CN117034729A - Method, device and storage medium for predicting service life of low-voltage electrical appliance - Google Patents

Abstract

The application provides a life prediction method and device of a low-voltage electrical appliance and a storage medium, and relates to the technical field of low-voltage electrical appliances. The method comprises the steps of obtaining switching-on and switching-off data of a piezoelectric device to be predicted in a historical time period in a plurality of switching-on and switching-off periods; determining the target class of the low-voltage electric appliance to be predicted based on a pre-classification model according to the opening and closing data, wherein the pre-classification model is obtained through training according to a first training sample data set; determining a target special prediction model according to the target class of the low-voltage electric appliance to be predicted; according to the switching-on and switching-off data, the residual life parameters of the low-voltage electric appliance to be predicted are predicted based on the target special prediction model, so that the target class of the low-voltage electric appliance to be predicted can be determined according to the pre-classification model, the residual life parameters of the low-voltage electric appliance to be predicted can be predicted by selecting the target special prediction model matched with the target class according to the target class, and the accuracy of a prediction result can be improved.

Description

Method, device and storage medium for predicting service life of low-voltage electrical appliance

Technical Field

The application relates to the technical field of low-voltage appliances, in particular to a life prediction method and device of a low-voltage appliance and a storage medium.

Background

The low-voltage electric appliance is usually an electric appliance working under 1200V ac voltage or 1500V dc voltage, and is widely applied to industrial production and life, and in some application scenarios with high requirements on equipment operation stability and safety, such as elevators, motor cars, new energy sources, etc., if the low-voltage electric appliance cannot be replaced in time before failure, great loss and inconvenience can be caused to enterprise production and people life, so that it is necessary to predict the current remaining electric life of the low-voltage electric appliance.

In the prior art, the estimated remaining service life of the low-voltage electrical appliance is generally estimated approximately according to the used time and the expected use time of the low-voltage electrical appliance when leaving the factory.

It can be seen that the existing method for estimating the residual life of the piezoelectric device is simpler, and the problem of inaccurate estimation exists.

Disclosure of Invention

The application aims to overcome the defects in the prior art and provide a life prediction method, a life prediction device and a storage medium for a low-voltage electrical appliance, which can improve the accuracy of a prediction result.

In order to achieve the above purpose, the technical scheme adopted by the embodiment of the application is as follows:

in a first aspect, the present application provides a life prediction method for a low-voltage electrical appliance, including:

Acquiring switching-on and switching-off data of the piezoelectric device to be predicted in a historical time period in a plurality of switching-on and switching-off periods;

determining a target class to which the low-voltage electrical appliance to be predicted belongs based on a pre-classification model according to the opening and closing data, wherein the pre-classification model is obtained through training according to a first training sample data set, the first training sample data set comprises first sample opening and closing data corresponding to a plurality of first sample electrical appliances in a first historical time period, and corresponding classes are marked;

determining a target special prediction model according to the target class of the to-be-predicted piezoelectric device;

and predicting the residual life parameter of the low-voltage electrical appliance to be predicted based on the target special prediction model according to the opening and closing data, wherein the target special prediction model is obtained through training according to a second training sample data set, the second training sample data set comprises second sample opening and closing data corresponding to a plurality of second sample electrical appliances belonging to the target class in a second historical time period, and corresponding residual life parameters are marked.

In an optional implementation manner, the determining a target-specific prediction model according to the target class to which the low-voltage electrical appliance to be predicted belongs includes:

Determining a target special prediction model in at least one special prediction model according to the target category to which the to-be-predicted piezoelectric device belongs and a preset mapping relation, wherein the preset mapping relation comprises a mapping relation between at least one category and the special prediction model.

In an alternative embodiment, the method further comprises:

acquiring a first training sample data set of a plurality of first sample electric appliances in a plurality of opening and closing periods in a first historical time period, wherein the first training sample data set comprises a plurality of first sample opening and closing data, and each first sample opening and closing data is marked with a total service life state corresponding to the first sample electric appliances;

and training to obtain the pre-classification model according to the first training sample data set.

In an alternative embodiment, the method further comprises:

acquiring a second training sample data set of a plurality of second sample electric appliances belonging to the target class in a second historical time period in a plurality of opening and closing periods, wherein the second training sample data set comprises a plurality of second sample opening and closing data, and each second sample opening and closing data is marked with a residual life parameter corresponding to the second sample electric appliance;

And training to obtain the target special prediction model according to the second training sample data set.

In an optional embodiment, the acquiring the switching on/off data of the low-voltage electrical appliance to be predicted in the historical time period in a plurality of switching on/off periods includes:

acquiring voltage data and/or current data of the piezoelectric device to be predicted in a plurality of opening and closing periods in the historical time period through an acquisition device;

and acquiring the electric life characteristic of the low-voltage electrical appliance to be predicted according to the voltage data and/or the current data, and taking the electric life characteristic as the switching-on/off data of the low-voltage electrical appliance to be predicted.

In an alternative embodiment, after training to obtain the pre-classification model according to the first training sample data set, the method further includes:

acquiring a third training sample data set of a plurality of first sample electric appliances in a plurality of opening and closing periods in a third historical time period, wherein the third training sample data set comprises a plurality of third sample opening and closing data, each third sample opening and closing data is marked with a total service life state corresponding to the first sample electric appliances, and the third historical time period and the first historical time period are different time periods;

And updating the pre-classification model according to the third training sample data set.

In an alternative embodiment, at least one of the categories includes: a first total life span and a second total life span, wherein the first total life span is less than the second total life span.

In an alternative embodiment, the electrical lifetime characteristics of the electrical device to be predicted include at least one of: contact voltage, contact current, contact resistance, arcing time, arcing energy, and arcing power.

