Refrigerator fault diagnosis methods, devices, media, and equipment

Abstract

This invention relates to a refrigerator fault calibration method, apparatus, medium, and device. The method includes: acquiring data sequences of multiple state parameters uploaded by multiple refrigerators; plotting a curve showing the change of each state parameter over time based on the data sequence uploaded by each refrigerator; providing the curves corresponding to each state parameter of a refrigerator to experts so that experts can identify faults and abnormal curve ranges in the curves; extracting abnormal curve segments from the curves; grouping the abnormal curve segments corresponding to the multiple state parameters of multiple refrigerators; and clustering the abnormal curve segments in each group to obtain the calibration fault curves for the fault and state parameter corresponding to that group. This invention digitizes the experience of experts in refrigerator fault diagnosis, which will improve the efficiency of computer-based fault diagnosis in the future.

Description

Technical Field

[0001] This invention relates to the field of fault handling technology, and in particular to a method, apparatus, medium, and equipment for calibrating refrigerator faults. Background Technology

[0002] With the development of society and the economy and the improvement of people's living standards, refrigerators have become a necessity in every household. Refrigerators typically connect to a network and report their status data, such as the refrigeration temperature. Generally, the status data reported by the refrigerator is within the normal range. However, in real-world environments, refrigerators may encounter various malfunctions, which can be accompanied by abnormal status data. Therefore, we can determine whether a refrigerator has malfunctioned based on abnormalities or fluctuations in the status data.

[0003] To improve the accuracy of fault diagnosis, it is necessary to provide some fault calibration data and compare it with the status data uploaded by the refrigerator to determine whether the refrigerator has malfunctioned or what kind of malfunction has occurred. Therefore, it is necessary to provide a fault calibration scheme. Summary of the Invention

[0004] In order to solve the above-mentioned technical problems, or at least partially solve the above-mentioned technical problems, the present invention provides a refrigerator fault calibration method, apparatus, medium, and equipment.

[0005] In a first aspect, embodiments of the present invention provide a refrigerator fault calibration method, comprising:

[0006] Acquire the collection data sequences of multiple status parameters uploaded by multiple refrigerators, wherein the multiple refrigerators are refrigerators that have reported faults and require repair;

[0007] Based on the data sequence of each state parameter uploaded by each refrigerator, a curve showing the change of that state parameter over time is plotted, and the curves corresponding to each state parameter of a refrigerator are provided to experts so that experts can identify the fault and abnormal curve range in the curves.

[0008] Based on the abnormal range of each curve graph, extract the abnormal curve segments of each curve from the curve graph.

[0009] Based on the state parameters and the faults marked in the curves, the abnormal curve segments of each curve corresponding to the multiple state parameters of the multiple refrigerators are grouped to obtain multiple groups, and each group corresponds to a fault and a state parameter.

[0010] Each abnormal curve segment in each group is clustered to form an abnormal change trajectory, which is then used as the calibration fault curve for the fault and status parameters corresponding to that group.

[0011] Secondly, embodiments of the present invention provide a refrigerator fault calibration device, comprising:

[0012] The data acquisition module is used to acquire the collected data sequences of multiple status parameters uploaded by multiple refrigerators, wherein the multiple refrigerators are refrigerators that have reported faults and are undergoing repairs.

[0013] The curve plotting module is used to plot the curve of the state parameter of each refrigerator changing over time based on the collected data sequence of each state parameter uploaded by each refrigerator, and to provide the curve graph of each state parameter of a refrigerator to experts so that experts can identify the fault and abnormal curve range in the curve graph.

[0014] The anomaly extraction module is used to extract the abnormal curve segments from each curve graph based on the abnormal range of each curve graph.

[0015] The curve grouping module is used to group the abnormal curve segments of each curve corresponding to the multiple state parameters of the multiple refrigerators according to the state parameters and the faults marked in the curve graph, to obtain multiple groups, each group corresponding to a fault and a state parameter.

[0016] The curve clustering module is used to cluster the abnormal curve segments in each group to form an abnormal change trajectory, which is then used as the calibration fault curve for the fault and status parameters corresponding to that group.

[0017] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method provided in the first aspect.

[0018] Fourthly, embodiments of the present invention provide a computing device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the method provided in the first aspect.

