Dual-system refrigerator freezing chamber door not closed detection method and system

Abstract

This application discloses a method and system for detecting an open freezer door in a dual-system refrigerator, belonging to the field of smart home appliance fault diagnosis technology. Addressing the problems of existing refrigerator door-open-door detection schemes being susceptible to defrosting interference, not covering the entire fault lifecycle, and lacking adaptability to dual-system refrigerators, this application first acquires the refrigerator's operating data, divides non-defrosting cycles based on the defrosting heater's operating status to isolate defrosting interference, and filters out valid detection data. The open freezer door fault is divided into a temperature rise stage, a steady-state stage, and a recovery stage, and corresponding detection rules are constructed for each stage. The detection logic is dynamically adjusted according to the number of non-defrosting cycles, and after rule matching, a diagnostic result including fault status, fault stage, and fault time interval is generated. This application can achieve accurate detection throughout the entire fault lifecycle, effectively reducing false alarms and false negatives, adapting to the operating characteristics of dual-system refrigerators, and improving the accuracy and reliability of fault diagnosis.

Description

Technical Field

[0001] This application relates to the field of intelligent home appliance fault diagnosis technology, and more specifically to a method and system for detecting that the freezer door of a dual-system refrigerator is not closed properly. Background Technology

[0002] As a widely used refrigeration and preservation appliance in the home, the operational stability of refrigerators directly affects the quality of food storage and the user experience. A freezer door not being closed properly is a common abnormal situation during refrigerator use. If not detected and addressed in time, it can lead to an abnormal increase in freezer temperature, causing food to thaw and spoil, while also increasing the compressor's workload, resulting in increased energy consumption and a shortened lifespan for the appliance. Therefore, accurate detection of a freezer door not being closed properly has significant practical value.

[0003] Currently, common refrigerator door-closing detection solutions can be divided into three categories. The first category is a physical detection solution based on mechanical contacts or infrared sensors, which determines the door's closure status by directly collecting the signal. The second category is a threshold judgment solution based on temperature data, which identifies temperature anomalies by setting a fixed temperature threshold. The third category is a detection solution based on big data analysis, which builds a feature model through historical fault data and identifies faults using methods such as similarity matching.

[0004] In the above solutions, the physical detection solution is susceptible to environmental interference and equipment aging, and the threshold judgment solution does not consider dynamic factors such as defrosting cycle, resulting in limitations in overall detection adaptability and accuracy. At the same time, the existing big data analysis solution does not divide the entire life cycle of the fault into stages, lacks refined mining of data characteristics at different fault stages, is difficult to adapt to the complex operating characteristics of dual-system refrigerators, and cannot achieve complete monitoring from fault occurrence to recovery. Summary of the Invention

[0005] To address the limitations in adaptability and accuracy of the aforementioned detection methods, and the lack of detailed data mining of data characteristics at different fault stages due to the absence of phase division in big data analysis methods, this application provides a method for detecting an open freezer door in a dual-system refrigerator, comprising the following steps: Acquire the refrigerator's operating data, which includes at least the freezer temperature, freezer evaporator temperature, freezer set temperature, and defrost heater operating status; Based on the operating status of the defrosting heater, the operating data is divided into multiple non-defrosting cycles to eliminate interference from the defrosting process on the detection. For each of the non-defrosting cycles, feature parameters are extracted to characterize different stages of the fault; According to the preset multi-stage fault detection rules, the feature parameters are matched. The multi-stage fault detection rules include at least the temperature rise stage rule for identifying the occurrence of the fault, the steady state stage rule for identifying the continuation of the fault, and the recovery stage rule for identifying the end of the fault. Based on the matching results, a diagnostic result is generated, which includes the fault status, fault stage, and time interval of the fault occurrence.

[0006] In one feasible implementation, the step of dividing the operating data into multiple non-defrosting cycles based on the operating state of the defrosting heater includes: The time interval between two defrosting heater start times is taken as a complete cycle, and the portion of each complete cycle from the start time of the defrosting heater after a preset time to the start time of the next defrosting heater is determined as the non-defrosting cycle.

[0007] In one feasible implementation, the step of matching the feature parameters according to a preset multi-stage fault detection rule includes: The detection logic is dynamically adjusted based on the number of non-defrosting cycles. When the first non-defrost cycle is detected, only the matching of the temperature rise phase rule and the steady-state phase rule is performed; When a second or more non-defrosting cycles are detected, matching is performed sequentially according to the priority order of the temperature rise phase rule, the steady state phase rule, and the recovery phase rule.

[0008] In one feasible implementation, the temperature rise phase rule includes: If, within a non-defrosting cycle, there exists a moment in which the rise in the freezer compartment temperature exceeds a first threshold within a preset duration prior to that moment, and the average deviation between the freezer compartment temperature and the set freezer compartment temperature from the start of the non-defrosting cycle to that moment is less than a second threshold, and the difference between the freezer compartment temperature and the set freezer compartment temperature is greater than a third threshold and the difference between the freezer compartment temperature and the freezer evaporator temperature is greater than a fourth threshold at the end of the non-defrosting cycle, and the duration of this state exceeds a fifth threshold, then it is determined that the temperature rise phase has begun.

[0009] In one feasible implementation, the steady-state phase rule includes: within a non-defrosting cycle, if a preset first condition group or a second condition group is met, and the duration of meeting the conditions exceeds a sixth threshold, then it is determined to be the steady-state phase. The first condition group includes: The percentage of data where the difference between the freezer temperature and the freezer set temperature is greater than the seventh threshold is greater than the first percentage threshold. The temperature of the freezer compartment first decreases and then increases within the cycle; The temperature of the refrigeration evaporator generally showed a downward trend; The percentage of data where the difference between the set temperature of the freezer compartment and the temperature of the freezer evaporator is less than the eighth threshold is greater than the second percentage threshold. The minimum temperature of the freezer compartment is within a first temperature range; The second condition group includes: the proportion of data where the difference between the freezer temperature and the freezer set temperature is greater than the seventh threshold is greater than the first proportion threshold; The temperature of the freezer compartment first decreases and then increases within the cycle; The slope of the temperature decrease trend in the freeze evaporator is within a preset slope range; The percentage of data where the difference between the set temperature of the freezer compartment and the temperature of the freezer evaporator is less than the eighth threshold is greater than the second percentage threshold. The minimum temperature of the freezer compartment is less than or equal to the ninth threshold.

[0010] In one feasible implementation, the recovery phase rules include: If, within a non-defrost cycle, there exists a moment in which the temperature drop in the freezer compartment exceeds the tenth threshold within a preset duration prior to that moment, and the overall temperature of the freezer compartment shows a downward trend during the non-defrost cycle, and the duration of this state exceeds the eleventh threshold, and at the end of the non-defrost cycle there exists a moment in which the difference between the temperature of the freezer compartment at that moment and the set temperature of the freezer compartment is less than or equal to the twelfth threshold, then it is determined to be the recovery phase.

[0011] Another aspect of this application provides a dual-system refrigerator freezer door not fully closed detection system, used to implement the dual-system refrigerator freezer door not fully closed detection method described in any of the above claims, including: The data acquisition module is used to acquire the refrigerator's operating data, which includes at least the freezer temperature, the freezer evaporator temperature, the freezer set temperature, and the defrost heater's working status. A data preprocessing module, connected to the data acquisition module, is used to receive the operating data and, based on the working status of the defrosting heater, divide the operating data into multiple non-defrosting cycles to eliminate interference from the defrosting process on the detection. The feature extraction module, connected to the data preprocessing module, is used to receive the non-defrosting cycle and extract feature parameters from it to characterize different stages of the fault. The rule matching module, connected to the feature extraction module, is used to receive the feature parameters and match the feature parameters according to the preset multi-stage fault detection rules. The multi-stage fault detection rules include at least the temperature rise stage rule for identifying the occurrence of the fault, the steady state stage rule for identifying the continuation of the fault, and the recovery stage rule for identifying the end of the fault. The result output module, connected to the rule matching module, is used to receive the matching results and generate diagnostic results, which include the fault status, fault stage, and time interval of the fault occurrence.

