adjusted based on the first deviation. For the second type of position to be adjusted, calculate the second deviation between the predicted temperature difference and the spatial temperature difference threshold, and determine the speed adjustment of the fan and the opening adjustment of the corresponding cooling water valve for the second type of position to be adjusted based on the second deviation.

[0010] As a preferred embodiment of the adaptive model-based temperature control method for air-cooled units according to the present invention, the method further includes: For a spatial position that simultaneously belongs to both the first type of adjustment position and the second type of adjustment position, compare the magnitudes of the first deviation and the second deviation. If the first deviation is greater than the second deviation, the speed adjustment of the fan corresponding to the spatial position and the opening adjustment of the corresponding cooling water valve are determined based on the first deviation. If the second deviation is greater than or equal to the first deviation, the speed adjustment of the fan corresponding to the spatial position and the opening adjustment of the corresponding cooling water valve are determined based on the second deviation.

[0011] As a preferred embodiment of the adaptive model-based temperature control method for air-cooled units according to the present invention, the step of determining the corresponding fan speed adjustment and the corresponding cooling water valve opening adjustment based on the magnitude by which the predicted temperature exceeds the temperature threshold or the magnitude by which the predicted temperature difference exceeds the spatial temperature difference threshold includes: Obtain the preset fan adjustment parameters and water valve adjustment parameters; The fan speed adjustment is determined based on the magnitude of the predicted temperature exceeding the temperature threshold and the fan adjustment parameters; the cooling water valve opening adjustment is determined based on the magnitude of the predicted temperature exceeding the temperature threshold and the water valve adjustment parameters; or: The fan speed adjustment amount is determined based on the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold and the fan adjustment parameters. The opening adjustment amount of the corresponding cooling water valve is determined based on the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold and the water valve adjustment parameters.

[0012] As a preferred embodiment of the adaptive model-based temperature control method for air-cooled units according to the present invention, wherein: calculating the predicted temperature difference between adjacent spatial locations includes: For each spatial location, determine all adjacent spatial locations that are adjacent to the stated spatial location; Calculate multiple temperature deviations between the predicted temperature at the spatial location and the predicted temperature at each of the adjacent spatial locations; The largest temperature deviation among the plurality of temperature deviations is selected as the predicted temperature difference corresponding to the spatial location.

[0013] As a preferred embodiment of the adaptive model-based temperature control method for air-cooled units according to the present invention, the step of inputting the air temperature and water temperature at the multiple spatial locations into the prediction model to obtain the predicted temperature of each spatial location at a future time includes: Obtain the air temperature and water temperature at the multiple spatial locations over a past time period; By inputting the air and water temperatures from the past time period, and the air and water temperatures at the current moment into the prediction model, the predicted temperature for each spatial location at a future time can be obtained.

[0014] As a preferred embodiment of the adaptive model-based temperature control method for air-cooled units according to the present invention, the method further includes: Obtain the cooling load of the air-cooled unit; The cooling load is input as an additional input along with the air and water temperatures at the multiple spatial locations into the prediction model to obtain the predicted temperature for each spatial location at a future time.

[0015] Secondly, the present invention provides an air-cooled unit temperature control system based on adaptive model prediction, including a data acquisition module for acquiring air temperature and water temperature at multiple spatial locations of the air-cooled unit; wherein each spatial location corresponds to an independent fan and cooling water valve, and a preset temperature threshold and spatial temperature difference threshold are obtained; The prediction module is used to input the air temperature and water temperature of the multiple spatial locations into the prediction model, obtain the predicted temperature of each spatial location at a future time, and calculate the predicted temperature difference between adjacent spatial locations. The identification module is used to compare the predicted temperature of each spatial location with the temperature threshold, compare the predicted temperature difference with the spatial temperature difference threshold, and identify spatial locations where the predicted temperature exceeds the temperature threshold or the predicted temperature difference exceeds the spatial temperature difference threshold as locations to be adjusted. The control quantity determination module is used to determine the speed adjustment amount of the corresponding fan and the opening amount of the corresponding cooling water valve for the position to be adjusted based on the magnitude by which the predicted temperature exceeds the temperature threshold or the magnitude by which the predicted temperature difference exceeds the spatial temperature difference threshold. The control output module is used to output the speed adjustment amount to the fan at the corresponding spatial position and the opening adjustment amount to the cooling water valve at the corresponding spatial position.

[0016] The beneficial effects of this invention are as follows: First, by constructing a predictive model to obtain the predicted temperature of each spatial location at a future time, and calculating the predicted temperature difference between adjacent spatial locations, the control basis is expanded from the current measured temperature to two dimensions: future temperature and spatial temperature difference. This overcomes the limitation of traditional feedback control, which relies solely on the current temperature signal. Compared to existing technologies that only generate control actions after the temperature exceeds a threshold, this invention can identify temperature anomalies in more locations before spatial heat conduction causes them, achieving a shift from passive response to proactive prevention and avoiding the adjustment lag caused by the controller chasing after problems.

[0017] Secondly, by simultaneously setting two judgment criteria—temperature threshold and spatial temperature difference threshold—this invention can identify a hidden non-uniformity where the temperature at a certain spatial location does not exceed the absolute threshold but the temperature difference with adjacent locations is too large. Traditional methods only judge whether the absolute temperature exceeds the limit, and cannot detect early signs of spatial non-uniformity. However, this invention, by judging spatial temperature difference, can capture the worsening trend of spatial non-uniformity that is difficult to identify by conventional control methods, significantly advancing the timing of control intervention and preventing local anomalies from spreading into large-scale temperature runaway.

[0018] Third, the adjustment amounts for the corresponding fans and cooling water valves are calculated separately for each identified location requiring adjustment, instead of applying the same control command to all fans and cooling water valves, thus achieving differentiated control. Traditional unified control enhances overall cooling capacity to correct local overheating, leading to undercooling in normal areas and increasing total energy consumption. This invention applies control actions only to the spatial locations requiring adjustment, while other locations maintain their original operating states. This ensures that the total cooling capacity matches the actual demand distribution, fundamentally eliminating the energy waste caused by unified control. While reducing total cooling energy consumption, the uniformity of spatial temperature distribution is improved, and thermal fatigue caused by temperature fluctuations in the equipment is also reduced. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a temperature control method for air-cooled units based on adaptive model prediction. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "example" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the invention. The appearance of an embodiment in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0024] Example 1, referring to Figure 1 This is one embodiment of the present invention, which provides a method for temperature control of an air-cooled unit based on adaptive model prediction, including the following steps: In step S100, the control device collects air temperature and water temperature data through temperature sensors distributed at different spatial locations within the air-cooled unit. Air temperature sensors are located at the air inlet at each spatial location to measure the temperature of the air entering the heat exchanger. Water temperature sensors are located on the cooling water pipes at each spatial location to measure the temperature of the cooling water flowing through that location. The collected temperature data is transmitted to the control device in digital signal form. The acquisition frequency is determined based on the thermal response characteristics of the air-cooled unit, typically set to once every 10 to 60 seconds.

