Refrigerator abnormal cooling detection method and system

Abstract

The application discloses a refrigerator refrigeration anomaly detection method and system. The method first determines that the current refrigerator meets the normal refrigeration condition based on the operation data collected at the refrigerator end, calculates the actual temperature difference between the target compartment sensor temperature and the set temperature, and triggers the cloud verification process when the cumulative duration of the actual temperature difference continuously exceeding the reference temperature difference threshold reaches the first duration. The cloud is based on the historical operation data of the same type of refrigerator group, and the proportion of refrigerators meeting the same abnormal judgment condition is calculated. If the proportion is less than the abnormal proportion threshold, it is determined that the target compartment refrigeration is abnormal, otherwise the reference temperature difference threshold is updated based on the start-up parameters and returned to re-detect. A double judgment mechanism of the reference high temperature threshold is added for the freezer. The application can dynamically adapt to the operation characteristics of different types of refrigerators, effectively improve the refrigeration anomaly judgment accuracy, and reduce the false alarm probability.

Description

Technical Field

[0001] This application relates to the field of refrigerator anomaly detection technology, specifically providing a method and system for detecting refrigerator cooling anomalies. Background Technology

[0002] With the gradual application of IoT technology in the field of smart home appliances, the ability to collect, store and analyze refrigerator operating data in the cloud is constantly improving, providing a technical foundation for optimizing online detection and judgment standards for refrigeration anomalies. As a result, the industry's demand for accurate identification of refrigeration faults and low false alarm detection is also increasing.

[0003] Existing refrigerator refrigeration fault detection technologies mostly involve installing temperature sensors inside each compartment of the refrigerator to collect real-time compartment temperatures. The collected temperature values ​​are then compared with pre-set fixed thresholds. When the detected temperature exceeds the preset threshold and remains above it for a set duration, an alert is sent to the user indicating an abnormal compartment temperature. Some detection solutions upload refrigerator operating data to a cloud server and rely on a pre-set multi-compartment diagnostic model in the cloud to determine and output the refrigeration fault results.

[0004] Because different refrigerator models have different structural designs and temperature sensor installation locations, using a uniform fixed threshold to determine cooling anomalies cannot be adapted to the inherent operating characteristics of different refrigerator models. This results in a high false alarm rate for cooling anomalies, which not only affects the normal user experience but also increases unnecessary after-sales maintenance costs. Summary of the Invention

[0005] This application addresses the problem that existing methods for determining refrigeration anomalies using a uniform fixed threshold cannot adapt to the inherent operating characteristics of different refrigerator models, resulting in a high false alarm rate. One approach is to provide a refrigerator refrigeration anomaly detection method, comprising the following steps: Based on the operating data collected from the refrigerator, it is determined that the refrigerator currently meets the preset normal cooling conditions. The operating data includes at least the sensor temperature of each compartment, the set temperature of each compartment, the door open / close status, and the load switch status. The actual temperature difference between the sensor temperature of the target chamber and the set temperature of the target chamber is calculated based on the operating data. The actual temperature difference is compared with a preset reference temperature difference threshold for the target room, and the cloud verification process is triggered when the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches a preset first duration. The cloud-based verification process includes: Based on the historical operating data of a group of refrigerators of the same model as the current refrigerator under normal cooling conditions stored in the cloud, the percentage of refrigerators in the group of refrigerators of the same model that meet the first anomaly judgment condition is calculated. The first anomaly judgment condition includes the cumulative duration for which the actual temperature difference of the group of refrigerators of the same model continuously exceeds the benchmark temperature difference threshold, reaching the first duration. The refrigerator percentage is compared with a preset abnormal percentage threshold. If the refrigerator percentage is less than the abnormal percentage threshold, the target compartment of the current refrigerator is determined to be refrigerating abnormally. If the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, the reference temperature difference threshold is adaptively updated based on the current refrigerator's startup parameters, and the updated reference temperature difference threshold is used as the new reference temperature difference threshold to return to the actual temperature difference calculation and comparison steps.

[0006] In one feasible implementation, the target compartment includes a cold storage compartment; The normal refrigeration conditions include the refrigerator door being closed, the defrost heater being off, and the compressor being on. The adaptive update of the reference temperature difference threshold based on the current refrigerator's operating parameters specifically includes: The original reference temperature difference threshold for the refrigerator compartment is added to the current refrigerator start-up parameters to obtain the updated new reference temperature difference threshold.

[0007] In one feasible implementation, the target compartment includes a variable temperature compartment; The adaptive update of the reference temperature difference threshold based on the current refrigerator's operating parameters specifically includes: The original reference temperature difference threshold for the variable temperature compartment is added to the current variable temperature start-up parameters of the refrigerator to obtain the updated new reference temperature difference threshold.

[0008] In one feasible implementation, the target compartment includes a freezer compartment; After the step of comparing the actual temperature difference with a preset reference temperature difference threshold for the target room, and before triggering the cloud verification process, the method further includes: During a continuous second time period, monitor whether the sensor temperature of the freezer compartment is continuously greater than a preset high temperature threshold for the freezer compartment; If not, then the freezer compartment is deemed to be cooling normally; If so, the cloud verification process will be triggered. The cloud-based verification process also includes: If the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, the proportion of refrigerators in the same model group that meet the second abnormality judgment condition is further counted. The second abnormality judgment condition includes the cumulative duration during which the freezer compartment sensor temperature of the same model group continuously exceeds the benchmark high temperature threshold and reaches the second duration. If the percentage of refrigerators that meet the second anomaly determination condition is greater than or equal to the anomaly percentage threshold, then the reference high temperature threshold is adaptively updated based on the current refrigerator's startup parameters, and the updated reference high temperature threshold is used as the new reference high temperature threshold to return to the freezer compartment sensor temperature monitoring step.

[0009] In one feasible implementation, the adaptive update of the benchmark high-temperature threshold based on the current refrigerator's operating parameters specifically includes: The original baseline high temperature threshold for the freezer compartment is added to the current freezer start-up parameters of the refrigerator to obtain the updated new baseline high temperature threshold.

[0010] In one feasible implementation, before the step of determining that the current refrigerator meets the preset normal cooling conditions based on the operating data collected at the refrigerator end, the method further includes: Obtain the refrigerator's model information and operating data, and upload the model information and operating data to the cloud; The cloud platform categorizes and stores the received operational data according to the model information for use in the cloud verification process.

[0011] Another aspect of this application provides a refrigerator cooling anomaly detection system for performing a refrigerator cooling anomaly detection method as described in any of the above claims, including refrigerator-side equipment and a cloud server; The refrigerator terminal device is configured to collect the refrigerator's operating data. The operating data includes at least the sensor temperature of each compartment, the set temperature of each compartment, the door open / close status, and the load switch status. Based on the operating data, it is determined whether the refrigerator currently meets the preset normal cooling conditions. The refrigerator terminal device is also configured to calculate the actual temperature difference between the sensor temperature of the target compartment and the set temperature of the target compartment, compare the actual temperature difference with a preset reference temperature difference threshold for the target compartment, and send a verification request to the cloud server when it is determined that the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches a preset first duration. The cloud server is configured to receive the verification request and, based on its own stored historical operating data of a group of refrigerators of the same model as the current refrigerator under normal cooling conditions, calculate the percentage of refrigerators in the group of refrigerators of the same model that meet the first anomaly determination condition. The first anomaly determination condition includes the cumulative duration during which the actual temperature difference of the group of refrigerators of the same model continuously exceeds the benchmark temperature difference threshold, reaching the first duration. The cloud server is also configured to compare the refrigerator percentage with a preset abnormal percentage threshold. If the refrigerator percentage is less than the abnormal percentage threshold, a judgment result is generated to indicate that the target compartment of the current refrigerator is refrigerating abnormally. If the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, the reference temperature difference threshold is adaptively updated based on the current refrigerator's startup parameters, and the updated reference temperature difference threshold is sent back to the refrigerator-side device so that the refrigerator-side device can re-execute the actual temperature difference calculation and comparison steps based on the new reference temperature difference threshold.

