Electric meter box fault prediction method and device, computer device and storage medium

Abstract

The application relates to an electric meter box fault prediction method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a plurality of groups of historical acquisition data of electric meter box terminals; determining an initial value of a linear change rate threshold parameter and an initial value of an elastic network hybrid parameter based on the plurality of groups of historical acquisition data and a target function; determining an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network hybrid parameter and the target function; acquiring current acquisition data of the electric meter box terminals; determining an initial prediction result according to the current acquisition data and the initial fault prediction function; adjusting the initial value of the elastic network hybrid parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function; and determining a fault prediction result of the electric meter box based on the adjusted fault prediction function. The method can improve the accuracy of the electric meter box fault prediction result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power detection, in particular to an electric meter box fault prediction method and device, computer equipment, storage medium and computer program product. BACKGROUND

[0002] The electric meter box is a basic infrastructure for installing electric meters, switches, wires and other equipment, which plays an important role in power transmission, metering and centralized control and management of electric meters, and can effectively guarantee the safety and convenience of electricity use. Due to the high frequency of use of electric meter box terminals for electric energy metering and other work requirements, the internal volume of the electric meter box is small and the environment is complex, which leads to the risk of aging and fire of the electric meter box terminals due to work heating and external environment, causing serious fire consequences, affecting normal electric energy metering and energy scheduling, and threatening the safety of residents' lives. Therefore, it is urgent to accurately predict the fault of the electric meter box and timely find safety hazards.

[0003] The traditional electric meter box fault prediction method mainly detects the state data of the electric meter box terminals and predicts the fault of the electric meter box according to artificial experience. However, the internal environment of the electric meter box is complex, and the existing method of predicting the fault of the electric meter box according to artificial experience cannot adapt to the dynamic environment inside the electric meter box, resulting in low accuracy of the fault prediction result. SUMMARY

[0004] Therefore, it is necessary to provide an electric meter box fault prediction method, device, computer equipment, computer readable storage medium and computer program product capable of improving the accuracy of the electric meter box fault prediction result to solve the technical problem of low accuracy of the traditional electric meter box fault prediction result.

[0005] In a first aspect, the present application provides an electric meter box fault prediction method. The method comprises:

[0006] obtaining a plurality of groups of historical acquisition data of electric meter box terminals;

[0007] determining an initial value of the linear change rate threshold parameter and an initial value of the elastic network hybrid parameter based on the plurality of groups of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic network hybrid parameter; and determining an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network hybrid parameter and the target function;

[0008] obtaining current acquisition data of the electric meter box terminals;

[0009] determining an initial prediction result according to the current acquisition data and the initial fault prediction function; and adjusting the initial value of the elastic network hybrid parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function;

[0010] determine a fault prediction result of the meter box based on the adjusted fault prediction function.

[0011] In one of the embodiments, the current acquisition data of the terminals of the meter box is obtained, including:

[0012] acquiring real-time data of each time within a preset time length of the terminals of the meter box;

[0013] acquiring a first linear change rate between the real-time data of each time within the preset time length relative to the real-time data of a starting time within the preset time length, and taking the first linear change rate of an ending time within the preset time length relative to the starting time as a target change rate;

[0014] if the absolute value of the difference between any first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter, updating the real-time data corresponding to the corresponding first linear change rate as preset data;

[0015] determining the current acquisition data based on the updated real-time data.

[0016] In one of the embodiments, the initial value of the elastic network mixing parameter and the linear change rate threshold parameter is adjusted based on the initial prediction result, to obtain an adjusted fault prediction function, including:

[0017] adjusting the initial value of the elastic network mixing parameter based on the initial prediction result and the initial value of the linear change rate threshold parameter, to obtain a first fault prediction function;

[0018] obtaining a first prediction result of the next time of the ending time according to the first fault prediction function and the current acquisition data, adjusting the initial value of the linear change rate threshold parameter based on the first prediction result, to obtain the adjusted fault prediction function.