[0025] Multiple spatial locations refer to different areas along the length or width of the air-cooled unit, with each spatial location equipped with an independent cooling unit. Each spatial location corresponds to an independent fan and cooling water valve. The fan drives airflow across the heat exchanger surface at that location, and the cooling water valve regulates the flow rate of cooling water through that location. The fans and cooling water valves at different spatial locations can be controlled independently, providing the capability for differentiated regulation.

[0026] The temperature threshold and spatial temperature difference threshold are obtained from the parameter configuration module of the control device. The temperature threshold refers to the upper limit of the predicted temperature for a spatial location, set according to process requirements and the safe operating range of the equipment. The spatial temperature difference threshold refers to the maximum allowable temperature deviation between adjacent spatial locations, set according to the structural characteristics and temperature uniformity requirements of the air-cooled unit. Both the temperature threshold and spatial temperature difference threshold are loaded when the control device is started and can be adjusted according to operating conditions.

[0027] Furthermore, in step S100, air temperature and water temperature at multiple spatial locations over past time periods are obtained so that the prediction model can use past data to improve prediction accuracy.

[0028] The control device reads historical data of air and water temperatures at multiple spatial locations over past time periods from the data storage module. The length of the past time period is determined based on the input requirements of the prediction model, typically set to 30 to 120 minutes from the past. The historical data is organized in time series format, containing air and water temperature values ​​for each spatial location at every acquisition moment within the past time period.

[0029] By acquiring air and water temperatures from past time periods, the prediction model can identify the temporal evolution of temperature and capture the periodic and inertial characteristics of temperature changes, thereby improving the accuracy of temperature predictions for future moments. Past data and current data are input into the prediction model together, providing complete input information for temperature prediction in step S200.

[0030] Furthermore, in step S100, the cooling load of the air-cooled unit is obtained so that the cooling load can be used as additional input information for prediction.

[0031] The control unit acquires cooling load data from the air-cooled unit via a communication interface. Cooling load refers to the current heat load power borne by the air-cooled unit, reflecting the amount of heat that needs to be removed through the cooling system. The cooling load data is read from the supervisory control system or process control system and transmitted to the control unit in the form of digital signals.

[0032] Cooling load directly affects the heat absorption of cooling water at various spatial locations, thus influencing the trends in air and water temperature changes. Incorporating cooling load as additional input information into the prediction model allows it to correlate the physical relationship between load and temperature changes, improving the accuracy of predictions under fluctuating load conditions. Cooling load data, along with air and water temperature data, constitute the set of input variables for the prediction model.

[0033] It should be noted that the control device refers to a computer control device used to execute the air-cooled unit temperature control method of the present invention. It is usually a programmable logic controller (PLC) or a distributed control system (DCS) with data acquisition, calculation and processing, and control command output functions. The control device collects operating data and outputs control commands through communication interfaces with devices such as temperature sensors, fans, and cooling water valves to complete the automatic control of the air-cooled unit temperature.

[0034] Preferably, step S100 acquires basic data reflecting the spatial temperature distribution of the air-cooled unit by collecting air and water temperatures from multiple spatial locations, providing a data source for identifying uneven spatial temperature phenomena. Simultaneously acquiring both temperature thresholds and spatial temperature difference thresholds provides a basis for dual limit-crossing judgments in subsequent steps. Obtaining air and water temperatures over past time periods provides temporal evolution information for the prediction model, enabling it to capture the inertial characteristics and periodic patterns of temperature changes, thus improving prediction accuracy. Acquiring cooling load data provides load change information for the prediction model, allowing it to correlate the physical relationship between load and temperature, improving its adaptability under fluctuating load conditions. Step S100 provides a complete and accurate input data foundation for the entire temperature control method.

[0035] Step S200: Input the air temperature and water temperature of multiple spatial locations into the prediction model to obtain the predicted temperature of each spatial location at a future time, and calculate the predicted temperature difference between adjacent spatial locations.

[0036] Specifically, step S200 includes the following sub-steps: Step S201: Input the air temperature and water temperature of multiple spatial locations into the prediction model to obtain the predicted temperature of each spatial location at multiple future moments within a preset time window.

[0037] In step S201, the control device inputs the air temperature and water temperature data from multiple spatial locations collected in step S100 into the prediction model. The prediction model is a pre-trained computational model deployed in the control device, capable of predicting temperature changes at each spatial location in the future based on the input temperature data.

[0038] The input variables for the prediction model include air temperature and water temperature at each spatial location. The input data is organized as a multidimensional array, where the first dimension corresponds to the spatial location number and the second dimension corresponds to the temperature type (air temperature or water temperature). The prediction model processes the input data and outputs the predicted temperature for each spatial location at multiple future moments within a preset time window.

[0039] A preset time window refers to a time range extending into the future from the current moment. The length of the time window is determined based on the thermal inertia and control response requirements of the air-cooled unit, and is typically set to 5 to 30 minutes. Multiple future moments refer to discrete time points within the preset time window divided at fixed time intervals, typically set to 30 seconds to 5 minutes.

[0040] The predicted temperatures output by the prediction model are organized in time series format, containing the predicted temperature values ​​for each spatial location at future times within a preset time window. The predicted temperature series is stored in the data buffer of the control device for use in subsequent sub-steps.

[0041] Furthermore, in step S201, the air and water temperatures from past time periods, as well as the current air and water temperatures, are input into the prediction model to obtain the predicted temperature for each spatial location at a future time.

[0042] The control device combines the air and water temperature data from past time periods acquired in step S100 with the air and water temperature data at the current moment, and inputs them into the prediction model. The temperature data from past time periods includes air and water temperature values ​​at various spatial locations at multiple historical moments, while the temperature data at the current moment refers to the latest temperature value collected in step S100.

[0043] The combined input data is organized in time series format, extending from the start of past time periods to the present, fully covering the historical information required by the prediction model. Based on the temporal evolution of temperature, the prediction model extracts trend information and periodic characteristics of temperature changes from past data, combines this with the current temperature status, and predicts the future temperature for each spatial location.

[0044] By incorporating past temperature data into the prediction model, the model can identify the inertial characteristics of temperature changes and capture the influence of factors such as load fluctuations and changes in ambient temperature on temperature evolution, thereby improving the accuracy and stability of temperature prediction.