[0012] In one feasible implementation, the refrigerator-side device includes an electronic control board and a WiFi module; The electronic control board is communicatively connected to the WiFi module, and the electronic control board is configured to drive the refrigerator load and collect the operating data. The WiFi module is configured to receive the operating data and model information transmitted by the electronic control board, and upload the operating data and model information to the cloud server; The cloud server is also configured to classify and store the received operational data according to the model information.

[0013] In one feasible implementation, the target compartment includes a freezer compartment; The refrigerator terminal device is further configured to: when the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches the first duration, further monitor whether the sensor temperature of the freezer compartment is continuously greater than the preset reference high temperature threshold for the freezer compartment within a continuous second duration; The cloud server is further configured to: when the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, further count the proportion of refrigerators in the same model group that meet the second abnormality judgment condition, the second abnormality judgment condition including the cumulative duration for which the freezer compartment sensor temperature of the same model group continuously exceeds the benchmark high temperature threshold to reach the second duration. If the proportion of refrigerators that meet the second anomaly determination condition is greater than or equal to the anomaly proportion threshold, the cloud server is further configured to adaptively update the benchmark high temperature threshold based on the current refrigerator's startup parameters, and send the updated benchmark high temperature threshold back to the refrigerator terminal device, so that the refrigerator terminal device can re-execute the monitoring step of the freezer compartment sensor temperature based on the new benchmark high temperature threshold.

[0014] In one feasible implementation, the refrigerator end device further includes a temperature sensor, a door switch sensor, a compressor, and a defrost heater, all electrically connected to the electronic control board. The temperature sensor is configured to detect the sensor temperature of each of the compartments; The door switch sensor is configured to detect the door switch status; The electronic control board regulates the refrigeration process of the refrigerator by controlling the on / off state of the compressor and the defrost heater.

[0015] As described above, the refrigerator cooling anomaly detection method and system provided in this application, through a two-level detection architecture combining local initial judgment and cloud-based group verification, can dynamically adapt to the actual operating characteristics of different refrigerator models, effectively improving the accuracy of cooling anomaly detection and reducing the probability of false alarms. The solution eliminates temperature interference from non-faulty operating conditions such as door opening, defrosting, and compressor shutdown through pre-screening under normal cooling conditions, and filters instantaneous temperature fluctuations by combining continuous-duration temperature difference judgment rules, reducing the causes of false alarms from the source of detection. For the freezer compartment, a dual judgment dimension of temperature difference and absolute temperature is adopted, which can cover more cooling anomaly scenarios and ensure the storage safety of frozen food. The threshold dynamic update mechanism based on group operating data of refrigerators of the same model can continuously optimize the judgment criteria as cloud data accumulates, solving the problem that fixed thresholds cannot adapt to differences in the structure and sensor installation of different refrigerator models. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the implementation of the invention and, together with the description, serve to explain the principles of the embodiments of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0017] Figure 1 This is a schematic flowchart illustrating a refrigerator cooling anomaly detection method according to an embodiment of this application; Figure 2 This is a flowchart illustrating a method for detecting refrigerator cooling anomalies when the target compartment is a freezer compartment, as shown in an embodiment of this application. Detailed Implementation

[0018] 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, these embodiments are provided so that the embodiments of the invention will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of how embodiments of the invention are carried out.

[0019] Current refrigerator refrigeration fault detection technologies typically rely on temperature sensors installed inside the refrigerator compartment to collect real-time temperatures. These temperatures are then compared to pre-set thresholds. If the detected temperature exceeds the threshold, an abnormal compartment temperature is alerted to the user. However, in real-world applications, different refrigerator models have varying structural designs, resulting in different installation locations for the compartment temperature sensors. Furthermore, some refrigerators may have inherent design flaws or insufficient prior verification, leading to significant inherent temperature deviations within the compartments. Applying a uniform preset threshold to all refrigerators for refrigeration fault detection would drastically increase the false alarm rate, impacting the user experience and causing unnecessary waste of after-sales resources.

[0020] To address the aforementioned issues, this embodiment provides a method for detecting refrigerator cooling anomalies. Based on operational data collected from the refrigerator itself and operational data of the same model stored in the cloud, the method determines the refrigerator's cooling status, effectively improving the accuracy of cooling anomaly detection and reducing the probability of false alarms. (Refer to...) Figure 1 As shown, the method includes the following steps: S100: Based on the operating data collected at the refrigerator end, determine that the current refrigerator meets the preset normal cooling conditions. The operating data includes at least the sensor temperature of each compartment, the set temperature of each compartment, the door open / close status, and the load switch status.

[0021] In practice, the refrigerator can acquire various operational data in real time through its own configured detection and status acquisition components. The refrigerator's cooling effect can be affected by various non-fault factors, such as outside hot air entering the compartment when the door is opened, the heater causing the compartment temperature to rise during defrosting, and the compartment temperature naturally rising when the compressor is off. Temperature increases in these scenarios do not fall under the category of cooling anomalies. If anomaly detection is performed directly without condition screening, a large number of false alarms will occur. Therefore, before conducting cooling anomaly detection, it is necessary to determine whether the refrigerator is currently operating under normal cooling conditions based on the collected operational data. Only when the preset normal cooling conditions are met will the subsequent anomaly detection results be of reference value.

[0022] In one implementation, the data acquisition process can be continuously executed by the refrigerator's local control unit. The acquisition frequency can be set to match the refrigerator's cooling control cycle to ensure that the data reflects the refrigerator's operating status in real time. The temperature of each compartment sensor included in the operating data directly reflects the real-time temperature level of the corresponding compartment and is the core basis for determining the cooling effect. The set temperature of each compartment corresponds to the user-set target cooling temperature and is the benchmark parameter for calculating temperature deviation. The door open / closed status reflects the opening and closing of the refrigerator door, used to eliminate temperature fluctuation interference caused by door opening. The load switch status reflects the working status of the refrigerator's core cooling components and auxiliary components, used to confirm whether the refrigerator is in the cooling operation phase. Through the acquisition of the above multi-dimensional operating data, various operating conditions affecting cooling determination can be comprehensively covered, providing sufficient data support for determining normal cooling conditions.

[0023] Furthermore, normal cooling conditions can be set according to the working logic of the refrigerator's cooling system. Their core function is to filter out non-cooling operation, external heat input, or auxiliary heating conditions, ensuring that anomaly detection is performed when the refrigerator is in a stable cooling state. In one implementation, normal cooling conditions can be set to the refrigerator being in a cooling cycle without additional heat input. The specific judgment logic can be achieved through a combination of door open / closed status and load open / closed status. When the refrigerator is determined to meet the preset normal cooling conditions, the subsequent temperature difference calculation and comparison stage can begin; if the normal cooling conditions are not met, the operating data is continuously monitored until the conditions are met before initiating the anomaly detection process. This approach can eliminate a large number of non-faulty temperature fluctuations at the source of detection, reducing the possibility of misjudgment from a fundamental level, while also reducing unnecessary subsequent calculation processes and lowering the system's operating load.

[0024] S200: Calculate the actual temperature difference between the sensor temperature of the target chamber and the set temperature of the target chamber based on the operating data.

[0025] After confirming that the refrigerator is in normal cooling mode, the cooling status of the target compartment can be quantitatively determined. The real-time sensor temperature value for the target compartment and the user-set temperature value for that compartment can be extracted from the collected operational data. The actual temperature difference is obtained by calculating the difference between the two. It is understandable that different compartments have different temperature ranges, and directly using absolute temperature for determination cannot adapt to different set temperature scenarios. Using the difference between the sensor temperature and the set temperature as the determination indicator can eliminate the benchmark differences caused by different set temperatures, giving the determination indicator a unified measurement standard and more intuitively reflecting the degree of deviation between the refrigerator's cooling capacity and the target requirements.