In a second aspect, the present invention provides a life predicting apparatus for a low-voltage electrical appliance, comprising:

the acquisition module is used for acquiring switching-on/off data of the piezoelectric device to be predicted in a historical time period in a plurality of switching-on/off periods;

the first determining module is used for determining the target class to which the low-voltage electrical appliance to be predicted belongs based on a pre-classification model according to the opening and closing data, the pre-classification model is obtained through training according to a first training sample data set, the first training sample data set comprises first sample opening and closing data corresponding to a plurality of first sample electrical appliances in a first historical time period, and the corresponding class is marked;

the second determining module is used for determining a target special prediction model according to the target category of the to-be-predicted piezoelectric device;

The prediction module is used for predicting the residual life parameter of the low-voltage electrical appliance to be predicted based on the target special prediction model according to the opening and closing data, the target special prediction model is obtained through training according to a second training sample data set, the second training sample data set comprises second sample opening and closing data corresponding to a plurality of second sample electrical appliances belonging to the target class in a second historical time period, and the corresponding residual life parameter is marked.

In an optional implementation manner, the second determining module is specifically configured to determine, in at least one dedicated prediction model, a target dedicated prediction model according to a target class to which the low-voltage electrical appliance to be predicted belongs and a preset mapping relationship, where the preset mapping relationship includes a mapping relationship between at least one class and the dedicated prediction model.

In an alternative embodiment, the lifetime prediction apparatus further includes: the first training module is used for acquiring a first training sample data set of a plurality of first sample electric appliances in a plurality of opening and closing periods in a first historical time period, wherein the first training sample data set comprises a plurality of first sample opening and closing data, and each first sample opening and closing data is marked with a total service life state corresponding to the first sample electric appliances;

And training to obtain the pre-classification model according to the first training sample data set.

In an alternative embodiment, the lifetime prediction apparatus further includes: the second training module is used for acquiring a second training sample data set of a plurality of second sample electric appliances belonging to the target class in a second historical time period in a plurality of opening and closing periods, wherein the second training sample data set comprises a plurality of second sample opening and closing data, and each second sample opening and closing data is marked with a residual life parameter corresponding to the second sample electric appliance;

and training to obtain the target special prediction model according to the second training sample data set.

In an optional embodiment, the acquiring module is specifically configured to acquire, by using an acquiring device, voltage data and/or current data of the to-be-predicted piezoelectric device in a plurality of switching-on and switching-off periods in the historical time period;

and acquiring the electric life characteristic of the low-voltage electrical appliance to be predicted according to the voltage data and/or the current data, and taking the electric life characteristic as the switching-on/off data of the low-voltage electrical appliance to be predicted.

In an optional embodiment, the first training module is further configured to obtain a third training sample data set of the plurality of first sample appliances in a third historical time period in a plurality of opening and closing periods, where the third training sample data set includes a plurality of third sample opening and closing data, each third sample opening and closing data is labeled with a total lifetime state corresponding to the first sample appliance, and the third historical time period is a different time period from the first historical time period;

And updating the pre-classification model according to the third training sample data set.

In an alternative embodiment, at least one of the categories includes: a first total life span and a second total life span, wherein the first total life span is less than the second total life span.

In an alternative embodiment, the electrical lifetime characteristics of the electrical device to be predicted include at least one of: contact voltage, contact current, contact resistance, arcing time, arcing energy, and arcing power.

In a third aspect, the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method for lifetime prediction of a low voltage electrical appliance according to any of the previous embodiments.

In a fourth aspect, an embodiment of the present application provides an electronic device, including: the device comprises a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device runs, the processor and the storage medium are communicated through the bus, and the processor executes the machine-readable instructions to execute the steps of the life prediction method of the low-voltage electrical appliance according to any one of the previous embodiments.

The beneficial effects of the application are as follows:

in the life prediction method, the device and the storage medium of the low-voltage electrical appliance, the switching-on/off data of the low-voltage electrical appliance to be predicted in a plurality of switching-on/off periods in a historical time period are obtained; determining the target class of the low-voltage electric appliance to be predicted based on a pre-classification model according to the opening and closing data, wherein the pre-classification model is obtained through training according to a first training sample data set; determining a target special prediction model according to the target class of the low-voltage electric appliance to be predicted; according to the switching-on and switching-off data, the residual life parameters of the low-voltage electric appliance to be predicted are predicted based on the target special prediction model, the target special prediction model is obtained through training according to the second training sample data set, the target class of the low-voltage electric appliance to be predicted can be determined according to the pre-classification model, the residual life parameters of the low-voltage electric appliance to be predicted can be predicted according to the target class by selecting the target special prediction model matched with the target class, and accuracy of a prediction result can be improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a life prediction method of a low-voltage electrical apparatus according to an embodiment of the present application;

fig. 2 is a flow chart of another life prediction method of a low-voltage electrical apparatus according to an embodiment of the present application;

fig. 3 is a flow chart of a life prediction method of another low-voltage electrical apparatus according to an embodiment of the present application;

fig. 4 is a flow chart of another life prediction method of a low-voltage electrical apparatus according to an embodiment of the present application;

fig. 5 is a flow chart of a life prediction method of another low-voltage electrical apparatus according to an embodiment of the present application;

fig. 6 is a schematic diagram of an electrical life prediction result of a to-be-predicted piezoelectric device after classification based on a first opening and closing data set according to an embodiment of the present application;

fig. 7 is a schematic diagram of an electrical life prediction result of a to-be-predicted piezoelectric device after classification based on a second opening and closing data set according to an embodiment of the present application;

fig. 8 is a schematic functional block diagram of a life prediction device for a low-voltage electrical appliance according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.