[0045] Furthermore, in step S201, the cooling load is input as an additional input along with the air temperature and water temperature at multiple spatial locations into the prediction model to obtain the predicted temperature of each spatial location at a future time.

[0046] The control device uses the cooling load data obtained in step S100 as an additional input variable, and inputs it into the prediction model along with air temperature and water temperature data from multiple spatial locations. The cooling load data includes the cooling load value at the current moment, as well as records of cooling load changes over past time periods.

[0047] The predictive model treats cooling load as an independent input dimension and establishes a correlation with temperature data at each spatial location. Based on the physical coupling between cooling load and temperature, the model analyzes the impact of load changes on the temperature at each spatial location and predicts the future temperature of each location.

[0048] Incorporating cooling load into the prediction model input enables the model to correlate the causal relationship between process load changes and temperature response, maintaining prediction accuracy under fluctuating load conditions and avoiding prediction lag caused by relying solely on historical temperature data.

[0049] Step S202: Select the predicted temperature corresponding to the target time from the predicted temperatures corresponding to multiple future times as the predicted temperature of each spatial location at future times.

[0050] In step S202, the control device selects the predicted temperature value corresponding to the target time from the predicted temperature sequence output in step S201. The target time refers to a specific future time selected within a preset time window, used to determine temperature exceedances and calculate control variables. The selection of the target time is determined according to the control strategy, and usually a time within the preset time window that is relatively close to the current time is selected to balance prediction accuracy and control lead time.

[0051] The selection operation is completed by indexing the data elements in the predicted temperature sequence corresponding to the target time. The selected predicted temperature contains the predicted temperature value for each spatial location at the target time, organized as a one-dimensional array, with the array length equal to the total number of spatial locations. The selected predicted temperature is stored as the predicted temperature for each spatial location at a future time, for use in step S203 and subsequent steps.

[0052] By selecting the predicted temperature corresponding to the target time, the complete time series output by the prediction model is simplified into a single-time prediction result, which facilitates subsequent judgment and calculation operations while retaining the predictive capability of the prediction model.

[0053] Step S203: For each spatial location, determine multiple adjacent spatial locations, and calculate the temperature deviation between the predicted temperature of the spatial location at a future time and the predicted temperature of each adjacent spatial location at a future time. Use the temperature deviation as the predicted temperature difference between adjacent spatial locations.

[0054] In step S203, the control device determines multiple adjacent spatial locations for each spatial location based on the spatial layout information of the air-cooled unit. The spatial layout information is stored in the control device's configuration database, recording the physical location relationships and adjacency relationships of each spatial location. Adjacent spatial locations refer to other spatial locations that are physically in direct contact with or close to the spatial location.

[0055] After determining adjacent spatial locations, the control device reads the predicted temperature of the spatial location at a future time and the predicted temperature of each adjacent spatial location at a future time, and calculates the temperature deviation between the predicted temperature of the spatial location and the predicted temperature of each adjacent spatial location. The temperature deviation is calculated by taking the absolute value between the two predicted temperatures, reflecting the degree of difference in the distribution of predicted temperatures between the two spatial locations.

[0056] The calculated temperature deviation is used as the predicted temperature difference between adjacent spatial locations. The predicted temperature difference is stored in the form of an array, the length of which is equal to the total number of pairs of adjacent spatial locations. The predicted temperature difference reflects the uniformity of the spatial temperature distribution of the air-cooled unit at future times, providing a basis for judging whether the spatial temperature difference exceeds the limit in step S300.

[0057] Furthermore, the specific method for calculating the predicted temperature difference between adjacent spatial locations in step S203 is as follows: for each spatial location, determine all adjacent spatial locations; calculate multiple temperature deviations between the predicted temperature of the spatial location and the predicted temperature of each adjacent spatial location; select the largest temperature deviation from the multiple temperature deviations as the predicted temperature difference corresponding to the spatial location.

[0058] For each spatial location, the control device reads a list of all adjacent spatial locations from the spatial layout information. The number of adjacent spatial locations varies depending on the location of the spatial location within the air-cooled unit; spatial locations at the boundaries have fewer adjacent locations, while those in the middle have more.

[0059] The control device reads the predicted temperature of the spatial location at a future time and the predicted temperature of each adjacent spatial location at a future time, and calculates the temperature deviation between the spatial location and each adjacent spatial location, resulting in multiple temperature deviations. The number of temperature deviations is equal to the number of adjacent spatial locations.

[0060] The control device selects the temperature deviation with the largest value from multiple temperature deviations and uses this largest temperature deviation as the predicted temperature difference corresponding to the spatial location. By selecting the largest temperature deviation, the predicted temperature difference can reflect the most severe temperature inhomogeneity between the spatial location and the surrounding environment, thereby improving the sensitivity of identifying abnormal spatial temperature distribution.

[0061] The selected predicted temperature difference, together with the predicted temperature in step S202, constitutes the output result of step S200, which is used for the dual limit judgment in step S300.

[0062] Preferably, step S200 obtains the predicted temperature of each spatial location at future times by inputting air and water temperatures from multiple spatial locations into the prediction model. This expands the control basis from the current measured temperature to the future predicted temperature, providing forward-looking information for early identification of temperature anomalies. By calculating the predicted temperature difference between adjacent spatial locations, the degree of unevenness in spatial temperature distribution is quantified, providing a judgment indicator for identifying the trend of hidden spatial unevenness worsening. By incorporating temperature data from past time periods into the prediction model input, the prediction model can capture the temporal evolution of temperature changes, improving the accuracy of temperature prediction. By using cooling load as an additional input, the prediction model can correlate the physical relationship between load changes and temperature response, improving the predictive model's adaptability under load fluctuation conditions. By selecting the maximum temperature deviation as the predicted temperature difference, the sensitivity to identifying spatial temperature distribution anomalies is improved, enabling the control device to intervene and adjust in advance when the degree of spatial unevenness is slight, preventing spatial temperature unevenness from worsening into large-scale temperature runaway. Step S200 provides a complete and accurate judgment basis for the dual limit judgment in step S300.

[0063] Step S300: Compare the predicted temperature of each spatial location with the temperature threshold, compare the predicted temperature difference with the spatial temperature difference threshold, and identify spatial locations where the predicted temperature exceeds the temperature threshold or the predicted temperature difference exceeds the spatial temperature difference threshold as locations to be adjusted.