[0019] The refrigerator fault calibration method, apparatus, medium, and equipment provided in this embodiment of the invention, when combined, have the following beneficial effects:

[0020] (1) The collected data sequence of each state parameter of a refrigerator is converted into curves, and the curves of a refrigerator form a graph. This graph is provided to experts, who can then perform fault calibration and curve anomaly range calibration within the graph. The curve anomaly range based on the graph can be obtained by extracting abnormal curve segments from each curve. The abnormal curve segments of each curve of each refrigerator are grouped according to fault and state parameter. In each group, the abnormal curve segments correspond to the same fault and the same state parameter. Then, each group is clustered to obtain the corresponding calibration fault curve. This process digitizes the experts' understanding of refrigerator faults and their R&D experience, transforming human experience into computer data. In the future, computers will be able to handle a large number of refrigerator fault detection and diagnosis tasks, solving the problem that experts with limited manpower cannot handle a large number of equipment fault diagnosis tasks, reducing diagnostic costs, and improving diagnostic efficiency. Furthermore, since all the curves of a refrigerator are plotted on the same graph, this is because when a refrigerator malfunctions, there may be interrelationships between certain state parameters. Therefore, when experts determine the fault, they can improve the accuracy of the calibration by using multiple state parameters together.

[0021] (2) In one embodiment, the data sequences of various state parameters related to refrigerator faults are extracted from the data sequences of various state parameters uploaded by multiple refrigerators. Since the refrigerators upload data sequences of various state parameters collected by various sensors, but some state parameters are not useful for fault determination or calibration, these data sequences of state parameters are removed, and only the data sequences of state parameters that are useful for fault determination and calibration are retained, thereby improving the efficiency of subsequent fault calibration.

[0022] (3) In one embodiment, the slope change of each abnormal curve segment in a group can reflect the changing trend of each abnormal curve segment in the group. Therefore, according to the slope change, clustering each abnormal curve segment in a group can improve the accuracy of the calibration fault curve obtained by clustering. Attached Figure Description

[0023] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating a refrigerator fault calibration method in one embodiment of the present invention;

[0026] Figure 2 This is a schematic diagram of a refrigerator in one embodiment of the present invention;

[0027] Figure 3 This is a schematic diagram illustrating the abnormal range of a curve for a refrigerator in one embodiment of the present invention.

[0028] Figure 4 This is a schematic diagram of the various abnormal curve segments in a group and the aggregated calibration fault curve in one embodiment of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] In a first aspect, embodiments of the present invention provide a refrigerator fault calibration method.

[0031] See Figure 1 The method includes the following steps S110 to S150:

[0032] S110. Obtain the collection data sequence of multiple status parameters uploaded by multiple refrigerators respectively, wherein the multiple refrigerators are refrigerators that have reported fault repairs;

[0033] For example, a smart refrigerator uploads status data collected by various sensors to a big data platform via its own IoT communication module. Each smart refrigerator uploads multiple data sequences corresponding to multiple status parameters.

[0034] For each state parameter, the corresponding sensor will collect multiple data in chronological order to form a data sequence. That is, the data sequence of a state parameter includes multiple data of that state parameter, and these data are arranged in chronological order to form the aforementioned data sequence.

[0035] The IoT communication module can be any mobile communication protocol module, such as a Wi-Fi module or an NB-IoT module.

[0036] Among them, a big data platform can be any big data system or environment, whether open source or closed source, capable of storing and processing massive amounts of data and running custom filtering programs.

[0037] It is understood that, since the purpose of this embodiment of the invention is to perform fault identification, the refrigerator mentioned above is the refrigerator that reported the repair fault, that is, the above-mentioned data collection sequence is the data collection sequence reported by each faulty refrigerator.

[0038] In one embodiment, after executing S110 and before executing S120, the method may further include: extracting a sequence of collected data on various state parameters related to refrigerator malfunctions from a sequence of collected data on various state parameters uploaded by multiple refrigerators respectively.

[0039] Understandably, refrigerators upload data sequences of various status parameters collected by various sensors. However, some status parameters are not useful for fault determination or calibration. Therefore, these status parameter data sequences are removed, and only the data sequences of status parameters that are useful for fault determination or calibration are retained.