Fig. 1 is a schematic flow chart of a method for predicting the lifetime of a low-voltage electrical apparatus according to an embodiment of the present application, and an execution subject of the method may be a processing unit, such as a processor, with a data processing function in the low-voltage electrical apparatus, which is not limited herein. Alternatively, the execution body of the method may be an electronic device such as a computer or a server, which may be different according to the actual application scenario. As shown in fig. 1, the method may include:

s101, acquiring switching-on and switching-off data of the piezoelectric device to be predicted in a historical time period in a plurality of switching-on and switching-off periods.

The low-voltage electrical appliance to be predicted may be any electrical appliance such as a switch, a contactor, a relay and the like working under 1200V ac voltage or 1500V dc voltage, and is not limited herein. Alternatively, the historical time period may be one week, one month, etc., and the value may be different according to the historical operating time of the piezoelectric device to be predicted. Of course, in some embodiments, the values of the historical time periods may also be customized by the user.

In some embodiments, a data acquisition unit may be set in the low-voltage to be predicted, and when the low-voltage to be predicted is switched on/off each time in the historical period, switching on/off data in each switching on/off period may be acquired by the data acquisition unit.

S102, determining the target class of the low-voltage electric appliance to be predicted based on a pre-classification model according to the opening and closing data, wherein the pre-classification model is obtained through training according to a first training sample data set.

The first training sample data set comprises first sample switching-on and switching-off data corresponding to a plurality of first sample electric appliances in a first historical time period, and corresponding categories are marked.

The type of the first sample electric appliance can be the same as the type of the low-voltage electric appliance to be predicted, for example, when the type of the low-voltage electric appliance to be predicted is a relay, the type of the first sample electric appliance can be a relay, and when the type of the low-voltage electric appliance to be predicted is a contactor, the type of the first sample electric appliance can be a contactor, so that accurate prediction results can be obtained when the target type of the low-voltage electric appliance is determined based on the trained pre-classification model.

Alternatively, the first history period may be one week of history, one month of history, three months of history, etc., without limitation. The class of the first sample electrical appliance label can represent the working condition state, the total service life state and the like of the first sample electrical appliance. Alternatively, the operating conditions may include: the working condition is normal and the working condition is abnormal, wherein the working condition is normal and can represent that the switching-on and switching-off data of the first sample electric appliance accords with preset conditions, and the working condition is abnormal and can represent that the switching-on and switching-off data of the first sample electric appliance does not accord with preset conditions. The total state of life may include: the first life state and the second life state are two states, wherein the first life state can represent that the total switching-on and switching-off times of the first sample electric appliance are smaller than a first preset threshold value; the second life state may represent that the total switching times of the first sample electrical appliance is larger than a first preset threshold, that is, the first life state may be understood as a short life state and the second life state may be understood as a long life state by comparing the two total life states. Alternatively, the value of the first preset threshold may be 800, 1000, etc., which is not limited herein, and may be different according to the actual application scenario.

It can be understood that the opening and closing data can reflect the working condition parameters of the low-voltage electric appliance to be predicted to a certain extent, so that a pre-classification model can be obtained according to the first sample opening and closing data training corresponding to a plurality of first sample electric appliances in a first historical time period, the obtained opening and closing data of the low-voltage electric appliance to be predicted is input into the pre-classification model, and the target class of the low-voltage electric appliance to be predicted is determined through the pre-classification model.

In some embodiments, if the class of the first sample electrical label includes: the working condition is normal and the working condition is abnormal, and optionally, the target class to which the low-voltage electric appliance to be predicted belongs can be the working condition abnormality. In some embodiments, if the class of the first sample electrical label includes: the first life state and the second life state, alternatively, the target class to which the piezoelectric device to be predicted belongs may be the first life state, that is, the short life state.

S103, determining a target special prediction model according to the target category of the to-be-predicted piezoelectric device.

Based on the above description, after the target class to which the low-voltage electrical appliance to be predicted belongs is determined, the target-specific prediction model can be determined according to the target class, so that the target-specific prediction model matched with the target class can be determined based on the target class, and preliminary division can be performed.

S104, predicting the residual life parameter of the low-voltage electric appliance to be predicted based on the target special prediction model according to the opening and closing data.

The target-specific prediction model can be obtained through training according to a second training sample data set, wherein the second training sample data set comprises second sample switching-on/off data corresponding to a plurality of second sample electric appliances belonging to a target class in a second historical time period, and corresponding residual life parameters are marked. The second sample electric appliance may be of the same type as the to-be-predicted electric appliance, and the second history period may be one week of history, one month of history, three months of history, or the like, which is not limited herein. For this part, reference is made to the above description of the first sample electrical apparatus, and no further description is given here. The remaining life parameter of the second sample electrical apparatus label may indicate a remaining operable duration, a remaining operable number of times, etc. of the second sample electrical apparatus, which is not limited herein.

Optionally, the second training sample data set may include a portion of the first sample data in the first training sample data set, which may increase the efficiency of obtaining the second training sample data set. Of course, in some embodiments, the second training sample data set and the first training sample data set may also be different, which is not limited herein. In addition, referring to the above description, it can be further seen that, because the target-specific prediction model is obtained through training according to the second sample switching-on/off data corresponding to the plurality of second sample electric appliances belonging to the target class, the feature data corresponding to the target class can be fully utilized, and further, when the residual life parameter of the low-voltage electric appliance to be predicted is predicted according to the target-specific prediction model, the accuracy of the prediction result can be improved, that is, the more accurate residual life parameter can be obtained.

Based on the above description, it can be understood that, since the target class to which the low-voltage electrical appliance to be predicted belongs can be determined according to the pre-classification model, the target-specific prediction model matched with the target class can be selected to predict the residual life parameter of the low-voltage electrical appliance to be predicted according to the target class, and thus, a relatively accurate prediction result can be obtained.