[0064] In step S300, the control device performs a dual limit-crossing judgment operation, identifying the spatial locations requiring adjustment through two independent judgment criteria. The first criterion compares the predicted temperature of each spatial location with a temperature threshold to determine whether the predicted temperature exceeds the permissible absolute temperature range. The second criterion compares the predicted temperature difference with a spatial temperature difference threshold to determine whether the temperature distribution between adjacent spatial locations exceeds the permissible degree of unevenness.

[0065] Dual limit detection enables the control device to identify both spatial locations where the absolute temperature exceeds the safe operating range and spatial locations where the temperature, while not exceeding the safe range, exhibits uneven spatial distribution. This provides identification capabilities at different stages of temperature anomaly development. An absolute temperature exceeding the limit indicates that the temperature at that location has reached a critical state for process or equipment safety, requiring immediate adjustment. A spatial temperature difference exceeding the limit indicates that the temperature difference between that location and adjacent locations is too large. Although the temperature at an individual location may not yet have reached a critical state, the spatial unevenness is showing a worsening trend, requiring early intervention to prevent the unevenness from spreading.

[0066] Specifically, step S300 includes the following sub-steps: Step S301: For each spatial location, if the predicted temperature exceeds the temperature threshold, mark the spatial location as a first-class location to be adjusted.

[0067] In step S301, the control device reads the predicted temperature selected in step S202 and the temperature threshold obtained in step S100 for each spatial location, and performs a comparison and judgment operation. The comparison and judgment is completed by calculating the relationship between the predicted temperature and the temperature threshold. If the value of the predicted temperature is greater than the value of the temperature threshold, it is determined that the predicted temperature exceeds the temperature threshold.

[0068] When the predicted temperature at a certain spatial location is determined to exceed a temperature threshold, the control device marks the spatial location as a first-class location to be adjusted. This marking is accomplished by adding a number to the list of locations to be adjusted and attaching a first-class identifier to that number. The first-class identifier distinguishes the reason for marking, indicating that the spatial location is marked because the predicted temperature exceeds the temperature threshold.

[0069] The control device traverses all spatial locations, performs comparison and marking operations for each location, and completes the identification of all spatial locations where the predicted temperature exceeds the temperature threshold. The marking results are stored in the control device's data buffer for use in step S303.

[0070] The first type of adjustable position corresponds to a spatial location where the absolute temperature will exceed the limit in the future, indicating that the temperature at that location will exceed the process requirements or the safe operating range of the equipment. By marking the predicted temperature exceeding the temperature threshold in advance, the control device can take adjustment measures before the actual temperature exceedance occurs, thus avoiding the impact of temperature exceedance on process stability and equipment safety.

[0071] Step S302: For each spatial location, if the predicted temperature difference exceeds the spatial temperature difference threshold, mark the spatial location as a second type of location to be adjusted.

[0072] In step S302, the control device reads the predicted temperature difference calculated in step S203 and the spatial temperature difference threshold obtained in step S100 for each spatial location, and performs a comparison and judgment operation. The comparison and judgment is completed by calculating the relationship between the predicted temperature difference and the spatial temperature difference threshold. If the value of the predicted temperature difference is greater than the value of the spatial temperature difference threshold, it is determined that the predicted temperature difference exceeds the spatial temperature difference threshold.

[0073] When the predicted temperature difference at a certain spatial location exceeds a spatial temperature difference threshold, the control device marks the spatial location as a second-category location to be adjusted. This marking is accomplished by adding a number to the list of locations to be adjusted and attaching a second-category identifier to that number. The second-category identifier distinguishes the reason for marking, indicating that the spatial location is marked because the predicted temperature difference exceeds the spatial temperature threshold.

[0074] The control device traverses all spatial locations, performs comparison and marking operations for each location, and completes the identification of all spatial locations where the predicted temperature difference exceeds the spatial temperature difference threshold. The marking results are stored in the control device's data buffer for use in step S303.

[0075] The second type of location to be adjusted corresponds to a spatial location where excessive temperature distribution differences will occur between the location and its adjacent locations in the future, reflecting that the unevenness of spatial temperature distribution exceeds the allowable range. By marking in advance when the predicted temperature difference exceeds the spatial temperature difference threshold, the control device can identify situations where the spatial temperature has not exceeded the absolute threshold but the spatial distribution has shown a worsening trend of unevenness. Adjustment measures can be taken before the spatial unevenness spreads into a large-scale temperature runaway, avoiding the shortcomings of a uniform control method that passively responds to problems.

[0076] Step S303: Select the first type of position to be adjusted and the second type of position to be adjusted as the positions to be adjusted.

[0077] In step S303, the control device reads the first type of adjustable positions marked in step S301 and the second type of adjustable positions marked in step S302, and merges the two types of marking results into an adjustable position list. The merging operation is completed by combining all elements of the first type of adjustable position list and the second type of adjustable position list into the same list.

[0078] Each element in the list of locations to be adjusted contains a spatial location number and a labeling type identifier. The labeling type identifier is used to record whether the spatial location is labeled due to a predicted temperature exceeding the limit (Category 1), a predicted temperature difference exceeding the limit (Category 2), or both conditions are met (labeled as both Category 1 and Category 2). The labeling type identifier provides a basis for determining the differential control quantity in step S400.

[0079] The merged list of locations to be adjusted is the output of step S300 and is used by step S400. The list of locations to be adjusted contains all spatial locations that need to be adjusted in the future, including spatial locations where the absolute temperature is about to exceed the limit, as well as spatial locations where the uneven distribution of spatial temperature is about to worsen, covering different manifestations and development stages of temperature anomalies.

[0080] Preferably, step S300 employs a dual limit-crossing judgment mechanism to identify both spatial locations where the predicted temperature exceeds the temperature threshold and spatial locations where the predicted temperature difference exceeds the spatial temperature difference threshold, enabling the control device to monitor temperature anomalies in two dimensions. Absolute temperature limit crossing judgment ensures the control device can identify spatial locations where the temperature is about to exceed the process or safety range, preventing temperature runaway from adversely affecting production and equipment. Spatial temperature difference limit crossing judgment ensures the control device can identify hidden non-uniformities where the spatial temperature, although not exceeding the absolute threshold, has a large temperature difference with adjacent locations, allowing for early intervention before spatial heat conduction leads to more temperature anomalies, preventing spatial non-uniformity from worsening into large-scale temperature runaway. By merging the results of the two judgment criteria into a list of locations to be adjusted, the control device can apply differentiated adjustments for different types of temperature anomalies, avoiding the energy waste caused by excessive adjustments to all fans and cooling water valves in traditional uniform control methods to correct local problems. The dual limit-crossing judgment mechanism in step S300 embodies the core innovation of this invention, providing a precise adjustment target for the differentiated control in step S400.

