阅读说明：本技术 冰箱化霜系统异常检测方法及装置 (Refrigerator defrosting system abnormity detection method and device ) 是由 刘录 何艳 任飞 王辰 于 2021-08-17 设计创作，主要内容包括：本发明提供一种冰箱化霜系统异常检测方法及装置,包括：通过冰箱内的物联网通讯模块获取在预设时间段内冰箱化霜系统的运行数据；根据所述移动时间窗口内每一天对应的运行数据计算所述每一天对应的运行特征统计量,将所述每一天对应的运行特征统计量输入至预先训练的化霜系统状态分类模型中,得到所述移动时间窗口内每一天对应的化霜系统状态；统计所述移动时间窗口内所述化霜系统状态为异常的天数,计算所述化霜系统状态为异常的天数与所述移动时间窗口之间的比值,判断所述比值是否大于预设阈值,若是则确定所述化霜系统出现异常。本发明可以实现对化霜系统是否异常进行检测,而且通过移动时间窗口的方式可以提高检测准确率。(The invention provides a method and a device for detecting the abnormality of a refrigerator defrosting system, which comprises the following steps: the method comprises the steps that operation data of a refrigerator defrosting system in a preset time period are obtained through an internet of things communication module in the refrigerator; calculating the operation characteristic statistic corresponding to each day according to the operation data corresponding to each day in the moving time window, and inputting the operation characteristic statistic corresponding to each day into a pre-trained defrosting system state classification model to obtain the defrosting system state corresponding to each day in the moving time window; counting the number of days in the moving time window when the state of the defrosting system is abnormal, calculating the ratio of the number of days in the moving time window when the state of the defrosting system is abnormal to the moving time window, judging whether the ratio is greater than a preset threshold value, and if so, determining that the defrosting system is abnormal. The invention can realize the detection of whether the defrosting system is abnormal or not, and can improve the detection accuracy rate by moving the time window.)

1. A refrigerator defrosting system abnormity detection method is characterized by comprising the following steps:

the method comprises the steps that operation data of a refrigerator defrosting system in a preset time period are obtained through an internet of things communication module in the refrigerator; wherein, the refrigerator defrosting system includes an evaporator and a defrosting heater, and the operation data includes: the temperature parameter of the sensor of the evaporator and the state parameter of the defrosting heater;

selecting a moving time window within the preset time period, calculating the operation characteristic statistic corresponding to each day according to the operation data corresponding to each day within the moving time window, and inputting the operation characteristic statistic corresponding to each day into a pre-trained defrosting system state classification model to obtain the defrosting system state corresponding to each day within the moving time window, wherein the defrosting system state is normal or abnormal;

counting the number of days in which the state of the defrosting system is abnormal in the moving time window according to the state of the defrosting system corresponding to each day in the moving time window, calculating the ratio of the number of days in which the state of the defrosting system is abnormal to the moving time window, judging whether the ratio is greater than a preset threshold value or not, and if so, determining that the defrosting system is abnormal; otherwise, moving the moving time window according to a preset step length, and detecting whether the defrosting system is abnormal in the next moving time window.

2. The method of claim 1, wherein the status parameters of the defrosting heater comprise an on time and an off time of the defrosting heater each day within a preset time period;

correspondingly, the calculating the operation feature statistic corresponding to each day according to the operation data corresponding to each day in the moving time window includes: and calculating the turn-on times and the single average turn-on time of the defrosting heater in each day according to the turn-on time and the turn-off time of the defrosting heater in each day, and taking the turn-on times and the single average turn-on time corresponding to each day as the running characteristic statistics of the defrosting heater in each day.

3. The method of claim 2, wherein the single average on-time is calculated using a first formula comprising:

in the formula, T_{Duration of time}For the said single average on-time period,for the ith off time of the defrosting heater on the day,and N is the starting time of the ith time of the defrosting heater on each day, and is the starting times of the defrosting heater on each day.