In summary, an embodiment of the present application provides a method for predicting a lifetime of a low-voltage electrical apparatus, including: acquiring switching-on and switching-off data of the piezoelectric device to be predicted in a historical time period in a plurality of switching-on and switching-off periods; determining the target class of the low-voltage electric appliance to be predicted based on a pre-classification model according to the opening and closing data, wherein the pre-classification model is obtained through training according to a first training sample data set; determining a target special prediction model according to the target class of the low-voltage electric appliance to be predicted; according to the switching-on and switching-off data, the residual life parameters of the low-voltage electric appliance to be predicted are predicted based on the target special prediction model, the target special prediction model is obtained through training according to the second training sample data set, the target class of the low-voltage electric appliance to be predicted can be determined according to the pre-classification model, the residual life parameters of the low-voltage electric appliance to be predicted can be predicted according to the target class by selecting the target special prediction model matched with the target class, and accuracy of a prediction result can be improved.

Optionally, determining the target-specific prediction model according to the target class to which the to-be-predicted piezoelectric device belongs includes:

and determining a target special prediction model in at least one special prediction model according to the target category to which the low-voltage electric appliance to be predicted belongs and a preset mapping relation, wherein the preset mapping relation comprises the mapping relation between the at least one category and the special prediction model.

In the life prediction method provided by the embodiment of the application, a special prediction model set and a preset mapping relation can be provided, the special prediction model set can include at least one special prediction model, the preset mapping relation can include a mapping relation between at least one category and the special prediction model, that is, after the target category of the low-voltage electric appliance to be predicted is determined, the target special prediction model matched with the low-voltage electric appliance can be determined in the at least one special prediction model according to the preset mapping relation.

For example, if the target class to which the to-be-predicted low-voltage apparatus belongs is a first lifetime state, that is, a short lifetime state, the remaining lifetime parameter of the to-be-predicted low-voltage apparatus may be determined according to a first dedicated prediction model matched with the first lifetime state, that is, a short lifetime model.

Fig. 2 is a flow chart of another life prediction method of a low-voltage electrical apparatus according to an embodiment of the present application. Optionally, as shown in fig. 2, the method further includes:

s201, acquiring a first training sample data set of a plurality of first sample electric appliances in a first historical time period in a plurality of opening and closing periods.

The first training sample data set comprises a plurality of first sample switching-on/off data, and each first sample switching-on/off data is marked with a total service life state corresponding to the first sample electric appliance. Wherein the total state of life may include: the first life state and the second life state are two states, and for this part, reference is made to the above related description, and no further description is given here.

S202, training to obtain a pre-classification model according to the first training sample data set.

Alternatively, in specific training, according to the first training sample data set, the implementation may be based on a recurrent neural network (Recurent Neural Networks, RNN), a deep neural network (Deep Neural Networks, DNN), a convolutional neural network (Convolutional Neural Networks, CNN), a Long Short-Term Memory (LSTM) network, and the like, which is not limited herein.

Fig. 3 is a flowchart of a life prediction method of a low-voltage electrical apparatus according to another embodiment of the present application. Optionally, as shown in fig. 3, the method further includes:

S301, acquiring a second training sample data set of a plurality of second sample electric appliances belonging to the target class in a second historical time period in a plurality of opening and closing periods.

The second training sample data set comprises a plurality of second sample switching-on/off data, and each second sample switching-on/off data is marked with a residual life parameter corresponding to a second sample electrical appliance.

S302, training to obtain a target special prediction model according to the second training sample data set.

In some embodiments, the second sample appliance may be the same sample appliance as the first sample appliance. The first sample opening and closing data belonging to the target class in the first training sample data set can be used as second sample opening and closing data in the second training sample data set, so that the first sample opening and closing data can be reused, and the efficiency of acquiring the first training sample data set by a user is improved. It can be understood that if the target class to which the to-be-predicted piezoelectric device belongs is the second lifetime state, that is, the long lifetime state, then a second training sample data set of a plurality of second sample electric devices belonging to the second lifetime state in a plurality of opening and closing periods may be obtained for training the target-specific prediction model.

Alternatively, the target-specific predictive model, the specific predictive model corresponding to the other categories may be implemented based on a RNN, DNN, CNN, LSTM network or the like, and the specific implementation is not limited herein. In some embodiments, the pre-classification model and the target-specific prediction model can be realized based on the same network model, so that multiplexing of the network model can be realized, the process of constructing each model by a user is simplified, and the construction efficiency of the model is improved. Of course, it should be noted that, according to different application scenarios, the pre-classification model and the target-specific prediction model may be implemented based on different network models, so that the characteristics of each network model may be fully exerted, and the accuracy of the prediction result of the present application is improved.

Of course, it should be noted that, for the specific prediction model corresponding to the other category, reference may be made to the training process of the target specific prediction model, which is not described herein. In addition, referring to the description of the relationship between the first training sample data set and the second training sample data set, optionally, the first sample switching-on/off data in the first training sample data set belonging to each category may be used as sample switching-on/off data in the corresponding training sample data set of each type of special prediction model.

Table 1 is a table of target categories to which the piezoelectric device to be predicted belongs according to the embodiment of the present application. As shown in table 1, the low-voltage electric appliance to be predicted includes 19 pieces of corresponding numbers 1 to 19, respectively, wherein the third low-voltage electric appliance to be predicted (corresponding number 3) and the seventh low-voltage electric appliance to be predicted (corresponding number 7) are taken as examples for illustration.