[0019] In one of the embodiments, the initial value of the elastic network mixing parameter is adjusted based on the initial prediction result and the initial value of the linear change rate threshold parameter, to obtain a first fault prediction function, including:

[0020] acquiring a second linear change rate between the initial prediction result and the real-time data of the starting time within the preset time length;

[0021] determining the absolute value of the difference between the second linear change rate and the target change rate as a change rate difference value;

[0022] if the change rate difference value is greater than the initial value of the linear change rate threshold parameter, adjusting the initial value of the elastic network mixing parameter based on the change rate difference value and the initial value of the linear change rate threshold parameter, to obtain the first fault prediction function.

[0023] In one of the embodiments, based on the first prediction result, the initial value of the linear change rate threshold parameter is adjusted to obtain an adjusted fault prediction function, including:

[0024] Based on the current acquisition data and the first prediction result, a first prediction sample data is determined, which is used to predict the data of the next moment of the moment corresponding to the first prediction result;

[0025] Based on the first prediction sample data and the first fault prediction function, the prediction process is continued until the number of predictions reaches a first preset number, and a first prediction number of first target prediction results is obtained;

[0026] A second linear change rate between each first target prediction result and the real-time data of the starting moment in the first prediction sample data is obtained respectively;

[0027] The second linear change rates corresponding to the respective first prediction results are averaged to obtain an average change rate;

[0028] Based on the average change rate and the target change rate, the initial value of the linear change rate threshold parameter is adjusted to obtain an adjusted fault prediction function.

[0029] In one of the embodiments, based on the adjusted fault prediction function, the fault prediction result of the electric meter box is determined, including:

[0030] According to the adjusted fault prediction function and the current acquisition data, a second prediction result is determined;

[0031] Based on the current acquisition data and the second prediction result, a second prediction sample data is determined;

[0032] Based on the second prediction sample data and the adjusted fault prediction function, the prediction process is continued until the number of predictions reaches a second preset number, and a second prediction number of prediction results is obtained;

[0033] It is determined whether the second prediction number of prediction results meets a preset fault condition;

[0034] If all meet, it is determined that the fault prediction result of the electric meter box is the electric meter box fault.

[0035] In a second aspect, the present application also provides an electric meter box fault prediction device. The device comprises:

[0036] A historical data acquisition module is configured to acquire a plurality of sets of historical acquisition data of the terminals of the electric meter box;

[0037] An initial function determining module is configured to determine an initial value of the linear change rate threshold parameter and an initial value of the elastic network mixing parameter based on a plurality of sets of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic network mixing parameter, and determine an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the target function.

[0038] A current data obtaining module is configured to obtain current acquisition data of the terminal of the meter box.

[0039] An adjusted function obtaining module is configured to determine an initial prediction result according to the current acquisition data and the initial fault prediction function, and adjust the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function.

[0040] A result determining module is configured to determine a fault prediction result of the meter box based on the adjusted fault prediction function.

[0041] In a third aspect, the present application further provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0042] Obtain a plurality of sets of historical acquisition data of the terminal of the meter box.

[0043] Determine an initial value of the linear change rate threshold parameter and an initial value of the elastic network mixing parameter based on the plurality of sets of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic network mixing parameter, and determine an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the target function.

[0044] Obtain current acquisition data of the terminal of the meter box.

[0045] Determine an initial prediction result according to the current acquisition data and the initial fault prediction function, and adjust the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function.

[0046] Determine a fault prediction result of the meter box based on the adjusted fault prediction function.

[0047] In a fourth aspect, the present application further provides a computer readable storage medium. The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0048] Obtain a plurality of sets of historical acquisition data of the terminal of the meter box.

[0049] determine the initial value of the linear change rate threshold parameter and the initial value of the elastic network mixing parameter based on the multiple sets of historical acquisition data and the objective function containing the linear change rate threshold parameter and the elastic network mixing parameter; and determine an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the objective function;

[0050] acquire current acquisition data of the terminal of the electric meter box;

[0051] determine an initial prediction result according to the current acquisition data and the initial fault prediction function; and adjust the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function;

[0052] determine the fault prediction result of the electric meter box based on the adjusted fault prediction function.