[0012] In one feasible implementation, the data preprocessing module is further configured to: take the time interval between two defrosting heater start times as a complete cycle, and determine the portion of each complete cycle between the time interval shifted backward by a preset time from the defrosting heater start time to the next defrosting heater start time as the non-defrosting cycle.

[0013] In one feasible implementation, the rule matching module is further configured to: dynamically adjust the detection logic according to the number of non-defrosting cycles; when the first non-defrosting cycle is detected, only the matching of the temperature rise phase rule and the steady-state phase rule is performed; when the second or more non-defrosting cycles are detected, the matching is performed sequentially according to the priority order of the temperature rise phase rule, the steady-state phase rule, and the recovery phase rule.

[0014] In one feasible implementation, the result output module is further configured to: mark and display the time interval of the fault occurrence in the diagnostic results on the visualization curve of the running data with different colors.

[0015] As described above, the dual-system refrigerator freezer door not fully closed detection method and system provided in this application, through a non-defrost cycle division mechanism, can effectively isolate the interference of the defrosting process on temperature data, providing a clean and effective data foundation for fault diagnosis and reducing the risk of misjudgment caused by defrosting conditions from the data source. The solution divides the freezer door not fully closed fault into three stages: temperature rise, steady state, and recovery, constructing differentiated quantitative detection rules that can fully cover the entire lifecycle of the fault from occurrence and persistence to recovery. The solution dynamically adjusts the detection logic based on the number of non-defrost cycles. In single-cycle scenarios, it focuses on the core fault stage to ensure accuracy; in multi-cycle scenarios, it performs full-stage detection according to priority to ensure coverage. Simultaneously, a cross-cycle correlation judgment mechanism filters isolated misjudgment results, further improving the reliability of diagnostic results. Two sets of composite judgment conditions are set in the steady-state stage to adapt to different degrees of cold leakage and environmental conditions, effectively reducing missed diagnoses. 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 flowchart illustrating the method for detecting an open freezer door in a dual-system refrigerator, as shown in an embodiment of this application. Figure 2 This is an architectural diagram of a dual-system refrigerator freezer compartment door not closed detection system 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] This embodiment provides a method for detecting an open freezer door in a dual-system refrigerator. It addresses the problems of existing freezer door detection solutions, such as the susceptibility of physical sensors to environmental interference and aging, the lack of consideration for the dynamic development of faults in threshold determination schemes, and the inability of existing big data solutions to divide the entire fault lifecycle into stages, resulting in insufficient detection accuracy and incompatibility with the complex operating characteristics of dual-system refrigerators. This method employs a technical approach of periodically isolating interference and using phased matching rules to achieve accurate, full-process fault detection. (Refer to...) Figure 1 As shown, the method includes the following steps: S100: Obtain the refrigerator's operating data, which includes at least the freezer temperature, freezer evaporator temperature, freezer set temperature, and defrost heater operating status.

[0020] Operational data can be collected in real time through the refrigerator's built-in temperature sensor and status detection unit, or historical operational data can be retrieved from the distributed database of the refrigerator's cloud-based fault diagnosis platform. The data sampling interval can be flexibly set according to the detection accuracy requirements, such as using a 1-minute sampling interval.

[0021] Understandably, the four types of operational data selected cover four dimensions: freezer environment status, refrigeration core component status, user-set parameters, and system operating conditions. This provides multi-dimensional data support for subsequent fault diagnosis, eliminating the need for additional dedicated door detection sensors. The detection function can be implemented using the refrigerator's existing hardware configuration, effectively reducing deployment and hardware modification costs. Furthermore, the combined use of multi-dimensional data avoids the limitations of relying on a single temperature indicator, reduces misjudgments caused by sensor malfunctions and transient interference, and improves the reliability of the detection results.

[0022] S200: Based on the working status of the defrosting heater, the operating data is divided into multiple non-defrosting cycles to eliminate the interference of the defrosting process on the detection.

[0023] Defrosting is an essential part of the operation of a frost-free refrigerator. When the defrosting heater is working, it releases a lot of heat, which causes the temperature of the freezer evaporator and the freezer compartment to rise abnormally. This kind of temperature fluctuation is generated by the system and is not related to the leakage of cold air caused by the door not being closed tightly. If the defrosting stage data is directly included in the detection range, it will inevitably lead to misdiagnosis of faults.

[0024] Based on the above issues, this step uses the working status of the defrosting heater as the basis for time division, dividing the continuous operating data into several independent non-defrosting cycles, and excluding the temperature data corresponding to the defrosting process from the effective detection range.

[0025] In one implementation, the time points of all defrosting events can be located by identifying the on / off status markers of the defrosting heater, and multiple non-defrosting cycles can be obtained by dividing the data around these defrosting events. It is understood that by dividing the data into non-defrosting cycles, temperature interference from the defrosting process can be completely isolated at the data level, ensuring that the data used in subsequent analysis corresponds to the normal cooling conditions of the refrigerator, thus improving the accuracy of fault detection from the data source. This step is the core foundation for ensuring the robustness of the detection solution, effectively adapting to the periodic defrosting operating characteristics of dual-system refrigerators and eliminating the negative impact of defrosting conditions on the detection results.

[0026] S300: For each of the aforementioned non-defrosting cycles, extract characteristic parameters used to characterize different stages of the fault.

[0027] The fault of a freezer door not being closed properly exhibits differentiated temperature characteristics at different stages of its development. The initial stage is characterized by a rapid temperature rise; the sustained stage by stable high-level temperature fluctuations; and the recovery stage by a rapid temperature drop. A single-dimensional feature parameter cannot simultaneously cover the diagnostic requirements of all stages. Based on these issues, this step extracts multi-dimensional feature parameters from the raw operating data for each independent non-defrost cycle. These feature parameters can include parameters related to temperature change amplitude, temperature difference, temperature trend slope, proportion of characteristic data, and extreme temperature values.

[0028] Feature parameter extraction can be achieved through trend fitting algorithms, proportion statistics algorithms, and extreme value solving algorithms. All sampling points within each non-defrosting cycle are processed to obtain the corresponding quantified feature values ​​for each dimension. It is understandable that by extracting multi-dimensional feature parameters, the original time-series temperature data can be transformed into quantified feature vectors that can be used for rule matching. This enables accurate quantification of the characteristics of different stages of a fault, providing a clear basis for subsequent staged rule matching and improving the accuracy and specificity of fault stage identification.

[0029] S400: Match the feature parameters according to the preset multi-stage fault detection rules. The multi-stage fault detection rules include at least the temperature rise stage rule for identifying the occurrence of the fault, the steady-state stage rule for identifying the continuation of the fault, and the recovery stage rule for identifying the end of the fault.

[0030] Most existing detection solutions employ only a single fault determination rule, which can only determine whether a door is not closed properly, but cannot distinguish the development stage of the fault, making it difficult to achieve complete monitoring of the fault's entire lifecycle, nor can it provide users with dynamic information on the fault's evolution. Based on these issues, this step pre-defines three independent detection rules, corresponding to the three core development stages of the fault. The extracted feature parameters are then matched and verified against each stage rule to determine the fault stage corresponding to the current non-defrosting cycle.

[0031] All three types of detection rules employ multi-condition composite judgment logic. Each rule contains several quantified judgment conditions, and only when all corresponding conditions are met is it determined to meet the characteristics of that stage. It is understandable that using phased detection rules for matching can fully cover the entire lifecycle of a fault, from its occurrence and persistence to its recovery, achieving accurate identification of the dynamic development process of the fault. This overcomes the shortcomings of existing solutions that can only determine the presence or absence of a fault but cannot distinguish its development stage, thus adapting to the fault diagnosis needs of complex operating scenarios in dual-system refrigerators.