[0081] Step S400: For the position to be adjusted, determine the speed adjustment amount of the corresponding fan and the opening amount of the corresponding cooling water valve based on the magnitude of the predicted temperature exceeding the temperature threshold or the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold.

[0082] In step S400, the control device performs a differentiated control quantity calculation operation for the adjustment positions identified in step S300. The differentiated control quantity calculation determines the corresponding fan speed adjustment and the corresponding cooling water valve opening adjustment based on the marking type and over-limit range of the adjustment position, ensuring that different adjustment positions receive control quantities that match their degree of temperature anomaly.

[0083] The magnitude by which the predicted temperature exceeds the temperature threshold reflects the degree of absolute temperature exceedance at a spatial location; a larger magnitude indicates a more severe temperature anomaly. The magnitude by which the predicted temperature difference exceeds the spatial temperature difference threshold reflects the degree of temperature unevenness between the spatial location and its adjacent locations; a larger magnitude indicates more severe spatial unevenness. Control variables are determined based on the magnitude of these exceedances, ensuring that the magnitude of the control variable is proportional to the severity of the temperature anomaly. This avoids applying excessive regulation to minor anomalies, leading to temperature fluctuations, and also avoids applying insufficient regulation to severe anomalies, leading to temperature runaway.

[0084] Specifically, step S400 includes the following sub-steps: Step S401: For the first type of position to be adjusted, calculate the first deviation between the predicted temperature and the temperature threshold, and determine the speed adjustment of the fan and the opening adjustment of the corresponding cooling water valve based on the first deviation.

[0085] In step S401, the control device reads the predicted temperature of the corresponding spatial location and the temperature threshold obtained in step S100 for the first type of adjustment position marked in step S300, and calculates the first deviation between the predicted temperature and the temperature threshold. The first deviation is calculated by subtracting the temperature threshold from the predicted temperature. When the calculation result is positive, it indicates the magnitude of the predicted temperature exceeding the temperature threshold, reflecting the severity of the absolute temperature exceeding the limit.

[0086] After calculating the first deviation, the control device determines the fan speed adjustment and the corresponding cooling water valve opening adjustment for the first type of adjustable position based on the first deviation. This determination is completed by establishing a mapping between the first deviation and a preset control relationship, which describes the correspondence between the deviation and the control quantity. The fan speed adjustment represents the required increase in fan speed, and the opening adjustment represents the required increase in cooling water valve opening; both are positive values, indicating enhanced cooling capacity to lower the temperature.

[0087] The control device iterates through all the first-class adjustable positions, performs a first deviation calculation and control quantity determination operation for each first-class adjustable position, and completes the control quantity calculation for all spatial positions that need to be adjusted due to predicted temperature exceeding the limit. The calculation results are stored in the data buffer of the control device for use in step S500.

[0088] By calculating the first deviation for the first type of adjustable position and determining the control quantity, the control device can apply corresponding cooling adjustment according to the severity of the absolute temperature exceeding the limit of the spatial position, so as to avoid the temperature from exceeding the process requirements or the safe operating range of the equipment.

[0089] Step S402: For the second type of position to be adjusted, calculate the second deviation between the predicted temperature difference and the spatial temperature difference threshold, and determine the speed adjustment of the fan and the opening adjustment of the corresponding cooling water valve based on the second deviation.

[0090] In step S402, the control device reads the predicted temperature difference of the corresponding spatial location and the spatial temperature difference threshold obtained in step S100 for the second type of adjustment position marked in step S300, and calculates the second deviation between the predicted temperature difference and the spatial temperature difference threshold. The second deviation is calculated by subtracting the spatial temperature difference threshold from the predicted temperature difference. A positive result indicates the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold, reflecting the severity of uneven spatial temperature distribution.

[0091] After calculating the second deviation, the control device determines the fan speed adjustment and the corresponding cooling water valve opening adjustment for the second type of adjustable position based on the second deviation. This determination is completed by establishing a mapping between the second deviation and a preset control relationship, which describes the correspondence between the deviation and the control quantity. The signs of the speed and opening adjustments are determined based on the temperature relative to adjacent spatial positions. If the predicted temperature of the spatial position is higher than the predicted temperature of the adjacent spatial position, the speed and opening adjustments are positive, indicating enhanced cooling; if the predicted temperature of the spatial position is lower than the predicted temperature of the adjacent spatial position, the speed and opening adjustments are negative, indicating weakened cooling.

[0092] The control device iterates through all the second type of adjustable positions, performs a second deviation calculation and control quantity determination operation for each second type of adjustable position, and completes the control quantity calculation for all spatial positions that need to be adjusted due to predicted temperature difference exceeding the limit. The calculation results are stored in the data buffer of the control device for use in step S500.

[0093] By calculating the second deviation for the second type of adjustable position and determining the control quantity, the control device can apply corresponding differential adjustment according to the degree of temperature distribution non-uniformity between the spatial position and the adjacent spatial positions, and eliminate the temperature distribution difference before the spatial non-uniformity spreads into a large-scale temperature runaway.

[0094] Furthermore, for a spatial position that simultaneously belongs to the first type of adjustable position and the second type of adjustable position, the magnitudes of the first deviation and the second deviation are compared; if the first deviation is greater than the second deviation, the speed adjustment amount of the fan corresponding to the spatial position and the opening adjustment amount of the corresponding cooling water valve are determined based on the first deviation; if the second deviation is greater than or equal to the first deviation, the speed adjustment amount of the fan corresponding to the spatial position and the opening adjustment amount of the corresponding cooling water valve are determined based on the second deviation.

[0095] When executing steps S401 and S402, the control device identifies spatial locations that simultaneously belong to both the first type of adjustable position and the second type of adjustable position. A spatial location that simultaneously belongs to both types indicates that the location exhibits both an absolute temperature exceeding the predicted temperature threshold and a spatial non-uniformity where the predicted temperature difference exceeds the spatial temperature difference threshold.

[0096] For spatial locations that simultaneously belong to two categories, the control device reads the first deviation calculated in step S401 and the second deviation calculated in step S402, and performs a comparison operation to determine which temperature anomaly is more severe. The comparison operation is completed by calculating the relationship between the magnitudes of the first and second deviations.

[0097] If the first deviation is greater than the second deviation, it indicates that the degree of absolute temperature exceeding the limit at the spatial location exceeds the degree of unevenness in spatial temperature distribution, and the absolute temperature exceeding the limit is the primary problem. The control device determines the speed adjustment of the fan corresponding to the spatial location and the opening adjustment of the corresponding cooling water valve based on the first deviation, prioritizing the resolution of the absolute temperature exceeding the limit problem.