[0026] In one implementation, the actual temperature difference can be calculated by subtracting the set temperature value from the sensor temperature value. A positive result indicates that the actual temperature of the room is higher than the user-set target temperature, and the larger the difference, the more severe the cooling deviation. A negative result indicates that the actual temperature of the room is lower than the set temperature, and the cooling effect exceeds the target requirements. By determining the sign and magnitude of the temperature difference, the degree of cooling deviation in the target room can be quantitatively characterized, providing standardized judgment parameters for subsequent threshold comparison. Furthermore, the calculation of the actual temperature difference can be performed synchronously with the acquisition of operational data. That is, the current actual temperature difference is calculated synchronously every time operational data is acquired to ensure the real-time nature of the temperature difference data and to promptly capture the changing trend of room temperature. This step, by constructing a standardized deviation index, transforms the originally discrete temperature data into a directly comparable judgment quantity, providing a unified judgment basis for subsequent anomaly initial judgment.

[0027] S300: Compare the actual temperature difference with a preset reference temperature difference threshold for the target room, and trigger the cloud verification process when the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches a preset first duration.

[0028] After obtaining the real-time actual temperature difference, this value is compared with a pre-set baseline temperature difference threshold for the target compartment. The baseline temperature difference threshold serves as the initial benchmark for determining whether the cooling deviation exceeds the normal range. It is understandable that the temperature within a refrigerator compartment exhibits certain fluctuations, and the start-stop control of the cooling system also causes periodic temperature changes within a small range. If a single temperature difference exceeding the threshold is automatically considered abnormal, normal temperature fluctuations would be misinterpreted as cooling malfunctions, leading to a higher false alarm rate. Therefore, this step introduces a continuous duration judgment condition. Only when the cumulative duration for which the actual temperature difference continuously exceeds the baseline temperature difference threshold reaches a preset first duration is it considered that the current compartment may have a cooling abnormality, thus triggering the subsequent cloud-based verification process.

[0029] In one implementation, the reference temperature difference threshold can be initially set according to the design cooling accuracy of the corresponding compartment, serving as the initial reference value for anomaly detection. The first duration setting can be matched according to the thermal inertia of the compartment and the cooling cycle, its function being to filter out short-term temperature fluctuations and ensure that only continuous temperature deviations are identified as potential anomalies.

[0030] Specifically, the duration of temperature difference exceeding the threshold can be accumulated through the timing unit. If the actual temperature difference falls below the baseline temperature difference threshold at a certain moment during the accumulation process, the accumulated duration is reset and the timing restarts. The cloud verification process is only triggered when the duration of continuous and uninterrupted exceeding the threshold reaches the first duration.

[0031] This processing method effectively eliminates interference from instantaneous temperature fluctuations, improves the accuracy of the initial judgment, and avoids frequent cloud verification triggered by occasional temperature fluctuations, reducing the computational pressure and data interaction volume on the cloud. Simultaneously, the continuous duration of the judgment better aligns with the fault characteristics of refrigeration anomalies. A decrease in cooling effect caused by a refrigeration system failure typically manifests as a sustained increase in temperature, rather than instantaneous temperature fluctuations. Therefore, this judgment logic matches the actual fault characteristics, improving the reliability of the initial judgment results. When the triggering conditions are met in the initial judgment stage, a verification request is initiated to the cloud, starting the cloud verification process. The initial judgment results are then re-verified using group data, further enhancing the accuracy of the judgment.

[0032] The cloud-based verification process includes: S410: Based on the historical operating data of a group of refrigerators of the same model as the current refrigerator under normal cooling conditions stored in the cloud, the percentage of refrigerators in the group of refrigerators of the same model that meet the first anomaly determination condition is calculated. The first anomaly determination condition includes the cumulative duration for which the actual temperature difference of the group of refrigerators of the same model continuously exceeds the benchmark temperature difference threshold.

[0033] The core of the cloud-based verification process is to use the operational data of a group of refrigerators of the same model to verify the initial judgment result of an individual refrigerator. The underlying logic is that refrigerators of the same model have the same structural design, the same sensor installation locations, and the same refrigeration system configuration. Therefore, under normal operating conditions, the temperature deviation characteristics of a group of refrigerators of the same model are consistent. If a large number of refrigerators of the same model are judged as abnormal at a certain temperature difference threshold, it indicates that the threshold setting is unreasonable, rather than a fault in a single refrigerator; if only a few refrigerators are judged as abnormal, it indicates an individual fault in a single refrigerator.

[0034] The cloud pre-stores a large amount of historical operating data for refrigerators, and all data is associated with the corresponding refrigerator model information. When a verification request for the current refrigerator is received, the cloud first filters out the operating data of all refrigerators of the same model based on the current refrigerator's model information, and then further filters out the historical operating data of refrigerators under the same normal cooling conditions as the current refrigerator, forming a sample dataset of refrigerators of the same model. Subsequently, based on this sample dataset, the cloud counts the number of refrigerators that meet the first anomaly judgment condition. The first anomaly judgment condition is consistent with the local initial judgment condition, that is, the cumulative duration for which the actual temperature difference continuously exceeds the benchmark temperature difference threshold reaches a first duration, to ensure that the judgment standard for the group statistics is completely consistent with the standard for the initial judgment of a single unit.

[0035] In one implementation, cloud-based data statistics can be updated at a fixed period or performed in real time each time a verification request is received. During the statistical process, historical operating data for each refrigerator of the same model can be analyzed to calculate the number of times it meets the first anomaly judgment condition under the same cooling conditions. Finally, the proportion of refrigerators meeting the condition out of the total number of refrigerators of the same model is counted, i.e., the refrigerator percentage. It is understandable that the larger the sample size of the refrigerator group of the same model, the more the statistical percentage result reflects the general operating characteristics of that model of refrigerator, and the higher the reliability of the verification result. Through statistical analysis of group data, the operating status of a single refrigerator can be measured at the overall level of the same model group, avoiding the limitations of a single threshold and providing an objective group reference for threshold rationality judgment and anomaly judgment.

[0036] S420: Compare the refrigerator percentage with a preset abnormal percentage threshold. If the refrigerator percentage is less than the abnormal percentage threshold, then determine that the target compartment of the current refrigerator is refrigerating abnormally.

[0037] After obtaining the percentage of refrigerators in a group of refrigerators of the same model that meet the first anomaly judgment condition, this percentage is compared with a preset anomaly percentage threshold. The anomaly percentage threshold is the criterion for distinguishing between "individual faults" and "unreasonable thresholds." The anomaly percentage threshold can be set according to the product's failure rate design indicators, representing the upper limit of a reasonable proportion of similar anomalies occurring in a group of refrigerators of the same model. When the percentage of refrigerators is less than the anomaly percentage threshold, it indicates that only a very small number of refrigerators in the group of refrigerators of the same model will experience the same temperature difference continuously exceeding the threshold. This phenomenon is a low-probability individual anomaly, rather than a widespread situation. Therefore, it can be inferred that the current temperature deviation of the refrigerator is not caused by an unreasonable threshold setting, but by a fault in the refrigerator's own refrigeration system. Thus, it can be determined that the target compartment of the current refrigerator is experiencing a refrigeration anomaly.

[0038] This judgment logic uses the normal level of the group as a reference standard. Compared with the traditional fixed threshold judgment, it fully considers the characteristics of different refrigerator models and can effectively avoid misjudgments caused by inherent model deviations. When a cooling abnormality is judged, a corresponding abnormality prompt message can be generated and sent to the refrigerator or user terminal to remind the user to take timely action and avoid food spoilage and other losses. This step, by comparing the group's proportion, completes a secondary verification of the initial judgment result, which can significantly improve the accuracy of abnormality judgment and reduce the probability of false alarms.

[0039] S430: If the refrigerator ratio is greater than or equal to the abnormal ratio threshold, the reference temperature difference threshold is adaptively updated based on the current refrigerator startup parameters, and the updated reference temperature difference threshold is used as the new reference temperature difference threshold to return to the actual temperature difference calculation and comparison steps.

[0040] If the percentage of refrigerators with abnormal operating rates is greater than or equal to the abnormal operating rate threshold, it indicates that a significant number of refrigerators of the same model are experiencing the same persistent temperature difference exceeding the threshold. This phenomenon is not an isolated fault, but rather indicates that the currently used baseline temperature difference threshold is set too low and does not conform to the actual operating characteristics of this refrigerator model. Continuing to use the original baseline temperature difference threshold for judgment would lead to a large number of normally operating refrigerators being mistakenly identified as abnormal. Therefore, the baseline temperature difference threshold needs to be adaptively adjusted to match the actual operating level of this refrigerator model.