4. The method of claim 2, wherein the sensor temperature parameter of the evaporator comprises a temperature detected by the sensor of the evaporator during respective on periods of the defrosting heater on each day for a preset period of time;

correspondingly, the calculating the operation feature statistic corresponding to each day according to the operation data corresponding to each day in the moving time window further includes: and calculating a maximum temperature value, a minimum temperature value, an average temperature value and/or a variance value of the temperature in each day according to the temperature detected by the sensor of the evaporator in each turn-on period of the defrosting heater in each day, and taking the maximum temperature value, the minimum temperature value, the average temperature value and/or the variance value corresponding to each day as the running characteristic statistic of the evaporator.

5. The method of claim 4, wherein prior to calculating the operational characteristic statistics for each day from the operational data for each day within the moving time window, the method further comprises: pre-processing the temperature detected by the sensor of the evaporator in each on-period of the defrosting heater in each day, wherein the pre-processing process comprises the following steps:

sequencing the temperatures detected by the sensors of the evaporator in each starting time period of the defrosting heater in each day from large to small to obtain a temperature sequence corresponding to each day;

acquiring temperature values at 1/5, 2/5 and 3/5 positions in the temperature sequence, and calculating standard deviation and mean value of the temperature sequence;

determining a temperature confidence interval based on the temperature values at the 1/5, 2/5, and 3/5 positions and the standard deviation and mean;

and according to the temperature confidence interval, rejecting temperature values outside the temperature confidence interval in the temperature sequence.

6. The method of claim 5, wherein determining a temperature confidence interval based on the temperature values at the 1/5, 2/5, and 3/5 locations and the standard deviation and mean comprises:

determining a corresponding interval size adjusting value N according to the standard deviation delta and the mean value mu, wherein the interval size adjusting value N can enable 95% -98% of temperature values in the temperature sequence to fall within an interval [ mu-a 1 XNXdelta, mu + a2 XNXdelta ], a1 and a2 are preset coefficients and are both located within a range of [0.8, 1 ];

calculating temperature gap parameters according to the temperature values at the 1/5 position, the 2/5 position and the 3/5 position;

and determining the temperature confidence interval according to the interval size adjusting value and the temperature gap parameter.

7. The method of claim 6, wherein the temperature gap parameter is calculated using a first formula comprising:

M＝min{(Q₂-Q₁)/5,(Q₃-Q₂)/5,2*(Q₃-Q₁)/5}

wherein M is the temperature difference parameter, Q₁Is the temperature value at position 1/5, Q₂Is the temperature value at position 2/5, Q₃Is the temperature value at position 3/5.

8. The method of claim 6, wherein determining the temperature confidence interval based on the interval size adjustment value and the temperature gap parameter comprises: calculating the temperature confidence interval using a second formula, the second formula comprising:

P＝[(Q1+Q2)/2-N×M,(Q3+Q2)/2+N×M]

wherein P is the temperature confidence interval, M is the temperature gap parameter, N is the interval size adjustment value, and Q₁Is the temperature value at position 1/5, Q₂Is the temperature value at position 2/5, Q₃Is the temperature value at position 3/5.

9. The method of claim 2, further comprising:

and reporting abnormal reminding information of the defrosting system to an after-sales service platform when the defrosting system is determined to be abnormal so as to enable personnel to check faults on the door, and if the defrosting system is determined to be abnormal by checking, carrying out maintenance, otherwise, feeding back information without faults to the after-sales service platform so as to carry out parameter adjustment on the state classification model of the defrosting system.

10. An abnormality detection device for a defrosting system of a refrigerator, comprising:

the data acquisition module is used for acquiring the operation data of the refrigerator defrosting system in a preset time period through the internet of things communication module in the refrigerator; wherein, the refrigerator defrosting system includes an evaporator and a defrosting heater, and the operation data includes: the temperature parameter of the sensor of the evaporator and the state parameter of the defrosting heater;

the model identification module is used for selecting a moving time window in the preset time period, calculating the running characteristic statistic corresponding to each day according to the running data corresponding to each day in the moving time window, and inputting the running characteristic statistic corresponding to each day into a pre-trained defrosting system state classification model to obtain the state of the defrosting system corresponding to each day in the moving time window, wherein the state of the defrosting system is normal or abnormal;

the abnormality judgment module is used for counting the days in the moving time window when the state of the defrosting system is abnormal according to the state of the defrosting system corresponding to each day in the moving time window, calculating the ratio of the days in the moving time window when the state of the defrosting system is abnormal to the moving time window, judging whether the ratio is greater than a preset threshold value or not, and if so, determining that the defrosting system is abnormal; otherwise, moving the moving time window according to a preset step length, and detecting whether the defrosting system is abnormal in the next moving time window.