[0040] In one embodiment, after executing S110 and before executing S120, the method may further include:

[0041] A1. Filter out duplicate data in each collected data sequence and remove the duplicate data;

[0042] A2. Count the number of missing values ​​in each collected data sequence;

[0043] A3. If the proportion of missing values ​​in a data sequence exceeds a preset proportion, the data sequence is deleted; if the proportion of missing values ​​in a data sequence does not exceed the preset proportion, the missing values ​​are filled; for numerical data sequences, missing values ​​are filled using the mean; for categorical data sequences, missing values ​​are filled using the mode.

[0044] In other words, there may be duplicate data in the collected data sequence. For example, if a sensor uploads data twice for a certain moment, then the two uploads are duplicates, and one of the uploads can be deleted.

[0045] Understandably, there will be duplicate data, and of course, there may also be missing data. For example, a sensor might not upload data at a certain moment, or the uploaded data might be lost during transmission. This would result in a missing data point in the data acquisition sequence. Missing data can be filled in to address this. However, if too many data points are missing in a data acquisition sequence, its usability will be significantly reduced. Therefore, if the percentage of missing values ​​in a data acquisition sequence exceeds a preset percentage, the sequence will be deleted. The preset percentage can be set as needed, for example, 60%.

[0046] For example, if the collected data sequence is a temperature sequence, and temperature values ​​are missing at certain points in the sequence, the system checks if the percentage of missing temperature values ​​reaches 60%. If it does, the temperature sequence is deleted. If the percentage of missing temperature values ​​is less than 60%, the sequence can be retained, but the missing values ​​need to be filled in. Since temperature values ​​are numerical data, the average value can be used for filling; for example, the temperature value at the missing location can be the average of several surrounding temperature values.

[0047] Some data collection sequences are not numerical but categorical. For categorical data collection sequences, if the proportion of missing values ​​is relatively small, the data collection sequence is retained. Specifically, the mode can be used to fill in the missing data.

[0048] The mode refers to the value that has a clear central tendency in the statistical distribution. It represents the general level of the data and is the value that appears most frequently in a set of data. In other words, in a categorical data collection sequence, the state value that appears most frequently can be selected to fill in the missing state position.

[0049] In one embodiment, before S120 and after missing value processing, the method provided by this embodiment of the invention may further include:

[0050] B1. Sort each collected data sequence in ascending order to obtain the corresponding sequence;

[0051] B2. Calculate the difference between the data at three-quarters position and the data at one-quarter position in the sequential sequence, and determine the outlier distribution range based on the difference;

[0052] B3. Remove data that falls within the outlier distribution range;

[0053] The outlier distribution intervals include: [-∞, Q1-3*QR] and [Q3+3*QR, +∞], where Q1 is the data at the one-quarter position, Q3 is the data at the three-quarter position, and IQR is the difference.

[0054] Understandably, for numerical data sequences, after missing value handling, steps B1 through B3 can be executed directly. However, for categorical data sequences, after missing value handling, the categorical data needs to be converted to numerical data. For example, categorical data sequences such as compressor status, damper status, fan status, and heater status need to be converted to numerical data sequences. This can be achieved using one-hot encoding.

[0055] One-hot encoding, also known as one-bit valid encoding, uses a multi-bit state register to encode multiple states. Each state has its own independent register bit, and at any given time, only one bit is valid.

[0056] For example, for a temperature data sequence collected by a sensor, the temperature values ​​in the sequence are arranged in ascending order to obtain a sequential sequence. Then, the temperature values ​​at three-quarters and one-quarter positions are taken from this sequential sequence. If the one-quarter and / or three-quarter positions are not integers, they can be determined by rounding up or down. The difference between the temperature values ​​at the three-quarters and one-quarter positions is calculated, and then the outlier distribution range is determined based on this difference. Temperature values ​​within the outlier distribution range are removed as outliers, thus achieving outlier handling.

[0057] The intervals [-∞, Q1-3*QR] and [Q3+3*QR, +∞] are not the ranges where most temperature values ​​are located, but rather the ranges where individual temperatures are located. This method can be used to identify individual special temperature values ​​and then remove them as outliers.

[0058] The above method can be simply referred to as the quartile method.

[0059] After the above processing, the data acquisition sequence for each state parameter can be stored. Subsequent steps will use data sequences obtained after the aforementioned processes of duplicate filtering, missing value imputation, and outlier removal.