[0041] The threshold update is based on the current refrigerator's startup parameters. These parameters correspond to the inherent characteristics of the refrigerator's cooling operation and reflect the temperature regulation capability and normal fluctuation range of the refrigeration system. Adjusting the original reference temperature difference threshold based on the startup parameters ensures that the adjusted threshold conforms to the operating characteristics of the refrigerator model, avoiding excessive or insufficient adjustments. In one implementation, the threshold update calculation can be completed in the cloud. After the update, the cloud sends the new reference temperature difference threshold to the refrigerator. Upon receiving the new threshold, the refrigerator replaces the original reference temperature difference threshold and re-executes the actual temperature difference calculation and comparison steps based on the new threshold to continue detecting cooling anomalies.

[0042] This dynamic threshold update mechanism automatically adjusts the judgment threshold based on actual operating data of refrigerators of the same model, ensuring that the threshold always matches the normal operating level of that model. As operating data stored in the cloud accumulates, the threshold adjustment becomes increasingly precise, ultimately achieving a judgment standard highly adapted to the characteristics of each model, fundamentally solving the problem of fixed thresholds being incompatible with different refrigerator models. Furthermore, re-testing after threshold updates ensures the continuity of the testing process without interruption, improving its smoothness and reliability.

[0043] In summary, the refrigerator cooling anomaly detection method provided in this embodiment addresses the high false alarm rate caused by the use of a uniform fixed threshold in existing technologies. It employs a two-level judgment architecture—local initial judgment plus cloud-based group verification—to dynamically adjust the judgment threshold based on the operating data of a group of refrigerators of the same model. First, local continuous temperature difference comparison performs initial screening to filter out interference from instantaneous temperature fluctuations. Then, statistical analysis of the proportion of data from the same model group in the cloud verifies the rationality of the initial judgment, distinguishing between individual faults and threshold deviations. When the judgment threshold is unreasonable, it automatically performs an adaptive update and returns for re-detection. This method can dynamically adjust the judgment threshold according to the actual operating conditions of different refrigerator models, effectively eliminating judgment errors caused by structural differences and sensor installation locations between different refrigerator models. This significantly improves the accuracy of cooling anomaly detection, reduces the false alarm rate, and also reduces unnecessary after-sales work orders, improving the user experience.

[0044] In some embodiments of this application, the target compartment includes a refrigerator compartment. The normal refrigeration conditions include the refrigerator compartment door being closed, the defrost heater being off, and the compressor being on. In the implementation scenario where the target compartment is a refrigerator compartment, the normal refrigeration conditions specifically include the three conditions of the refrigerator compartment door being closed, the defrost heater being off, and the compressor being on.

[0045] Specifically, by collecting door opening / closing status data from the refrigerator, it can be determined whether the refrigerator door is closed. If the refrigerator door is open, hot outside air will continuously enter the refrigerator compartment, causing the compartment temperature to rise. This temperature rise is normal and does not meet the criteria for determining a cooling malfunction. By limiting the condition to a closed refrigerator door, temperature interference from door opening can be eliminated, ensuring that the test is conducted under closed refrigerator conditions.

[0046] Secondly, the defrosting heater being off is another necessary condition for normal cooling. During refrigerator operation, a defrosting process occurs periodically. The defrosting heater activates during defrosting, causing a slight temperature rise in the refrigerator compartment. This temperature rise is a normal phenomenon in the defrosting process and does not indicate a malfunction in the cooling system. Therefore, it is necessary to confirm that the defrosting heater is off by checking the load switch status data to eliminate interference from the defrosting process and avoid misinterpreting the temperature rise caused by defrosting as a cooling abnormality.

[0047] Third, the compressor being on is a crucial prerequisite for refrigeration operation. The compressor is the core power component of the refrigerator's refrigeration system. Only when the compressor is on can the refrigeration system operate, and only then will the refrigerator compartment temperature decrease under the action of the refrigeration system. If the compressor is off, the refrigerator compartment temperature will naturally rise again, which is normal temperature control logic. Therefore, it is necessary to confirm that the compressor is on by checking the load switch status to ensure that the test is conducted under normal operating conditions of the refrigeration system.

[0048] In one implementation, all three conditions must be met simultaneously for the refrigerator compartment to be determined to be under normal cooling conditions, allowing the subsequent anomaly detection process to proceed. If any one condition is not met, the status is continuously monitored until all conditions are met before detection is initiated. By setting three specific operating conditions for the refrigerator compartment, various non-fault-related temperature fluctuation factors in the refrigerator compartment scenario can be accurately filtered out, providing a stable and reliable operating condition basis for subsequent temperature difference determination and further reducing the probability of misjudging refrigerator compartment cooling anomalies.

[0049] For the refrigerator compartment scenario, the actual temperature difference is the difference between the refrigerator compartment sensor temperature and the refrigerator compartment set temperature, and the reference temperature difference threshold is a preset reference value for the refrigerator compartment. In determining the duration of continuous exceedance of the threshold, the first duration can be adapted to the thermal inertia and cooling cycle of the refrigerator compartment to match the temperature change rate of the refrigerator compartment. When the cumulative duration of the actual temperature difference exceeding the refrigerator compartment reference temperature difference threshold reaches the first duration, the cloud verification process is triggered. This initial judgment logic is consistent with that of Embodiment 1 and is adapted to the operating characteristics of the refrigerator compartment, which can effectively screen out potential abnormal cooling conditions in the refrigerator compartment.

[0050] Therefore, in step S430 of the cloud verification process, the adaptive update of the reference temperature difference threshold based on the current refrigerator's startup parameters specifically includes: S431: Add the original reference temperature difference threshold for the refrigerator compartment to the current refrigerator start-up parameters to obtain the updated new reference temperature difference threshold.

[0051] Specifically, in the cloud-based verification process, the cloud statistically analyzes the percentage of refrigerators of the same model whose refrigerator compartments meet the first anomaly detection criteria under the same normal cooling conditions, and compares this percentage with the anomaly percentage threshold. If the percentage is less than the anomaly percentage threshold, the refrigerator compartment is determined to be malfunctioning; if the percentage is greater than or equal to the anomaly percentage threshold, the baseline temperature difference threshold for the refrigerator compartment is adaptively updated.

[0052] The refrigerator start-up parameters are parameters that match the refrigeration operation characteristics of the refrigerator compartment, reflecting the normal temperature fluctuation range and refrigeration regulation capability of the refrigerator compartment's refrigeration system. Adjusting the threshold based on the refrigerator start-up parameters ensures that the adjustment range is compatible with the operating characteristics of the refrigerator compartment, avoiding excessively high or low thresholds after adjustment. In one implementation, the refrigerator start-up parameters can be initially set according to the refrigerator compartment's design parameters, and the refrigerator start-up parameters for refrigerators of the same model are kept consistent to ensure uniformity in threshold adjustment for refrigerators of the same model. The updated new reference temperature difference threshold is sent from the cloud to the refrigerator. After the refrigerator replaces the original threshold with the new threshold, it recalculates the actual temperature difference and compares it, continuing the refrigerator compartment's refrigeration anomaly detection process.

[0053] Understandably, using dedicated refrigeration start-up parameters as the adjustment step size ensures that each threshold adjustment aligns with the refrigeration characteristics of the refrigerator compartment, avoiding issues such as excessively large adjustments leading to lenient judgment standards or insufficient adjustments failing to resolve false alarms. As cloud data continues to accumulate, the threshold will gradually converge to the value best suited to this refrigerator compartment model through multiple iterations, ensuring that the judgment standard always remains consistent with the actual operating level of the product.

