Distributed dynamic defrosting control method for evaporative air coolers

Abstract

This invention discloses a distributed dynamic defrosting control method for evaporative air coolers, relating to the field of distributed defrosting technology for evaporative air coolers. The method includes the following steps: acquiring the spatial location and airflow direction of the evaporative air cooler in the refrigeration environment, as well as the defrosting exhaust status of adjacent evaporative air coolers; constructing a spatiotemporal perception map of the heat exchange status based on spatial coordinate relationships to determine whether the evaporative air cooler is disturbed by the defrosting airflow from neighboring evaporative air coolers; collecting temperature, humidity, current, and airflow data for the evaporative air cooler identified as being disturbed by the defrosting airflow from neighboring evaporative air coolers; constructing a heat exchange characteristic curve of the target evaporative air cooler in the current time period; and performing residual fitting calculations with a standard heat exchange curve under historical non-defrosting conditions to determine the local heat exchange parameter distortion caused by the disturbance of the evaporative air cooler by the defrosting airflow from neighboring evaporative air coolers. This invention solves the problem of falsely triggered defrosting caused by the disturbance of defrosting airflow from evaporative air coolers, achieving accurate identification and dynamic control optimization of defrosting needs.

Description

Technical Field

[0001] This invention relates to the field of distributed defrosting technology for evaporative air coolers, and more specifically to a distributed dynamic defrosting control method for evaporative air coolers. Background Technology

[0002] Distributed dynamic defrosting control for evaporative air coolers is an intelligent defrosting management method applied to multi-point evaporative air cooling systems. Its core lies in deploying multiple distributed control nodes within the evaporative air cooler system. These nodes collect real-time data such as temperature and humidity, surface frost status, and fan operating parameters within their respective areas. The data is then uploaded to a central control module or edge computing node via a communication network (such as CAN bus, Modbus, or wireless communication). Based on the environmental and operational status feedback from each node, the system uses preset control algorithms (such as fuzzy control, PID control, or adaptive algorithms) to determine whether a defrosting procedure needs to be initiated. This allows for on-demand, dynamic, and localized defrosting of frost-covered areas. The entire process mainly includes five key stages: data acquisition, data transmission, intelligent judgment, defrosting execution, and feedback adjustment. First, distributed sensor nodes continuously collect temperature, humidity, and frost information and upload it in real time. Second, the system performs data fusion analysis to determine whether each evaporative air cooler has reached the defrosting threshold. Third, defrosting control commands are issued to evaporative air cooler nodes that meet the defrosting conditions (such as switching heat exchange modes, activating heating devices, or stopping defrosting). Then, the defrosting effect is continuously monitored during the defrosting process, and the defrosting strategy is dynamically adjusted. Finally, the defrosting process and effect feedback are used for subsequent strategy optimization. Compared to traditional timed or unified control defrosting methods, this method improves defrosting efficiency and system energy efficiency, avoids unnecessary energy consumption and performance loss, and is particularly suitable for evaporative air cooler systems in complex environments with large variations in operating load.

[0003] The existing technology has the following shortcomings:

[0004] In refrigerated environments employing distributed dynamic defrosting control methods for evaporative air coolers, the air coolers are distributed across areas, each independently performing dynamic defrosting control by sensing local parameters such as temperature, humidity, airflow, and current. When an air cooler is not in defrosting mode, if its area is disturbed by hot and humid airflow from adjacent air coolers that are defrosting, it may experience short-term temperature rises and humidity disturbances in that area, causing abnormal fluctuations in the heat exchange performance parameters sensed by that air cooler. Because current distributed dynamic defrosting control technology primarily relies on node self-sensing and self-judgment mechanisms, it lacks the ability to identify external interference sources causing abnormalities in locally sensed data. This results in the system being unable to distinguish whether such parameter fluctuations are caused by frost accumulation on the air cooler itself or by the defrosting exhaust behavior of external air coolers. Therefore, in the above situation, existing distributed dynamic defrosting control technology for evaporative air coolers cannot accurately determine whether a defrosting operation is truly necessary based on the distortion of local heat exchange parameters caused by the defrosting airflow interference from neighboring air coolers. Such misjudgments will cause evaporative coolers that do not require defrosting to be frequently triggered in the defrosting process. This will not only disrupt their normal operation rhythm, but also affect the collaborative judgment of other nodes due to the erroneous state being included in the system calculation model. This will lead to a chain of false triggers in the defrosting control chain, resulting in increased system energy consumption, reduced heat exchange efficiency, and aggravated regional temperature fluctuations. This will seriously affect the stability and reliability of the dynamic defrosting control of the distributed evaporative cooler network.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a distributed dynamic defrosting control method for evaporative air coolers to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a distributed dynamic defrosting control method for evaporative air coolers, specifically comprising the following steps:

[0008] S1. Obtain the spatial location, airflow direction and other information of the evaporative cooler in the refrigeration environment, as well as the defrosting and exhaust status of adjacent evaporative coolers. Construct a spatiotemporal perception map of the heat exchange status based on the spatial coordinate relationship to determine whether the evaporative cooler is disturbed by the defrosting airflow of the adjacent evaporative cooler.

[0009] S2. Collect temperature, humidity, current and air volume data for the air cooler that has been identified as being disturbed by the defrosting airflow of the adjacent air cooler, construct the heat exchange characteristic curve of the target air cooler in the current period, and perform residual fitting calculation with the standard heat exchange curve under historical non-defrosting conditions to determine the local heat exchange parameter distortion caused by the air cooler being disturbed by the defrosting airflow of the adjacent air cooler.

[0010] S3. The local heat exchange parameter distortion is fused with the real-time heat exchange data of the air cooler. The fusion result is corrected by interference weight. The corrected fused data is input into the defrost reliability assessment model to generate defrost reliability assessment parameters. These parameters are used to determine whether the air cooler really needs to perform defrost operation based on the local heat exchange parameter distortion caused by the airflow interference from the defrost air of the neighboring air cooler.

[0011] S4. Based on the defrosting reliability assessment parameters, combined with the historical operating characteristics of the air cooler and the load fluctuation of the cold storage area, a judgment label is generated to classify the air cooler into three states: needing immediate defrosting, needing delayed observation, and not needing defrosting.

[0012] S5. Based on the judgment label corresponding to the evaporative air cooler, a differentiated defrosting control process is executed, and the data generated by each evaporative air cooler in the current cycle is used to update the spatiotemporal perception map of the heat exchange status, so as to realize the optimization of defrosting control under dynamic regulation.

[0013] Preferably, S1 specifically includes the following steps:

[0014] S101. By deploying a spatial identification unit in the cold storage environment, the spatial position of each air cooler is obtained. Combined with the wind direction and wind speed detection components configured at the air cooler outlet, the air direction and flow information of the air cooler under different operating conditions is obtained. At the same time, based on the air cooler operation control logic and heating start signal, it is identified whether the adjacent air cooler is in the defrosting and exhaust state.

[0015] S102. Based on the spatial location and airflow direction information of the evaporative cooler, establish a spatial coordinate relationship, take the evaporative cooler in the defrost and exhaust state as the source of heat and humidity disturbance, call the three-dimensional disturbance diffusion prediction model to calculate its influence radius and directional propagation path, and then construct a spatiotemporal perception map of heat transfer state that includes the airflow interaction relationship between evaporative coolers.

[0016] S103. By analyzing the spatial angle between the air cooler and the disturbance source, the mapping relationship between the airflow overlap area and the disturbance intensity, the disturbed area is identified in the spatiotemporal perception map of the heat exchange state, and the air cooler is determined to be disturbed by the defrosting airflow of the adjacent air cooler according to the preset disturbance judgment rules.

[0017] Preferably, S103 is as follows:

[0018] Calculate the spatial angle between the air cooler and the disturbance source point, and use the overlap between the angle range and the main direction of airflow as the first interference index. Screen potential interference directions through spatial paths with an angle smaller than the set interference threshold.

[0019] Extract the spatial volume of the airflow overlap region, construct a disturbance intensity mapping relationship between the volume ratio and the superposition intensity of airflow velocity, and use it as the second disturbance index. Mark the overlapping region of airflow propagation trajectory and intensity distribution in the heat transfer state spatiotemporal perception map, and identify the disturbed region that meets the disturbance index weight threshold.

[0020] The preset interference judgment rules are based on the angle threshold judgment, the disturbance intensity coverage ratio judgment, and the airflow interference duration judgment. The angle threshold judgment is based on whether the angle between the air outlet direction of the evaporator and the direction of the disturbance source is lower than the set interference threshold. The disturbance intensity coverage ratio judgment is based on whether the proportion of the disturbed area on the heat exchange surface of the evaporator exceeds the preset limit ratio. The airflow interference duration judgment is based on whether the duration of the disturbance effect exceeds the preset time window. An evaporator that meets at least two of the above three judgment conditions is identified as an evaporator affected by the defrosting airflow of a neighboring evaporator.

[0021] Preferably, S2 specifically includes the following steps:

[0022] S201. By deploying temperature sensors, humidity sensors, current sensors, and airflow detection components on the surface of the heat exchanger inside the evaporative cooler and along its air inlet and outlet paths, the system collects in real time the temperature, humidity, current, and airflow data of the evaporative cooler that has been identified as being affected by the defrosting airflow of a neighboring evaporative cooler during its current operating period. The system also performs time alignment and data loss repair on different types of sensor data to ensure the continuity and consistency of the input data.

[0023] S202. The processed temperature, humidity, current and air volume data are integrated into the heat exchange behavior modeling engine according to the time series. The heat exchange characteristic factors are extracted by the multivariate feature extraction algorithm. The heat exchange characteristic curve of the target air cooler in the current period is constructed in the same time dimension, so that it fully reflects the linkage between heat load change and operating status.

[0024] S203. Perform residual fitting calculation on the heat exchange characteristic curve of the target air cooler in the current period and the standard heat exchange curve selected in the historical data under non-defrost conditions. By analyzing the offset amplitude, trend and duration of the residual interval, extract the local heat exchange parameter distortion caused by the air cooler being disturbed by the defrost airflow of the adjacent air cooler, and use it as the input for subsequent defrost judgment logic.

[0025] Preferably, S202 specifically refers to:

[0026] Temperature, humidity, current and air volume data are aligned sequentially on the same time axis. A sliding window mechanism is used to extract sensor data segments of continuous time periods at equal intervals. Bidirectional interpolation is used to complete and correct missing or abnormally fluctuating data to ensure that multi-source data in the time series maintains synchronization and data integrity at each time point.

[0027] The time-series aligned data is input into the heat exchange behavior modeling engine. Through the joint principal component analysis algorithm and cluster decomposition algorithm, heat exchange characteristic factors are extracted from the temperature, humidity, current and air volume data, and redundant and low-correlation factors are removed to form a set of feature vectors to describe the heat exchange behavior state of the target air cooler.

[0028] The feature vector set is mapped according to the time series to construct the heat exchange characteristic curve of the target air cooler in the current period. The vertical axis is the refined heat exchange characteristic factor group, and the horizontal axis is the time axis consistent with the sensor data. The heat exchange characteristic curve reflects the response law of the target air cooler's current heat exchange capacity and operating status under dynamic changes in heat load.