[0032] S500: Based on the matching results, generate diagnostic results, which include fault status, fault stage, and time interval of fault occurrence.

[0033] If only a binary judgment result indicating whether a fault exists is output, it cannot support maintenance personnel in tracing the fault evolution process, nor is it conducive to users promptly grasping the progress and effectiveness of fault handling, thus limiting the practical value of the detection results. Based on the above issues, this step integrates the results obtained from rule matching to generate a complete diagnostic result. The fault status indicates whether a freezer door not being closed properly exists within the current cycle; the fault stage indicates the current development stage of the fault; and the fault occurrence time interval is used to locate the specific start and end time range corresponding to the fault.

[0034] The diagnostic results can be stored and transmitted using standardized data formats, facilitating integration with subsequent early warning systems, visualization systems, and operation and maintenance management systems. Therefore, this step can output comprehensive and clearly defined fault diagnosis results, providing end-users with timely fault warnings to reduce food loss and energy waste, and offering maintenance personnel traceable fault data to help quickly locate problems, analyze fault causes, and improve the efficiency and accuracy of fault handling.

[0035] The detection method provided in this embodiment isolates defrosting interference by dividing the non-defrosting cycle and covers the entire life cycle of the fault through multi-stage detection rules. It can effectively improve the detection accuracy of the fault of the freezer door not being closed tightly in a dual-system refrigerator, realize dynamic and full-process monitoring of the fault, and has strong environmental adaptability and working condition adaptability.

[0036] In some embodiments of this application, the specific division method of the non-defrosting cycle is further defined. In view of the problem that the temperature drop stage after defrosting may still interfere with the detection results, the rules for determining the cycle boundary are refined to further improve the purity of the effective detection data.

[0037] Based on the operating status of the defrosting heater, the step of dividing the operating data into multiple non-defrosting cycles includes: The time interval between two defrosting heater start times is taken as a complete cycle, and the portion of each complete cycle from the start time of the defrosting heater after a preset time to the start time of the next defrosting heater is determined as the non-defrosting cycle.

[0038] In one implementation, when dividing the cycle, the start-up times of all defrost heaters are first identified from the operating data. The complete time interval between the start-up times of two adjacent defrost heaters is defined as a complete operating cycle. Each complete operating cycle includes one complete defrosting process and the subsequent cooling operation. It is understood that using the start-up time of the defrost heaters as the cycle boundary ensures that the starting point of each complete cycle corresponds to the starting point of the defrosting process, maintaining a consistent benchmark for cycle division and avoiding deviations in cycle division due to fluctuations in defrosting duration.

[0039] For example, the system automatically triggers a diagnostic task at 2:00 AM daily, retrieving at least D hours (this duration can be preset) of raw operating data from each refrigerator. The system first performs data cleaning: based on the defrost heater activation signal. Identify all defrosting events .

[0040] Define the i-th non-defrosting cycle for: .

[0041] The above formula defines the "non-defrosting period"—that is, the effective analysis time period excluding interference from the defrosting process. In the formula, , These are the start times of the defrosting heater for the (i-1)th and i-th defrosting cycles, respectively. This is the threshold duration for removing items after the defrosting process begins.

[0042] Because the evaporator and freezer compartment temperatures rise abnormally for approximately 2 hours after the defrost heater starts, the starting point for the current effective analysis period is set 2 hours after the last defrost start time; the ending point is the next defrost start time (excluding that point). This ensures that subsequent analysis data only reflects temperature behavior under normal cooling conditions, avoiding misinterpretations.

[0043] For example, if a refrigerator starts defrosting at 08:00 on November 25 and 06:00 on November 27, the effective analysis period includes: Period 1: 10:00 on November 25 to 06:00 on November 27 (if there is no other defrosting during this period).

[0044] Furthermore, after the defrost heater is turned on, it continues to heat for a period of time. After heating stops, the temperature of the evaporator and freezer compartment does not immediately drop back to the normal cooling temperature, but requires a buffer period to gradually decrease to the stable cooling range. The temperature change during this buffer period is also not within the normal cooling operating condition. Including it in the detection range may still interfere with fault diagnosis. Based on the above problems, in this embodiment, a preset time is shifted backward from the moment the defrost heater is turned on each time. The shifted time is taken as the start time of the non-defrost cycle, and the moment the next defrost heater is turned on is taken as the end time of the non-defrost cycle. The time period between the start time and the end time is the final non-defrost cycle. It can be set so that the preset time can be flexibly adjusted according to the refrigerator's defrost power, evaporator volume, and cooling capacity. The adjustment standard is based on the complete elimination of the temperature impact caused by defrosting and the refrigerator entering a stable cooling state.

[0045] Specifically, in this embodiment, for each non-defrosting cycle Calculate the following two basic parameters: Low-temperature percentage in the freezer compartment: ; Low-temperature percentage of evaporators: ; In the formula, This represents the total number of sampling points within the i-th non-defrosting cycle (for example, if the cycle length is 24 hours and the sampling interval is 1 minute, then...). =1440); It is an indicator function: the function value is 1 when the condition in parentheses is true (such as the freezer temperature being below 0°C); otherwise, it is 0.

[0046] therefore, This indicates the percentage of time the freezer compartment temperature is below 0°C during the entire non-defrost cycle. If this percentage is too low, it indicates that the refrigerator may be experiencing a power outage, high ambient temperature, or compressor malfunction, and therefore lacks cooling capacity. In this case, the door not being closed properly should not be used for diagnosis.

[0047] Similarly, This indicates whether the evaporator is working properly—if the evaporator remains above 0°C for an extended period, the refrigeration system is malfunctioning, and the diagnostic results are unreliable.

[0048] This embodiment requires and Only within a certain range can the subsequent diagnostic process begin.

[0049] The cycle will only be included in subsequent diagnostics if both of the above conditions are met simultaneously; otherwise, it will be considered a non-cooling state and skipped.

[0050] Understandably, this classification method not only eliminates high-temperature data corresponding to the defrosting heater's operation but also allows for a buffer period of temperature drop after defrosting. This completely eliminates the impact of temperature fluctuations throughout the defrosting process, ensuring that all data outside the defrosting cycle corresponds to stable cooling conditions in the refrigerator. Compared to schemes that only eliminate the defrosting heating period, this classification method further reduces the probability of misjudgments caused by defrosting interference, providing a cleaner and more reliable data foundation for subsequent feature extraction and rule matching. Therefore, this embodiment, by refining the boundary determination rules for the non-defrosting cycle, can further improve the anti-interference capability and judgment accuracy of the detection scheme, and enhance its adaptability to the defrosting operation characteristics of dual-system refrigerators.

[0051] In some embodiments of this application, the execution logic of rule matching is further defined, and a differentiated detection process is adopted for different numbers of non-defrosting cycles to solve the problems of easy misjudgment in the recovery stage in single-cycle scenarios and easy conflict in stage judgment in multi-cycle scenarios, thereby improving the scenario adaptability of the detection logic.

[0052] The step of matching the feature parameters according to the preset multi-stage fault detection rules includes: The detection logic is dynamically adjusted based on the number of non-defrosting cycles. When the first non-defrost cycle is detected, only the matching of the temperature rise phase rule and the steady-state phase rule is performed; When a second or more non-defrosting cycles are detected, matching is performed sequentially according to the priority order of the temperature rise phase rule, the steady state phase rule, and the recovery phase rule.