[0054] In some embodiments of this application, the target chamber may further include a variable temperature chamber. In implementation scenarios where the target chamber is a variable temperature chamber, normal cooling conditions can be set with reference to the general operating conditions of cooling operation to ensure that the variable temperature chamber is in a closed, non-defrosting, cooling operation state to eliminate temperature interference from non-fault factors. The actual temperature difference is the difference between the sensor temperature of the variable temperature chamber and the set temperature of the variable temperature chamber. The reference temperature difference threshold is a preset reference value for the variable temperature chamber. The first duration can be adapted and set according to the thermal inertia and cooling rate of the variable temperature chamber. When the actual temperature difference continuously exceeds the reference temperature difference threshold of the variable temperature chamber for a cumulative duration reaching the first duration, the cloud verification process is triggered to complete the initial judgment of the cooling anomaly of the variable temperature chamber.

[0055] The set temperature of the variable temperature compartment can be adjusted by the user within a certain range. Different set temperatures correspond to different cooling loads, but using temperature difference as the judgment index can eliminate the influence of different set temperature references, making the judgment standard consistent across different temperature levels. The initial value of the reference temperature difference threshold can be set according to the average cooling accuracy of the variable temperature compartment to cover the normal fluctuation range of most operating scenarios.

[0056] Therefore, in step S430 of the cloud verification process, the adaptive update of the reference temperature difference threshold based on the current refrigerator's startup parameters further includes: S432: Add the original reference temperature difference threshold for the variable temperature compartment to the current variable temperature start-up parameters of the refrigerator to obtain the updated new reference temperature difference threshold.

[0057] Specifically, in the cloud-based verification process, the cloud statistically analyzes the percentage of refrigerators of the same model whose variable temperature compartments meet the first anomaly detection criteria under the same normal cooling conditions, and compares this percentage with an anomaly percentage threshold. If the percentage is less than the anomaly percentage threshold, the variable temperature compartment is determined to be malfunctioning; if the percentage is greater than or equal to the anomaly percentage threshold, the baseline temperature difference threshold for the variable temperature compartment is adaptively updated.

[0058] The variable temperature start-up parameters are designed to match the cooling operation characteristics of the variable temperature compartment, adapting to its wide temperature range and reflecting the normal temperature fluctuation range of the cooling system under different set temperatures. Adjusting the threshold based on these parameters ensures that the adjusted threshold matches the operating characteristics of the variable temperature compartment, maintaining reasonable accuracy regardless of the user-set temperature.

[0059] In one implementation, the variable temperature start-up parameters can be initially set based on the design cooling capacity and temperature adjustment range of the variable temperature compartment. The variable temperature start-up parameters of refrigerators of the same model are kept consistent to ensure the uniformity of threshold adjustment for the same model of equipment. After the update is completed, the cloud sends the new reference temperature difference threshold to the refrigerator. The refrigerator replaces the original threshold, recalculates the actual temperature difference, performs a comparison, and continues to execute the cooling anomaly detection process of the variable temperature compartment.

[0060] Understandably, variable temperature chambers have a wide temperature setting range. If a fixed temperature difference threshold is used, the accuracy of judgment will vary at different temperature settings. However, a dynamic adjustment mechanism based on variable temperature start-up parameters ensures that the threshold always adapts to the current operating state of the variable temperature chamber, guaranteeing consistency in judgment standards across different temperature settings. Furthermore, the verification logic combined with data from a group of similar models effectively eliminates judgment errors caused by differences in the installation location of the variable temperature chamber sensors, improving the accuracy of anomaly detection.

[0061] In some embodiments of this application, reference is made to Figure 2 As shown, the target compartment also includes a freezer compartment; After step S300, which compares the actual temperature difference with a preset reference temperature difference threshold for the target room, and before triggering the cloud verification process, the following step is also included: S301: During a continuous second time period, monitor whether the sensor temperature of the freezer compartment is continuously greater than a preset high temperature threshold for the freezer compartment; S302: If not, then the freezer compartment is determined to be cooling normally; S303: If so, then the cloud verification process is triggered.

[0062] In scenarios where the target compartment is the freezer, the system first confirms that the refrigerator meets normal cooling conditions based on operational data collected from the refrigerator itself. Then, it calculates the actual temperature difference between the freezer sensor temperature and the set freezer temperature, comparing this actual temperature difference with a preset baseline temperature difference threshold for the freezer. If the actual temperature difference continuously exceeds the baseline threshold for a cumulative period exceeding a preset first duration, the cloud verification process is not directly triggered; instead, an additional step of determining the absolute temperature is added.

[0063] If the freezer compartment sensor temperature does not consistently exceed the baseline high temperature threshold during the second time period, it indicates that the absolute temperature of the freezer compartment is still within a reasonable range, with only a brief deviation in temperature difference, which is not considered a cooling anomaly, and the freezer compartment is judged to be cooling normally. If the freezer compartment sensor temperature consistently exceeds the baseline high temperature threshold during the second consecutive time period, it indicates that the freezer compartment not only exceeds the temperature difference standard, but also that the absolute temperature has exceeded the safe range, and there is a possibility of a cooling anomaly. At this time, the cloud verification process is triggered.

[0064] This embodiment adds a benchmark high-temperature threshold determination step, which enables secondary screening of the freezer's cooling status from an absolute temperature perspective, avoiding misjudgments caused by excessive temperature differences due to factors such as temperature setting adjustments, even when the actual temperature remains normal. In one implementation, the benchmark high-temperature threshold is the highest permissible safe temperature of the freezer; exceeding this temperature compromises the safety of stored food. This threshold can be set according to the storage requirements of frozen foods. The second duration can be set based on the thermal inertia of the freezer to filter short-term absolute temperature fluctuations and ensure the stability of the determination.

[0065] By using a dual preliminary judgment mechanism of temperature difference and absolute temperature, the refrigeration status of the freezer can be comprehensively judged from two dimensions: relative deviation and absolute level. This makes the preliminary judgment results more consistent with the actual use needs of the freezer, effectively improving the reliability of the preliminary judgment process and reducing misjudgment and omission caused by a single judgment dimension.

[0066] Furthermore, the cloud verification process also includes: S401: If the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, further calculate the proportion of refrigerators in the same model group that meet the second abnormality judgment condition. The second abnormality judgment condition includes the cumulative duration during which the freezer compartment sensor temperature of the same model group continuously exceeds the benchmark high temperature threshold to reach the second duration. S402: If the proportion of refrigerators that meet the second abnormality determination condition is greater than or equal to the abnormality proportion threshold, then the reference high temperature threshold is adaptively updated based on the current refrigerator's start-up parameters, and the updated reference high temperature threshold is used as the new reference high temperature threshold to return to the step of monitoring the freezer compartment sensor temperature.

[0067] Specifically, after triggering the cloud verification process, the cloud first uses historical operating data of refrigerators of the same model under the same normal cooling conditions to calculate the percentage of refrigerators that meet the first anomaly determination condition. The first anomaly determination condition is that the cumulative duration for which the actual temperature difference continuously exceeds the benchmark temperature difference threshold reaches a certain duration. This percentage of refrigerators is then compared with the anomaly percentage threshold. If the percentage of refrigerators is less than the anomaly percentage threshold, the cloud continues to calculate the percentage of refrigerators of the same model that meet the second anomaly determination condition. The second anomaly determination condition includes that the cumulative duration for which the freezer compartment sensor temperature of the same model refrigerator group continuously exceeds the benchmark high temperature threshold reaches a certain duration.

[0068] If the percentage of refrigerators meeting the second anomaly detection condition is also less than the anomaly percentage threshold, it indicates that both excessive temperature difference and excessive absolute temperature are low-probability individual phenomena within the same model group, thus determining that the freezer compartment of the current refrigerator is malfunctioning. If the percentage of refrigerators meeting the first anomaly detection condition is greater than or equal to the anomaly percentage threshold, it indicates that the baseline temperature difference threshold is set unreasonably and needs to be updated adaptively. If the percentage of refrigerators meeting the second anomaly detection condition is greater than or equal to the anomaly percentage threshold, it indicates that the baseline high temperature threshold is set unreasonably and needs to be updated adaptively.