[0029] Preferably, S203 is as follows:

[0030] In the historical operating data, the time period that is closest to the current cold storage environment load status and operating conditions is selected, and the standard heat exchange curve of the target air cooler during the time period is extracted. It is ensured that the standard curve does not contain data interference during the defrosting period, and it is used as the reference for the current heat exchange characteristic curve.

[0031] The heat exchange characteristic curve of the target air cooler in the current period is compared with the standard heat exchange curve on the time axis. The residual fitting calculation is performed by using the sliding residual window method. The numerical deviation and directional change of the characteristic factor are calculated in each window to obtain the residual time series and remove abnormal peak disturbance values.

[0032] Dynamic clustering and segmentation analysis is performed on the residual time series to extract residual interval segments whose offset exceeds the warning interval threshold. Combining the unidirectionality and duration characteristics of the offset change trend in these interval segments, it is determined that they represent the local heat exchange parameter distortion caused by the interference of the defrosting airflow of the adjacent air cooler. This distortion is then passed on to the subsequent defrosting judgment logic.

[0033] Preferably, S3 specifically includes the following steps:

[0034] S301. The local heat exchange parameter distortion caused by the interference of the defrosting airflow of the adjacent air cooler is fused with the real-time heat exchange data collected by the air cooler in the current period according to the time dimension. In the fusion process, the feature vector weighting operation is used to superimpose the features in the same dimension to form fused heat exchange feature data that can characterize the heat exchange state of the air cooler.

[0035] S302. Calculate the interference weight based on the spatial relationship between the air cooler and the disturbance source, the airflow superposition intensity and the interference coverage ratio. Apply the interference weight to the feature dimensions selected by the interference index weight threshold in the fused heat exchange feature data. Perform weighted correction on each feature dimension and output the corrected fused data after interference compensation.

[0036] S303. Input the corrected fusion data into the defrost reliability assessment model, and use the reliability classification algorithm and heat exchange performance deviation identification algorithm trained in the assessment model to generate defrost reliability assessment parameters. The defrost reliability assessment parameters include the degree of heat exchange performance degradation, interference offset factor and duration index. Based on whether the degree of heat exchange performance degradation continuously exceeds the threshold of the normal operation range, whether the interference offset factor exceeds the threshold of the interference source, and whether the duration index exceeds the preset time threshold, determine that the air cooler that meets at least two of the index conditions is the air cooler that really needs to perform defrost operation.

[0037] Preferably, S302 is as follows:

[0038] The spatial positional relationship parameters between the air cooler and the disturbance source point in the current time period are obtained. The angle between the air cooler's air outlet direction and the disturbance source propagation direction is calculated. The airflow superposition intensity parameters are determined by combining the spatial intersection ratio of the airflow superposition area and the wind speed information of the disturbance source point propagation path. The volume ratio parameters of the interference area covered by the heat exchange surface of the air cooler are collected to form the input dataset used to construct the interference weight.

[0039] A weighting function is constructed based on the spatial location angle, airflow superposition intensity parameter and interference coverage ratio parameter to generate interference weight value that characterizes the degree of interference. The interference weight value is applied to each feature dimension in the fused heat transfer feature data, and the feature dimensions whose interference sensitivity coefficient exceeds the interference index weight threshold are extracted to form a set of feature dimensions to be corrected.

[0040] For each feature dimension in the set of features to be corrected, a weighted correction operation is performed. The interference weight value is weighted and superimposed with the original value of the corresponding feature dimension to generate corrected feature data to replace the original value. Finally, the corrected fusion data after interference compensation is output as the input data source for the subsequent defrosting reliability assessment model.

[0041] Preferably, S4 is as follows:

[0042] Based on the joint state of the heat exchange performance degradation degree, interference offset factor and duration index in the defrosting reliability assessment parameters, the numerical characteristics of various heat exchange parameters of the evaporative cooler in the historical operating cycle are extracted. A historical operating feature vector including the heat exchange efficiency reduction, current load response amplitude and humidity fluctuation stability is constructed to describe the heat exchange behavior of the evaporative cooler under different load conditions.

[0043] Collect load fluctuation data of the current cold storage area, extract feature factors including spatial temperature and humidity distribution, cargo circulation frequency and transient cold load change rate, input the historical operation feature vector and the current load fluctuation factor into the defrost status judgment engine, and generate the current defrost judgment label of the air cooler through the trained judgment model.

[0044] Based on the output results of the judgment label, the air cooler is divided into three states: immediate defrosting required, delayed observation required, and no defrosting required. The state of immediate defrosting is characterized by the heat exchange performance degradation level continuously exceeding the threshold of the normal operating range, the interference offset factor exceeding the threshold of the interference source, and the duration index exceeding the preset time threshold. The state of delayed observation is characterized by the heat exchange performance degradation level being within the upper limit of the threshold of the normal operating range, and either the interference offset factor or the duration index being satisfied. The state of no defrosting is characterized by the heat exchange performance degradation level, the interference offset factor, and the duration index not exceeding their respective limit thresholds.

[0045] Preferably, S5 is as follows:

[0046] Based on the judgment tag corresponding to the evaporative air cooler, a differentiated defrosting control process is executed. Evaporative air coolers with the judgment tag indicating that they need to be defrosted immediately are included in the instant defrosting control sequence and a defrosting execution signal is sent. Evaporative air coolers with the judgment tag indicating that they need to be observed later are marked as key monitoring objects and the heat exchange data acquisition frequency is increased. Evaporative air coolers with the judgment tag indicating that they do not need to be defrosted maintain their original operating parameters and retain the observation task records.

[0047] The data generated by each air cooler in the current cycle, including heat exchange characteristic data, interference identification indicators and operating status parameters, are input into the heat exchange status spatiotemporal perception map update engine according to the time index and spatial location mapping method. The status labels of air cooler nodes, the edge weights of adjacent airflow interference relationships and the spatial heat load distribution results in the map are updated synchronously.

[0048] Based on the updated spatiotemporal perception map of heat exchange status, the heat exchange behavior correlation network of each air cooler in the current operating cycle is reconstructed. The heat load distribution status, defrosting interference path and control strategy adaptability among the air coolers are comprehensively analyzed to generate the defrosting control priority ranking result for the next round, thereby realizing the optimization of defrosting control under dynamic regulation.

[0049] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0050] 1. This invention constructs a spatiotemporal sensing map of heat exchange status, integrating information on the spatial location of the evaporative cooler, airflow direction, and defrosting exhaust status. This breaks through the limitations of traditional single-point sensing mechanisms, achieving for the first time spatial identification and propagation path prediction of airflow interference sources in the evaporative cooler's operating environment. By combining multi-source parameters such as temperature, humidity, current, and airflow to construct heat exchange characteristic curves, and introducing residual fitting and interference offset analysis algorithms, it effectively extracts the local heat exchange parameter distortion caused by external hot and humid airflow interference, providing highly reliable basic data for determining whether defrosting is truly required. This method significantly enhances the system's ability to identify and filter false defrosting signals, reducing the risk of energy consumption fluctuations and operational instability caused by misjudged defrosting.

[0051] 2. This invention introduces a multi-dimensional decision-making mechanism for reliable defrosting assessment parameters and constructs a defrosting judgment label system by combining historical operating characteristics and cold storage area load fluctuation information. This supports graded control of evaporative coolers, classifying them as "requiring immediate defrosting," "requiring delayed observation," and "not requiring defrosting," further improving the accuracy and rhythm matching of defrosting regulation. Based on this, the constructed data closed-loop mechanism can feed back the core sensing data, correction data, and judgment results from each round of decision-making to the spatiotemporal sensing map of the heat exchange status in real time. This enables dynamic learning and optimization of the heat load distribution, defrosting interference paths, and control strategy adaptability in the evaporative cooler network, giving the defrosting control strategy continuous evolution capabilities and significantly improving the energy efficiency, heat exchange performance, and operational reliability of the distributed evaporative cooler system under complex operating conditions. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0053] Figure 1 This is a flowchart illustrating the distributed dynamic defrosting control method for air coolers according to the present invention. Detailed Implementation

[0054] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0055] This invention provides, for example Figure 1 The distributed dynamic defrosting control method for evaporative air coolers shown includes the following steps:

[0056] S1. Obtain the spatial location, airflow direction and other information of the evaporative cooler in the refrigeration environment, as well as the defrosting and exhaust status of adjacent evaporative coolers. Construct a spatiotemporal perception map of the heat exchange status based on the spatial coordinate relationship to determine whether the evaporative cooler is disturbed by the defrosting airflow of the adjacent evaporative cooler.

[0057] In this embodiment, S1 specifically includes the following steps:

[0058] S101. By deploying a spatial identification unit in the cold storage environment, the spatial position of each air cooler is obtained. Combined with the wind direction and wind speed detection components configured at the air cooler outlet, the air direction and flow information of the air cooler under different operating conditions is obtained. At the same time, based on the air cooler operation control logic and heating start signal, it is identified whether the adjacent air cooler is in the defrosting and exhaust state.

[0059] Spatial identification units are deployed in the refrigerated environment to obtain the spatial position of each air cooler. This can be achieved by pre-setting positioning reference markers within the refrigerated area and integrating an ultra-wideband positioning chip, lidar unit, or high-precision inertial navigation device into the air cooler itself. The precise coordinates of the air cooler in three-dimensional space are determined through ranging feedback or coordinate fusion algorithms. Wind direction and speed detection components are installed at the air cooler's outlet, employing a hot-film anemometer and a differential pressure wind direction sensor. These components synchronously collect wind speed vector data and outlet direction angle information when the air cooler is in different operating states such as air supply, defrosting, and standby, constructing a database of airflow behavior parameters for each state. The air cooler's operation control logic obtains its operating mode switching signals, such as air supply and defrosting start control signals, through connection to the main controller. The heating start signal serves as a key trigger signal for the defrosting process; by collecting the on / off status of the electric heater or heat pump reversing device, it identifies whether the air cooler is currently in the defrosting and exhaust process. By logically mapping spatial location, wind direction, and heating signals, it is possible to determine in real time whether the evaporative cooler is acting as a source of disturbance to the hot and humid airflow during defrosting. This method provides fundamental dynamic input data for subsequently constructing airflow propagation paths and identifying interference areas, avoiding misjudgments or delays caused by relying solely on static deployment information.

[0060] S102. Based on the spatial location and airflow direction information of the evaporative cooler, establish a spatial coordinate relationship, take the evaporative cooler in the defrost and exhaust state as the source of heat and humidity disturbance, call the three-dimensional disturbance diffusion prediction model to calculate its influence radius and directional propagation path, and then construct a spatiotemporal perception map of heat transfer state that includes the airflow interaction relationship between evaporative coolers.