[0053] In one implementation, before performing rule matching, the total number of non-defrosting cycles in the current data to be tested is first counted. This number of cycles is used as the basis for adjusting the detection logic, automatically adapting to the corresponding diagnostic process. It is understood that different numbers of non-defrosting cycles correspond to different data coverage ranges, and the dimensions for determining the fault stage that can be supported vary. Using a unified detection logic cannot simultaneously guarantee accuracy and comprehensiveness. Dynamic adjustment can achieve optimal matching between the detection logic and data conditions.

[0054] When the first non-defrost cycle is detected, meaning there is only a single non-defrost cycle, the fault is usually in the occurrence or persistence stage, not yet in the complete recovery process. Furthermore, the data length of a single cycle is insufficient to support a complete judgment in the recovery stage. Forcing rule matching for the recovery stage could easily lead to misjudgments. Based on these issues, in this scenario, only the rules for the temperature rise stage and the steady-state stage are matched; the matching of rules for the recovery stage is temporarily suspended. It is understandable that skipping the recovery stage detection in the first cycle avoids invalid and erroneous judgments due to insufficient data coverage, improves the reliability of diagnostic results in single-cycle scenarios, and reduces unnecessary computation, thus improving detection efficiency.

[0055] When a second or subsequent non-defrost cycle is detected, multiple consecutive cycles can cover the complete fault chain from occurrence and persistence to recovery, providing the data foundation for performing full-stage detection. Based on the above conditions, in this scenario, matching is performed sequentially according to the priority order of temperature rise stage rules, steady-state stage rules, and recovery stage rules. Furthermore, adopting a progressive priority matching order ensures that each cycle prioritizes matching the core fault stage. If the current cycle meets the rule of a higher priority stage, the judgment result of the corresponding stage is directly output, and the matching of rules for subsequent lower priority stages is not executed. It can be understood that the priority control mechanism can avoid the situation of matching multiple stages simultaneously in the same cycle, prevent conflicts in multi-stage judgment results, and ensure the uniqueness and determinism of the diagnostic result of each cycle.

[0056] Understandably, dynamically adjusted detection logic can adapt to both single-cycle and multi-cycle detection scenarios. In single-cycle scenarios, it prioritizes accuracy while considering comprehensiveness in multi-cycle scenarios, achieving a balance between detection precision and coverage. Therefore, this embodiment, by introducing dynamic detection logic driven by the number of cycles, effectively improves the adaptability of the detection scheme to different data scenarios. Simultaneously, it avoids decision conflicts through a priority mechanism, optimizing detection efficiency and result stability.

[0057] In some embodiments of this application, the specific judgment conditions of the temperature rise phase rule are further defined, and a multi-dimensional composite judgment logic is adopted to solve the problems of easy misjudgment and inability to accurately locate the fault initiation point of a single temperature rise threshold, thereby improving the accuracy of temperature rise phase detection.

[0058] The rules for the temperature rise phase include: If, within a non-defrosting cycle, there exists a moment in which the rise in the freezer compartment temperature exceeds a first threshold within a preset duration prior to that moment, and the average deviation between the freezer compartment temperature and the set freezer compartment temperature from the start of the non-defrosting cycle to that moment is less than a second threshold, and the difference between the freezer compartment temperature and the set freezer compartment temperature is greater than a third threshold and the difference between the freezer compartment temperature and the freezer evaporator temperature is greater than a fourth threshold at the end of the non-defrosting cycle, and the duration of this state exceeds a fifth threshold, then it is determined that the temperature rise phase has begun.

[0059] In one implementation, when determining the temperature rise phase, all sampling times within the non-defrosting cycle are iterated through, and each time point is checked to see if all determination conditions are met. Only when all conditions are met simultaneously is that time point determined as the starting point of the temperature rise phase. It is understandable that this multi-condition composite determination logic can verify the fault attributes of the temperature rise from multiple dimensions, effectively filter out various interference factors, and improve the accuracy of the determination.

[0060] Specifically, iterate through all time points t within the period and check if there exists a time that satisfies the following condition. : The conditions for a sudden temperature rise are: ; This formula is used to detect whether the freezer temperature rises sharply in a short period of time. In the formula, The first threshold is set, since the data sampling interval is 1 minute. Use 180 minutes (representing 3 hours). If the current freezer temperature is 6°C or more higher than 3 hours ago (i.e., ...), the freezer will be frozen for 3 hours. If the temperature is 6℃, it indicates that there may be a rapid leakage of cold air, typically caused by the door being opened or not closed properly.

[0061] Temperature difference stability before ascent for: ; This formula calculates starting from the period. When the temperature rises sharply Between, the actual temperature of the freezer compartment With user-set temperature (e.g.) The average absolute deviation of (=-18℃), where, This is the second threshold. If the average deviation is less than 3℃ (i.e. The temperature rise was 3℃, indicating that the refrigerator had been operating stably and was in normal cooling mode, thus ruling out the possibility of a temperature rise caused by a previous malfunction.

[0062] Relationship between end-of-cycle temperature difference and evaporator temperature: ; ; The above two inequalities are used to verify the rationality of the state after the temperature rise. In the formula, The third threshold, The fourth threshold is defined as follows: The former requires the freezer temperature to be significantly higher than the set value at the end of the cycle, indicating a continuous loss of cooling capacity; the latter requires the freezer temperature not to be much lower than the evaporator temperature, in order to rule out interference such as sensor malfunction or abnormal compressor shutdown.

[0063] Duration constraint: (like Take 60 minutes. (for the fifth threshold) After a sudden temperature rise, the abnormal state must persist for at least one hour to be confirmed as a genuine malfunction, rather than a momentary door opening disturbance.

[0064] If all the above conditions are met, the fault is determined to be in the "temperature rise phase," and the fault area is... .

[0065] Among them, the freezer temperature rise exceeding the first threshold within a preset time period. The conditions are used to characterize the rapid rise in freezer temperature over a short period of time, corresponding to the initial stage of rapid cold air leakage after the door is not closed tightly. In one embodiment, the preset duration and the first threshold are... The settings can be configured based on the refrigerator's freezer volume, cooling capacity, and temperature fluctuation range under normal usage conditions. This ensures that the condition can effectively distinguish between normal temperature fluctuations and rapid temperature rises caused by malfunctions. Understandably, this condition is a core characteristic of the temperature rise phase, enabling rapid identification of the time point of abnormal temperature increases.

[0066] The average deviation between the freezer temperature and the set temperature from the start of the non-defrost cycle to this point is less than the second threshold. The conditions are used to verify the refrigerator's operating status before the temperature rise. If the average deviation is within a small range, it indicates that the refrigerator was in a stable and normal cooling state before the temperature rise occurred, and the temperature rise was a sudden abnormal situation, rather than a continuous evolution of a pre-existing fault. Understandably, this condition can rule out scenarios where a fault already existed at the initial stage of the cycle, ensuring that the identified temperature rise point corresponds to the true start time of the fault, achieving accurate location of the fault occurrence point, and avoiding misjudging the intermediate temperature rise process of a continuous fault as the fault initiation stage.

[0067] At the end of the non-defrost cycle, the difference between the freezer compartment temperature and the set temperature exceeds the third threshold. The temperature difference between the freezer compartment and the evaporator exceeds the fourth threshold. The conditions are used to verify the rationality of the operating conditions after the temperature rise. The former verifies that the freezer compartment remains at a sustained high temperature after the temperature rise, and that cold air leakage is ongoing; the latter verifies that the temperature relationship conforms to the operating condition of a door not being closed tightly, ruling out other faults such as sensor malfunction or abnormal compressor shutdown that could cause the temperature rise. Understandably, introducing the refrigeration evaporator temperature for joint verification can distinguish between two different types of faults: faults within the refrigeration system itself and cold air leakage from the door, further reducing the probability of false positives and improving the specificity of fault diagnosis.