[0069] In one implementation, the determination and updating of the two thresholds are independent and can be performed and adjusted separately. When the proportion of one threshold exceeds the standard, only that threshold is updated, while the other threshold remains unchanged; if the proportion of both thresholds exceeds the standard, both thresholds are updated simultaneously. After the threshold update is completed, the updated threshold is sent to the refrigerator, which then re-executes the steps of monitoring the freezer compartment sensor temperature and comparing the temperature difference based on the new threshold to continue detecting refrigeration anomalies in the freezer compartment.

[0070] In this embodiment, the cloud verification process also employs a dual-judgment logic, verifying and adjusting the rationality of the two judgment thresholds separately. This ensures that the judgment criteria in both dimensions match the actual operating characteristics of the same model of refrigerator. As cloud data accumulates, both thresholds will be gradually optimized to the level best suited to this model of freezer compartment, thereby comprehensively improving the accuracy of freezer compartment cooling anomaly detection.

[0071] In some embodiments of this application, in an implementation scenario where the target compartment is a freezer compartment, step S430, which adaptively updates the reference high-temperature threshold based on the current refrigerator's operating parameters, specifically includes: The original baseline high temperature threshold for the freezer compartment is added to the current freezer start-up parameters of the refrigerator to obtain the updated new baseline high temperature threshold.

[0072] In the cloud-based verification process for detecting refrigeration anomalies in the freezer compartment, when it is determined that the baseline high-temperature threshold setting is unreasonable and needs to be updated adaptively, the freezer start-up parameters are used as the adjustment step size, and the parameters are superimposed on the original baseline high-temperature threshold to obtain the updated new baseline high-temperature threshold.

[0073] The freezer start-up parameters are designed to match the refrigeration operation characteristics of the freezer compartment. They reflect the normal temperature fluctuation range and low-temperature maintenance capability of the freezer's refrigeration system. Adjusting the threshold based on these parameters ensures that the adjustment range aligns with the freezer's operating patterns. Because the freezer compartment has a lower temperature range, the rate and magnitude of temperature change differ from those of the refrigerator and variable temperature compartments. Therefore, using dedicated freezer start-up parameters for adjustment avoids excessively large or small adjustments, ensuring that the updated baseline high-temperature threshold is neither too lenient (leading to missed detections) nor too strict (leading to false detections).

[0074] After the update is completed, the cloud sends the new benchmark high temperature threshold to the refrigerator. The refrigerator replaces the original value with the new benchmark high temperature threshold and then restarts the temperature monitoring process of the freezer compartment. During the second consecutive time period, it monitors whether the temperature of the freezer compartment sensor is continuously greater than the new benchmark high temperature threshold, and continues to carry out refrigeration anomaly detection based on the temperature difference.

[0075] In some embodiments of this application, before step S100, which determines that the current refrigerator meets preset normal cooling conditions based on the operating data collected at the refrigerator end, the following step is further included: S101: Obtain the model information and operating data of the refrigerator, and upload the model information and operating data to the cloud; S102: The cloud platform classifies and stores the received operating data according to the model information for use in the cloud verification process.

[0076] Before determining whether the refrigerator meets the preset normal cooling conditions based on the operational data collected at the refrigerator's end, the first step is to perform data collection and uploading / storage. Specifically, the refrigerator's end first obtains the refrigerator's model information and various operational data. The model information is the refrigerator's identifier, corresponding to its structural design, hardware configuration, and refrigeration system parameters, and is the core basis for distinguishing different refrigerator products. The operational data includes data required for detecting various cooling anomalies, such as the sensor temperatures of each compartment, the set temperatures of each compartment, door open / close status, and load switch status.

[0077] In one implementation, model information can be pre-stored in the refrigerator's local storage unit and written during manufacturing, with each refrigerator having a unique model identifier. Operating data is collected in real-time by various sensors and status detection components located within the refrigerator, and then aggregated by the refrigerator's control unit. After data collection, the refrigerator uploads both the model information and operating data to a cloud server via its configured communication module. The upload frequency can be set to match the data collection frequency or to be set for batch uploads at fixed intervals, reducing communication load while ensuring data timeliness.

[0078] After receiving the model information and operating data uploaded by the refrigerator, the cloud classifies and stores the received operating data according to the model information. That is, all operating data uploaded by refrigerators of the same model are collected into the same data set, while data of refrigerators of different models are stored separately and independently.

[0079] Categorizing and storing data by model is fundamental to cloud-based group verification. Different refrigerator models vary in structure, sensor location, and cooling capacity. Mixing and statistically analyzing these models would render the group data unusable and make it impossible to accurately determine the reasonableness of thresholds. Categorizing and storing data by model ensures the organization and accuracy of cloud data, providing reliable data support for subsequent cloud verification processes. Furthermore, categorized data can be used for other purposes such as product operation status analysis and fault pattern statistics, enhancing the value of the data.

[0080] After the data upload and categorized storage are completed, the cloud can quickly retrieve the group data of the corresponding refrigerator model for statistical analysis when the verification process is triggered, improving the response speed of the verification process. On the refrigerator side, while the data is being uploaded, the local normal cooling condition determination and anomaly preliminary judgment processes are carried out simultaneously. Data upload and local testing can be executed in parallel without interference, ensuring the efficiency of the testing process.

[0081] Another aspect of this application provides a refrigerator cooling anomaly detection system for executing the refrigerator cooling anomaly detection methods described in the above method embodiments. The system includes two parts: a refrigerator-side device and a cloud server. The two are connected wirelessly to collaboratively complete the entire process of detecting, verifying, and adjusting thresholds for cooling anomalies.

[0082] The refrigerator-side device is configured to collect the refrigerator's operating data, which includes at least the sensor temperature of each compartment, the set temperature of each compartment, the door open / close status, and the load switch status. Based on the operating data, it is determined whether the refrigerator currently meets the preset normal cooling conditions.

[0083] The refrigerator-side equipment, deployed on the refrigerator itself, serves as the local core for refrigerator operation control and data acquisition. By connecting to various detection components, the refrigerator-side equipment can acquire real-time status data during refrigerator operation and perform preliminary data processing and analysis.

[0084] The refrigerator-side device is also configured to calculate the actual temperature difference between the sensor temperature of the target compartment and the set temperature of the target compartment, compare the actual temperature difference with a preset benchmark temperature difference threshold for the target compartment, and send a verification request to the cloud server when the cumulative duration for which the actual temperature difference continuously exceeds the benchmark temperature difference threshold reaches a preset first duration. Temperature difference calculation, threshold comparison, and continuous duration statistics are all completed locally on the refrigerator side, constituting the initial judgment stage of anomaly detection. Local initial judgment can quickly identify potential anomalies without frequent interaction with the cloud, reducing communication overhead and improving the real-time performance of detection.

[0085] When the triggering conditions are met, the refrigerator sends a verification request to the cloud server. This request can include essential information such as the refrigerator's model, current operating parameters, and baseline temperature difference threshold, enabling cloud-based verification. The verification request can be automatically triggered by the refrigerator's control unit without manual intervention, ensuring automated anomaly detection.

[0086] The cloud server is configured to receive the verification request and, based on its own stored historical operating data of a group of refrigerators of the same model as the current refrigerator under normal cooling conditions, calculate the percentage of refrigerators in the group of refrigerators of the same model that meet the first anomaly determination condition. The first anomaly determination condition includes the cumulative duration for which the actual temperature difference of the group of refrigerators of the same model continuously exceeds the benchmark temperature difference threshold, reaching the first duration.

[0087] In one implementation, the cloud server can employ a distributed storage architecture to store operational data, supporting the data access and storage needs of massive numbers of devices. Statistical calculations can utilize parallel computing methods to improve processing efficiency in scenarios with multiple concurrent requests, ensuring real-time feedback of verification results.

[0088] The cloud server is also configured to compare the percentage of refrigerators with a preset abnormal percentage threshold. If the percentage of refrigerators is less than the abnormal percentage threshold, a determination result indicating an abnormal cooling condition in the target compartment of the current refrigerator is generated. This determination result can be sent to the refrigerator-side device, triggering an abnormality alert, or it can be directly pushed to the user's terminal device to remind the user to handle the issue promptly. The abnormality determination result can also be synchronized to the after-sales system, providing data support for after-sales maintenance and improving troubleshooting efficiency.