[0061] Based on the spatial location and wind direction information of the evaporative air coolers, a spatial coordinate grid can be constructed to map the location data of each evaporative air cooler into a three-dimensional spatial model. Combined with wind direction angle and wind speed vector information, the main propagation direction and initial velocity field of the airflow are marked within the coordinate model. For evaporative air coolers currently in defrosting and exhaust mode, they can be used as disturbance sources. A three-dimensional disturbance diffusion prediction model optimized based on CFD (Computational Fluid Dynamics) simulation is used to simulate disturbances, taking into account the airflow rate, wind direction angle, and ambient temperature and humidity background fields. This model is based on the physical diffusion characteristics of hot and humid air masses propagating in a refrigerated environment, considering parameters such as turbulent diffusion, temperature difference convection, and airflow obstruction structures, to calculate the influence radius and propagation path of the disturbance source within a certain time window. For example, when an evaporative air cooler discharges high-humidity and high-temperature airflow forward during defrosting, the model can predict that this airflow will form a disturbance field within a 2-meter radius in front, generating a certain intensity of residual wave propagation in the wind direction deflection area, ultimately resulting in localized temperature rise and humidity anomalies on the heat exchange surface. By superimposing the airflow propagation paths of multiple air coolers, a spatiotemporal perception map containing airflow overlap, interaction, and boundary reflection effects can be generated, which can be used to further analyze the range and intensity distribution of interference.

[0062] The spatial location of the evaporative air cooler is a core parameter for constructing the three-dimensional coordinate foundation of the refrigeration environment, and can be obtained in real time through spatial positioning equipment. Airflow direction information describes the directional characteristics of the airflow under different operating conditions of the evaporative air cooler, serving as a crucial basis for modeling disturbance paths. Evaporative air coolers in defrost and exhaust mode are selected as the source of thermal and humidity disturbances because the high-temperature, high-humidity gases released in this state are most likely to affect the heat exchange perception of adjacent equipment. The three-dimensional disturbance diffusion prediction model is an algorithm model trained based on the principles of gas dynamics in the refrigeration environment, capable of calculating the disturbance propagation boundary and path direction in real time. The influence radius represents the boundary range within which the disturbance produces significant heat exchange parameter disturbances in space, and the directional propagation path represents the mainstream movement direction of the disturbing airflow and its path line. The spatiotemporal perception map of heat exchange status is a multi-dimensional visual data structure that integrates spatial coordinates, airflow direction, disturbance path, and intensity annotations. It can be used to accurately analyze the airflow interaction relationships between evaporative air coolers and serves as a basis for subsequent judgment on whether disturbances have occurred.

[0063] S103. By analyzing the spatial angle between the air cooler and the disturbance source, the mapping relationship between the airflow overlap area and the disturbance intensity, the disturbed area is identified in the spatiotemporal perception map of the heat exchange state, and the air cooler is determined to be disturbed by the defrosting airflow of the adjacent air cooler according to the preset disturbance judgment rules.

[0064] The reason for analyzing the spatial angle between the evaporative cooler and the disturbance source, the mapping relationship between the airflow overlap area and the disturbance intensity, to identify the disturbed area in the spatiotemporal perception map of the heat exchange state, and to determine whether the evaporative cooler is disturbed by the defrosting airflow of neighboring evaporative coolers based on preset disturbance judgment rules, is that in the distributed dynamic defrosting control process of evaporative coolers, each evaporative cooler needs to rely on local sensing parameters to make defrosting judgments. However, when a neighboring evaporative cooler is defrosting and releasing hot and humid airflow, this airflow will propagate along the spatial path and overlap with the airflow field of the undefrosted evaporative cooler, causing abnormal fluctuations in local temperature, humidity, and other sensing data, thereby interfering with the accuracy of defrosting judgments. By constructing a spatiotemporal perception map of the heat exchange state and comprehensively considering dimensional features such as spatial angle, airflow overlap degree, and disturbance intensity, the area affected by external disturbances can be accurately identified, and the source of abnormal sensing data can be effectively distinguished. Furthermore, by combining preset interference judgment rules based on multiple conditions, the reliability of identification can be further improved, avoiding misjudgments of "frost state" due to short-term external disturbances, and fundamentally enhancing the stability and energy efficiency of distributed dynamic defrosting control of evaporative air coolers. This strategy helps the system abstract interference sources and influence paths from complex spatial interaction relationships, providing a reliable and precise basis for subsequent defrosting decisions.

[0065] In this embodiment, S103 specifically refers to:

[0066] Calculate the spatial angle between the air cooler and the disturbance source point, and use the overlap between the angle range and the main direction of airflow as the first interference index. Screen potential interference directions through spatial paths with an angle smaller than the set interference threshold.

[0067] To determine whether an evaporative air cooler is potentially affected by interference, we can first obtain the airflow direction vector of the cooler and the main propagation direction vector of the airflow from the disturbance source. Then, we calculate the spatial angle between these two vectors using a three-dimensional coordinate system. This spatial angle reflects the geometric relationship between the cooler and the disturbance source along the airflow propagation path. The smaller the angle, the more collinear they are, and the more likely the cooler's direction is to be directly impacted by the airflow disturbance. To quantify this relationship, the degree of overlap between the angle range and the main propagation direction of the airflow from the disturbance source can be used as the first interference indicator. A fixed angle threshold can be set to filter potentially affected spatial paths. For example, setting the angle threshold to 15 degrees indicates that the cooler is located on the main channel of the airflow propagation from the disturbance source, with a high probability of being affected. This angle threshold was set after multiple rounds of simulation and regression analysis of measured data on airflow behavior in the refrigerated environment, effectively distinguishing between significantly disturbed and non-disturbed airflow paths. The angle calculation employs a three-dimensional vector angle formula, involving vector projection operations between the airflow direction of the evaporative cooler and the diffusion path of the disturbance source. By filtering out all paths with angles less than a threshold, evaporative coolers in the direction of the disturbed airflow can be identified, thereby narrowing the scope of subsequent interference analysis and improving judgment accuracy. This method can accurately characterize the spatial directionality of airflow propagation, which is crucial for determining whether an evaporative cooler is truly disturbed by the defrosting airflow from neighboring evaporative coolers.

[0068] Extract the spatial volume of the airflow overlap region, construct a disturbance intensity mapping relationship between the volume ratio and the superposition intensity of airflow velocity, and use it as the second disturbance index. Mark the overlapping region of airflow propagation trajectory and intensity distribution in the heat transfer state spatiotemporal perception map, and identify the disturbed region that meets the disturbance index weight threshold.

[0069] To further determine whether the evaporative air cooler is actually disturbed by the defrosting airflow from neighboring evaporative air coolers, it is necessary to extract the spatial volume of the overlapping airflow region from the 3D spatial model. This spatial volume represents the overlapping portion between the current heat exchange influence range of the evaporative air cooler and the airflow propagation path of the disturbance source. By establishing a heat exchange volume model for each evaporative air cooler and an airflow propagation trajectory model for the disturbance source, the overlapping spatial region in the refrigeration environment can be extracted using 3D volume Boolean intersection operations. To measure the potential disturbance level of this overlapping region, the proportion of the overlapping region within the heat exchange surface area of ​​the evaporative air cooler can be used as a volume percentage indicator. This is combined with a weighted sum of the airflow velocity at the disturbance source within the overlapping region to construct a disturbance intensity mapping relationship, used to measure the actual impact of the disturbance intensity within the affected area. For example, when the overlapping region accounts for 40% of the evaporative air cooler's heat exchange area volume, and the disturbance airflow velocity remains high in this region, the system will mark this region as a high-risk disturbance area. In the spatiotemporal perception map of the heat exchange status, these high-intensity disturbance areas are visually marked with different colors or symbols, providing an intuitive basis for judgment. The interference index weighting threshold is a reference standard set through experimental statistics and comparative analysis of energy efficiency changes under numerous airflow interference scenarios. It is typically defined as a validly interfered area only when both the volume percentage exceeds 30% and the superposition intensity exceeds a set value. This threshold is used to filter out low-intensity, occasional airflow overlap areas, ensuring that interference determination has high confidence and practical response value. This method avoids misjudgments caused by edge contact of airflows, while focusing on key interference areas that may truly affect heat exchange performance, providing accurate basic data support for subsequent determination of whether the evaporative cooler needs defrosting.

[0070] The preset interference judgment rules are based on the angle threshold judgment, the disturbance intensity coverage ratio judgment, and the airflow interference duration judgment. The angle threshold judgment is based on whether the angle between the air outlet direction of the evaporator and the direction of the disturbance source is lower than the set interference threshold. The disturbance intensity coverage ratio judgment is based on whether the proportion of the disturbed area on the heat exchange surface of the evaporator exceeds the preset limit ratio. The airflow interference duration judgment is based on whether the duration of the disturbance effect exceeds the preset time window. An evaporator that meets at least two of the above three judgment conditions is identified as an evaporator affected by the defrosting airflow of a neighboring evaporator.

[0071] To accurately determine whether an evaporative air cooler is affected by the defrosting airflow from a neighboring evaporative air cooler, three preset interference judgment rules can be used in conjunction: angle threshold judgment, disturbance intensity coverage ratio judgment, and airflow interference duration judgment. The angle threshold judgment analyzes the spatial angle formed between the air cooler's outlet direction and the propagation direction of the disturbance source airflow. If the angle is less than the preset interference angle threshold, it indicates that the two airflows have a consistent height direction, increasing the possibility of overlapping interference. The angle threshold is generally determined based on the refrigeration space structure and fan layout characteristics; for example, 15 degrees is used as an empirical reference. The disturbance intensity coverage ratio judgment uses the proportion of the affected area within the evaporative air cooler's heat exchange surface as the core indicator. It is usually calculated automatically using 3D modeling software as the percentage of the overlapping volume to the total heat exchange volume. If this proportion exceeds a preset limit, the evaporative air cooler is considered to be substantially affected during heat exchange. This limit is selected based on a sensitive threshold for the decrease in the evaporative air cooler's heat exchange efficiency and is usually set to 30% or higher. The duration of airflow interference is determined by continuously monitoring the exhaust status of the disturbance source and the duration of its continuous effect on the target air cooler's spatial area. This is combined with data acquisition cycles to accumulate statistics over a time window. When the duration exceeds a preset time window, such as 5 or 10 minutes, it indicates that the interference is not an occasional, instantaneous phenomenon but has a sustained impact. These three judgment indicators are independent yet complementary. The included angle threshold reflects the spatial trend, the coverage ratio reflects the degree of influence, and the duration reflects the intensity of the effect. The system is set to consider interference valid if at least two of these indicators are met, effectively avoiding misjudgment based on a single indicator and improving the accuracy and reliability of interference identification. This joint judgment mechanism not only enhances anti-interference capabilities but also provides a more reliable data foundation for subsequent dynamic defrosting strategies.

[0072] S2. Collect temperature, humidity, current and air volume data for the air cooler that has been identified as being disturbed by the defrosting airflow of the adjacent air cooler, construct the heat exchange characteristic curve of the target air cooler in the current period, and perform residual fitting calculation with the standard heat exchange curve under historical non-defrosting conditions to determine the local heat exchange parameter distortion caused by the air cooler being disturbed by the defrosting airflow of the adjacent air cooler.

[0073] In this embodiment, S2 specifically includes the following steps:

[0074] S201. By deploying temperature sensors, humidity sensors, current sensors, and airflow detection components on the surface of the heat exchanger inside the evaporative cooler and along its air inlet and outlet paths, the system collects in real time the temperature, humidity, current, and airflow data of the evaporative cooler that has been identified as being affected by the defrosting airflow of a neighboring evaporative cooler during its current operating period. The system also performs time alignment and data loss repair on different types of sensor data to ensure the continuity and consistency of the input data.