[0068] The duration of the state exceeds the fifth threshold. The conditions are designed to filter temperature fluctuations caused by short-term disturbances. Normal user opening of the freezer door to retrieve food can cause a brief temperature rise, but this rise is short-lived and does not constitute a door not being closed properly. Based on this, a duration constraint is set; only when the temperature condition is met for a certain duration is the condition considered faulty. Understandably, this duration constraint effectively eliminates interference from momentary door opening operations, ensuring the judgment corresponds to continuous cold air leakage caused by a door not being closed properly, thus reducing false alarms.

[0069] Therefore, this embodiment uses a composite judgment rule based on four dimensions: temperature rise amplitude, early stability, final state, and duration. This allows for accurate identification of the initial stage of a fault, effective filtering of normal operation interference and other fault interference, significantly improving the accuracy of temperature rise detection, and enabling precise location of the fault's initiation moment.

[0070] In some embodiments of this application, the specific judgment conditions of the steady-state stage rules are further defined, and two sets of composite judgment conditions are set to adapt to different working conditions with different degrees of cold leakage, so as to solve the problem that the coverage of a single judgment condition is limited and easy to miss, and improve the comprehensiveness of steady-state stage detection.

[0071] The steady-state phase rule includes: within a non-defrosting cycle, if a preset first condition group or second condition group is met, and the duration of meeting the conditions exceeds a sixth threshold, then it is determined to be a steady-state phase. The first condition group includes: The percentage of data where the difference between the freezer temperature and the freezer set temperature is greater than the seventh threshold is greater than the first percentage threshold. The temperature of the freezer compartment first decreases and then increases within the cycle; The temperature of the refrigeration evaporator generally showed a downward trend; The percentage of data where the difference between the set temperature of the freezer compartment and the temperature of the freezer evaporator is less than the eighth threshold is greater than the second percentage threshold. The minimum temperature of the freezer compartment is within a first temperature range; The second set of conditions includes: The percentage of data where the difference between the freezer temperature and the freezer set temperature is greater than the seventh threshold is greater than the first percentage threshold. The temperature of the freezer compartment first decreases and then increases within the cycle; The slope of the temperature decrease trend in the freeze evaporator is within a preset slope range; The percentage of data where the difference between the set temperature of the freezer compartment and the temperature of the freezer evaporator is less than the eighth threshold is greater than the second percentage threshold. The minimum temperature of the freezer compartment is less than or equal to the ninth threshold.

[0072] In one implementation, the steady-state phase is the continuous phase of the door not being closed properly. During this phase, the compressor continues to run and cool, but the amount of cold leakage and the cooling capacity reach a relative balance, and the freezer temperature is maintained at a level higher than the set value. Due to differences in factors such as ambient temperature, door gap size, and food loading, the temperature performance during the steady-state phase varies considerably. A single judgment condition cannot cover all operating conditions. Therefore, two independent sets of conditions are set up to adapt to different degrees of cold leakage scenarios.

[0073] Both sets of conditions have a duration constraint as a prerequisite. Only when the duration of the state in the corresponding condition set exceeds the sixth threshold will the subsequent condition set be evaluated. Understandably, the duration constraint can filter out false judgments caused by short-term temperature fluctuations, ensuring that the judgment result corresponds to a continuous fault steady state and improving the reliability of the steady-state stage judgment.

[0074] Specifically, calculate the following key parameters: 1. Percentage of sustained high temperatures: ; 2. Set the percentage of abnormal temperature differences in the evaporator: ; This formula determines whether the evaporator temperature is frequently higher than the set temperature (i.e., ...). Under normal circumstances, the evaporator should be colder than the freezer compartment. If its temperature is higher than the set value, it indicates a serious drop in cooling efficiency. This is commonly seen when the door is not closed tightly, causing the compressor to run at a high load continuously but failing to cool the room.

[0075] 3. Evaporator cooling trend slope (only for the second condition group): If the slope is within a certain threshold range, it indicates that the evaporator is still providing weak cooling, but the efficiency is extremely low. This corresponds to a special operating condition where the freezer temperature is extremely low, but the set value still cannot be maintained.

[0076] 4. Lowest temperature in the freezer compartment: ; This formula determines the lowest temperature reached by the freezer compartment during the cycle. Based on this value, the system distinguishes between two types of steady-state scenarios: a) If Applicable to the first condition group (medium-high temperature steady state); b) If The second condition group (low temperature steady state) applies.

[0077] This distinction can cover fault performance under different ambient temperatures or door opening degrees.

[0078] If any set of composite conditions is met and the duration is greater than or equal to a certain time threshold, it is determined to be a "steady-state phase fault," and the fault area is the entire cycle. .

[0079] As can be seen from the above, the first condition group is suitable for medium-high temperature steady-state scenarios, corresponding to conditions where the door is not closed tightly and the amount of cold air leakage is large. The condition that the percentage of data where the difference between the freezer temperature and the set temperature is greater than the seventh threshold is greater than the first percentage threshold is used to measure the persistence of the high-temperature state in the freezer. If the percentage meets the standard, it indicates that the freezer is in a state above the set temperature for most of the period, which meets the core characteristic of a persistent fault. The condition that the freezer temperature first decreases and then increases within the period is used to characterize the temporal evolution of the fault. This trend corresponds to the process where the refrigerator first cools normally and reaches the minimum temperature, and then the temperature gradually rises due to cold air leakage, which is a typical temporal manifestation of a steady-state fault. It can be understood that this trend characteristic can distinguish the persistent high temperature caused by the door not being closed tightly from the static high temperature caused by other faults, improving the specificity of the judgment.

[0080] In the first condition group, the condition that the evaporator temperature generally shows a downward trend is used to verify the working status of the refrigeration system. If the evaporator temperature generally decreases, it indicates that the compressor is running continuously, the refrigeration system itself is not faulty, and the high temperature is caused by cold air leakage. The condition that the percentage of data where the difference between the freezer compartment set temperature and the evaporator temperature is less than the eighth threshold is greater than the second percentage threshold is used to further verify the evaporator's cooling capacity. If the temperature difference percentage meets the standard, it indicates that the evaporator temperature is close to the set temperature, and the cooling capacity is within the normal range. The condition that the minimum freezer compartment temperature is within the first temperature range is used to define the medium-high temperature steady-state temperature range, corresponding to the condition of severe cold leakage and the freezer compartment being unable to drop to a lower temperature.

[0081] The second condition group is adapted for low-temperature steady-state scenarios, corresponding to conditions where the door is not fully closed and the amount of cold air leakage is small. In this condition group, the three conditions—the proportion of high temperature in the freezer compartment, the temperature trend of the freezer compartment, and the proportion of the temperature difference between the set temperature and the evaporator—are consistent with the first condition group, ensuring the consistency of the core judgment criteria in both scenarios. Based on this, the overall downward trend of the evaporator is replaced with a slope check of the evaporator's downward trend. If the slope of the evaporator's temperature downward trend is within the preset slope range, it indicates that the evaporator is still continuously cooling, but at a low rate, corresponding to a condition where the amount of cold air leakage is small and the cooling capacity and leakage are slightly balanced. Simultaneously, the condition that the minimum freezer compartment temperature is less than or equal to the ninth threshold is used to define the temperature range of the low-temperature steady state, corresponding to a condition where the degree of cold air leakage is small and the freezer compartment can still drop to a low temperature but cannot reach the set temperature.

[0082] It is understandable that the two sets of conditions are logically ORed; satisfying either set of conditions for a specified duration is sufficient to determine a steady-state phase. This setting significantly expands the detection coverage of steady-state faults while ensuring accuracy, adapting to fault behavior under different ambient temperatures and varying degrees of cold leakage, and effectively reducing missed detections. Therefore, this embodiment, through a composite steady-state determination rule with dual condition sets, can accurately identify the fault duration phase, cover multiple types of steady-state fault conditions, and significantly improve the accuracy and environmental adaptability of steady-state phase detection.