[0089] If the percentage of refrigerators is greater than or equal to the abnormal percentage threshold, the cloud server is configured to adaptively update the baseline temperature difference threshold based on the current refrigerator's power-on parameters, and then send the updated baseline temperature difference threshold back to the refrigerator-side device. This allows the refrigerator-side device to re-execute the actual temperature difference calculation and comparison steps based on the new baseline temperature difference threshold. The threshold update calculation is performed in the cloud, ensuring unified management of the update rules. After the updated threshold is sent to the local storage, the refrigerator-side device can continue detection using the new standard, achieving dynamic iteration of the threshold.

[0090] In some embodiments of this application, the refrigerator-side device includes an electronic control board and a WiFi module; the electronic control board is communicatively connected to the WiFi module, and the electronic control board is configured to drive the refrigerator load to operate and collect the operating data.

[0091] The control board can be connected to various loads and detection components of the refrigerator via electrical connection. On the one hand, it outputs control signals to drive loads such as the compressor and defrost heater to operate according to the refrigeration logic, thereby realizing the refrigeration control function of the refrigerator. On the other hand, it receives feedback signals from detection components such as temperature sensors and door switch sensors, and collects various operating data of the refrigerator, including the temperature of each compartment, the door switch status, and the working status of the loads.

[0092] In one implementation, the control board can integrate circuit modules such as a microcontroller unit, a storage unit, and input / output interfaces. The microcontroller unit is used to run the refrigeration control program and local logic for anomaly detection. The storage unit is used to store preset threshold parameters, refrigerator model information, and temporary operating data. The input / output interfaces are used to connect various peripheral components. The control board can independently complete the basic refrigeration control of the refrigerator and the local anomaly judgment function without relying on the cloud, ensuring that the refrigerator can still operate normally and perform basic tests when the network is disconnected.

[0093] The WiFi module is configured to receive the operating data and model information transmitted by the electronic control board, and upload the operating data and model information to the cloud server. The WiFi module acts as a bridge for wireless communication between the refrigerator and the cloud, responsible for uploading local data to the cloud, and simultaneously receiving verification results, updated threshold parameters, and other information from the cloud, and forwarding them to the electronic control board for execution.

[0094] The cloud server is also configured to classify and store the received operational data according to the model information. After receiving the operational data carrying model information uploaded by the WiFi module, the cloud server categorizes the data into the corresponding model's dataset based on the model identifier, achieving standardized classification and management of the data. The classified and stored data can be used to respond to verification requests from the refrigerator and for subsequent product data analysis, enhancing the data's reusability.

[0095] Understandably, this data management approach, which categorizes and stores data by model, ensures that the group data accessed during cloud verification consists entirely of refrigerators of the same model, guaranteeing consistency in the reference benchmark and forming the foundation for accurate and reliable cloud verification results. Simultaneously, categorized storage facilitates separate management and maintenance of data for different refrigerator models in the cloud, improving data management efficiency.

[0096] In some embodiments of this application, the target compartment includes a freezer compartment; the refrigerator end device is further configured to: when the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches the first duration, further monitor whether the sensor temperature of the freezer compartment is continuously greater than a preset reference high temperature threshold for the freezer compartment within a continuous second duration.

[0097] In scenarios where the target compartment is the freezer, the refrigerator employs a dual-judgment initial judgment mechanism for the freezer. First, it performs a continuous duration judgment on the temperature difference dimension. Once the temperature difference judgment meets the triggering conditions, it then performs a continuous duration judgment on the absolute temperature dimension.

[0098] Specifically, after the electronic control board completes the continuous duration statistics of temperature difference exceeding the threshold, if the first duration is reached, it starts the monitoring timer for the baseline high temperature threshold, continuously collecting the sensor temperature of the freezer compartment to determine whether it remains above the baseline high temperature threshold. If the temperature drops below the baseline high temperature threshold within the second duration, it is determined that the freezer compartment is cooling normally, the timer is reset, and temperature difference detection restarts; if the temperature remains above the baseline high temperature threshold for the second consecutive duration, it is determined to be an initial anomaly, and a verification request is sent to the cloud server, carrying two judgment parameters and threshold information for the freezer compartment.

[0099] The local dual initial judgment mechanism can screen potential anomalies in the freezer from two dimensions: relative deviation and absolute level, thereby improving the comprehensiveness of the initial judgment and avoiding missed or misjudgments caused by a single dimension judgment.

[0100] The cloud server is further configured to: when the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, further calculate the proportion of refrigerators in the same model group that meet the second abnormality judgment condition, the second abnormality judgment condition including the cumulative duration for which the freezer compartment sensor temperature of the same model group continuously exceeds the benchmark high temperature threshold to reach the second duration; if the proportion of refrigerators that meet the second abnormality judgment condition is greater than or equal to the abnormal proportion threshold, the cloud server is further configured to: adaptively update the benchmark high temperature threshold based on the current refrigerator's power-on parameters, and send the updated benchmark high temperature threshold back to the refrigerator-side device, so that the refrigerator-side device re-executes the freezer compartment sensor temperature monitoring step based on the new benchmark high temperature threshold.

[0101] After receiving the verification request from the freezer compartment, the cloud server first counts and determines the percentage of refrigerators corresponding to the baseline temperature difference threshold. If the percentage exceeds the limit, the baseline temperature difference threshold is adjusted; if the percentage does not exceed the limit, the count and determination of the percentage of refrigerators corresponding to the baseline high temperature threshold continues. If the percentage of refrigerators meeting the second anomaly determination condition is greater than or equal to the anomaly percentage threshold, the baseline high temperature threshold is adaptively updated.

[0102] The updates to the two thresholds are independent, each judged and adjusted based on the corresponding group proportion results to ensure that each threshold matches the actual operating level of the same model of refrigerator. In one implementation, the cloud server can verify the rationality of the two thresholds separately. If the proportion of both thresholds exceeds the standard, both thresholds are updated simultaneously and sent back together; if only one exceeds the standard, only the corresponding item is updated. After receiving the updated thresholds, the refrigerator replaces the original parameters and restarts the dual detection process of the freezer compartment.

[0103] In some embodiments of this application, the refrigerator end device further includes a temperature sensor, a door switch sensor, a compressor, and a defrost heater, all electrically connected to the electronic control board.

[0104] The temperature sensors are configured to detect the sensor temperature of each compartment. In one embodiment, each compartment of the refrigerator may be equipped with at least one temperature sensor, installed at a corresponding position inside the compartment, for real-time acquisition of the air temperature inside the compartment and conversion of the temperature signal into an electrical signal for transmission to the electronic control board. The temperature sensor is the core component for acquiring compartment temperature data, and its detection accuracy directly affects the accuracy of temperature difference calculation and anomaly detection. Temperature sensors for different compartments can be selected according to the temperature range of the corresponding compartment to ensure detection accuracy and reliability.

[0105] The door switch sensor is configured to detect the door's open / closed status. Each door can be equipped with a corresponding door switch sensor to detect the door's open and closed status and transmit the status signal to the control board. The door switch sensor is the core basis for determining the door's open condition. Through its feedback status signal, the control board can accurately identify the door's open or closed status, thereby eliminating temperature fluctuation interference caused by door opening and ensuring the accuracy of normal cooling condition determination.

[0106] The electronic control board regulates the refrigerator's cooling process by controlling the on / off states of the compressor and the defrost heater. Both the compressor and the defrost heater are core loads in the refrigerator. The compressor is the power source of the refrigeration cycle; the electronic control board controls the compressor's start and stop to regulate the refrigeration output of the refrigeration system and control the compartment temperature. The defrost heater is used in the defrosting process; the electronic control board controls its start and stop to complete the defrosting operation of the evaporator, ensuring the heat exchange efficiency of the refrigeration system. The control signals from the electronic control board to these loads also serve as load on / off status data for determining normal cooling conditions.