[0075] In refrigerated environments, to achieve high-precision sensing of the heat exchange status of a target evaporative air cooler, thermocouple array temperature sensors are uniformly distributed on the surface of the heat exchanger fins. High-speed response humidity sensors and airflow detection components are installed on both sides of the air inlet and outlet, and a current acquisition module is installed at the power input terminal to obtain the target evaporative air cooler's operating parameters such as temperature, humidity, airflow, and current at different times. The data acquisition process uses a uniform sampling frequency and is synchronized via a timestamp mechanism through a data interface to align different types of sensor data along the time axis. To address data interruptions caused by network fluctuations or temporary sensor malfunctions, a linear regression interpolation algorithm based on a sliding window is used to fill in missing values. This is combined with slope constraints at neighboring points to eliminate abrupt outliers, ensuring the temporal continuity and numerical stability of the data. For example, if the airflow data is missing continuous values ​​within 3 seconds in a certain sampling, linear fitting interpolation can be performed using sliding window values ​​of 5 seconds before and after the missing values. The reasonableness of the fitted value is then judged based on the airflow change trend to prevent error propagation into the subsequent modeling process.

[0076] The temperature sensor is deployed at the center and transition areas on both sides of the heat exchanger fins, reflecting the temperature gradient distribution during the evaporative cooler's heat load response. The humidity sensor, employing a fast-response thin-film capacitor structure, is installed inside the cooler's outlet duct to effectively sense the humidity characteristics of the output air. The airflow detection component, either impeller speed-based or differential pressure-based, is positioned between the inlet and outlet to synchronously detect airflow fluctuations. The current sensor uses a Hall effect closed-loop measurement scheme to capture the current response characteristics of the cooler's fan under load changes. Data from these four types of sensors is recorded at millisecond intervals and enters a unified data buffer channel. High-precision timestamps ensure consistency in data timing across channels. During data repair, bidirectional moving averages and central difference methods are used to identify discontinuous segments. Conditional regression fitting and interpolation preserve true operational characteristics while minimizing data distortion, providing a high-quality, low-noise input foundation for subsequent heat exchange behavior modeling.

[0077] S202. The processed temperature, humidity, current and air volume data are integrated into the heat exchange behavior modeling engine according to the time series. The heat exchange characteristic factors are extracted by the multivariate feature extraction algorithm. The heat exchange characteristic curve of the target air cooler in the current period is constructed in the same time dimension, so that it fully reflects the linkage between heat load change and operating status.

[0078] S203. Perform residual fitting calculation on the heat exchange characteristic curve of the target air cooler in the current period and the standard heat exchange curve selected in the historical data under non-defrost conditions. By analyzing the offset amplitude, trend and duration of the residual interval, extract the local heat exchange parameter distortion caused by the air cooler being disturbed by the defrost airflow of the adjacent air cooler, and use it as the input for subsequent defrost judgment logic.

[0079] The fundamental purpose of this approach is to distinguish between fluctuations in heat exchange performance parameters caused by airflow interference from neighboring evaporative air coolers and genuine defrosting needs arising from frost buildup within the refrigerated environment where the evaporative air cooler operates, thereby improving the accuracy of defrosting control decisions. Under complex airflow interference, the heat exchange parameters of the evaporative air cooler may experience short-term temperature rises, current changes, or abnormal airflow. Traditional distributed defrosting control, relying on a single-node self-sensing mechanism, struggles to identify whether these disturbances are caused by external interference or actual frost blockage. By fitting the residuals of the current evaporative air cooler's heat exchange characteristic curve with historical standard curves, the differences in actual operating performance can be quantified. The magnitude, trend, and duration of the residual shift reveal whether these differences are systematic and persistent, and whether they conform to typical characteristics of airflow interference. This analysis not only provides a clear decision-making basis for whether to trigger defrosting operations but also constructs a reliable judgment chain connecting sensing data and control execution, avoiding cascading false triggers of the defrosting chain due to misjudgment, and ensuring stable temperature control and energy efficiency in the refrigerated environment.

[0080] In this embodiment, S202 specifically refers to:

[0081] Temperature, humidity, current and air volume data are aligned sequentially on the same time axis. A sliding window mechanism is used to extract sensor data segments of continuous time periods at equal intervals. Bidirectional interpolation is used to complete and correct missing or abnormally fluctuating data to ensure that multi-source data in the time series maintains synchronization and data integrity at each time point.

[0082] To achieve sequence alignment of temperature, humidity, current, and airflow data on a unified time axis, the data from the four types of sensors should first be standardized and synchronized based on a unified timestamp system. All sensor data should be remapped to a millisecond-level time axis, and a sampling period reference point should be set as the alignment benchmark. Using a sliding window mechanism, continuous, fixed-length data segments can be divided at a set time step, for example, every 30 seconds as a window, sliding for 10 seconds each time, thus generating overlapping data segments across multiple time periods, facilitating local trend judgment and anomaly identification. Within the window, if the airflow or current value is missing at a certain time point, or if a sudden jump occurs, bidirectional interpolation can be applied to repair the value at that point. Bidirectional interpolation involves selecting several valid data points before and after the target point, calculating predicted values ​​from both the forward and backward directions using linear interpolation or cubic spline interpolation, and then taking a weighted average to fill in the missing data, thereby improving the smoothness and robustness of the interpolation. For example, if the airflow is missing at t=80s due to signal fluctuations during the operation of a certain air cooler, valid data points can be extracted between t=75s and t=79s and between t=81s and t=85s to construct forward and backward interpolation models. The weighted average of these two models is then used as the complete value at t=80s. When correcting abnormal fluctuation values, the mean and standard deviation information within the sliding window are used to determine whether they are abnormal peak values. Correction is then performed based on the trend of adjacent points to ensure the synchronicity, continuity, and integrity of multi-source data at the same time point, avoiding noise interference with subsequent heat transfer behavior modeling. The sliding window mechanism can dynamically adapt to local changes in data, and the bidirectional interpolation method can take into account contextual trend information. The combination of the two can significantly improve the temporal fusion quality of multi-source sensor data.

[0083] The time-series aligned data is input into the heat exchange behavior modeling engine. Through the joint principal component analysis algorithm and cluster decomposition algorithm, heat exchange characteristic factors are extracted from the temperature, humidity, current and air volume data, and redundant and low-correlation factors are removed to form a set of feature vectors to describe the heat exchange behavior state of the target air cooler.

[0084] To extract heat transfer characteristic factors that accurately characterize the heat transfer behavior of evaporative air coolers from temperature, humidity, current, and airflow data, the time-series aligned data is first input into the heat transfer behavior modeling engine. A joint principal component analysis (PCA) algorithm and cluster decomposition algorithm are then used for feature dimensionality reduction and structure identification. PCA can transform multiple high-dimensional variables into a small number of unrelated principal components while maximizing information, thus highlighting the main trends in the heat transfer process. For example, when the input includes multiple time-series indicators such as inlet air temperature, outlet air humidity, fan power, current load, and air velocity, PCA can identify the main cause with the greatest impact on heat load changes, such as the "high temperature, high humidity, high load" operating condition. The cluster decomposition algorithm further classifies the distribution of these principal components across different time periods, grouping periods with similar operating states into the same category. The most representative statistical features (such as mean change, fluctuation amplitude, and periodic waveform) within this category are extracted as heat transfer characteristic factors to characterize the heat transfer capacity of the evaporative air cooler under the current operating conditions. For example, by identifying clustered samples where current amplitude increases synchronously across multiple window periods, accompanied by stable wind speed and a sudden increase in humidity, the heat transfer characteristic factor of "decreased evaporative heat transfer efficiency" can be extracted. By combining these two algorithms, not only can redundant information and low-correlation variables be eliminated, but key factors reflecting changes in the heat exchange capacity of the evaporative cooler can also be retained. This ultimately forms a feature vector set containing multiple high-weight heat transfer characteristic factors, providing a strong model foundation for subsequent heat transfer curve construction and anomaly identification. This modeling approach not only improves the accuracy of heat transfer state characterization but also enhances the ability to compare the responses of evaporative coolers under dynamic environments.

[0085] The feature vector set is mapped according to the time series to construct the heat exchange characteristic curve of the target air cooler in the current period. The vertical axis is the refined heat exchange characteristic factor group, and the horizontal axis is the time axis consistent with the sensor data. The heat exchange characteristic curve reflects the response law of the target air cooler's current heat exchange capacity and operating status under dynamic changes in heat load.

[0086] To construct the heat transfer characteristic curve of the target evaporative air cooler for the current time period, it is necessary to first serialize and map the set of feature vectors composed of heat transfer feature factors extracted by principal component analysis and cluster decomposition algorithms according to their corresponding timestamps. During the mapping process, each feature vector represents the heat transfer behavior state of the evaporative air cooler at a certain time point. By sequentially arranging the feature vectors of multiple consecutive time points, a heat transfer characteristic curve can be formed in a two-dimensional coordinate system. The horizontal axis of this coordinate system is a standard time axis consistent with the timestamps of the sensing data, and the vertical axis is the multi-dimensional feature projection value composed of the heat transfer feature factor set corresponding to each time point. This construction method can fully express the response rhythm and state change pattern of the target evaporative air cooler in the current operating period when facing changes in environmental heat load and airflow interference from adjacent fans. For example, if the heat load in the refrigerated area suddenly increases, and the evaporative air cooler can quickly respond by increasing airflow and reducing evaporation temperature, its heat transfer characteristic curve will show a gradual stabilization after a rapid shift in feature factors over a specific time period. However, if the response is delayed due to external disturbances, the curve will show a delayed shift and increased fluctuations. By constructing this curve, not only can the instantaneous fluctuation trend of heat exchange performance be intuitively captured, but it also provides a clear structured input for subsequent comparison with standard heat exchange behavior, enabling a clear analysis of the causal relationship between the operating status of the evaporative cooler and interfering factors in the time dimension. This method emphasizes the temporal and structural aspects of the data during the construction process, ensuring that the resulting characteristic curve possesses high-resolution behavioral characterization capabilities.

[0087] In this embodiment, S203 specifically refers to:

[0088] In the historical operating data, the time period that is closest to the current cold storage environment load status and operating conditions is selected, and the standard heat exchange curve of the target air cooler during the time period is extracted. It is ensured that the standard curve does not contain data interference during the defrosting period, and it is used as the reference for the current heat exchange characteristic curve.