[0083] In some embodiments of this application, the specific judgment conditions of the recovery phase rules are further defined, and a multi-dimensional composite judgment logic is used to identify the fault recovery process, which solves the problems that existing solutions cannot determine the fault recovery status and the detection life cycle is incomplete, and realizes closed-loop monitoring of the entire fault process.

[0084] The recovery phase rules include: If, within a non-defrost cycle, there exists a moment in which the temperature drop in the freezer compartment exceeds the tenth threshold within a preset duration prior to that moment, and the overall temperature of the freezer compartment shows a downward trend during the non-defrost cycle, and the duration of this state exceeds the eleventh threshold, and at the end of the non-defrost cycle there exists a moment in which the difference between the temperature of the freezer compartment at that moment and the set temperature of the freezer compartment is less than or equal to the twelfth threshold, then it is determined to be the recovery phase.

[0085] In one implementation, the recovery phase corresponds to the process where, after the door is closed, cold air leakage stops, and the refrigeration system reduces the freezer compartment temperature from the abnormally high temperature back to the normal range. The recovery phase rule is only activated in the second and subsequent non-defrost cycles to ensure sufficient data length to support a complete determination of the recovery process. The recovery phase employs multi-condition composite judgment logic, verifying the results from four dimensions: temperature drop magnitude, overall trend, duration, and final state, ensuring the accuracy of the judgment results.

[0086] Check if time exists satisfy: Rapid cooling rate: ; This formula tests whether the freezer compartment cools down rapidly within 3 hours (eleventh threshold). (Tenth threshold) indicates that the user has closed the door and the system is recovering from the abnormal high temperature state.

[0087] Final temperature difference requirement: , ; This condition requires that there exists a certain moment at the end of the period. The temperature deviation between its freezer compartment and the set temperature shall not exceed [the specified value]. (The twelfth threshold) indicates that the system has basically recovered its normal cooling capacity.

[0088] If the following conditions are met simultaneously: temperature drop magnitude, overall downward trend, duration greater than or equal to a certain threshold hour, and final temperature difference requirement, then it is determined to be a "recovery phase fault," and the fault area is... .

[0089] Among them, the freezer temperature drop exceeding the tenth threshold within the preset time period. The conditions are used to characterize the rapid temperature drop in the freezer compartment, corresponding to the initial recovery performance of the refrigeration system after the door is closed, where cold air leakage ceases and the refrigeration system cools down rapidly. In one embodiment, a preset time (e.g., 3 hours) and a tenth threshold are used. The settings can be configured based on the refrigerator's cooling capacity to ensure that this condition effectively distinguishes between normal cooling fluctuations and rapid temperature recovery after a malfunction. Understandably, this condition is a core characteristic of the recovery phase, enabling rapid identification of the point at which the temperature begins to drop quickly.

[0090] The condition that the freezer compartment temperature shows an overall downward trend during the non-defrost cycle is used to verify the temperature trend throughout the entire cycle. If there is an overall downward trend, it indicates that the cycle is in a continuous recovery process, rather than a localized cooling caused by short-term temperature fluctuations. It can be understood that the overall trend constraint can exclude small temperature fluctuations in normal refrigeration cycles, ensuring that the judgment result corresponds to a complete fault recovery process, rather than temperature adjustment under normal refrigeration conditions.

[0091] The condition that the duration of a state exceeds the eleventh threshold is used to filter out short-term cooling interference. Normal adjustments to the refrigeration system may also cause short-term temperature drops, but these drops are limited in duration and do not constitute a fault recovery process. Based on this, a duration constraint is set; only when a state meeting the cooling condition persists for a certain duration is it considered part of the recovery phase. Understandably, the duration constraint effectively eliminates interference from normal refrigeration adjustments, ensuring that the judgment result corresponds to a continuous recovery process after a fault, thus reducing false alarms.

[0092] At the end of the non-defrost cycle, there exists a moment where the difference between the freezer temperature and the set temperature is less than or equal to the twelfth threshold. The conditions are used to verify the final effect of the recovery process. If the temperature at the end of the cycle can return to the normal range near the set temperature, it indicates that the cooling status of the freezer compartment has basically returned to normal after the recovery process, and the fault has indeed been eliminated. Understandably, the final temperature condition can verify the effectiveness of the recovery process, avoid misjudging the incomplete cooling process as the recovery stage, and ensure the rigor of the recovery stage determination.

[0093] Finally, after completing the single-cycle phase matching, the final result stored in the fault interval needs to be confirmed according to the cross-cycle association rules. The result includes the following situations: 1. The non-defrosting cycle is a single cycle: it only detects faults in the steady-state stage, and the faults are stored in the fault range.

[0094] 2. The number of non-defrosting cycles is greater than or equal to k cycles: 1) First cycle: First check for temperature rise faults, and only mark the faults; if there are no faults, check for steady state, and store the faults in the interval.

[0095] 2) Subsequent cycle: full-stage detection, only steady-state single faults are stored in the interval; temperature rise marking status, recovery marking initialization.

[0096] 3. Association determination: 1) A temperature rise fault can only coexist with a subsequent steady-state fault if it is a fault that is not present.

[0097] 2) A steady-state fault requires a subsequent steady-state or recovery fault, or a steady-state or recovery fault occurs after an interval without a steady-state fault.

[0098] 3) Fault recovery is not output separately; it must be associated with a temperature rise or steady-state fault.

[0099] Therefore, this embodiment uses a composite judgment rule based on four dimensions: cooling range, overall trend, duration, and final state, to accurately identify the fault recovery process and achieve closed-loop monitoring of the entire lifecycle of the fault from occurrence to recovery. This allows users to know the fault handling effect in a timely manner and provides maintenance personnel with complete fault evolution data, significantly improving the completeness and practical value of the detection solution.

[0100] Another aspect of this application provides a dual-system refrigerator freezer door not fully closed detection system, used to implement the dual-system refrigerator freezer door not fully closed detection method described in the foregoing embodiments. Addressing the problems of high hardware cost, limited functionality, and inability to achieve phased fault identification in existing detection systems, this system adopts a modular architecture design, achieving multi-stage accurate fault detection through the coordinated operation of various modules. (Refer to...) Figure 2 As shown, the system includes: The data acquisition module is used to acquire the refrigerator's operating data, which includes at least the freezer temperature, the freezer evaporator temperature, the freezer set temperature, and the defrost heater's working status. A data preprocessing module, connected to the data acquisition module, is used to receive the operating data and, based on the working status of the defrosting heater, divide the operating data into multiple non-defrosting cycles to eliminate interference from the defrosting process on the detection. The feature extraction module, connected to the data preprocessing module, is used to receive the non-defrosting cycle and extract feature parameters from it to characterize different stages of the fault. The rule matching module, connected to the feature extraction module, is used to receive the feature parameters and match the feature parameters according to the preset multi-stage fault detection rules. The multi-stage fault detection rules include at least the temperature rise stage rule for identifying the occurrence of the fault, the steady state stage rule for identifying the continuation of the fault, and the recovery stage rule for identifying the end of the fault. The result output module, connected to the rule matching module, is used to receive the matching results and generate diagnostic results, which include the fault status, fault stage, and time interval of the fault occurrence.

[0101] In this embodiment, the system adopts a layered modular architecture, with each module connected through standardized data interfaces. Data flow is clear, logical connections are tight, and the system possesses good scalability and maintainability. The system can be deployed on the refrigerator's local main control board or on a cloud-based big data platform, adapting to different application scenarios.