[0098] If the second deviation is greater than or equal to the first deviation, it indicates that the unevenness of the spatial temperature distribution at the spatial location exceeds or is equal to the degree of absolute temperature exceeding the limit. Spatial unevenness is the primary problem, or both are equally important. Based on the second deviation, the control device determines the speed adjustment of the fan and the opening adjustment of the corresponding cooling water valve at the spatial location, prioritizing the resolution of the spatial unevenness problem or simultaneously addressing both issues.

[0099] By comparing the first deviation and the second deviation and selecting the larger deviation as the control basis, the control device can focus on the most severe type of anomaly when multiple temperature anomalies exist in the same spatial location, avoiding conflicts or mutual cancellations in the calculation of control quantities, and ensuring that the control quantities can effectively resolve the main contradictions.

[0100] Further, in step S400, the specific method for determining the speed adjustment amount of the corresponding fan and the opening adjustment amount of the corresponding cooling water valve based on the magnitude of the predicted temperature exceeding the temperature threshold or the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold is as follows: obtain preset fan adjustment parameters and water valve adjustment parameters; determine the speed adjustment amount of the corresponding fan based on the magnitude of the predicted temperature exceeding the temperature threshold and the fan adjustment parameters, and determine the opening adjustment amount of the corresponding cooling water valve based on the magnitude of the predicted temperature exceeding the temperature threshold and the water valve adjustment parameters; or determine the speed adjustment amount of the corresponding fan based on the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold and the fan adjustment parameters, and determine the opening adjustment amount of the corresponding cooling water valve based on the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold and the water valve adjustment parameters.

[0101] The control device obtains preset fan adjustment parameters and water valve adjustment parameters from the parameter configuration module. The fan adjustment parameters describe the mapping relationship between the deviation and the fan speed adjustment, reflecting the fan speed adjustment range corresponding to a unit deviation. The water valve adjustment parameters describe the mapping relationship between the deviation and the cooling water valve opening adjustment, reflecting the valve opening adjustment range corresponding to a unit deviation. The fan and water valve adjustment parameters are determined based on physical parameters such as the heat exchange characteristics of the air-cooled unit, the fan performance curve, and the water valve flow characteristics, and are loaded when the control device starts.

[0102] For the first type of adjustable position, the control device determines the corresponding fan speed adjustment amount based on the predicted temperature exceeding the temperature threshold and the fan adjustment parameters. This determination is performed by associating the predicted temperature exceeding the temperature threshold (i.e., the first deviation) with the fan adjustment parameters, following a mapping relationship defined for the fan adjustment parameters. Similarly, the control device determines the corresponding cooling water valve opening adjustment amount based on the predicted temperature exceeding the temperature threshold and the water valve adjustment parameters. This determination is also performed by associating the first deviation with the water valve adjustment parameters.

[0103] For the second type of adjustable position, the control device determines the corresponding fan speed adjustment amount based on the predicted temperature difference exceeding the spatial temperature difference threshold and the fan adjustment parameters. This determination is completed by correlating the predicted temperature difference exceeding the spatial temperature difference threshold (i.e., the second deviation) with the fan adjustment parameters. Similarly, the control device determines the corresponding cooling water valve opening adjustment amount based on the predicted temperature difference exceeding the spatial temperature difference threshold and the water valve adjustment parameters. This determination is also completed by correlating the second deviation with the water valve adjustment parameters.

[0104] By introducing fan and water valve regulation parameters, the conversion process from deviation to control quantity has a clear physical basis and adjustability. These parameters can be adjusted according to the characteristics and operating conditions of different air-cooled units, enabling the control method to adapt to different equipment and operating conditions.

[0105] Preferably, step S400 calculates the deviation and determines the control quantity for both the first and second types of adjustable locations, achieving differentiated control based on different temperature anomaly types. For the first type of adjustable location, the control quantity is proportional to the degree of absolute temperature exceeding the limit, enabling the control device to apply corresponding cooling adjustments based on the severity of the temperature exceeding the safe range, preventing adverse effects of temperature runaway on the process and equipment. For the second type of adjustable location, the control quantity is proportional to the degree of unevenness in spatial temperature distribution, enabling the control device to apply corresponding differentiated adjustments based on the severity of spatial unevenness, eliminating temperature distribution differences before the unevenness spreads. By comparing the deviations of spatial locations belonging to both types and selecting the larger deviation as the control basis, the control device can focus on the most severe anomaly type when multiple temperature anomalies exist in the same spatial location, avoiding conflicts in control quantity calculations. By introducing fan adjustment parameters and water valve adjustment parameters to establish a mapping relationship between the deviation and the control quantity, the control quantity calculation has a clear physical basis and adjustability, improving the adaptability of the control method to different equipment and operating conditions. The differential control quantity calculation in step S400 enables each position to obtain a control quantity that matches its temperature anomaly level, avoiding over-adjustment and energy waste caused by applying the same control command to all fans and cooling water valves in the traditional unified control method, and providing a control command basis for the precise control execution in step S500.

[0106] Step S500: Output the speed adjustment amount to the fan in the corresponding spatial position, and output the opening adjustment amount to the cooling water valve in the corresponding spatial position.

[0107] In step S500, the control device executes a control command output operation based on the speed adjustment and opening adjustment amounts calculated in step S400. The control command output operation is completed through the communication interface between the control device and the fan controller and the cooling water valve actuator, so that the speed adjustment and opening adjustment amounts are converted into actual equipment actions, thereby completing the temperature adjustment of the position to be adjusted.

[0108] The control device reads the speed adjustment and opening adjustment values ​​of all positions to be adjusted stored in step S400, and determines the fan and cooling water valve corresponding to the spatial position based on the spatial position number. Each fan and cooling water valve at each spatial position has a unique device address and communication parameters in the control device's equipment configuration database, enabling the control device to accurately send control commands to the target device.

[0109] For each position to be adjusted, the control device converts the speed adjustment amount into a fan control command. The fan control command includes a target speed setpoint, which is calculated by superimposing the current fan speed with the speed adjustment amount. The sign handling of the superposition calculation ensures that a positive speed adjustment amount increases the fan speed, while a negative speed adjustment amount decreases the fan speed. The control device sends the fan control command to the fan controller at the corresponding spatial location via a communication interface. Upon receiving the control command, the fan controller drives the fan to perform the speed adjustment.