[0089] To extract the time period closest to the current refrigeration environment's load and operating conditions from historical operational data, a system of operating condition characteristic indicators for similarity matching must first be established. This indicator system can include parameters such as the heat load level of the area served by the target air cooler within the refrigerated space, ambient temperature and humidity, air cooler operating frequency, airflow output curve, and current variation trend. Using a multi-dimensional operating condition similarity matching algorithm, such as Mahalanobis distance combined with a time window weighting strategy, multiple time periods closest to the current operating conditions can be selected from historical data as candidates. Subsequently, defrosting status filtering is performed on each candidate time period. By identifying heating current peaks, control signal markers, or defrosting log records, segments containing defrosting interference are eliminated. Finally, a historical time period with the highest operating condition similarity and no defrosting interference is determined, and the heat transfer characteristic curve of the target air cooler is extracted from this time period as a standard heat transfer curve. This standard curve is used to construct a reference for the current heat transfer characteristic curve, ensuring that the residual fitting is compared to the operating state under ideal conditions, thereby improving the accuracy of subsequent error analysis and the reliability of interference identification. This process not only ensures the consistency of the baseline curve in terms of heat load, operational stability, and absence of defrosting interference, but also effectively enhances the ability to distinguish between abnormal heat exchange behavior and actual frosting phenomena.

[0090] The heat exchange characteristic curve of the target air cooler in the current period is compared with the standard heat exchange curve on the time axis. The residual fitting calculation is performed by using the sliding residual window method. The numerical deviation and directional change of the characteristic factor are calculated in each window to obtain the residual time series and remove abnormal peak disturbance values.

[0091] To effectively compare the heat transfer characteristic curve of the target evaporative cooler with the standard heat transfer curve in the current time period, it is first necessary to ensure that the two curves are perfectly aligned in the time dimension, meaning that each time point corresponds to a unique feature factor vector. After alignment, a sliding residual window of fixed width is used to progressively traverse the time axis, with each window covering several consecutive time points. For the corresponding feature factor group at each time point, the difference between the window and the corresponding time point in the standard heat transfer curve is calculated, including numerical deviation and trend differences in the direction of change. Based on the mean and variance of the residuals of all feature factors within the window, a continuous residual time series is formed. To eliminate misjudgments caused by occasional fluctuations, median filtering and abrupt change detection algorithms, such as the IQR outlier identification strategy, can be combined to clean up prominent peak values ​​in the residual series, thereby retaining continuous error segments that truly reflect the abnormal trend of heat transfer. The entire residual fitting process not only captures local trends and overall deviations, but also uses a sliding window to control the separation of short-term fluctuations and long-term biases, enhancing the sensitivity and discriminative power for abnormal operating conditions of the evaporative cooler. This provides a high-resolution data foundation for subsequent error trend identification and interference distortion judgment. This method is particularly suitable for heat exchange processes in refrigerated environments with frequent nonlinear fluctuations, effectively filtering out short-term irregular fluctuations and highlighting persistent deviation characteristics.

[0092] Dynamic clustering and segmentation analysis is performed on the residual time series to extract residual interval segments whose offset exceeds the warning interval threshold. Combining the unidirectionality and duration characteristics of the offset change trend in these interval segments, it is determined that they represent the local heat exchange parameter distortion caused by the interference of the defrosting airflow of the adjacent air cooler. This distortion is then passed on to the subsequent defrosting judgment logic.

[0093] To identify key segments from residual time series that characterize abnormal heat exchange behavior of evaporative coolers, a dynamic clustering segmentation analysis method is first introduced to divide the entire residual sequence into multiple segments with similar characteristic change patterns. Dynamic clustering segmentation analysis is typically implemented using a sliding distance metric algorithm and a trend direction similarity index, adaptively dividing the residual curve into several intervals with unified trends. The data within each segment exhibit high consistency in numerical offset direction and rate of change, helping to eliminate non-persistent disturbances caused by local noise. Simultaneously, combined with a set warning interval threshold, the average offset amplitude of each residual segment is compared; if it exceeds the threshold, it is considered a potential interference area. This threshold is usually set based on the distribution range of historical normal state data to ensure statistical significance of anomaly identification. Furthermore, for segments meeting the threshold requirements, it is necessary to verify whether their offset trend is unidirectional and continuous, and whether the duration reaches a set time length threshold, ultimately filtering out distorted intervals that meet all characteristic requirements. These intervals are identified as localized heat transfer parameter distortions caused by the airflow interference from neighboring evaporative air coolers during defrosting. These distortions can be used as key variables input into the subsequent defrosting logic, effectively controlling the risk of false triggering. This method improves the accuracy of anomaly identification and the rigor of the judgment logic by performing behavioral trend modeling and statistical significance testing on the residual data.

[0094] S3. The local heat exchange parameter distortion is fused with the real-time heat exchange data of the air cooler. The fusion result is corrected by interference weight. The corrected fused data is input into the defrost reliability assessment model to generate defrost reliability assessment parameters. These parameters are used to determine whether the air cooler really needs to perform defrost operation based on the local heat exchange parameter distortion caused by the airflow interference from the defrost air of the neighboring air cooler.

[0095] In this embodiment, S3 specifically includes the following steps:

[0096] S301. The local heat exchange parameter distortion caused by the interference of the defrosting airflow of the adjacent air cooler is fused with the real-time heat exchange data collected by the air cooler in the current period according to the time dimension. In the fusion process, the feature vector weighting operation is used to superimpose the features in the same dimension to form fused heat exchange feature data that can characterize the heat exchange state of the air cooler.

[0097] To accurately characterize the current heat exchange behavior of the evaporative air cooler, it is necessary to fuse the localized heat exchange parameter distortion caused by the interference of defrosting airflow from neighboring evaporative air coolers with the real-time heat exchange data collected by the evaporative air cooler in the current time period. The fusion process first aligns the two types of data along a unified time dimension, ensuring synchronization of each feature data point by constructing time series with the same time granularity. Based on this, heat exchange features of the same dimension are numerically superimposed. A feature vector weighting algorithm is used to combine the real-time heat exchange feature values ​​and the distortion feature values ​​according to preset fusion weights, forming fused heat exchange feature data with temporal continuity. For example, for the feature dimension of evaporative air cooler outlet airflow, if the real-time monitoring value is 2.8 m³ / min, the distortion correction value is -0.3 m³ / min, and the fusion weights are set to 0.7 and 0.3, the fused value is a weighted sum of 2.59 m³ / min. This fusion method effectively preserves the original characteristics of the actual operating conditions of the evaporative cooler, while introducing parameter drift information caused by airflow interference, so that the fused data can comprehensively reflect the heat exchange state of the evaporative cooler under disturbance.

[0098] Distortion of localized heat transfer parameters caused by the airflow interference from nearby evaporative air coolers refers to phenomena such as decreased heat transfer efficiency, abnormal airflow fluctuations, or deviations in temperature and humidity parameters under specific interference. These data typically originate from the anomaly feature identification results during residual fitting analysis. Real-time heat transfer data collected by the evaporative air cooler at the current time includes multi-dimensional sensor information such as temperature, humidity, current, and airflow, with sampling frequency and timestamps uniformly set at the minute level. Data fusion processing refers to performing feature-level fusion calculations on real-time heat transfer data and distorted performance data at the same time point, using the timestamp as a benchmark. Feature vector weighting is an algorithm that linearly combines feature values ​​of the same dimension using weighting coefficients. Its purpose is to enhance the expressive power of interference effects in the fused data and maintain the smoothness of overall numerical changes. Fusion heat transfer feature data refers to a time-series data set containing multiple heat transfer feature dimensions, used as input for subsequent interference correction and reliability assessment models, ensuring the consistency and integrity of data during model training and inference.

[0099] S302. Calculate the interference weight based on the spatial relationship between the air cooler and the disturbance source, the airflow superposition intensity and the interference coverage ratio. Apply the interference weight to the feature dimensions selected by the interference index weight threshold in the fused heat exchange feature data. Perform weighted correction on each feature dimension and output the corrected fused data after interference compensation.

[0100] S303. Input the corrected fusion data into the defrost reliability assessment model, and use the reliability classification algorithm and heat exchange performance deviation identification algorithm trained in the assessment model to generate defrost reliability assessment parameters. The defrost reliability assessment parameters include the degree of heat exchange performance degradation, interference offset factor and duration index. Based on whether the degree of heat exchange performance degradation continuously exceeds the threshold of the normal operation range, whether the interference offset factor exceeds the threshold of the interference source, and whether the duration index exceeds the preset time threshold, determine that the air cooler that meets at least two of the index conditions is the air cooler that really needs to perform defrost operation.

[0101] To determine whether an evaporative air cooler truly requires defrosting, the corrected and fused data obtained from the preprocessing stage needs to be input into a reliable defrosting assessment model. This model, built during the training phase, includes a reliable classification algorithm and a heat exchange performance deviation identification algorithm, capable of comprehensively evaluating the differences between the current heat exchange state and historical heat exchange performance of the evaporative air cooler. Specifically, the feature vectors representing dimensions such as heat exchange efficiency, temperature response, current change, and airflow stability from the corrected and fused data are first input into the model. Internally, the model uses algorithms such as cluster boundary identification, probability distribution fitting, and error deviation regression to extract three types of evaluation parameters: the degree of heat exchange performance degradation, the disturbance offset factor, and the duration index. For example, the degree of heat exchange performance degradation can be scored by comparing the deviation of the heat exchange curve under standard conditions with the current state curve in key features; the disturbance offset factor is calculated based on the weighted fitting result of the rate of change of the feature curve within the disturbance influence area; and the duration index is calculated by statistically determining the length of the time window during which the abnormal heat exchange state continues beyond the reference curve, ultimately forming a complete set of reliable defrosting assessment parameters.

[0102] When determining whether to perform a defrost operation, a joint judgment is required based on the three parameters mentioned above and the set threshold standards. The normal operating range threshold limits the maximum permissible deviation range of the heat exchange performance degradation index, and is typically determined based on statistical analysis of the characteristic fluctuation range collected under multi-cycle stable operation conditions of the evaporative cooler. The interference source impact threshold defines the maximum acceptable deviation of the disturbed characteristic factor, set by comparing the residual range of data fitting between defrost and non-defrost periods. The preset time threshold determines whether the duration of the abnormal heat exchange state exceeds the normal window of natural fluctuations in heat exchange behavior, and is generally determined based on experience with refrigeration conditions and unit response cycles. During the judgment process, the system calculates whether each of these three indicators meets the judgment conditions. When at least two of the indicators exceed the corresponding threshold standards, it is determined that the current evaporative cooler has experienced a substantial decline in heat exchange performance due to interference, and a defrost operation is required. This joint judgment strategy effectively avoids misjudgments caused by local interference or short-term fluctuations, thereby ensuring the accuracy of the defrost operation and the high efficiency of the refrigeration system.

[0103] In this embodiment, S302 specifically refers to:

[0104] The spatial positional relationship parameters between the air cooler and the disturbance source point in the current time period are obtained. The angle between the air cooler's air outlet direction and the disturbance source propagation direction is calculated. The airflow superposition intensity parameters are determined by combining the spatial intersection ratio of the airflow superposition area and the wind speed information of the disturbance source point propagation path. The volume ratio parameters of the interference area covered by the heat exchange surface of the air cooler are collected to form the input dataset used to construct the interference weight.