Technical Field

One or more embodiments of the present disclosure relate to the technical field of refrigerators, and in particular, to a method and an apparatus for detecting an abnormality of a defrosting system of a refrigerator.

Background

The refrigerator is an essential household appliance in family life, and in the use process of the refrigerator, because the humidity of air and the water content of food can inevitably generate water vapor, the evaporator of the refrigerator can be frosted. When the frosting amount of the evaporator reaches a certain degree, the refrigerating effect of the refrigerator is affected. A corresponding defrosting heater is designed for the purpose, and defrosting treatment is carried out on the evaporator. However, the continuous high temperature of the defrosting heater in the using process can cause faults, such as fusing, so that the defrosting cannot be normally carried out, the refrigerating capacity of the refrigerator is reduced, even the refrigerator cannot be refrigerated, and the faults are not easy to detect. It is necessary to check whether the defrosting system of the refrigerator is normal.

Disclosure of Invention

One or more embodiments of the invention describe a refrigerator defrosting system abnormity detection method and device.

According to a first aspect, there is provided a refrigerator defrosting system abnormality detection method, including:

the method comprises the steps that operation data of a refrigerator defrosting system in a preset time period are obtained through an internet of things communication module in the refrigerator; wherein, the refrigerator defrosting system includes an evaporator and a defrosting heater, and the operation data includes: the temperature parameter of the sensor of the evaporator and the state parameter of the defrosting heater;

selecting a moving time window within the preset time period, calculating the operation characteristic statistic corresponding to each day according to the operation data corresponding to each day within the moving time window, and inputting the operation characteristic statistic corresponding to each day into a pre-trained defrosting system state classification model to obtain the defrosting system state corresponding to each day within the moving time window, wherein the defrosting system state is normal or abnormal;

counting the number of days in which the state of the defrosting system is abnormal in the moving time window according to the state of the defrosting system corresponding to each day in the moving time window, calculating the ratio of the number of days in which the state of the defrosting system is abnormal to the moving time window, judging whether the ratio is greater than a preset threshold value or not, and if so, determining that the defrosting system is abnormal; otherwise, moving the moving time window according to a preset step length, and detecting whether the defrosting system is abnormal in the next moving time window.

According to a second aspect, there is provided a refrigerator defrosting system abnormality detecting apparatus including:

the data acquisition module is used for acquiring the operation data of the refrigerator defrosting system in a preset time period through the internet of things communication module in the refrigerator; wherein, the refrigerator defrosting system includes an evaporator and a defrosting heater, and the operation data includes: the temperature parameter of the sensor of the evaporator and the state parameter of the defrosting heater;

the model identification module is used for selecting a moving time window in the preset time period, calculating the running characteristic statistic corresponding to each day according to the running data corresponding to each day in the moving time window, and inputting the running characteristic statistic corresponding to each day into a pre-trained defrosting system state classification model to obtain the state of the defrosting system corresponding to each day in the moving time window, wherein the state of the defrosting system is normal or abnormal;

the abnormality judgment module is used for counting the days in the moving time window when the state of the defrosting system is abnormal according to the state of the defrosting system corresponding to each day in the moving time window, calculating the ratio of the days in the moving time window when the state of the defrosting system is abnormal to the moving time window, judging whether the ratio is greater than a preset threshold value or not, and if so, determining that the defrosting system is abnormal; otherwise, moving the moving time window according to a preset step length, and detecting whether the defrosting system is abnormal in the next moving time window.