[0110] For each position to be adjusted, the control device converts the opening adjustment amount into a cooling water valve control command. The cooling water valve control command includes a target opening setpoint, which is calculated by superimposing the current cooling water valve opening with the opening adjustment amount. The sign handling of the superposition calculation ensures that a positive opening adjustment amount increases the valve opening, while a negative opening adjustment amount decreases the valve opening. The control device sends the cooling water valve control command to the corresponding cooling water valve actuator via a communication interface. Upon receiving the control command, the cooling water valve actuator drives the valve to perform the opening adjustment.

[0111] The control commands are transmitted using a digital communication protocol that supports point-to-point addressing, enabling the control device to send independent control commands to fans and cooling water valves at specific spatial locations. Fans and cooling water valves at different spatial locations may receive different control commands, reflecting the implementation of differentiated control. Fans and cooling water valves at spatial locations not marked as requiring adjustment do not receive control commands and remain in their current operating state.

[0112] After receiving control commands, the fan controller and cooling water valve actuator execute the equipment action response. The fan controller adjusts the fan's power supply frequency or voltage, causing the fan speed to gradually change to the target speed setpoint. The cooling water valve actuator drives the valve opening to gradually change to the target opening setpoint. The time constant of the equipment action response is determined based on the mechanical characteristics of the fan and cooling water valve, and is typically completed within the range of several seconds to tens of seconds.

[0113] Adjusting the fan speed changes the airflow at the corresponding spatial location. This change in airflow alters the convective heat transfer intensity on the heat exchanger surface, thereby changing the air and water temperature at that location. Adjusting the cooling water valve opening changes the cooling water flow at the corresponding spatial location. This change in cooling water flow alters the cooling capacity of the heat exchanger, thus changing the water temperature at that location. The combined effect of adjusting the fan speed and cooling water valve opening directs the temperature at the desired location towards the expected direction, eliminating predicted temperature deviations or spatial unevenness.

[0114] After outputting the control command, the control device continues to execute the cyclical operation from steps S100 to S500. In the next control cycle, the control device collects the adjusted air temperature and water temperature, and re-executes the prediction, judgment, and control quantity calculation operations to form a closed-loop control process. The closed-loop control process continuously monitors the temperature status of the air-cooled unit, continuously identifies temperature anomalies, and continuously applies differentiated adjustments to keep the spatial temperature distribution of the air-cooled unit within the allowable range.

[0115] Preferably, step S500 completes the conversion from control quantity calculation to equipment action by outputting the speed adjustment and opening adjustment quantities to the fans and cooling water valves at the corresponding spatial locations, realizing the actual execution of differentiated control. By employing a point-to-point addressing digital communication protocol, the control device can send independent control commands to fans and cooling water valves at different spatial locations. Different spatial locations receive different adjustment quantities based on their temperature anomaly levels, avoiding over-adjustment caused by applying the same control command to all fans and cooling water valves in traditional unified control methods. By outputting control commands only to the location to be adjusted, spatial locations not marked as requiring adjustment maintain their current operating state, avoiding unnecessary interference with normally operating locations, reducing the number of equipment actions, and lowering mechanical wear and energy consumption. Through the synergistic effect of fan speed adjustment and cooling water valve opening adjustment, the location to be adjusted simultaneously receives cooling capacity adjustments from both the air and water sides, improving the response speed and adjustment accuracy of temperature regulation. By continuously executing the cyclical operation from step S100 to step S500, a closed-loop control process is formed, enabling the control device to continuously monitor the temperature status and apply differentiated adjustments. This ensures that the spatial temperature distribution of the air-cooled unit remains within the allowable range, achieving the beneficial effects of improved spatial temperature distribution uniformity, reduced total cooling energy consumption, and reduced equipment thermal fatigue.

[0116] This embodiment also provides an air-cooled unit temperature control system based on adaptive model prediction, including: The data acquisition module is used to collect air temperature and water temperature at multiple spatial locations of the air-cooled unit; each spatial location corresponds to an independent fan and cooling water valve, and obtains preset temperature thresholds and spatial temperature difference thresholds; The prediction module is used to input the air temperature and water temperature of multiple spatial locations into the prediction model, obtain the predicted temperature of each spatial location at a future time, and calculate the predicted temperature difference between adjacent spatial locations. The identification module is used to compare the predicted temperature of each spatial location with the temperature threshold, compare the predicted temperature difference with the spatial temperature difference threshold, and identify spatial locations where the predicted temperature exceeds the temperature threshold or the predicted temperature difference exceeds the spatial temperature difference threshold as locations to be adjusted. The control quantity determination module is used to determine the speed adjustment of the corresponding fan and the opening adjustment of the corresponding cooling water valve for the position to be adjusted based on the predicted temperature exceeding the temperature threshold or the predicted temperature difference exceeding the spatial temperature difference threshold. The control output module is used to output the speed adjustment to the fan at the corresponding spatial position and the opening adjustment to the cooling water valve at the corresponding spatial position.

[0117] This embodiment also provides a computer device applicable to the temperature control adjustment method of air-cooled units based on adaptive model prediction, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the temperature control adjustment method of air-cooled units based on adaptive model prediction as proposed in the above embodiment.

[0118] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0119] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the method for temperature control of an air-cooled unit based on adaptive model prediction as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0120] In summary, this invention can identify a hidden trend of uneven temperature deterioration in a spatial location, where the temperature difference with adjacent locations is too large even though the temperature at a certain location has not exceeded the limit. It intervenes in advance before spatial heat diffusion causes more locations to exceed the limit, avoiding the phenomenon of uniform control chasing after problems. This invention avoids over-adjusting all actuators to correct local problems through differentiated control, thereby reducing rather than increasing the total cooling energy consumption, breaking the traditional perception that differentiated control equals high energy consumption. In addition, this invention improves the uniformity of spatial temperature distribution and reduces the thermal fatigue of equipment.

[0121] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for temperature control of air-cooled units based on adaptive model prediction, characterized in that: include, The system collects air and water temperatures at multiple locations within the air-cooled unit; each location corresponds to an independent fan and cooling water valve, and preset temperature thresholds and spatial temperature difference thresholds are obtained. The air and water temperatures at the multiple spatial locations are input into the prediction model to obtain the predicted temperature of each spatial location at a future time, and the predicted temperature difference between adjacent spatial locations is calculated. The predicted temperature at each spatial location is compared with a temperature threshold, and the predicted temperature difference is compared with a spatial temperature difference threshold. Spatial locations where the predicted temperature exceeds the temperature threshold or the predicted temperature difference exceeds the spatial temperature difference threshold are identified as locations to be adjusted. For the position to be adjusted, the speed adjustment amount of the corresponding fan and the opening amount of the corresponding cooling water valve are determined based on the magnitude by which the predicted temperature exceeds the temperature threshold or the magnitude by which the predicted temperature difference exceeds the spatial temperature difference threshold. The speed adjustment is output to the fan in the corresponding spatial position, and the opening adjustment is output to the cooling water valve in the corresponding spatial position.