Based on the foregoing description, if the first preset threshold is 800, it can be seen from table 1 that the total lifetime of the third low-voltage electrical appliance to be predicted, which is predicted by using the pre-classification model, is 472.25, and the value is less than 800, so that it can be determined that the target class to which the third low-voltage electrical appliance to be predicted belongs is the second lifetime state, that is, the short lifetime class, further, a dedicated prediction model corresponding to the second lifetime state can be used to predict the remaining lifetime parameter of the third low-voltage electrical appliance, specifically, the dedicated prediction model corresponding to the second lifetime state can be used to predict and obtain the total lifetime number corresponding to the third low-voltage electrical appliance to be predicted, and then, according to the total lifetime number and the number of lifetime records in real time during the use of the third low-voltage electrical appliance to be predicted, the remaining lifetime parameter of the third low-voltage electrical appliance to be predicted can be calculated. As can be seen by combining table 1, compared with the total life time 472.25 of the third low-voltage electrical appliance to be predicted, which is predicted by using the pre-classification model, the total life time corresponding to the third low-voltage electrical appliance to be predicted, which is predicted and obtained by using the special prediction model corresponding to the second life state, is 633.85, which is closer to the actual total life time 641 of the third low-voltage electrical appliance to be predicted, which is recorded in the test process.

Furthermore, as can be seen from table 1, the total life time of the seventh low-voltage apparatus to be predicted, which is predicted by using the pre-classification model, is 1660.06, and the value is greater than 800, so that it can be determined that the target class to which the seventh low-voltage apparatus to be predicted belongs is the first life state, that is, the long life class, further, a dedicated prediction model corresponding to the first life state can be used to predict the remaining life parameter of the seventh low-voltage apparatus to be predicted, specifically, the dedicated prediction model corresponding to the first life state can be used to predict and obtain the total life time corresponding to the seventh low-voltage apparatus to be predicted first, and then, according to the total life time and the number of service lives recorded in real time in the use process of the seventh low-voltage apparatus to be predicted, the remaining life parameter of the seventh low-voltage apparatus to be predicted can be calculated. As can be seen by combining table 1, compared with the total life time 1660.06 of the seventh to-be-predicted low-voltage electric appliance predicted by the pre-classification model, the total life time of the seventh to-be-predicted low-voltage electric appliance predicted by the special prediction model corresponding to the first life state is 1348.94, which is closer to the actual total life time 1390 of the seventh to-be-predicted low-voltage electric appliance recorded in the test process.

In summary, it can be seen that the difference between the prediction results of the prediction performed by the pre-classification model and the target-specific prediction model is large, so that it is necessary to improve the accuracy of the prediction results by the cooperation of the pre-classification model and the target-specific prediction model. In the test, compared with the actual total life times of the low-voltage appliances to be predicted recorded in the test process, the average error of the total life times of the low-voltage appliances to be predicted, which are predicted by adopting a single pre-classification model, is 23.45%, and the average error of the total life times of the low-voltage appliances to be predicted, which are predicted by adopting the method provided by the embodiment of the application (i.e. predicting by combining the pre-classification model with the target special prediction model), is 14.25%, so that the accuracy of the prediction result can be effectively improved by adopting the embodiment of the application.

TABLE 1

Fig. 4 is a flow chart of another life prediction method of a low-voltage electrical apparatus according to an embodiment of the present application. Optionally, as shown in fig. 4, the acquiring the switching-on/off data of the piezoelectric device to be predicted in the historical time period in a plurality of switching-on/off periods includes:

s401, acquiring voltage data and/or current data of the piezoelectric device to be predicted in a plurality of opening and closing periods through an acquisition device.

Wherein, this collection system can include: the voltage acquisition device and/or the current acquisition device can acquire and acquire voltage data of the piezoelectric device to be predicted in a historical time period in a plurality of switching-on and switching-off periods, and can acquire and acquire current data of the piezoelectric device to be predicted in the historical time period in a plurality of switching-on and switching-off periods. According to the practical application scene, the voltage acquisition device and/or the current acquisition device can be flexibly selected to acquire voltage data and/or current data.

Optionally, the voltage acquisition device may include: the system comprises at least one acquisition end, an overvoltage protection circuit, an amplifying circuit and a signal processing circuit which are sequentially connected in series, wherein one end of the at least one acquisition end is electrically connected with a to-be-detected end of a to-be-predicted piezoelectric device, the other end of the at least one acquisition end is electrically connected with one end of the overvoltage protection circuit, and the other end of the overvoltage protection circuit is electrically connected with the amplifying circuit. The low-voltage electrical apparatus to be predicted is taken as a breaker or a contactor as an example for explanation, and the to-be-detected end of the low-voltage electrical apparatus to be predicted can be a contact or a busbar.

In some embodiments, the current collecting device may be connected in series to a loop in which the low-voltage apparatus to be predicted is located, for collecting loop current data. Alternatively, the current collecting device may include a transformer, which may be an open type transformer or a Rogowski hollow transformer, which may also be referred to as a Rogowski Coil (Rogowski Coil), and a signal operational amplifier circuit. Of course, it should be noted that, according to an actual application scenario, the transformer may also be a sensor, for example, the current collecting device may include: the hall current sensor and the signal operational amplifier circuit can be flexibly selected according to actual application scenes, and are not limited herein.

And S402, acquiring the electric life characteristics of the low-voltage electrical appliance to be predicted according to the voltage data and/or the current data, and taking the electric life characteristics as the switching-on/off data of the low-voltage electrical appliance to be predicted.

Based on the above description, after acquiring the voltage data and/or the current data, the electrical life characteristic of the low-voltage electrical appliance to be predicted can be acquired, so that the working state parameter of the low-voltage electrical appliance to be predicted can be reflected in a multi-dimensional manner based on the electrical life characteristic, the electrical life characteristic is used as the opening and closing data of the low-voltage electrical appliance to be predicted, and then a more accurate prediction result can be obtained when the target class of the low-voltage electrical appliance to be predicted and the residual life parameter of the low-voltage electrical appliance to be predicted are determined based on the opening and closing data.