The method and the device for detecting the abnormality of the refrigerator defrosting system, provided by the embodiment of the specification, are used for obtaining the operation data of the defrosting system, converting the operation data in a moving time window into operation characteristic statistics, inputting the operation characteristic statistics of each day into a state classification model of the defrosting system to obtain the state of the defrosting system of each day, calculating the ratio of the number of days with abnormal defrosting in the moving time window to the moving time window, considering that the defrosting system is abnormal if the ratio is greater than a preset threshold value, and otherwise, performing abnormal detection on a next moving time window, so that the abnormal detection of the defrosting system is realized. Due to the adoption of the mode of moving the time window, if the abnormality is not found after the detection of one moving time window, the moving time window is moved to the next position, so that the abnormality can be found in time. And the detection accuracy can be improved by comparing the ratio of the number of days with abnormal defrosting in a moving time window to the moving time window with a preset threshold value instead of simply judging whether the refrigerator defrosting system is normal in a one-day state.

Drawings

In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present specification, and other drawings can be obtained by those skilled in the art without creative efforts.

Fig. 1 is a schematic flow chart of an abnormality detection method for a refrigerator defrosting system according to an embodiment of the present invention.

Detailed Description

The scheme provided by the specification is described below with reference to the accompanying drawings.

In a first aspect, the present invention provides a method for detecting an abnormality of a defrosting system of a refrigerator, as shown in fig. 1, the method may include the following steps S110 to S130:

s110, acquiring operation data of a refrigerator defrosting system in a preset time period through an internet of things communication module in the refrigerator;

wherein, the refrigerator defrosting system includes evaporimeter and defrosting heater, the operating data that corresponds includes: the temperature parameter of the sensor of the evaporator and the state parameter of the defrosting heater.

The refrigerator comprises a refrigerator body, a defrosting system and an internet of things communication module, wherein the refrigerator body is provided with the internet of things communication module, and the defrosting system is used for detecting the defrosting system. And the defrosting system comprises an evaporator and a defrosting heater, so the acquired operation data can comprise a sensor temperature parameter of the evaporator, a state parameter of the defrosting heater and the like. When the defrosting heater is turned on, the temperature detected by the sensor of the evaporator gradually increases.

The specific duration can be determined according to the requirement within a preset time period, for example, one month, two months, one week, and the like.

S120, selecting a moving time window within the preset time period, calculating the operation characteristic statistic corresponding to each day according to the operation data corresponding to each day within the moving time window, and inputting the operation characteristic statistic corresponding to each day into a pre-trained defrosting system state classification model to obtain the defrosting system state corresponding to each day within the moving time window, wherein the defrosting system state is normal or abnormal;

the window size of the moving time window and the moving step size may be set as required, for example, the size of one moving time window is one week, and the moving step size may be selected to be 2 days. Other window sizes and step sizes may of course be selected.

It can be understood that the operation data of a day is firstly converted into the operation characteristic statistics, and then the operation characteristic vector corresponding to the day is input into the model, so that the model can output whether the state of the defrosting system corresponding to the day is normal or abnormal. When the model is trained, each adopted training sample is the operation characteristic statistic corresponding to one day, the training sample is subjected to normal or abnormal marking, the training sample marked as normal is a positive sample, and the training sample marked as abnormal is a negative sample.

For example, the window size of a moving time window is 5 days, the operation data of each of the 5 days is converted into the operation feature statistics and then input into the defrosting system state classification model, so that the corresponding defrosting system state of the day can be obtained, and a total of 5 defrosting system states can be obtained through the method.

In specific implementation, the state parameters of the defrosting heater may include the turn-on time and the turn-off time of the defrosting heater on each day within a preset time period; correspondingly, the calculating the operation feature statistic corresponding to each day according to the operation data corresponding to each day in the moving time window may include: and calculating the turn-on times and the single average turn-on time of the defrosting heater in each day according to the turn-on time and the turn-off time of the defrosting heater in each day, and taking the turn-on times and the single average turn-on time corresponding to each day as the running characteristic statistics of the defrosting heater in each day.