[0053] In a fifth aspect, the present application further provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the following steps:

[0054] acquire multiple sets of historical acquisition data of the terminal of the electric meter box;

[0055] determine the initial value of the linear change rate threshold parameter and the initial value of the elastic network mixing parameter based on the multiple sets of historical acquisition data and the objective function containing the linear change rate threshold parameter and the elastic network mixing parameter; and determine an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the objective function;

[0056] acquire current acquisition data of the terminal of the electric meter box;

[0057] determine an initial prediction result according to the current acquisition data and the initial fault prediction function; and adjust the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function;

[0058] determine the fault prediction result of the electric meter box based on the adjusted fault prediction function.

[0059] The meter box fault prediction method, device, computer equipment, storage medium and computer program product can determine the initial value of the linear change rate threshold parameter and the initial value of the elastic network hybrid parameter based on the multiple sets of historical collection data and a target function containing the linear change rate threshold parameter and the elastic network hybrid parameter, and determine the initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network hybrid parameter and the target function. The initial fault prediction function determined based on the historical collection data can be used for fault prediction of the meter box, which is beneficial to improving the accuracy of the fault prediction result. The current collection data of the meter box terminal is obtained, and the initial prediction result is determined according to the current collection data and the initial fault prediction function. The initial value of the elastic network hybrid parameter and the linear change rate threshold parameter is adjusted based on the initial prediction result, and the adjusted fault prediction function is obtained. The adjusted fault prediction function is obtained by adjusting the initial value of the elastic network hybrid parameter and the linear change rate threshold parameter based on the initial prediction result predicted by the initial fault prediction function. The accuracy of the fault prediction result is further improved based on the adjusted fault prediction function. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 An application environment diagram of the meter box fault prediction method in an embodiment;

[0061] Figure 2 A flowchart of the meter box fault prediction method in an embodiment;

[0062] Figure 3 A sub-flowchart of S203 in an embodiment;

[0063] Figure 4 A sub-flowchart of S204 in an embodiment;

[0064] Figure 5 A sub-flowchart of S402 in an embodiment;

[0065] Figure 6 A sub-flowchart of S404 in an embodiment;

[0066] Figure 7 A sub-flowchart of S205 in an embodiment;

[0067] Figure 8 A general flowchart of the meter box fault prediction method in an embodiment;

[0068] Figure 9 A diagram of the step of obtaining the current collection data in an embodiment;

[0069] Figure 10 A structure block diagram of the electric meter box fault prediction device in an embodiment;

[0070] Figure 11 An internal structure diagram of the computer device in an embodiment. DETAILED DESCRIPTION

[0071] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0072] The electric meter box fault prediction method provided by the embodiments of the present application can be applied to an application environment as shown in Figure 1 . The terminal 102 communicates with the electric meter box 104 through a network. The electric meter box fault prediction method provided by the embodiments of the present application can be executed by the terminal 102 or the server alone, or can be executed by the terminal 102 and the server in cooperation. Taking the case of being executed by the terminal 102 alone as an example: the terminal 102 acquires a plurality of groups of historical acquisition data of the terminals of the electric meter box; based on the plurality of groups of historical acquisition data and a target function containing a linear change rate threshold parameter and an elastic network mixing parameter, the initial value of the linear change rate threshold parameter and the initial value of the elastic network mixing parameter are determined; according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the target function, an initial fault prediction function is determined; current acquisition data of the terminals of the electric meter box are acquired; according to the current acquisition data and the initial fault prediction function, an initial prediction result is determined; based on the initial prediction result, the initial value of the elastic network mixing parameter and the linear change rate threshold parameter is adjusted to obtain an adjusted fault prediction function; based on the adjusted fault prediction function, a fault prediction result of the electric meter box is determined. The terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things device can be a smart speaker, a smart television, a smart air conditioner, a smart vehicle-mounted device, etc. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, etc. The server can be implemented by an independent server or a server cluster composed of multiple servers.

[0073] In an embodiment, as shown in Figure 2 , an electric meter box fault prediction method is provided. Taking the terminal 102 in Figure 1 as an example, the method includes the following steps:

[0074] S201, a plurality of groups of historical acquisition data of the terminals of the electric meter box are acquired.