[0060] S120. Based on the data sequence of each state parameter uploaded by each refrigerator, draw the curve of the state parameter of the refrigerator changing over time, and provide the curve graph of each state parameter of a refrigerator to the experts so that the experts can mark the fault and the abnormal range of the curve in the curve graph.

[0061] In other words, the collected data sequence corresponding to each state parameter is plotted as a curve, that is, the collected data sequence is represented by a curve, thereby converting discrete data into continuous data.

[0062] In each curve, the positive axis represents time, and the ordinate represents the specific value of the corresponding state parameter.

[0063] Specifically, by plotting the curves corresponding to each state parameter of a refrigerator on the same graph, a graph corresponding to that refrigerator is obtained. See also Figure 2 This is a graph showing multiple curves corresponding to the collected data sequences of various status parameters uploaded for a faulty refrigerator. Figure 2 Each curve corresponds to a sequence of collected data for a state parameter, and a single curve can show how a state parameter changes or its trend over time. Then, a curve for a specific refrigerator is provided to experts so they can manually annotate the graph.

[0064] Plotting all the curves of a refrigerator on the same graph is necessary because when a refrigerator malfunctions, certain state parameters may interact with each other. Therefore, multiple state parameters are needed to determine the fault. This way, the graph defines the fault and the abnormal curve range for a single refrigerator. The fault refers to each individual curve on the graph, and the abnormal curve range also refers to each individual curve on the graph.

[0065] For example, see Figure 3 Experts used a graph of a refrigerator to define segments of curves within a specific time period. These defined time ranges can be considered the abnormal ranges of the curves. The abnormal ranges of all curves in a refrigerator's graph are the same.

[0066] In one embodiment, providing the graph to experts in S120 may specifically include:

[0067] The repair time of each refrigerator is obtained, and the curve within the first time range corresponding to that refrigerator is extracted from the curve graph of that refrigerator. The curve within the first time range is then provided to the expert personnel. Here, the first time range corresponding to a refrigerator refers to the time range set with the repair time of that refrigerator as the center.

[0068] For example, based on the warranty records of each refrigerator, the repair report time can be determined, and the time period during which the refrigerator malfunctioned can be identified. A first time range is defined centered on the repair report time of a refrigerator. This first time range encompasses the time period around the repair report time, meaning malfunctions generally occur within this first time range. Therefore, by extracting the curve corresponding to a refrigerator and sending the curve within the first time range to experts, it becomes easier for experts to identify and label malfunctions and anomalies in the curve, thus improving their labeling efficiency.

[0069] S130. Based on the abnormal range of each curve graph, extract the abnormal curve segments of each curve from the curve graph.

[0070] Understandably, based on the abnormal range of each curve chart marked by experts, line segments within this abnormal range can be extracted from each curve in the chart. The line segment extracted from each curve is called an abnormal curve segment, and one abnormal curve segment can be extracted from each curve.

[0071] S140. Based on the state parameters and the faults marked in the curve graph, the abnormal curve segments of each curve corresponding to the multiple state parameters of the multiple refrigerators are grouped to obtain multiple groups, and each group corresponds to a fault and a state parameter.

[0072] In other words, according to the fault and status parameters, the abnormal curve segments are grouped into multiple groups, and each group includes abnormal curve segments with the same fault and the same status parameters.

[0073] For example, in multiple graphs for multiple refrigerators, a total of 3 types of faults are marked. Each refrigerator's graph includes 5 curves, meaning there are 5 status parameters for each refrigerator. By grouping the abnormal curve segments, a total of 15 groups can be obtained. The faults and status parameters of the abnormal curve segments in a group are the same.

[0074] S150. Cluster the abnormal curve segments in each group to form an abnormal change trajectory, and use the abnormal change trajectory as the calibration fault curve for the fault and status parameters corresponding to that group.

[0075] In other words, clustering the abnormal curve segments in a group yields a fitted curve, which is called the abnormal change trajectory. This abnormal change trajectory can then be used as the calibration fault curve for the fault and state parameters corresponding to that group.

[0076] In one embodiment, S150 may specifically include:

[0077] Based on the slope changes of each abnormal curve segment in each group, clustering is performed on each abnormal curve segment in that group to obtain an abnormal change trajectory.