[0105] To accurately assess the degree of interference from the defrosting airflow of neighboring evaporative air coolers, it is necessary to obtain and quantify the spatial relationship parameters between the evaporative air cooler and the disturbance source. First, the coordinates of the evaporative air cooler and the disturbance source within the refrigerated space are obtained using a 3D positioning sensor. Based on a geometric model, the angle between the evaporative air cooler's outlet direction and the disturbance source's propagation direction is calculated. This angle reflects the likelihood of airflow overlap; a smaller angle indicates a higher risk of overlap. Next, the spatial extent of the airflow overlap area is identified based on airflow simulation results or a wind speed sensor array, and the intersection volume between this area and the evaporative air cooler's operating airflow path is calculated, thus determining the spatial intersection ratio of the airflow overlap. Simultaneously, combined with wind speed data along the disturbance source's propagation path, an airflow overlap intensity parameter is constructed to reflect the degree of interference energy impact. Furthermore, the spatial model of the evaporative air cooler's heat exchange surface needs to be analyzed to identify the area covered by the interfering airflow and calculate its volume percentage within the total heat exchange surface area, forming the interference area volume percentage parameter. By unifying the modeling of three physical quantities—spatial angle, superposition intensity, and interference coverage ratio—and constructing an input dataset, a foundation can be provided for subsequent interference weight calculations, thereby enabling a quantitative assessment of the interference intensity on the evaporative cooler. For example, when the angle between the evaporative cooler and the disturbance source is 15 degrees, the airflow overlap volume is 40%, the maximum wind speed at the superposition point is 3.2 m / s, and 25% of the evaporative cooler's heat exchange surface is within the overlap range, the input dataset constructed using these three parameters can fully reflect the spatial impact characteristics of airflow interference on the evaporative cooler's operating state, providing an accurate basis for subsequent data fusion correction.

[0106] A weighting function is constructed based on the spatial location angle, airflow superposition intensity parameter and interference coverage ratio parameter to generate interference weight value that characterizes the degree of interference. The interference weight value is applied to each feature dimension in the fused heat transfer feature data, and the feature dimensions whose interference sensitivity coefficient exceeds the interference index weight threshold are extracted to form a set of feature dimensions to be corrected.

[0107] To achieve accurate correction of the fusion heat transfer characteristic data of the evaporative air cooler, a weighting function needs to be constructed based on the spatial position angle, airflow superposition intensity parameter, and interference coverage ratio parameter. This function then generates interference weight values ​​to quantify the degree of interference. The spatial position angle measures the geometric alignment between the air cooler's outlet direction and the propagation direction of the disturbance source; a smaller angle indicates a more direct potential interference path. The airflow superposition intensity parameter reflects the energy overlap between the disturbance source airflow and the airflow during normal operation of the evaporative air cooler in space, which can be achieved through a wind speed distribution superposition model. The interference coverage ratio parameter represents the proportion of the heat transfer surface of the evaporative air cooler affected by the disturbed airflow, typically calculated through spatial distribution mapping and volumetric modeling. These three parameters are input into the weighting function, and a linear or nonlinear weighting model is used to generate interference weight values. These weight values ​​are used to measure the likelihood of disturbance to each feature dimension in the fusion heat transfer characteristic data. Subsequently, a disturbance sensitivity analysis is performed on each dimension within the fused heat transfer characteristic data. The response to airflow disturbances under historical operating conditions is calculated and compared with a pre-defined disturbance index weight threshold. This threshold, which can be trained using historical data or set by expert rules, is used to select the feature dimensions most sensitive to disturbance changes. For example, if the temperature gradient change rate dimension shows a significant shift exceeding the disturbance index weight threshold under disturbance conditions, it is identified as a target dimension requiring correction. This ultimately forms the set of feature dimensions to be corrected for subsequent weighted processing. This method avoids indiscriminate correction of all features, improving the targeting and efficiency of disturbance correction.

[0108] For each feature dimension in the set of features to be corrected, a weighted correction operation is performed. The interference weight value is weighted and superimposed with the original value of the corresponding feature dimension to generate corrected feature data to replace the original value. Finally, the corrected fusion data after interference compensation is output as the input data source for the subsequent defrosting reliability assessment model.

[0109] When processing each feature dimension in the set of features to be corrected, a weighted correction operation is required to perform targeted numerical adjustments to features in the evaporative air cooler heat exchange data affected by the defrosting airflow from neighboring evaporative air coolers. The weighted correction is achieved by weighting and superimposing the original value of each feature dimension with its corresponding interference weight value to calculate a new corrected value. This new corrected value is then used to replace the original data, ensuring that the fused data more accurately reflects the actual heat exchange state of the evaporative air cooler under interference conditions. For example, when the rate of temperature change is identified as a sensitive feature dimension, and its original curve exhibits abnormal fluctuations during the interference period, it is weighted with the interference weight value to suppress the offset trend caused by the interference, resulting in a more stable corrected rate of temperature change. The interference weight value is derived from the weighting function constructed in the preceding analysis, and its magnitude is determined by comprehensively evaluating the spatial angle between the evaporative air cooler and the disturbance source, the intensity of airflow overlap, and the proportion of heat exchange surface covered by interference. During the correction process, to avoid overcompensation or data distortion, a proportional limitation or error suppression mechanism can be used to control the weighting ratio within a certain range. All corrected feature data are ultimately integrated into a corrected fusion dataset, organized in a time-series format, and input into the defrost reliability assessment model. This provides a disturbance-corrected decision basis for determining whether the evaporative cooler truly needs defrosting. This processing strategy significantly improves the model's ability to distinguish between actual heat exchange performance degradation and misjudgments of disturbances, helping to reduce unnecessary defrosting and improve the operating efficiency of the refrigeration system.

[0110] S4. Based on the defrosting reliability assessment parameters, combined with the historical operating characteristics of the air cooler and the load fluctuation of the cold storage area, a judgment label is generated to classify the air cooler into three states: needing immediate defrosting, needing delayed observation, and not needing defrosting.

[0111] In this embodiment, S4 specifically refers to:

[0112] Based on the joint state of the heat exchange performance degradation degree, interference offset factor and duration index in the defrosting reliability assessment parameters, the numerical characteristics of various heat exchange parameters of the evaporative cooler in the historical operating cycle are extracted. A historical operating feature vector including the heat exchange efficiency reduction, current load response amplitude and humidity fluctuation stability is constructed to describe the heat exchange behavior of the evaporative cooler under different load conditions.

[0113] In the process of intelligently judging the defrosting status of evaporative air coolers, it is first necessary to construct a feature representation that reflects the historical trend of the heat exchange behavior of the evaporative air cooler. To achieve this goal, the historical operating data of the target evaporative air cooler in different time periods can be traced back using the reliable defrosting assessment parameters as a reference. The numerical evolution of key heat exchange parameters in sensor data such as temperature, humidity, current, and airflow can be extracted through windowed analysis. The decrease in heat exchange efficiency can be quantified by the gradient of the difference between the outlet and inlet air temperatures over a continuous time period. The current load response amplitude can be calculated by the synchronization index of the average current and load fluctuation. The stability of humidity fluctuation is evaluated based on the standard deviation of the inlet and outlet humidity difference at different load stages. By vectorizing and combining the above parameters such as heat exchange efficiency, current fluctuation, and humidity stability, and mapping them uniformly to the time dimension, a historical operating feature vector for characterizing the thermodynamic performance stability of the evaporative air cooler can be formed. This feature vector can be compared and matched with the real-time status in subsequent models, providing basic data support for the identification of abnormal heat exchange trends in evaporative air coolers.

[0114] The degree of heat exchange performance degradation measures the decrease in the heat exchange capacity of an evaporative cooler under a specific load compared to historical reference values. It is calculated by comparing the current outlet air temperature difference with the average heat exchange curve under historical normal conditions. The interference offset factor measures the deviation between the current heat exchange characteristic curve of the evaporative cooler and the curve after superimposing interference characteristics; it can be characterized using Euclidean distance or Mahalanobis distance within the feature dimension. The duration index is used to capture whether the deteriorated or offset state is stable and can be measured by the length of time the deviation remains greater than a threshold within a sliding time window. The current load response amplitude characterizes the current adjustment sensitivity of the evaporative cooler under changes in heat load and can be calculated using the correlation coefficient between the current curve and the rate of temperature change. The stability of humidity fluctuations can be described by the root mean square error of humidity change per unit time in the time series. These parameters, forming a historical operating feature vector, are standardized and input into the subsequent classification model, providing statistical and physical criteria for determining whether the evaporative cooler currently meets the requirements for defrosting.

[0115] Collect load fluctuation data of the current cold storage area, extract feature factors including spatial temperature and humidity distribution, cargo circulation frequency and transient cold load change rate, input the historical operation feature vector and the current load fluctuation factor into the defrost status judgment engine, and generate the current defrost judgment label of the air cooler through the trained judgment model.

[0116] To intelligently determine whether an evaporative air cooler needs to defrost, the first step is to collect load fluctuation data within the refrigerated area. This data can be simultaneously acquired through a multi-point temperature and humidity sensor array, access control sensors, and flow trajectory recording devices deployed within the refrigerated space. The spatial temperature and humidity distribution reflects the environmental heat distribution balance; the frequency of goods flow can be quantified through door opening / closing records and the frequency of goods handling tag readings; and the transient cooling load change rate can be calculated from the temperature and humidity change gradient per unit time. These multi-source data are fused along a time axis to form a load fluctuation factor vector, which is then jointly input with the historical operating feature vector of the evaporative air cooler. In the data processing flow, a defrosting status determination engine with feature dimensionality reduction and pattern recognition capabilities is introduced. Through feature cross-combination and temporal dependency modeling, a mapping relationship between current load behavior and the heat exchange trend of the evaporative air cooler is established, thereby classifying the operating status of the evaporative air cooler.

[0117] The defrost status determination engine is a data inference system built on machine learning methods, with its core component being a trained determination model. This model is trained using supervised learning on a large number of historical operating samples. The model structure can be an ensemble decision tree, support vector machine, or temporal neural network, all with strong generalization capabilities. During training, historical feature vectors of the evaporative air cooler under various load fluctuation scenarios are used as input, and defrost decision results labeled with expert experience are used as output labels to fit the model parameters. In practical applications, after inputting the joint feature data of the current evaporative air cooler, the model outputs defrost determination labels based on patterns identified from historical data. These labels include three states: immediate defrosting required, delayed observation required, and no defrosting needed. These labels guide the implementation of the system's defrost control strategy, improving the accuracy and energy efficiency of defrost operations.

[0118] Based on the output results of the judgment label, the air cooler is divided into three states: immediate defrosting required, delayed observation required, and no defrosting required. The state of immediate defrosting is characterized by the heat exchange performance degradation level continuously exceeding the threshold of the normal operating range, the interference offset factor exceeding the threshold of the interference source, and the duration index exceeding the preset time threshold. The state of delayed observation is characterized by the heat exchange performance degradation level being within the upper limit of the threshold of the normal operating range, and either the interference offset factor or the duration index being satisfied. The state of no defrosting is characterized by the heat exchange performance degradation level, the interference offset factor, and the duration index not exceeding their respective limit thresholds.