Optionally, the electrical life characteristics of the low-voltage electrical appliance to be predicted may include at least one of: contact voltage, contact current, contact resistance, arcing time, arcing energy, and arcing power.

Wherein, the contact voltage represents the voltage at two ends of the low-voltage electrical appliance during the closing period; contact current: the current across the low voltage electrical appliance during closing; contact resistance, which represents the ratio of the effective value of the contact voltage to the effective value of the contact current; the arcing time is the time from the opening of the movable contact to the extinction of the arc, and is illustrated by taking a relay as an example, and represents the moment when the fixed contact of the relay is separated to generate the arc; the arcing energy represents that the low-voltage electrical appliance generates an electric arc between contacts at the moment of power failure (such as the moment of switching off), and the integral of the arcing power in the arcing time is the arcing energy; arcing power, which is the product of the voltage across the electrical appliance and the current across the electrical appliance when the electrical appliance is in an arc during the arcing time.

Of course, it should be noted that, in some embodiments, other electrical lifetime characteristics may be obtained based on the voltage data and/or the current data, or may be obtained according to the type of the piezoelectric device, which is not limited herein.

Fig. 5 is a flowchart of a life prediction method of a low-voltage electrical apparatus according to another embodiment of the present application. Optionally, as shown in fig. 5, after training to obtain the pre-classification model according to the first training sample data set, the method further includes:

s501, acquiring a third training sample data set of a plurality of first sample electric appliances in a third historical time period in a plurality of opening and closing periods.

The third training sample data set comprises a plurality of third sample switching-on/off data, each third sample switching-on/off data is marked with a total life parameter corresponding to the first sample electric appliance, and the third historical time period and the first historical time period are different time periods. Alternatively, the third history period may be a period of time after or before the first history period, which is not limited herein. Of course, the length of the first history period and the third history period is not limited herein, alternatively, the first history period may be one week of history, the third history period may be one month of history, and the like, and may be flexibly set according to an actual application scenario.

S502, updating the pre-classification model according to the third training sample data set.

In some embodiments, it is contemplated that when the pre-classification model is obtained based only on training of the first training sample data set over the first historical period of time, the accuracy of the prediction of the obtained pre-classification model is not high. Optionally, a third training sample data set may be further obtained, and the pre-classification model is updated according to the third training sample data set, so as to implement correction of the pre-classification model. Based on the description, it can be understood that accuracy can be improved when the target class to which the low-voltage electric appliance to be predicted belongs is further determined according to the updated pre-classification model.

Of course, the application is not limited to the timing of each update, and may be updated periodically according to an actual application scenario, for example, after the first sample electrical apparatus performs 100 times of opening and closing operations each time, opening and closing data corresponding to the 100 times of opening and closing operations may be obtained as a third training sample data set to act on updating of the pre-classification model. Of course, the specific update method is not limited thereto, and may be a timing update, for example, update once every week.

It should be noted that, according to an actual application scenario, the switching-on/off data of the low-voltage electrical apparatus to be predicted can also be obtained in real time, at this time, the switching-on/off data of the low-voltage electrical apparatus to be predicted in a plurality of switching-on/off periods and the switching-on/off data obtained in real time in a historical period can be combined, and the classification condition of the low-voltage electrical apparatus to be predicted is determined through a pre-classification model, so that a more accurate target class can be obtained according to more and newer switching-on/off data, and further a more accurate prediction result can be obtained when the residual life parameter of the low-voltage electrical apparatus to be predicted is determined based on the target class. The following description is made in connection with specific experiments.

Fig. 6 is a schematic diagram of an electrical life prediction result of a to-be-predicted piezoelectric device after classification based on a first opening and closing data set according to an embodiment of the present application; fig. 7 is a schematic diagram of an electrical life prediction result of a to-be-predicted piezoelectric device after classification based on a second opening and closing data set according to an embodiment of the present application. The first opening and closing data set includes a plurality of first opening and closing data sets and a plurality of second opening and closing data sets, the plurality of first opening and closing data sets are opening and closing data corresponding to a low-voltage electrical appliance to be predicted in a first historical time period (for example, 8:00 to 12:00 am), and the plurality of second opening and closing data sets are opening and closing data corresponding to a low-voltage electrical appliance to be predicted after the first historical time period (for example, 1:00 to 3:00 pm) and can be understood as opening and closing data which are acquired recently. As shown in fig. 6 and 7, the abscissa indicates the number of times the low-voltage electrical appliance to be predicted has been used, the ordinate indicates the ratio of the number of times the low-voltage electrical appliance to be predicted remains to the total life time, the ratio is 0-1, S4 indicate the relationship between the first ratio of the number of times the low-voltage electrical appliance to be predicted remains to the total life time in the actual test and the number of times the low-voltage electrical appliance to be predicted has been used, and it can be seen from the graph that S1 and S4 are straight lines, that is, the ratio of the number of times the low-voltage electrical appliance to be predicted remains to the total life time and the number of times the low-voltage electrical appliance to be predicted has been used are in a linear relationship; s3, based on the first opening and closing data set in the application process, determining a second ratio of the residual times of the to-be-predicted low-voltage electric appliance to the total service life times and the relation between the residual times of the to-be-predicted low-voltage electric appliance used times by adopting the pre-classification model provided by the embodiment of the application; s2 is a fitting straight line obtained by fitting the S3 curve.