[0102] The data acquisition module, serving as the system's data input port, is responsible for acquiring various parameter data during the refrigerator's operation. In one implementation, the data acquisition module can be configured to connect to the refrigerator's main control board via a communication bus to collect operational data detected by the refrigerator's built-in sensors in real time; alternatively, it can be configured to connect to a distributed database on a cloud-based big data platform to retrieve historical operational data in batches. It is understood that the four types of operational data acquired by the data acquisition module cover four dimensions: freezer environment, core refrigeration components, user settings, and system operating conditions. This provides sufficient raw data support for subsequent processing, and all data originates from the refrigerator's existing hardware, eliminating the need for additional dedicated door detection sensors, effectively reducing system hardware costs and deployment complexity.

[0103] The data preprocessing module communicates with the data acquisition module. After receiving the raw operating data transmitted by the data acquisition module, it performs cycle division and data cleaning. Using the defrosting heater's operating status as the time base, the data preprocessing module divides the continuous operating data into multiple independent non-defrosting cycles, removing interfering data corresponding to the defrosting process. It can be understood that the data preprocessing module is a fundamental step in ensuring the system's detection accuracy. By isolating the temperature interference during the defrosting process, it provides clean and effective data for subsequent feature extraction and rule matching, reducing the risk of misjudgment from the source and improving the system's anti-interference capability.

[0104] The feature extraction module communicates with the data preprocessing module. After receiving the divided non-defrosting cycle data, it performs feature extraction operations for each non-defrosting cycle. The feature extraction module incorporates multiple algorithm units, including trend fitting, proportion statistics, extreme value solving, and slope calculation, enabling it to extract multi-dimensional quantitative feature parameters from the raw time-series temperature data. Essentially, the feature extraction module transforms the raw data into vectorized features, serving as a core intermediate link between data preprocessing and fault determination. It provides clear quantitative criteria for subsequent rule matching, ensuring the accuracy of fault stage identification.

[0105] The rule matching module communicates with the feature extraction module. After receiving feature parameters, it calls the built-in multi-stage fault detection rule library for matching and verification. The rule matching module has three independent detection rules, corresponding to the temperature rise stage, steady-state stage, and recovery stage, respectively, enabling intelligent identification of different stages of fault development. In one implementation, the rule matching module also has built-in priority control logic, which can execute rule matching according to a preset priority order to avoid conflicts in multi-stage judgment results. It can be understood that the rule matching module is the core decision-making unit of the system. By outputting the fault stage judgment result through multi-stage rule matching operations, it determines the system's detection accuracy and diagnostic capability.

[0106] The results output module communicates with the rule matching module, receives the matching results from the rule matching module, and integrates them to generate standardized diagnostic results. The diagnostic results include three core types of information: fault status, fault stage, and the time interval of the fault occurrence, comprehensively reflecting the relevant attributes of the fault. In one implementation, the results output module can interface with external user early warning systems and operation and maintenance management systems, synchronizing the diagnostic results to the corresponding systems to support subsequent early warning and operation and maintenance operations. Therefore, as the system's output port, the results output module realizes the standardized output and external interaction of diagnostic results, and is a key link in the implementation of system functions.

[0107] The detection system provided in this embodiment adopts a modular and layered architecture, with clear responsibilities for each module and smooth data transmission. It can stably realize multi-stage detection of the fault of the freezer door not being closed properly in a dual-system refrigerator, and has the advantages of low deployment cost, high detection accuracy and strong adaptability to different scenarios.

[0108] In some embodiments of this application, the data preprocessing module is further configured to: take the time period between two defrosting heater start times as a complete cycle, and determine the portion of each complete cycle between the time period after the defrosting heater start time shifted backward by a preset time and the next defrosting heater start time as the non-defrosting cycle.

[0109] Specifically, the data preprocessing module has a built-in cycle division unit. When performing cycle division, the cycle division unit first identifies the start-up time nodes of all defrosting heaters from the raw operating data, and defines the complete time period between two adjacent start-up times as a complete operating cycle. It can be understood that using the start-up time of the defrosting heater as the benchmark node for cycle division can ensure the uniformity of the cycle boundary, unaffected by fluctuations in the duration of defrosting, and improve the stability of cycle division.

[0110] Furthermore, after the defrosting process is complete, the temperatures of the evaporator and freezer compartment need a period of time to drop from their high temperatures back to normal cooling temperatures. This temperature change during this drop-off phase is not considered normal cooling operation and may still interfere with subsequent fault diagnosis. Based on these issues, the cycle division unit shifts a preset time forward from the moment the defrost heater is turned on each time. The shifted time is taken as the start time of the non-defrost cycle, and the next time the defrost heater turns on is taken as the end time of the non-defrost cycle, ultimately obtaining a complete non-defrost cycle range. This preset time can be flexibly configured according to the specific refrigerator model and defrosting parameters, with the configuration standard being that the temperature impact of defrosting is completely eliminated and the refrigerator enters a stable cooling state.

[0111] By employing the aforementioned refined data segmentation method, the data preprocessing module not only eliminates high-temperature data during the defrosting heater's operation but also removes interfering data from the post-defrosting temperature drop buffer phase, ensuring that all output non-defrosting cycle data corresponds to the refrigerator's stable cooling conditions. Compared to simply eliminating defrosting heating periods, this approach further reduces the probability of misjudgments caused by defrosting interference, providing a more reliable and effective data foundation for subsequent modules. Therefore, this embodiment, by refining the cycle segmentation logic of the data preprocessing module, effectively enhances the system's anti-interference capability and further guarantees the accuracy of the detection results.

[0112] In some embodiments of this application, the rule matching module is further configured to: dynamically adjust the detection logic according to the number of non-defrosting cycles; when the first non-defrosting cycle is detected, only the matching of the temperature rise phase rule and the steady-state phase rule is performed; when the second or more non-defrosting cycles are detected, the matching is performed sequentially according to the priority order of the temperature rise phase rule, the steady-state phase rule, and the recovery phase rule.

[0113] The rule matching module has a built-in diagnostic logic unit. Before executing rule matching, the diagnostic logic unit first counts the total number of non-defrosting cycles to be detected and automatically selects the corresponding detection strategy based on the number of cycles. It is understandable that different numbers of non-defrosting cycles correspond to different data coverage capabilities, and using dynamically adaptable detection logic can achieve an optimal balance between detection accuracy and detection range.

[0114] When only the first non-defrost cycle exists, the data from a single cycle cannot cover the complete fault recovery process, making it highly susceptible to misjudgments if recovery phase detection is performed. Based on this issue, the diagnostic logic unit's control rule matching module only calls the temperature rise phase rules and steady-state phase rules for matching, temporarily disabling recovery phase rules. Understandably, skipping recovery phase detection in a single-cycle scenario avoids invalid and erroneous judgments, improves the reliability of the first cycle's diagnostic results, reduces unnecessary computational overhead, and improves detection efficiency.

[0115] When there is a second or subsequent non-defrosting cycle, continuous multi-cycle data can cover the entire fault chain from occurrence to recovery, providing a data foundation for full-stage detection. Based on the above conditions, the diagnostic logic unit control rule matching module executes matching sequentially according to the priority order of temperature rise stage rules, steady-state stage rules, and recovery stage rules. Furthermore, the priority order setting ensures that the core fault stage is matched first in each cycle. If the judgment conditions of a high-priority stage are met, the corresponding result is directly output, and the matching of low-priority stages is not executed. It can be understood that the priority control mechanism can avoid matching multiple fault stages simultaneously in the same cycle, prevent conflicting judgment results, and ensure the uniqueness and determinism of the diagnostic results.

[0116] In some embodiments of this application, the result output module is further configured to: mark and display the time interval of the fault occurrence in the diagnostic result on the visualization curve of the running data with different colors.