[0078] Understandably, the slope changes of each abnormal curve segment within a group can reflect the changing trend of each abnormal curve segment within that group. Therefore, based on the slope changes, the abnormal curve segments within a group are clustered to obtain an abnormal change trajectory. Specifically, if the slope changes of most abnormal curve segments within a group are consistent, then the changing trend of the abnormal change trajectory is determined based on the changing trend of these majority of abnormal curve segments.

[0079] For example, see Figure 4 This shows that clustering the abnormal curve segments within a group yields an abnormal change trajectory. Figure 4 The light-colored curve segments represent multiple abnormal curve segments, while the dark-colored curve segment represents the abnormal change trajectory obtained from clustering.

[0080] At this point, a calibration fault curve can be generated for each fault and each state parameter.

[0081] In one embodiment, the method provided by this invention further includes:

[0082] The calibration fault curve corresponding to each fault and each status parameter is stored for fault diagnosis.

[0083] In other words, the calibration fault curve corresponding to a fault and a status parameter is stored, so that it can be applied to fault diagnosis later.

[0084] Understandably, the method provided in this embodiment of the invention comprises two main parts: data processing and data calibration. In the data processing stage, the collected data sequences reported by the refrigerators are collected through a big data platform. Then, the collected data sequences undergo duplicate filtering, missing value processing, outlier processing, and the removal of collected data sequences containing state parameters unrelated to the fault. The final collected data sequences are then stored. In the data calibration stage, the collected data sequences of each state parameter for each refrigerator are plotted on the same graph to generate corresponding curves, with one curve per refrigerator. Then, based on the repair reporting time of a refrigerator, the curve for that refrigerator is cropped, extracting curves near the repair reporting time. These curves are then sent to experts, who, based on mechanistic characteristics and R&D experience, annotate the curves with faults and anomaly ranges. Furthermore, based on the anomaly range of a curve, each curve in the graph is cropped to obtain multiple abnormal curve segments. These abnormal curve segments are then grouped according to the fault and state parameters, resulting in multiple groups. Each group includes abnormal curve segments corresponding to the same fault and the same state parameter. Then, the abnormal curve segments in each group are clustered to obtain the calibration fault curve corresponding to that group.

[0085] It is evident that this process digitizes experts' understanding of refrigerator malfunctions and their R&D experience, transforming human experience into computer data. This provides a training foundation and model for subsequent machine learning, enabling computers to handle a large number of refrigerator malfunction detection and diagnosis tasks in the future. This solves the problem that experts with limited manpower cannot handle a large number of equipment malfunction diagnosis tasks, reduces diagnostic costs, and improves diagnostic efficiency.

[0086] Secondly, embodiments of the present invention provide a refrigerator fault calibration device, comprising:

[0087] The data acquisition module is used to acquire the collected data sequences of multiple status parameters uploaded by multiple refrigerators, wherein the multiple refrigerators are refrigerators that have reported faults and are undergoing repairs.

[0088] The curve plotting module is used to plot the curve of the state parameter of each refrigerator changing over time based on the collected data sequence of each state parameter uploaded by each refrigerator, and to provide the curve graph of each state parameter of a refrigerator to experts so that experts can identify the fault and abnormal curve range in the curve graph.

[0089] The anomaly extraction module is used to extract the abnormal curve segments from each curve graph based on the abnormal range of each curve graph.

[0090] The curve grouping module is used to group the abnormal curve segments of each curve corresponding to the multiple state parameters of the multiple refrigerators according to the state parameters and the faults marked in the curve graph, to obtain multiple groups, each group corresponding to a fault and a state parameter.

[0091] The curve clustering module is used to cluster the abnormal curve segments in each group to form an abnormal change trajectory, which is then used as the calibration fault curve for the fault and status parameters corresponding to that group.

[0092] In one embodiment, it also includes:

[0093] The sequence extraction module is used to extract the acquisition data sequence of each state parameter related to the refrigerator fault from the acquisition data sequence of multiple state parameters uploaded by multiple refrigerators before the curve plotting module plots the curve of the state parameter of the refrigerator over time based on the acquisition data sequence of each state parameter uploaded by each refrigerator.

[0094] In one embodiment, it also includes:

[0095] The first processing module is used to perform preprocessing through the following unit before the curve plotting module plots the curve of the state parameter of each refrigerator changing over time based on the collected data sequence of each state parameter uploaded by each refrigerator:

[0096] The first filtering unit is used to filter out duplicate data in each collected data sequence and remove the duplicate data.