With continued reference to fig. 7, S6 represents a relationship between a third ratio of the number of times of remaining low-voltage electric appliances to be predicted to the total life time determined by using the pre-classification model provided by the embodiment of the present application and the number of times of used low-voltage electric appliances to be predicted based on the second opening and closing data set in the application process, and S5 is a fitting straight line obtained by fitting the S6 curve. As can be seen from fig. 6 and 7, the difference between the S3 curve and the S6 curve is larger, and the test proves that the S6 curve is closer to the real working condition, that is, the fitting straight line S5 corresponding to S6 is more accurate. That is, when the remaining life parameter of the low-voltage electrical apparatus to be predicted is determined based on the second opening and closing data set, the result is more accurate, that is, the remaining life parameter of the low-voltage electrical apparatus to be predicted is determined according to more and newer opening and closing data, so that the accuracy of the prediction result can be improved.

Optionally, the at least one category comprises: a first total life span and a second total life span, wherein the first total life span is less than the second total life span.

In some embodiments, each category may specifically correspond to a total lifetime range, and in some embodiments, at least one category may include: a first total life span and a second total life span, wherein the first total life span is less than the second total life span. For example, the first total lifetime may be the second lifetime state, i.e. the short lifetime state, and optionally, the first total lifetime may be 600-800 times; the second total lifetime may be the first lifetime state, i.e. the long lifetime state, and optionally, the second total lifetime may be 800-1500 times. Based on the description, that is to say, if the total service life number of a certain low-voltage electrical appliance to be predicted is 700 times based on the pre-classification model, the class corresponding to the low-voltage electrical appliance to be predicted is known to be the first total service life range through judgment.

Of course, the application is not limited to the number of categories herein, and alternatively, in some embodiments, may include multiple total life ranges of 3, 5, etc.

Fig. 8 is a schematic diagram of a functional module of a life prediction device for a low-voltage electrical appliance according to an embodiment of the present application, where the basic principle and the technical effects of the device are the same as those of the foregoing corresponding method embodiment, and for brevity, no parts are mentioned in this embodiment, and reference may be made to corresponding contents in the method embodiment. As shown in fig. 8, the lifetime prediction apparatus 100 includes:

an obtaining module 110, configured to obtain opening and closing data of the piezoelectric device to be predicted in a historical time period in a plurality of opening and closing periods;

the first determining module 120 is configured to determine, according to the opening and closing data, a target class to which the low-voltage electrical appliance to be predicted belongs based on a pre-classification model, where the pre-classification model is obtained by training according to a first training sample data set, and the first training sample data set includes first sample opening and closing data corresponding to a plurality of first sample electrical appliances in a first historical time period, and marks the corresponding class;

a second determining module 130, configured to determine a target-specific prediction model according to a target class to which the to-be-predicted piezoelectric device belongs;

The prediction module 140 is configured to predict, according to the opening and closing data, a remaining life parameter of the low-voltage electrical appliance to be predicted based on the target-specific prediction model, where the target-specific prediction model is obtained by training according to a second training sample data set, and the second training sample data set includes second sample opening and closing data corresponding to a plurality of second sample electrical appliances belonging to the target class in a second historical time period, and marks the corresponding remaining life parameter.

In an optional implementation manner, the second determining module 130 is specifically configured to determine, in at least one dedicated prediction model, a target dedicated prediction model according to a target class to which the low-voltage electrical appliance to be predicted belongs and a preset mapping relationship, where the preset mapping relationship includes a mapping relationship between at least one class and the dedicated prediction model.

In an alternative embodiment, the lifetime prediction apparatus further includes: the first training module is used for acquiring a first training sample data set of a plurality of first sample electric appliances in a plurality of opening and closing periods in a first historical time period, wherein the first training sample data set comprises a plurality of first sample opening and closing data, and each first sample opening and closing data is marked with a total service life state corresponding to the first sample electric appliances;

And training to obtain the pre-classification model according to the first training sample data set.

In an alternative embodiment, the lifetime prediction apparatus further includes: the second training module is used for acquiring a second training sample data set of a plurality of second sample electric appliances belonging to the target class in a second historical time period in a plurality of opening and closing periods, wherein the second training sample data set comprises a plurality of second sample opening and closing data, and each second sample opening and closing data is marked with a residual life parameter corresponding to the second sample electric appliance;

and training to obtain the target special prediction model according to the second training sample data set.

In an optional embodiment, the acquiring module 110 is specifically configured to acquire, by using an acquiring device, voltage data and/or current data of the to-be-predicted piezoelectric device in a plurality of switching-on/off periods in the historical time period;

and acquiring the electric life characteristic of the low-voltage electrical appliance to be predicted according to the voltage data and/or the current data, and taking the electric life characteristic as the switching-on/off data of the low-voltage electrical appliance to be predicted.

In an optional embodiment, the first training module is further configured to obtain a third training sample data set of the plurality of first sample appliances in a third historical time period in a plurality of opening and closing periods, where the third training sample data set includes a plurality of third sample opening and closing data, each third sample opening and closing data is labeled with a total lifetime state corresponding to the first sample appliance, and the third historical time period is a different time period from the first historical time period;

And updating the pre-classification model according to the third training sample data set.

In an alternative embodiment, at least one of the categories includes: a first total life span and a second total life span, wherein the first total life span is less than the second total life span.

In an alternative embodiment, the electrical lifetime characteristics of the electrical device to be predicted include at least one of: contact voltage, contact current, contact resistance, arcing time, arcing energy, and arcing power.

The foregoing apparatus is used for executing the method provided in the foregoing embodiment, and its implementation principle and technical effects are similar, and are not described herein again.