[0119] When intelligently classifying the defrosting status of evaporative air coolers, the system can categorize them into three states: requiring immediate defrosting, requiring delayed observation, and not requiring defrosting, based on the judgment labels output by the defrosting status judgment model. The judgment model comprehensively assesses the system based on three dimensions of evaluation parameters: the degree of heat exchange performance degradation, the interference offset factor, and the duration of the interference. If the degree of heat exchange performance degradation exceeds the upper limit of the normal operating range for multiple consecutive time periods, and the interference offset factor exceeds a preset interference threshold, and the duration of this interference exceeds a predefined time threshold, the evaporative air cooler is classified as requiring immediate defrosting. This indicates that the heat exchange capacity is severely limited, and immediate defrosting intervention is necessary. If only the interference offset factor or the duration meets the specified conditions, and the degree of heat exchange performance degradation is still near the upper limit but not exceeding it, the system will classify it as requiring delayed observation, indicating a potential latent defrosting need but no immediate action required. If none of the three indicators exceed their respective set threshold ranges, the system classifies it as not requiring defrosting to avoid unnecessary energy waste.

[0120] The degree of heat exchange performance degradation is determined by comparing the residual distribution between the current heat exchange characteristic curve of the evaporative cooler and the standard curve formed by historical normal heat exchange behavior. A higher index value indicates a more significant decrease in actual heat exchange efficiency. The disturbance offset factor measures the degree of influence of airflow disturbance on the heat exchange parameters of the evaporative cooler, typically calculated by quantifying the impact of superimposed airflow on key characteristic dimensions. The duration index is an assessment value recording the duration of the disturbance effect, statistically analyzed over the entire evaporative cooler operating cycle using a sliding time window. These three indices correspond to three risk sources: performance degradation, disturbance impact, and impact persistence. Judgment is based on a logical combination of whether each exceeds its respective threshold. This three-state classification helps the system distinguish defrosting needs of varying urgency, thereby improving the accuracy and response efficiency of defrosting decisions while ensuring system energy efficiency.

[0121] S5. Based on the judgment tag corresponding to the evaporative air cooler, a differentiated defrosting control process is executed, and the data generated by each evaporative air cooler in the current cycle is used to update the spatiotemporal perception map of the heat exchange status, so as to realize the optimization of defrosting control under dynamic regulation.

[0122] In this embodiment, S5 specifically refers to:

[0123] Based on the judgment tag corresponding to the evaporative air cooler, a differentiated defrosting control process is executed. Evaporative air coolers with the judgment tag indicating that they need to be defrosted immediately are included in the instant defrosting control sequence and a defrosting execution signal is sent. Evaporative air coolers with the judgment tag indicating that they need to be observed later are marked as key monitoring objects and the heat exchange data acquisition frequency is increased. Evaporative air coolers with the judgment tag indicating that they do not need to be defrosted maintain their original operating parameters and retain the observation task records.

[0124] To achieve precise control over the defrosting operation of evaporative air coolers, a differentiated defrosting control process is required based on the current status label of the air cooler. This can be achieved by constructing a defrosting task scheduling controller, binding the obtained status labels with the unique identifier of the air cooler, and classifying them during task scheduling. For air coolers marked as requiring immediate defrosting, a defrosting execution signal is sent to the corresponding air cooler through the control interface, activating the electric heating element or the heat pump in reverse mode, and recording the defrosting trigger time in the system log. For air coolers requiring delayed observation, the current status is recorded and added to the monitoring list. The system automatically increases the sensor data sampling frequency of the air cooler to capture subtle changes in its heat exchange performance, thereby determining whether further escalation to immediate defrosting is necessary. For air coolers not requiring defrosting, the current operating parameters remain unchanged, and operating data is recorded during the defrosting evaluation cycle for subsequent verification. This differentiated control method ensures that the response strategy for air coolers in different states is targeted, effectively improving the real-time performance and accuracy of defrosting control.

[0125] The judgment label is a classification result, generated based on the defrosting reliability assessment parameters output by the evaluation model. Specifically, it includes three states: immediate defrosting required, delayed observation required, and no defrosting needed. The defrosting control sequence refers to the list of air coolers that prioritize defrosting operations during task scheduling; their inclusion and exclusion are controlled by dynamic changes in the label status. Increasing the heat exchange data acquisition frequency is mainly achieved by adjusting the sampling intervals of sensors such as temperature, current, humidity, and airflow, typically set to more than twice the standard sampling frequency to improve detection sensitivity. The observation task record refers to the data buffer used in subsequent evaluations to trace the evolution of the air cooler's state. These records support model optimization and trend prediction analysis, providing data support for defrosting decisions in the next cycle.

[0126] The data generated by each air cooler in the current cycle, including heat exchange characteristic data, interference identification indicators and operating status parameters, are input into the heat exchange status spatiotemporal perception map update engine according to the time index and spatial location mapping method. The status labels of air cooler nodes, the edge weights of adjacent airflow interference relationships and the spatial heat load distribution results in the map are updated synchronously.

[0127] To achieve dynamic monitoring and intelligent control of the operating status of each air cooler in the refrigeration system, data generated by each air cooler in the current cycle, including heat exchange characteristic data, interference identification indicators, and operating status parameters, can be uniformly input into the heat exchange status spatiotemporal perception map update engine according to time indexing and spatial location mapping rules. In the implementation process, firstly, based on the physical layout location, air outlet direction, and numbering information of the air cooler, the data is located to the corresponding node in the map structure; then, the time dimension of the node is aligned based on the data timestamp, and data reflecting the heat exchange performance, disturbance status, and operating mode of the air cooler in the current cycle is written into the node attributes. For the identified airflow interference relationships between adjacent air coolers, the weight values ​​on their connection edges are updated to reflect the intensity and direction of the latest airflow interference. Simultaneously, by combining the heat exchange data of each air cooler node in the entire spatial area, the heat load distribution state in the spatial grid is recalculated, realizing the continuous evolution and dynamic reflection of the heat exchange status spatiotemporal perception map, thereby providing real-time data support for subsequent defrosting control strategies and cold chain energy efficiency optimization.

[0128] Heat exchange characteristic data is a heat exchange capacity assessment index constructed by integrating real-time temperature, humidity, current, and airflow data collected by the evaporative air cooler, reflecting its heat exchange level under current operating conditions. Interference identification indicators characterize whether the evaporative air cooler is affected by the defrosting airflow from neighboring evaporative air coolers, typically including interference offset factors and spatial airflow superposition. Operating status parameters include whether the evaporative air cooler is currently in cooling, standby, defrosting, or observation mode. The time index refers to the standard timestamp used to locate the time period to which the data belongs; the spatial location mapping method is based on the installation coordinates and spatial layout relationship of the evaporative air cooler within the refrigerated area for node positioning. The spatiotemporal perception map of heat exchange status is a dynamic state expression model of the evaporative air cooler constructed based on a graph structure. Its nodes represent evaporative air cooler entities, edges represent airflow interference relationships, and node attributes include heat exchange capacity, interference status, and operating status. Edge weights reflect the intensity of airflow interference between evaporative air coolers. Through continuous updating of the map, the operational coordination relationship and heat load distribution status of evaporative air coolers within the refrigerated area can be monitored in real time, thereby achieving more refined defrosting control and energy efficiency management.

[0129] Based on the updated spatiotemporal perception map of heat exchange status, the heat exchange behavior correlation network of each air cooler in the current operating cycle is reconstructed. The heat load distribution status, defrosting interference path and control strategy adaptability among the air coolers are comprehensively analyzed to generate the defrosting control priority ranking result for the next round, thereby realizing the optimization of defrosting control under dynamic regulation.

[0130] To achieve dynamic optimization of defrost control, a network relating the heat exchange behaviors of each air cooler in the refrigeration system during the current operating cycle can be reconstructed based on an updated spatiotemporal perception map of heat exchange status. This process involves modeling and fusing analysis of the heat load transfer relationships, interference propagation paths, and historical defrost execution effects between air cooler nodes. First, based on the heat exchange feature vectors and spatial location information of each node in the map, a graph mining algorithm is used to extract the heat exchange coupling relationships between air coolers, establishing a network relating the heat exchange behaviors. Then, paths in the network that cause significant heat exchange disturbances to other air coolers after defrost execution are identified, constructing a defrost interference path map. Simultaneously, based on the execution feedback of each air cooler under previous control strategies, including post-defrost heat recovery time, energy consumption changes, and refrigeration environment response results, the adaptability of the control strategy is evaluated. By comprehensively considering the heat load distribution, interference path intensity, and control strategy adaptability parameters, the priority scoring and ranking of defrost control execution for all air cooler nodes are determined, forming a defrost execution plan for the next round, thereby achieving defrost control optimization under dynamic operating conditions.

[0131] The heat exchange status spatiotemporal perception map is a graph structure model that records the heat exchange characteristics, interference relationships, and operating history of each air cooler node using both time and space as dual indexes. The heat exchange behavior association network is a directed graph constructed based on the map through the similarity of heat exchange characteristics and spatial proximity between nodes, used to express the heat exchange interaction patterns between air coolers. The heat load distribution status is expressed by spatially mapping the heat exchange capacity of multiple air coolers within the refrigeration area, showing the heat flow and concentration. The defrost interference path is the path representation of the direction and intensity of crosstalk caused by defrost airflow in the network, used to determine whether the defrosting of one air cooler may interfere with other air coolers. The control strategy adaptability is a multi-dimensional evaluation index generated based on the historical control records of the air coolers, used to measure the response effect of the existing strategy under specific operating conditions. The defrost control priority ranking result is the basis for deciding whether each air cooler should be prioritized for defrosting in subsequent control cycles, combining operating status, environmental feedback, and energy efficiency performance to guide the system in executing an orderly and efficient defrosting plan.