[0097] The first statistical unit is used to count the number of missing values ​​in each collected data sequence;

[0098] The missing value handling unit is used to delete a data sequence if the proportion of missing values ​​in a data sequence exceeds a preset proportion, and to fill in missing values ​​if the proportion of missing values ​​in a data sequence does not exceed the preset proportion. Specifically, for missing values ​​in numerical data sequences, the mean is used for filling; for missing values ​​in categorical data sequences, the mode is used for filling.

[0099] Furthermore, it also includes:

[0100] The second processing module is used to process data through the following unit before the curve plotting module plots the curve of the state parameter of each refrigerator over time based on the collected data sequence of each state parameter uploaded by each refrigerator, and after the first processing module performs missing value processing:

[0101] The first sorting unit is used to sort each collected data sequence in ascending order to obtain the corresponding sequence;

[0102] The first calculation unit is used to calculate the difference between the data at the three-quarters position and the data at the one-quarter position in the sequential sequence, and to determine the outlier distribution range based on the difference;

[0103] A data removal unit is used to remove data located within the outlier distribution range;

[0104] The outlier distribution intervals include [-∞, Q1-3*IQR] and [Q3+3*IQR, +∞], where Q1 is the data at the quarter position, Q3 is the data at the three-quarter position, and IQR is the difference.

[0105] In one embodiment, the step of providing the curve to the expert in the curve plotting module includes: obtaining the repair time of each refrigerator, extracting the curve within a first time range corresponding to the refrigerator from the curve of the refrigerator, and providing the curve within the first time range to the expert; wherein, the first time range corresponding to a refrigerator refers to a time range set with the repair time of the refrigerator as the center.

[0106] In one embodiment, the curve clustering module is specifically used to: cluster each abnormal curve segment in each group according to the slope change of each abnormal curve segment in each group, and obtain an abnormal change trajectory.

[0107] In one embodiment, the apparatus further includes:

[0108] The curve storage module is used to store the calibrated fault curve corresponding to each fault and each status parameter for fault diagnosis.

[0109] It is understood that explanations, examples, and beneficial effects of the relevant content in the apparatus provided in the embodiments of the present invention can be referred to the relevant content in the first aspect, and will not be repeated here.

[0110] Thirdly, embodiments of the present invention provide a computer-readable medium storing computer instructions, which, when executed by a processor, cause the processor to perform the method provided in the first aspect.

[0111] Specifically, a system or apparatus equipped with a storage medium may be provided, on which software program code implementing the functions of any of the embodiments described above is stored, and the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage medium.

[0112] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.

[0113] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0114] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system or other components operating on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0115] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion module connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion module execute some and all of the actual operations, thereby realizing the function of any of the above embodiments.

[0116] It is understood that explanations, specific implementation methods, beneficial effects, examples, etc. of the contents in the computer-readable medium provided in the embodiments of the present invention can be found in the corresponding parts of the method provided in the first aspect, and will not be repeated here.

[0117] Fourthly, one embodiment of this specification provides a computing device including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the method of any embodiment of the specification.

[0118] It is understood that explanations, specific implementation methods, beneficial effects, examples, etc. of the computing device provided in the embodiments of the present invention can be found in the corresponding parts of the method provided in the first aspect, and will not be repeated here.

[0119] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0120] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in this invention can be implemented using hardware, software, widgets, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium.

[0121] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solution of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calibrating refrigerator faults, characterized in that, include: Acquire the collection data sequences of multiple status parameters uploaded by multiple refrigerators, wherein the multiple refrigerators are refrigerators that have reported faults and require repair; Based on the data sequence of each status parameter uploaded by each refrigerator that reported a fault and is under repair, a curve showing the change of that status parameter over time is plotted. The curves corresponding to each status parameter of a refrigerator that reported a fault and is under repair are provided to the experts so that the experts can identify the fault and the range of curve abnormalities in the curves. Based on the abnormal range of each curve graph, extract the abnormal curve segments of each curve from the curve graph. Based on the status parameters and the faults marked in the curves, the abnormal curve segments of each curve corresponding to the multiple status parameters of the multiple refrigerators that have reported fault repairs are grouped to obtain multiple groups, and each group corresponds to a fault and a status parameter. Cluster the abnormal curve segments in each group to form an abnormal change trajectory, and use this abnormal change trajectory as the calibration fault curve for the fault and status parameters corresponding to that group. The system stores the calibration fault curve corresponding to each fault and each status parameter. This curve is then compared with the status data uploaded by a refrigerator to determine whether the refrigerator has a fault or what kind of fault it has, thus enabling fault diagnosis of the refrigerator.