That is, according to the on-time and the off-time of the defrosting heater in one day, the number of times of turning on the defrosting heater in the day and the single average on-time can be counted, and the two values are used as the running characteristic statistics of the defrosting heater in the day. Specifically, the single average open time may be calculated by using a first formula, where the first formula includes:

in the formula, T_{Duration of time}For the said single average on-time period,for the ith off time of the defrosting heater on the day,and N is the starting time of the ith time of the defrosting heater on each day, and is the starting times of the defrosting heater on each day.

In particular implementation, the sensor temperature parameter of the evaporator may include a temperature detected by the sensor of the evaporator during each on-period of the defrosting heater for each day within a preset time period; correspondingly, the calculating the operation feature statistic corresponding to each day according to the operation data corresponding to each day in the moving time window may further include: and calculating a maximum temperature value, a minimum temperature value, an average temperature value and/or a variance value of the temperature in each day according to the temperature detected by the sensor of the evaporator in each turn-on period of the defrosting heater in each day, and taking the maximum temperature value, the minimum temperature value, the average temperature value and/or the variance value corresponding to each day as the running characteristic statistic of the evaporator.

That is, the maximum temperature value, the minimum temperature value, the average temperature value and/or the variance value of the temperature during the day are calculated as the operation characteristic statistics of the evaporator according to the temperatures detected by the sensors of the evaporator during all the on-period of the defrosting heater during the day.

S130, counting the number of days in which the state of the defrosting system is abnormal in the moving time window according to the state of the defrosting system corresponding to each day in the moving time window, calculating the ratio of the number of days in which the state of the defrosting system is abnormal to the moving time window, judging whether the ratio is greater than a preset threshold value or not, and if so, determining that the defrosting system is abnormal; otherwise, moving the moving time window according to a preset step length, and detecting whether the defrosting system is abnormal in the next moving time window.

For example, a moving time window is 5 days, if 5 defrosting system states obtained after the defrosting system state classification model are normal, abnormal and normal respectively, the number of days that the defrosting system state is abnormal in the moving time window is 1 day, the ratio of the number of days that the defrosting system state is abnormal to the moving time window is 1/5, and if the preset threshold is 4/5 and 1/5 is smaller than 4/5, it is determined that the defrosting system is normal in the moving time window. And then, after the moving time window is moved by a certain step length to obtain a new moving time window, detecting whether the new moving time window is abnormal or not.

If the 5 defrosting system states obtained after the defrosting system state classification model are all abnormal, the ratio of the number of days for which the defrosting system state is abnormal to the moving time window is 1, and 1 is greater than a preset threshold value 4/5, it indicates that the defrosting system is abnormal in the moving time window, and at this moment, the defrosting system cannot carry out normal defrosting, so that the refrigerating capacity of the refrigerator is reduced, even the refrigerator is not refrigerated, and therefore, the defrosting system needs to be reported for maintenance.

It can be understood that, since the condition of using the refrigerator by the user is complicated, the state of the refrigerator is greatly influenced, and therefore, whether the defrosting system of the refrigerator is normal or not can not be simply judged in a state of one day, and therefore, the judgment needs to be carried out in combination with the state of the refrigerator in a continuous time period.

In specific implementation, the method provided by the invention can further comprise:

and S140, reporting abnormal reminding information of the defrosting system to an after-sales service platform when the defrosting system is determined to be abnormal so as to enable personnel to check for faults when going to the door, and if the defrosting system is determined to be abnormal by checking, carrying out maintenance, otherwise, feeding back information without faults to the after-sales service platform so as to carry out parameter adjustment on the state classification model of the defrosting system.

That is to say, when the state of the defrosting system corresponding to one moving time window is determined to be abnormal, the state is reported to the after-sales service platform, then the after-sales service platform informs the staff to go to the door for maintenance, and no problem exists in the defrosting system during the door-to-door maintenance, and no fault exists, which indicates that the accuracy of the state classification model of the defrosting system still does not meet the requirement, and at the moment, the parameter adjustment can be performed on the state classification model of the defrosting system, so as to improve the accuracy of the state classification. And after the workers find that the defrosting system really has problems, the defrosting system is maintained in time, and the use experience of users is improved.