[0132] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means (e.g., infrared, wireless, microwave, etc.). A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0133] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0134] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0135] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0136] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0137] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0138] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A distributed dynamic defrosting control method for evaporative air coolers, characterized in that, Specifically, the following steps are included: S1. Obtain the spatial location, airflow direction and other information of the evaporative cooler in the refrigeration environment, as well as the defrosting and exhaust status of adjacent evaporative coolers. Construct a spatiotemporal perception map of the heat exchange status based on the spatial coordinate relationship to determine whether the evaporative cooler is disturbed by the defrosting airflow of the adjacent evaporative cooler. S1 specifically includes the following steps: S101. By deploying a spatial identification unit in the cold storage environment, the spatial position of each air cooler is obtained. Combined with the wind direction and wind speed detection components configured at the air cooler outlet, the air direction and flow information of the air cooler under different operating conditions is obtained. At the same time, based on the air cooler operation control logic and heating start signal, it is identified whether the adjacent air cooler is in the defrosting and exhaust state. S102. Based on the spatial location and airflow direction information of the evaporative cooler, establish a spatial coordinate relationship, take the evaporative cooler in the defrosting and exhaust state as the source of heat and humidity disturbance, call the three-dimensional disturbance diffusion prediction model to calculate its influence radius and directional propagation path, and then construct a spatiotemporal perception map of heat transfer state that includes the airflow interaction relationship between evaporative coolers. S103. By analyzing the spatial angle between the air cooler and the disturbance source, the mapping relationship between the airflow overlap area and the disturbance intensity, the disturbed area is identified in the spatiotemporal perception map of the heat exchange state, and the air cooler is determined to be disturbed by the defrosting airflow of the adjacent air cooler according to the preset disturbance judgment rules. S2. Collect temperature, humidity, current and air volume data for the air cooler that has been identified as being disturbed by the defrosting airflow of the adjacent air cooler, construct the heat exchange characteristic curve of the target air cooler in the current period, and perform residual fitting calculation with the standard heat exchange curve under historical non-defrosting conditions to determine the local heat exchange parameter distortion caused by the air cooler being disturbed by the defrosting airflow of the adjacent air cooler. S3. The local heat exchange parameter distortion is fused with the real-time heat exchange data of the air cooler. The fusion result is corrected by interference weight. The corrected fused data is input into the defrost reliability assessment model to generate defrost reliability assessment parameters. These parameters are used to determine whether the air cooler really needs to perform defrost operation based on the local heat exchange parameter distortion caused by the airflow interference from the defrost air of the neighboring air cooler. S3 specifically includes the following steps: S301. The local heat exchange parameter distortion caused by the interference of the defrosting airflow of the adjacent air cooler is fused with the real-time heat exchange data collected by the air cooler in the current period according to the time dimension. In the fusion process, the feature vector weighting operation is used to superimpose the features in the same dimension to form fused heat exchange feature data that can characterize the heat exchange state of the air cooler. S302. Calculate the interference weight based on the spatial relationship between the air cooler and the disturbance source, the airflow superposition intensity and the interference coverage ratio. Apply the interference weight to the feature dimensions selected by the interference index weight threshold in the fused heat exchange feature data. Perform weighted correction on each feature dimension and output the corrected fused data after interference compensation. S303. Input the corrected fusion data into the defrost reliability assessment model, and use the reliability classification algorithm and heat exchange performance deviation identification algorithm trained in the assessment model to generate defrost reliability assessment parameters. The defrost reliability assessment parameters include the degree of heat exchange performance degradation, interference offset factor and duration index. Based on whether the degree of heat exchange performance degradation continuously exceeds the threshold of the normal operation range, whether the interference offset factor exceeds the threshold of the interference source, and whether the duration index exceeds the preset time threshold, determine that the air cooler that meets at least two of the index conditions is the air cooler that really needs to perform defrost operation. S4. Based on the defrosting reliability assessment parameters, combined with the historical operating characteristics of the air cooler and the load fluctuation of the cold storage area, a judgment label is generated to classify the air cooler into three states: needing immediate defrosting, needing delayed observation, and not needing defrosting. S5. Based on the judgment tag corresponding to the evaporative air cooler, a differentiated defrosting control process is executed, and the data generated by each evaporative air cooler in the current cycle is used to update the spatiotemporal perception map of the heat exchange status, so as to realize the optimization of defrosting control under dynamic regulation.

2. The distributed dynamic defrosting control method for air coolers according to claim 1, characterized in that, S103 specifically refers to: Calculate the spatial angle between the air cooler and the disturbance source point, and use the overlap between the angle range and the main direction of airflow as the first interference index. Screen potential interference directions through spatial paths with an angle smaller than the set interference threshold. Extract the spatial volume of the airflow overlap region, construct a disturbance intensity mapping relationship between the volume ratio and the superposition intensity of airflow velocity, and use it as the second disturbance index. Mark the overlapping region of airflow propagation trajectory and intensity distribution in the heat transfer state spatiotemporal perception map, and identify the disturbed region that meets the disturbance index weight threshold. The preset interference judgment rules are based on the angle threshold judgment, the disturbance intensity coverage ratio judgment, and the airflow interference duration judgment. The angle threshold judgment is based on whether the angle between the air outlet direction of the evaporator and the direction of the disturbance source is lower than the set interference threshold. The disturbance intensity coverage ratio judgment is based on whether the proportion of the disturbed area on the heat exchange surface of the evaporator exceeds the preset limit ratio. The airflow interference duration judgment is based on whether the duration of the disturbance effect exceeds the preset time window. An evaporator that meets at least two of the above three judgment conditions is identified as an evaporator affected by the defrosting airflow of a neighboring evaporator.

3. The distributed dynamic defrosting control method for air coolers according to claim 1, characterized in that, S2 specifically includes the following steps: S201. By deploying temperature sensors, humidity sensors, current sensors, and airflow detection components on the surface of the heat exchanger inside the evaporative cooler and along its air inlet and outlet paths, the system collects in real time the temperature, humidity, current, and airflow data of the evaporative cooler that has been identified as being affected by the defrosting airflow of a neighboring evaporative cooler during its current operating period. The system also performs time alignment and data loss repair on different types of sensor data to ensure the continuity and consistency of the input data. S202. The processed temperature, humidity, current and air volume data are integrated into the heat exchange behavior modeling engine according to the time series. The heat exchange characteristic factors are extracted by the multivariate feature extraction algorithm. The heat exchange characteristic curve of the target air cooler in the current period is constructed in the same time dimension, so that it fully reflects the linkage between heat load change and operating status. S203. Perform residual fitting calculation on the heat exchange characteristic curve of the target air cooler in the current period and the standard heat exchange curve selected in the historical data under non-defrost conditions. By analyzing the offset amplitude, trend and duration of the residual interval, extract the local heat exchange parameter distortion caused by the air cooler being disturbed by the defrost airflow of the adjacent air cooler, and use it as the input for subsequent defrost judgment logic.

4. The distributed dynamic defrosting control method for air coolers according to claim 3, characterized in that, S202 specifically refers to: Temperature, humidity, current and air volume data are aligned sequentially on the same time axis. A sliding window mechanism is used to extract sensor data segments of continuous time periods at equal intervals. Bidirectional interpolation is used to complete and correct missing or abnormally fluctuating data to ensure that multi-source data in the time series maintains synchronization and data integrity at each time point. The time-series aligned data is input into the heat exchange behavior modeling engine. Through the joint principal component analysis algorithm and cluster decomposition algorithm, heat exchange characteristic factors are extracted from the temperature, humidity, current and air volume data, and redundant and low-correlation factors are removed to form a set of feature vectors to describe the heat exchange behavior state of the target air cooler. The feature vector set is mapped according to the time series to construct the heat exchange characteristic curve of the target air cooler in the current period. The vertical axis is the refined heat exchange characteristic factor group, and the horizontal axis is the time axis consistent with the sensor data. The heat exchange characteristic curve reflects the response law of the target air cooler's current heat exchange capacity and operating status under dynamic changes in heat load.

5. The distributed dynamic defrosting control method for air coolers according to claim 3, characterized in that, S203 specifically refers to: In the historical operating data, the time period that is closest to the current cold storage environment load status and operating condition characteristics is selected, and the standard heat exchange curve of the target air cooler under the time period is extracted. It is ensured that the standard heat exchange curve does not contain data interference during the defrosting period, and it is used as the reference for the current heat exchange characteristic curve. The heat exchange characteristic curve of the target air cooler in the current period is compared with the standard heat exchange curve on the time axis. The residual fitting calculation is performed by using the sliding residual window method. The numerical deviation and directional change of the characteristic factor are calculated in each window to obtain the residual time series and remove abnormal peak disturbance values. Dynamic clustering and segmentation analysis is performed on the residual time series to extract residual interval segments whose offset exceeds the warning interval threshold. Combining the unidirectionality and duration characteristics of the offset change trend in these interval segments, it is determined that they represent the local heat exchange parameter distortion caused by the interference of the defrosting airflow of the adjacent air cooler. This distortion is then passed on to the subsequent defrosting judgment logic.

6. The distributed dynamic defrosting control method for air coolers according to claim 1, characterized in that, S302 specifically refers to: The spatial positional relationship parameters between the air cooler and the disturbance source point in the current time period are obtained. The angle between the air cooler's air outlet direction and the disturbance source propagation direction is calculated. The airflow superposition intensity parameters are determined by combining the spatial intersection ratio of the airflow superposition area and the wind speed information of the disturbance source point propagation path. The volume ratio parameters of the interference area covered by the heat exchange surface of the air cooler are collected to form the input dataset used to construct the interference weight. A weighting function is constructed based on the spatial location angle, airflow superposition intensity parameter and interference coverage ratio parameter to generate interference weight value that characterizes the degree of interference. The interference weight value is applied to each feature dimension in the fused heat transfer feature data, and the feature dimensions whose interference sensitivity coefficient exceeds the interference index weight threshold are extracted to form a set of feature dimensions to be corrected. For each feature dimension in the set of features to be corrected, a weighted correction operation is performed. The interference weight value is weighted and superimposed with the original value of the corresponding feature dimension to generate corrected feature data to replace the original value. Finally, the corrected fusion data after interference compensation is output as the input data source for the subsequent defrosting reliability assessment model.

7. The distributed dynamic defrosting control method for air coolers according to claim 1, characterized in that, S4 specifically refers to: Based on the joint state of the heat exchange performance degradation degree, interference offset factor and duration index in the defrosting reliability assessment parameters, the numerical characteristics of various heat exchange parameters of the evaporative cooler in the historical operating cycle are extracted. A historical operating feature vector including the heat exchange efficiency reduction, current load response amplitude and humidity fluctuation stability is constructed to describe the heat exchange behavior of the evaporative cooler under different load conditions. Collect load fluctuation data of the current cold storage area, extract feature factors including spatial temperature and humidity distribution, cargo circulation frequency and transient cold load change rate, input the historical operation feature vector and the current load fluctuation factor into the defrost status judgment engine, and generate the current defrost judgment label of the air cooler through the trained judgment model. Based on the output results of the judgment label, the air cooler is divided into three states: immediate defrosting required, delayed observation required, and no defrosting required. The state of immediate defrosting is characterized by the heat exchange performance degradation level continuously exceeding the threshold of the normal operating range, the interference offset factor exceeding the threshold of the interference source, and the duration index exceeding the preset time threshold. The state of delayed observation is characterized by the heat exchange performance degradation level being within the upper limit of the threshold of the normal operating range, and either the interference offset factor or the duration index being satisfied. The state of no defrosting is characterized by the heat exchange performance degradation level, the interference offset factor, and the duration index not exceeding their respective limit thresholds.

8. The distributed dynamic defrosting control method for air coolers according to claim 1, characterized in that, S5 specifically refers to: Based on the judgment tag corresponding to the evaporative air cooler, a differentiated defrosting control process is executed. Evaporative air coolers with the judgment tag indicating that they need to be defrosted immediately are included in the instant defrosting control sequence and a defrosting execution signal is sent. Evaporative air coolers with the judgment tag indicating that they need to be observed later are marked as key monitoring objects and the heat exchange data acquisition frequency is increased. Evaporative air coolers with the judgment tag indicating that they do not need to be defrosted maintain their original operating parameters and retain the observation task records. The data generated by each air cooler in the current cycle, including heat exchange characteristic data, interference identification indicators and operating status parameters, are input into the heat exchange status spatiotemporal perception map update engine according to the time index and spatial location mapping method. The status labels of air cooler nodes, the edge weights of adjacent airflow interference relationships and the spatial heat load distribution results in the map are updated synchronously. Based on the updated spatiotemporal perception map of heat exchange status, the heat exchange behavior correlation network of each air cooler in the current operating cycle is reconstructed. The heat load distribution status, defrosting interference path and control strategy adaptability among the air coolers are comprehensively analyzed to generate the defrosting control priority ranking result for the next round, thereby realizing the optimization of defrosting control under dynamic regulation.

Images