2. The method for temperature control of air-cooled units based on adaptive model prediction as described in claim 1, characterized in that: The step of inputting the air and water temperatures from the multiple spatial locations into the prediction model to obtain the predicted temperature for each spatial location at a future time, and calculating the predicted temperature difference between adjacent spatial locations, includes: The air temperature and water temperature at the multiple spatial locations are input into the prediction model to obtain the predicted temperature of each spatial location at multiple future moments within a preset time window. Select the predicted temperature corresponding to the target time from the predicted temperatures corresponding to the plurality of future times as the predicted temperature of each spatial location at the future time; For each spatial location, multiple adjacent spatial locations are determined, and the temperature deviation between the predicted temperature of the spatial location at a future time and the predicted temperature of each of the adjacent spatial locations at a future time is calculated. The temperature deviation is used as the predicted temperature difference between the adjacent spatial locations.

3. The air-cooled unit temperature control method based on adaptive model prediction as described in claim 2, characterized in that: The identification of spatial locations where the predicted temperature exceeds the temperature threshold or the predicted temperature difference exceeds the spatial temperature difference threshold as locations to be adjusted includes: For each spatial location, if the predicted temperature exceeds the temperature threshold, the spatial location is marked as a first type of location to be adjusted. For each spatial location, if the predicted temperature difference exceeds the spatial temperature difference threshold, the spatial location is marked as a second type of location to be adjusted. The first type of position to be adjusted and the second type of position to be adjusted are taken as the position to be adjusted.

4. The temperature control method for air-cooled units based on adaptive model prediction as described in claim 3, characterized in that: For the position to be adjusted, determining the corresponding fan speed adjustment and the corresponding cooling water valve opening adjustment based on the magnitude by which the predicted temperature exceeds the temperature threshold or the magnitude by which the predicted temperature difference exceeds the spatial temperature difference threshold includes: For the first type of position to be adjusted, calculate the first deviation between the predicted temperature and the temperature threshold, and determine the speed adjustment of the fan and the opening adjustment of the corresponding cooling water valve for the first type of position to be adjusted based on the first deviation. For the second type of position to be adjusted, calculate the second deviation between the predicted temperature difference and the spatial temperature difference threshold, and determine the speed adjustment of the fan and the opening adjustment of the corresponding cooling water valve for the second type of position to be adjusted based on the second deviation.

5. The temperature control method for air-cooled units based on adaptive model prediction as described in claim 4, characterized in that: The method further includes: For a spatial position that simultaneously belongs to both the first type of adjustment position and the second type of adjustment position, compare the magnitudes of the first deviation and the second deviation. If the first deviation is greater than the second deviation, the speed adjustment of the fan corresponding to the spatial position and the opening adjustment of the corresponding cooling water valve are determined based on the first deviation. If the second deviation is greater than or equal to the first deviation, the speed adjustment of the fan corresponding to the spatial position and the opening adjustment of the corresponding cooling water valve are determined based on the second deviation.

6. The temperature control method for air-cooled units based on adaptive model prediction as described in claim 5, characterized in that: The step of determining the corresponding fan speed adjustment and the corresponding cooling water valve opening adjustment based on the magnitude by which the predicted temperature exceeds the temperature threshold or the magnitude by which the predicted temperature difference exceeds the spatial temperature difference threshold includes: Obtain the preset fan adjustment parameters and water valve adjustment parameters; The fan speed adjustment is determined based on the magnitude of the predicted temperature exceeding the temperature threshold and the fan adjustment parameters; the cooling water valve opening adjustment is determined based on the magnitude of the predicted temperature exceeding the temperature threshold and the water valve adjustment parameters; or: The fan speed adjustment amount is determined based on the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold and the fan adjustment parameters. The opening adjustment amount of the corresponding cooling water valve is determined based on the magnitude of the predicted temperature difference exceeding the spatial temperature difference threshold and the water valve adjustment parameters.

7. The temperature control method for air-cooled units based on adaptive model prediction as described in claim 6, characterized in that: The calculation of the predicted temperature difference between adjacent spatial locations includes: For each spatial location, determine all adjacent spatial locations that are adjacent to the stated spatial location; Calculate multiple temperature deviations between the predicted temperature at the spatial location and the predicted temperature at each of the adjacent spatial locations; The largest temperature deviation among the plurality of temperature deviations is selected as the predicted temperature difference corresponding to the spatial location.

8. The method for temperature control of air-cooled units based on adaptive model prediction as described in claim 7, characterized in that: The step of inputting the air and water temperatures from the multiple spatial locations into the prediction model to obtain the predicted temperature for each spatial location at a future time includes: Obtain the air temperature and water temperature at the multiple spatial locations over a past time period; By inputting the air and water temperatures from the past time period, and the air and water temperatures at the current moment into the prediction model, the predicted temperature for each spatial location at a future time can be obtained.

9. The method for temperature control of air-cooled units based on adaptive model prediction as described in claim 8, characterized in that: The method further includes: Obtain the cooling load of the air-cooled unit; The cooling load is input as an additional input along with the air and water temperatures at the multiple spatial locations into the prediction model to obtain the predicted temperature for each spatial location at a future time.

10. A temperature control system for air-cooled units based on adaptive model prediction, based on any one of claims 1 to 9, characterized in that: include, The data acquisition module is used to collect air temperature and water temperature at multiple spatial locations within the air-cooled unit; Each spatial location corresponds to an independent fan and cooling water valve, and preset temperature thresholds and spatial temperature difference thresholds are obtained; The prediction module is used to input the air temperature and water temperature of the multiple spatial locations into the prediction model, obtain the predicted temperature of each spatial location at a future time, and calculate the predicted temperature difference between adjacent spatial locations. The identification module is used to compare the predicted temperature of each spatial location with the temperature threshold, compare the predicted temperature difference with the spatial temperature difference threshold, and identify spatial locations where the predicted temperature exceeds the temperature threshold or the predicted temperature difference exceeds the spatial temperature difference threshold as locations to be adjusted. The control quantity determination module is used to determine the speed adjustment amount of the corresponding fan and the opening amount of the corresponding cooling water valve for the position to be adjusted based on the magnitude by which the predicted temperature exceeds the temperature threshold or the magnitude by which the predicted temperature difference exceeds the spatial temperature difference threshold. The control output module is used to output the speed adjustment amount to the fan at the corresponding spatial position and the opening adjustment amount to the cooling water valve at the corresponding spatial position.

Images