2. The method according to claim 1, characterized in that, Before plotting the curve of the change of a state parameter over time for each refrigerator that reported a fault and was uploaded based on the collected data sequence for each state parameter, the method further includes: Extract the data sequences of various status parameters related to refrigerator malfunctions from the data sequences of multiple status parameters uploaded by multiple refrigerators.

3. The method according to claim 1, characterized in that, Before plotting the curve of the change of a state parameter over time for each refrigerator that reported a fault and was uploaded based on the collected data sequence for each state parameter, the method further includes: Duplicate data is identified in each collected data sequence and then filtered out. Count the number of missing values ​​in each collected data sequence; If the proportion of missing values ​​in a data sequence exceeds a preset proportion, the data sequence is deleted; if the proportion of missing values ​​in a data sequence does not exceed the preset proportion, the missing values ​​are filled. Specifically, for missing values ​​in numerical data sequences, the mean is used for filling; for missing values ​​in categorical data sequences, the mode is used for filling.

4. The method according to claim 3, characterized in that, Before plotting the curve of the change of the status parameter of the refrigerator reporting a fault repair based on the collected data sequence of each status parameter uploaded by each refrigerator reporting a fault repair, and after handling missing values, the method further includes: Each collected data sequence is sorted in ascending order to obtain the corresponding sequential sequence; Calculate the difference between the data at three-quarters position and the data at one-quarter position in the sequential sequence, and determine the outlier distribution interval based on the difference; Remove data that falls within the outlier distribution range; The outlier distribution range includes: Q1 is the data at the one-quarter position, Q3 is the data at the three-quarter position, and IQR is the difference.

5. The method according to claim 1, characterized in that, Providing the graph to experts includes: The repair time of each refrigerator is obtained, and the curve within the first time range corresponding to that refrigerator is extracted from the curve graph of that refrigerator. The curve within the first time range is then provided to the expert personnel. Here, the first time range corresponding to a refrigerator refers to the time range set with the repair time of that refrigerator as the center.

6. The method according to claim 1, characterized in that, The step of clustering the abnormal curve segments in each group to form an abnormal change trajectory includes: Based on the slope changes of each abnormal curve segment in each group, clustering is performed on each abnormal curve segment in that group to obtain an abnormal change trajectory.

7. A refrigerator fault calibration device, characterized in that, include: The data acquisition module is used to acquire the collected data sequences of multiple status parameters uploaded by multiple refrigerators, wherein the multiple refrigerators are refrigerators that have reported faults and are undergoing repairs. The curve plotting module is used to plot the curve of the status parameter of each refrigerator that reported a fault and is repaired over time based on the collected data sequence of each status parameter uploaded by each refrigerator that reported a fault and is repaired. The module also provides the curve graph of each status parameter of a refrigerator that reported a fault and is repaired to the experts so that the experts can identify the fault and the abnormal range of the curve in the curve graph. The anomaly extraction module is used to extract the abnormal curve segments from each curve graph based on the abnormal range of each curve graph. The curve grouping module is used to group the abnormal curve segments of each curve corresponding to the multiple status parameters of the multiple refrigerators that have reported fault repairs, according to the status parameters and the faults marked in the curve graph, to obtain multiple groups, each group corresponding to a fault and a status parameter. The curve clustering module is used to cluster the abnormal curve segments in each group to form an abnormal change trajectory, and use the abnormal change trajectory as the calibration fault curve for the fault and status parameters corresponding to that group. The calibration fault curve corresponding to each fault and each status parameter is stored so that it can be compared with the status data uploaded by a refrigerator to determine whether the refrigerator has a fault or what kind of fault it has, thereby realizing fault diagnosis of the refrigerator.

8. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method according to any one of claims 1 to 6.

9. A computing device comprising a memory and a processor, wherein the memory stores executable code, and the processor, when executing the executable code, implements the method according to any one of claims 1 to 6.

Images