The abnormal detection method of the refrigerator defrosting system provided by the invention comprises the steps of obtaining operation data of the defrosting system, converting the operation data in a moving time window into operation characteristic statistics, inputting the operation characteristic statistics of each day into a state classification model of the defrosting system to obtain the state of the defrosting system of each day, calculating the ratio of the number of days with abnormal defrosting in the moving time window to the moving time window, considering that the defrosting system is abnormal if the ratio is greater than a preset threshold value, and otherwise, performing abnormal detection on a next moving time dismantling window, thereby realizing the abnormal detection of the defrosting system. Due to the adoption of the mode of moving the time window, if the abnormality is not found after the detection of one moving time window, the moving time window is moved to the next position, so that the abnormality can be found in time. And the detection accuracy can be improved by comparing the ratio of the number of days with abnormal defrosting in a moving time window to the moving time window with a preset threshold value instead of simply judging whether the refrigerator defrosting system is normal in a one-day state.

It can be understood that, in practice, whether the defrosting system of the refrigerator is normal or abnormal, there is a possibility that abnormal data may occur due to detection errors of sensors and the like. The abnormality here means that the difference between the detected temperature and the true temperature is large. In order to avoid a state judgment error of the defrosting system caused by a large difference between the detected temperature value and the real temperature value, that is, to reduce or avoid a judgment error of an influence factor other than the defrosting system on the state of the defrosting system, before calculating the operation characteristic statistic corresponding to each day according to the operation data corresponding to each day in the moving time window in S120, the method provided by the present invention may further include:

s150, preprocessing the temperature detected by the sensor of the evaporator in each starting time interval of the defrosting heater in each day, and eliminating abnormal data in a preprocessing mode.

Wherein the preprocessing process may include:

s151, sequencing the temperatures detected by the sensors of the evaporator in each starting time period of the defrosting heater in each day from large to small to obtain a temperature sequence corresponding to each day;

namely, the corresponding temperature values in each day are sorted to obtain the temperature sequence corresponding to the day.

S152, acquiring temperature values at 1/5, 2/5 and 3/5 positions in the temperature sequence, and calculating standard deviation and mean value of the temperature sequence;

if the 1/5-, 2/5-or 3/5-positions in the temperature series are not integers, rounding is performed, and the rounded position is defined as the 1/5-, 2/5-or 3/5-position. For example, a total of 24 temperature values in the temperature sequence are obtained by rounding 24/5 to obtain 4, that is, the 4 th temperature value in the temperature sequence is used as the temperature value at the 1/5 position, and 24 × 2/5 is obtained by rounding 9, then the 9 th temperature value in the temperature sequence is used as the temperature value at the 2/5 position, and 24 × 3/5 is obtained by rounding 14, and then the 14 th temperature value in the temperature sequence is used as the temperature value at the 3/5 position.

The reason why the temperatures at the above three positions are adopted here is: the reporting frequency of the internet communication module of the refrigerator is relatively high in one day, and the reporting frequency can be basically reported once an hour, so that a large number of temperature values can be generated, the temperature sequence comprises 24 temperature values as an example, and actually the number of the temperature values is far larger than 24. The 1/5, 2/5 and 3/5 positions are closer to the middle part in the temperature sequence and can also show data at two ends, so that the 1/5, 2/5 and 3/5 positions are adopted. After a plurality of tests, the temperatures corresponding to one-fourth, one-half, three-fourth, etc. or other positions, namely the temperatures at the 1/5 position, the 2/5 position and the 3/5 position are used as the basis of subsequent calculation, and the effect is best, namely the reliability of the temperature confidence interval calculated based on the temperatures at the 1/5 position, the 2/5 position and the 3/5 position is higher under the condition that other parameters are inconvenient in the subsequent calculation.