The above modules may be one or more integrated circuits configured to implement the above methods, for example: one or more application specific integrated circuits (Application Specific Integrated Circuit, abbreviated as ASICs), or one or more microprocessors, or one or more field programmable gate arrays (Field Programmable Gate Array, abbreviated as FPGAs), etc. For another example, when a module above is implemented in the form of a processing element scheduler code, the processing element may be a general-purpose processor, such as a central processing unit (Central Processing Unit, CPU) or other processor that may invoke the program code. For another example, the modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

Fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application, where the electronic device may be integrated in the lifetime prediction apparatus described above. As shown in fig. 9, the electronic device may include: processor 210, storage medium 220, and bus 230, storage medium 220 storing machine-readable instructions executable by processor 210, processor 210 executing machine-readable instructions to perform steps of the method embodiments described above when the electronic device is operating, processor 210 communicating with storage medium 220 via bus 230. The specific implementation manner and the technical effect are similar, and are not repeated here.

Optionally, the present application further provides a storage medium, on which a computer program is stored, which when being executed by a processor performs the steps of the above-described method embodiments. The specific implementation manner and the technical effect are similar, and are not repeated here.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in hardware plus software functional units.

The integrated units implemented in the form of software functional units described above may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (english: processor) to perform part of the steps of the methods of the embodiments of the application. And the aforementioned storage medium includes: u disk, mobile hard disk, read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.

It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.

The above is only a preferred embodiment of the present application, and is not intended to limit the present application, but various modifications and variations can be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application. It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. The above is only a preferred embodiment of the present application, and is not intended to limit the present application, but various modifications and variations can be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims (10)

1. A life prediction method of a low-voltage electrical apparatus, comprising:

acquiring switching-on and switching-off data of the piezoelectric device to be predicted in a historical time period in a plurality of switching-on and switching-off periods;

determining a target class to which the low-voltage electrical appliance to be predicted belongs based on a pre-classification model according to the opening and closing data, wherein the pre-classification model is obtained through training according to a first training sample data set, the first training sample data set comprises first sample opening and closing data corresponding to a plurality of first sample electrical appliances in a first historical time period, and corresponding classes are marked;

determining a target special prediction model according to the target class of the to-be-predicted piezoelectric device;

and predicting the residual life parameter of the low-voltage electrical appliance to be predicted based on the target special prediction model according to the opening and closing data, wherein the target special prediction model is obtained through training according to a second training sample data set, the second training sample data set comprises second sample opening and closing data corresponding to a plurality of second sample electrical appliances belonging to the target class in a second historical time period, and corresponding residual life parameters are marked.

2. The method according to claim 1, wherein the determining the target-specific prediction model according to the target class to which the to-be-predicted piezoelectric device belongs comprises:

Determining a target special prediction model in at least one special prediction model according to the target category to which the to-be-predicted piezoelectric device belongs and a preset mapping relation, wherein the preset mapping relation comprises a mapping relation between at least one category and the special prediction model.

3. The method according to claim 1, wherein the method further comprises:

acquiring a first training sample data set of a plurality of first sample electric appliances in a plurality of opening and closing periods in a first historical time period, wherein the first training sample data set comprises a plurality of first sample opening and closing data, and each first sample opening and closing data is marked with a total service life state corresponding to the first sample electric appliances;

and training to obtain the pre-classification model according to the first training sample data set.

4. The method according to claim 1, wherein the method further comprises:

acquiring a second training sample data set of a plurality of second sample electric appliances belonging to the target class in a second historical time period in a plurality of opening and closing periods, wherein the second training sample data set comprises a plurality of second sample opening and closing data, and each second sample opening and closing data is marked with a residual life parameter corresponding to the second sample electric appliance;

And training to obtain the target special prediction model according to the second training sample data set.

5. The method of claim 1, wherein obtaining the switching on/off data of the piezoelectric device to be predicted in the historical period over a plurality of switching on/off cycles comprises:

acquiring voltage data and/or current data of the piezoelectric device to be predicted in a plurality of opening and closing periods in the historical time period through an acquisition device;

and acquiring the electric life characteristic of the low-voltage electrical appliance to be predicted according to the voltage data and/or the current data, and taking the electric life characteristic as the switching-on/off data of the low-voltage electrical appliance to be predicted.

6. The method of claim 3, wherein said training to obtain said pre-classification model based on said first training sample data set further comprises:

acquiring a third training sample data set of a plurality of first sample electric appliances in a plurality of opening and closing periods in a third historical time period, wherein the third training sample data set comprises a plurality of third sample opening and closing data, each third sample opening and closing data is marked with a total service life state corresponding to the first sample electric appliances, and the third historical time period and the first historical time period are different time periods;

And updating the pre-classification model according to the third training sample data set.

7. The method of claim 2, wherein at least one of the categories comprises: a first total life span and a second total life span, wherein the first total life span is less than the second total life span.

8. The method of claim 5, wherein the electrical lifetime characteristics of the electrical device to be predicted include at least one of: contact voltage, contact current, contact resistance, arcing time, arcing energy, and arcing power.

9. A life predicting apparatus for a low-voltage electric appliance, comprising:

the acquisition module is used for acquiring switching-on/off data of the piezoelectric device to be predicted in a historical time period in a plurality of switching-on/off periods;

the first determining module is used for determining the target class to which the low-voltage electrical appliance to be predicted belongs based on a pre-classification model according to the opening and closing data, the pre-classification model is obtained through training according to a first training sample data set, the first training sample data set comprises first sample opening and closing data corresponding to a plurality of first sample electrical appliances in a first historical time period, and the corresponding class is marked;

The second determining module is used for determining a target special prediction model according to the target category of the to-be-predicted piezoelectric device;

the prediction module is used for predicting the residual life parameter of the low-voltage electrical appliance to be predicted based on the target special prediction model according to the opening and closing data, the target special prediction model is obtained through training according to a second training sample data set, the second training sample data set comprises second sample opening and closing data corresponding to a plurality of second sample electrical appliances belonging to the target class in a second historical time period, and the corresponding residual life parameter is marked.

10. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of the lifetime prediction method of a low voltage electrical appliance according to any one of claims 1-8.