[0117] In one implementation, the result output module has a built-in visualization unit. After generating the diagnostic results, the visualization unit retrieves the operating data for the corresponding time period and generates time-series curves for parameters such as freezer temperature and evaporator temperature. Simultaneously, based on the fault stage type and fault time interval in the diagnostic results, fault area markers are overlaid at the corresponding positions on the time-series curves. Different fault stages are distinguished by different colors to ensure that the viewer can intuitively identify the different stages of fault development.

[0118] In one implementation, the markers can be in the form of a semi-transparent rectangular overlay, superimposed on the temperature curve. This clearly identifies the range of the fault area without obscuring the details of the original temperature curve, facilitating the viewer's analysis of fault characteristics in conjunction with the temperature curve trend. Understandably, using color to distinguish different fault stages allows maintenance personnel to quickly grasp the evolution process and time distribution of the fault, eliminating the need to manually verify text data and significantly improving the efficiency of fault tracing and analysis.

[0119] Furthermore, the visualization unit also supports interactive operations. Clicking on a marked fault area will pop up a details window, displaying key parameters, diagnostic basis, and other detailed information for the corresponding fault stage. Therefore, by adding a visualization marking function to the result output module, this embodiment can significantly improve the intuitiveness and readability of diagnostic results, helping maintenance personnel quickly locate faults, analyze the fault evolution process, and improve the efficiency of fault handling and maintenance management.

[0120] 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.

[0121] 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 an open freezer door in a dual-system refrigerator, characterized in that, Includes the following steps: Acquire the refrigerator's operating data, which includes at least the freezer temperature, freezer evaporator temperature, freezer set temperature, and defrost heater operating status; Based on the operating status of the defrosting heater, the operating data is divided into multiple non-defrosting cycles to eliminate interference from the defrosting process on the detection. For each of the non-defrosting cycles, feature parameters are extracted to characterize different stages of the fault; According to the preset multi-stage fault detection rules, the feature parameters are matched. The multi-stage fault detection rules include at least the temperature rise stage rule for identifying the occurrence of the fault, the steady state stage rule for identifying the continuation of the fault, and the recovery stage rule for identifying the end of the fault. Based on the matching results, a diagnostic result is generated, which includes the fault status, fault stage, and time interval of the fault occurrence.

2. The method for detecting an open freezer door in a dual-system refrigerator according to claim 1, characterized in that, The step of dividing the operating data into multiple non-defrosting cycles based on the operating status of the defrosting heater includes: The time interval between two defrosting heater start times is taken as a complete cycle, and the portion of each complete cycle from the start time of the defrosting heater after a preset time to the start time of the next defrosting heater is determined as the non-defrosting cycle.

3. The method for detecting an open freezer door in a dual-system refrigerator according to claim 1, characterized in that, The step of matching the feature parameters according to the preset multi-stage fault detection rules includes: The detection logic is dynamically adjusted based on the number of non-defrosting cycles. When the first non-defrost cycle is detected, only the matching of the temperature rise phase rule and the steady-state phase rule is performed; When a second or more non-defrosting cycles are detected, matching is performed sequentially according to the priority order of the temperature rise phase rule, the steady state phase rule, and the recovery phase rule.

4. The method for detecting an open freezer door in a dual-system refrigerator according to claim 1 or 3, characterized in that, The rules for the temperature rise phase include: If, within a non-defrosting cycle, there exists a moment in which the rise in the freezer compartment temperature exceeds a first threshold within a preset duration prior to that moment, and the average deviation between the freezer compartment temperature and the set freezer compartment temperature from the start of the non-defrosting cycle to that moment is less than a second threshold, and the difference between the freezer compartment temperature and the set freezer compartment temperature is greater than a third threshold and the difference between the freezer compartment temperature and the freezer evaporator temperature is greater than a fourth threshold at the end of the non-defrosting cycle, and the duration of this state exceeds a fifth threshold, then it is determined that the temperature rise phase has begun.

5. The method for detecting an open freezer door in a dual-system refrigerator according to claim 1 or 3, characterized in that, The steady-state phase rule includes: within a non-defrosting cycle, if a preset first condition group or second condition group is met, and the duration of meeting the conditions exceeds a sixth threshold, then it is determined to be a steady-state phase. The first condition group includes: The percentage of data where the difference between the freezer temperature and the freezer set temperature is greater than the seventh threshold is greater than the first percentage threshold. The temperature of the freezer compartment first decreases and then increases within the cycle; The temperature of the refrigeration evaporator generally showed a downward trend; The percentage of data where the difference between the set temperature of the freezer compartment and the temperature of the freezer evaporator is less than the eighth threshold is greater than the second percentage threshold. The minimum temperature of the freezer compartment is within a first temperature range; The second condition group includes: the proportion of data where the difference between the freezer temperature and the freezer set temperature is greater than the seventh threshold is greater than the first proportion threshold; The temperature of the freezer compartment first decreases and then increases within the cycle; The slope of the temperature decrease trend in the freeze evaporator is within a preset slope range; The percentage of data where the difference between the set temperature of the freezer compartment and the temperature of the freezer evaporator is less than the eighth threshold is greater than the second percentage threshold. The minimum temperature of the freezer compartment is less than or equal to the ninth threshold.

6. The method for detecting an open freezer door in a dual-system refrigerator according to claim 1 or 3, characterized in that, The recovery phase rules include: If, within a non-defrost cycle, there exists a moment in which the temperature drop in the freezer compartment exceeds the tenth threshold within a preset duration prior to that moment, and the overall temperature of the freezer compartment shows a downward trend during the non-defrost cycle, and the duration of this state exceeds the eleventh threshold, and at the end of the non-defrost cycle there exists a moment in which the difference between the temperature of the freezer compartment at that moment and the set temperature of the freezer compartment is less than or equal to the twelfth threshold, then it is determined to be the recovery phase.

7. A detection system for an open freezer door in a dual-system refrigerator, used to implement the detection method for an open freezer door in a dual-system refrigerator as described in any one of claims 1-6, characterized in that, include: The data acquisition module is used to acquire the refrigerator's operating data, which includes at least the freezer temperature, the freezer evaporator temperature, the freezer set temperature, and the defrost heater's working status. A data preprocessing module, connected to the data acquisition module, is used to receive the operating data and, based on the working status of the defrosting heater, divide the operating data into multiple non-defrosting cycles to eliminate interference from the defrosting process on the detection. The feature extraction module, connected to the data preprocessing module, is used to receive the non-defrosting cycle and extract feature parameters from it to characterize different stages of the fault. The rule matching module, connected to the feature extraction module, is used to receive the feature parameters and match the feature parameters according to the preset multi-stage fault detection rules. The multi-stage fault detection rules include at least the temperature rise stage rule for identifying the occurrence of the fault, the steady state stage rule for identifying the continuation of the fault, and the recovery stage rule for identifying the end of the fault. The result output module, connected to the rule matching module, is used to receive the matching results and generate diagnostic results, which include the fault status, fault stage, and time interval of the fault occurrence.

8. The dual-system refrigerator freezer door not fully closed detection system according to claim 7, characterized in that, The data preprocessing module is further configured to: take the time interval between two defrosting heater start times as a complete cycle, and determine the portion of each complete cycle between the start time of the defrosting heater and the start time of the next defrosting heater as the non-defrosting cycle.

9. The dual-system refrigerator freezer door not fully closed detection system according to claim 7, characterized in that, The rule matching module is further configured to: dynamically adjust the detection logic according to the number of non-defrosting cycles; when the first non-defrosting cycle is detected, only the matching of the temperature rise phase rule and the steady-state phase rule is performed; when the second or more non-defrosting cycles are detected, the matching is performed sequentially according to the priority order of the temperature rise phase rule, the steady-state phase rule, and the recovery phase rule.

10. The dual-system refrigerator freezer door not fully closed detection system according to claim 7, characterized in that, The result output module is also used to mark and display the time interval of the fault occurrence in the diagnostic results on the visualization curve of the running data with different colors.

Images