S153, determining a temperature confidence interval according to the temperature values at the 1/5 position, the 2/5 position and the 3/5 position and the standard deviation and the mean value;

in a specific implementation, S153 may specifically include the following steps:

s153a, determining a corresponding interval size adjusting value N according to the standard deviation delta and the mean value mu, wherein the interval size adjusting value N can enable 95% -98% of temperature values in the temperature sequence to fall within an interval [ mu-a 1 multiplied by N delta, mu + a2 multiplied by N multiplied by delta ], and a1 and a2 are preset coefficients and are both located within a range of [0.8, 1 ]. Therefore, 95% -98% of temperature values can be ensured to fall in the interval, and the temperature confidence interval calculated according to the N at the moment is not too wide or too narrow.

S153b, calculating temperature difference parameters according to the temperature values at the 1/5 position, the 2/5 position and the 3/5 position;

the temperature difference parameter is a parameter for representing the temperature difference in the temperature sequence. Specifically, the temperature difference parameter may be calculated by using a first formula, where the first formula includes:

M＝min{(Q2-Q1)/5,(Q3-Q2)/5,2*(Q3-Q1)/5}

wherein M is the temperature difference parameter, Q₁Is the temperature value at position 1/5, Q₂Is the temperature value at position 2/5, Q₃Is the temperature value at position 3/5.

S153c, determining the temperature confidence interval according to the interval size adjusting value and the temperature gap parameter.

In particular implementations, the temperature confidence interval may be calculated using a second formula, the second formula comprising:

P＝[(Q1+Q2)/2-N×M,(Q3+Q2)/2+N×M]

wherein P is the temperature confidence interval, M is the temperature gap parameter, N is the interval size adjustment value, and Q₁Is the temperature value at position 1/5, Q₂Is the temperature value at position 2/5, Q₃Is the temperature value at position 3/5.

Therefore, the temperature values at the three positions are used as the basis of interval adjustment, the size of the temperature confidence interval can be adjusted through the interval size adjustment value N and the temperature difference parameter M, the second formula is obtained through multiple tests, and the formula can ensure that most of temperature values fall in the temperature confidence interval and can exclude abnormal data outside the temperature confidence interval.

S154, according to the temperature confidence interval, temperature values outside the temperature confidence interval in the temperature sequence are removed. Namely, the temperature values outside the temperature confidence interval in the temperature sequence are removed, and the temperature data are preprocessed.

Through the steps S151 to S154, it is possible to realize that the abnormal detection temperature value caused by the detection error and the like cannot well represent the real temperature value, and if the state of the defrosting system is judged according to the abnormal detection temperature value, the judgment error may be caused, and the judgment accuracy is reduced. It can be seen that the sensor temperature parameter of the evaporator is preprocessed through S151-S154, and the subsequent judgment accuracy can be improved.

In a second aspect, the present invention provides an abnormality detection apparatus for a defrosting system of a refrigerator, comprising:

the data acquisition module is used for acquiring the operation data of the refrigerator defrosting system in a preset time period through the internet of things communication module in the refrigerator; wherein, the refrigerator defrosting system includes an evaporator and a defrosting heater, and the operation data includes: the temperature parameter of the sensor of the evaporator and the state parameter of the defrosting heater;

the model identification module is used for selecting a moving time window in the preset time period, calculating the running characteristic statistic corresponding to each day according to the running data corresponding to each day in the moving time window, and inputting the running characteristic statistic corresponding to each day into a pre-trained defrosting system state classification model to obtain the state of the defrosting system corresponding to each day in the moving time window, wherein the state of the defrosting system is normal or abnormal;

the abnormality judgment module is used for counting the days in the moving time window when the state of the defrosting system is abnormal according to the state of the defrosting system corresponding to each day in the moving time window, calculating the ratio of the days in the moving time window when the state of the defrosting system is abnormal to the moving time window, judging whether the ratio is greater than a preset threshold value or not, and if so, determining that the defrosting system is abnormal; otherwise, moving the moving time window according to a preset step length, and detecting whether the defrosting system is abnormal in the next moving time window.

It is understood that, for the device provided in the embodiment of the present invention, for the explanation, examples, and beneficial effects of the related contents, reference may be made to the corresponding parts in the foregoing method, and details are not described here.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

Those skilled in the art will recognize that, in one or more of the examples described above, the functions described in this disclosure may be implemented in hardware, software, hardware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.

The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the present invention should be included in the scope of the present invention.