[0107] Understandably, the aforementioned hardware components are not only the foundation for the refrigerator's cooling function, but also the data source and status determination basis for the cooling anomaly detection function. The temperature sensor provides temperature data, the door switch sensor provides door status data, and the control status of the compressor and defrost heater provides load status data. All data is aggregated to the electronic control board, which then completes the local anomaly detection logic.

[0108] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein.

[0109] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for detecting abnormal cooling in a refrigerator, characterized in that, Includes the following steps: Based on the operating data collected from the refrigerator, it is determined that the refrigerator currently meets the preset normal cooling conditions. The operating data includes at least the sensor temperature of each compartment, the set temperature of each compartment, the door open / close status, and the load switch status. The actual temperature difference between the sensor temperature of the target chamber and the set temperature of the target chamber is calculated based on the operating data. The actual temperature difference is compared with a preset reference temperature difference threshold for the target room, and the cloud verification process is triggered when the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches a preset first duration. The cloud-based verification process includes: Based on the historical operating data of a group of refrigerators of the same model as the current refrigerator under normal cooling conditions stored in the cloud, the percentage of refrigerators in the group of refrigerators of the same model that meet the first anomaly judgment condition is calculated. The first anomaly judgment condition includes the cumulative duration for which the actual temperature difference of the group of refrigerators of the same model continuously exceeds the benchmark temperature difference threshold, reaching the first duration. The refrigerator percentage is compared with a preset abnormal percentage threshold. If the refrigerator percentage is less than the abnormal percentage threshold, the target compartment of the current refrigerator is determined to be refrigerating abnormally. If the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, the reference temperature difference threshold is adaptively updated based on the current refrigerator's startup parameters, and the updated reference temperature difference threshold is used as the new reference temperature difference threshold to return to the actual temperature difference calculation and comparison steps.

2. The refrigerator cooling anomaly detection method according to claim 1, characterized in that, The target compartment includes a cold storage compartment; The normal refrigeration conditions include the refrigerator door being closed, the defrost heater being off, and the compressor being on. The adaptive update of the reference temperature difference threshold based on the current refrigerator's operating parameters specifically includes: The original reference temperature difference threshold for the refrigerator compartment is added to the current refrigerator start-up parameters to obtain the updated new reference temperature difference threshold.

3. The refrigerator cooling anomaly detection method according to claim 1, characterized in that, The target compartment includes a variable temperature compartment; The adaptive update of the reference temperature difference threshold based on the current refrigerator's operating parameters specifically includes: The original reference temperature difference threshold for the variable temperature compartment is added to the current variable temperature start-up parameters of the refrigerator to obtain the updated new reference temperature difference threshold.

4. The refrigerator cooling anomaly detection method according to claim 1, characterized in that, The target compartment includes a freezer compartment; After the step of comparing the actual temperature difference with a preset reference temperature difference threshold for the target room, and before triggering the cloud verification process, the method further includes: During a continuous second time period, monitor whether the sensor temperature of the freezer compartment is continuously greater than a preset high temperature threshold for the freezer compartment; If not, then the freezer compartment is deemed to be cooling normally; If so, the cloud verification process will be triggered. The cloud-based verification process also includes: If the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, the proportion of refrigerators in the same model group that meet the second abnormality judgment condition is further counted. The second abnormality judgment condition includes the cumulative duration during which the freezer compartment sensor temperature of the same model group continuously exceeds the benchmark high temperature threshold and reaches the second duration. If the percentage of refrigerators that meet the second anomaly determination condition is greater than or equal to the anomaly percentage threshold, then the reference high temperature threshold is adaptively updated based on the current refrigerator's startup parameters, and the updated reference high temperature threshold is used as the new reference high temperature threshold to return to the freezer compartment sensor temperature monitoring step.

5. The refrigerator cooling anomaly detection method according to claim 4, characterized in that, The adaptive update of the benchmark high temperature threshold based on the current refrigerator's operating parameters specifically includes: The original baseline high temperature threshold for the freezer compartment is added to the current freezer start-up parameters of the refrigerator to obtain the updated new baseline high temperature threshold.

6. The refrigerator cooling anomaly detection method according to claim 1, characterized in that, Before the step of determining that the current refrigerator meets the preset normal cooling conditions based on the operating data collected at the refrigerator end, the method further includes: Obtain the model information and operating data of the refrigerator, and upload the model information and operating data to the cloud; The cloud platform categorizes and stores the received operational data according to the model information for use in the cloud verification process.

7. A refrigerator cooling anomaly detection system, used to execute the refrigerator cooling anomaly detection method according to any one of claims 1 to 6, characterized in that, This includes refrigerator-based equipment and cloud servers; The refrigerator terminal device is configured to collect the refrigerator's operating data. The operating data includes at least the sensor temperature of each compartment, the set temperature of each compartment, the door open / close status, and the load switch status. Based on the operating data, it is determined whether the refrigerator currently meets the preset normal cooling conditions. The refrigerator terminal device is also configured to calculate the actual temperature difference between the sensor temperature of the target compartment and the set temperature of the target compartment, compare the actual temperature difference with a preset reference temperature difference threshold for the target compartment, and send a verification request to the cloud server when it is determined that the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches a preset first duration. The cloud server is configured to receive the verification request and, based on its own stored historical operating data of a group of refrigerators of the same model as the current refrigerator under normal cooling conditions, calculate the percentage of refrigerators in the group of refrigerators of the same model that meet the first anomaly determination condition. The first anomaly determination condition includes the cumulative duration during which the actual temperature difference of the group of refrigerators of the same model continuously exceeds the benchmark temperature difference threshold, reaching the first duration. The cloud server is also configured to compare the refrigerator percentage with a preset abnormal percentage threshold. If the refrigerator percentage is less than the abnormal percentage threshold, a judgment result is generated to indicate that the target compartment of the current refrigerator is refrigerating abnormally. If the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, the reference temperature difference threshold is adaptively updated based on the current refrigerator's startup parameters, and the updated reference temperature difference threshold is sent back to the refrigerator-side device so that the refrigerator-side device can re-execute the actual temperature difference calculation and comparison steps based on the new reference temperature difference threshold.

8. A refrigerator cooling anomaly detection system according to claim 7, characterized in that, The refrigerator-side equipment includes an electronic control board and a WiFi module; The electronic control board is communicatively connected to the WiFi module, and the electronic control board is configured to drive the refrigerator load and collect the operating data. The WiFi module is configured to receive the operating data and model information transmitted by the electronic control board, and upload the operating data and model information to the cloud server; The cloud server is also configured to classify and store the received operational data according to the model information.

9. A refrigerator cooling anomaly detection system according to claim 7, characterized in that, The target compartment includes a freezer compartment; The refrigerator terminal device is further configured to: when the cumulative duration for which the actual temperature difference continuously exceeds the reference temperature difference threshold reaches the first duration, further monitor whether the sensor temperature of the freezer compartment is continuously greater than the preset reference high temperature threshold for the freezer compartment within a continuous second duration; The cloud server is further configured to: when the proportion of refrigerators is greater than or equal to the abnormal proportion threshold, further count the proportion of refrigerators in the same model group that meet the second abnormality judgment condition, the second abnormality judgment condition including the cumulative duration for which the freezer compartment sensor temperature of the same model group continuously exceeds the benchmark high temperature threshold to reach the second duration. If the proportion of refrigerators that meet the second anomaly determination condition is greater than or equal to the anomaly proportion threshold, the cloud server is further configured to adaptively update the benchmark high temperature threshold based on the current refrigerator's startup parameters, and send the updated benchmark high temperature threshold back to the refrigerator terminal device, so that the refrigerator terminal device can re-execute the monitoring step of the freezer compartment sensor temperature based on the new benchmark high temperature threshold.

10. A refrigerator cooling anomaly detection system according to claim 7, characterized in that, The refrigerator terminal equipment also includes a temperature sensor, a door switch sensor, a compressor, and a defrost heater, all of which are electrically connected to the electronic control board. The temperature sensor is configured to detect the sensor temperature of each of the compartments; The door switch sensor is configured to detect the door switch status; The electronic control board regulates the refrigeration process of the refrigerator by controlling the on / off state of the compressor and the defrost heater.

Images