[0075] The meter box terminal refers to a terminal in the meter box. The historical acquisition data is historical data collected by the terminal in a historical time period. Each set of historical acquisition data includes at least one historical acquisition data. The type of the historical acquisition data includes at least one of a voltage flowing through the meter box terminal, a current flowing through the meter box terminal, or a temperature of the meter box terminal. The terminal obtains multiple sets of historical acquisition data of the meter box terminal through a sensor. For example, multiple sets of historical voltage acquisition data of the meter box terminal are obtained through a voltage sensor, multiple sets of historical current acquisition data of the meter box terminal are obtained through a current sensor, and multiple sets of historical temperature acquisition data of the meter box terminal are obtained through a temperature sensor.

[0076] In S202, initial values of the linear change rate threshold parameter and the elastic network mixing parameter are determined based on the multiple sets of historical acquisition data and a target function including the linear change rate threshold parameter and the elastic network mixing parameter. An initial fault prediction function is determined according to the initial values of the linear change rate threshold parameter and the elastic network mixing parameter and the target function.

[0077] The target function is a function for meter box fault prediction. The target function includes the linear change rate threshold parameter and the elastic network mixing parameter. The terminal inputs each set of historical acquisition data in the multiple sets of historical acquisition data into the target function respectively, and obtains the initial values of the linear change rate threshold parameter and the elastic network mixing parameter. The terminal inputs the obtained initial values of the linear change rate threshold parameter and the elastic network mixing parameter into the target function, and obtains the initial fault prediction function. The initial fault prediction function is determined based on the historical acquisition data and the target function, and is used for meter box fault prediction.

[0078] In S203, current acquisition data of the meter box terminal is obtained.

[0079] The current acquisition data is real-time data of the meter box terminal collected by the terminal in a current time period. The type of the current acquisition data includes at least one type in the types of the historical acquisition data. The terminal obtains the current acquisition data of the meter box terminal through a sensor.

[0080] In S204, an initial prediction result is determined according to the current acquisition data and the initial fault prediction function. The initial values of the elastic network mixing parameter and the linear change rate threshold parameter are adjusted based on the initial prediction result, and an adjusted fault prediction function is obtained.

[0081] The terminal determines the initial prediction result according to the current acquisition data and the initial fault prediction function. Specifically, the terminal inputs the current acquisition data into the initial fault prediction function, and obtains a result as the initial prediction result. The initial prediction result is prediction data of a next time of a collection time of the current acquisition data.

[0082] The terminal adjusts the initial values of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result, to obtain an adjusted elastic network mixing parameter and an adjusted linear change rate threshold parameter. The adjusted elastic network mixing parameter and the adjusted linear change rate threshold parameter are brought into the initial fault prediction function to obtain an adjusted fault prediction function. The adjusted fault prediction function is used to obtain the fault prediction result of the meter box.

[0083] S205, determining the fault prediction result of the meter box based on the adjusted fault prediction function.

[0084] The terminal performs fault prediction by using the adjusted fault prediction function to obtain prediction data. Illustratively, the terminal performs at least one fault prediction by using the adjusted fault prediction function to obtain at least one prediction data. The terminal performs big data statistics on the at least one prediction data, and determines the fault prediction result of the meter box based on the at least one prediction data. Alternatively, in a case where the at least one prediction data is greater than a preset value, the terminal determines that the fault prediction result of the meter box is a meter box fault. Alternatively, in a case where an average of the at least one prediction data is greater than a preset average value, the terminal determines that the fault prediction result of the meter box is a meter box fault. Alternatively, the fault prediction result of the meter box is determined based on a median and a mode of the at least one prediction data.

[0085] In the above meter box fault prediction method, a plurality of groups of historical acquisition data of terminals of a meter box are obtained, the initial value of the linear change rate threshold parameter and the initial value of the elastic network mixing parameter are determined based on the plurality of groups of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic network mixing parameter, and the initial fault prediction function is determined according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the target function. The initial fault prediction function determined based on the historical acquisition data can be used for fault prediction of the meter box, which is beneficial to improving the accuracy of the fault prediction result. The current acquisition data of the terminals of the meter box are obtained, the initial prediction result is determined according to the current acquisition data and the initial fault prediction function, the initial values of the elastic network mixing parameter and the linear change rate threshold parameter are adjusted based on the initial prediction result to obtain an adjusted fault prediction function. The adjusted fault prediction function is obtained by adjusting the initial values of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result predicted by the initial fault prediction function, the fault prediction result of the meter box is determined based on the adjusted fault prediction function, and the accuracy of the fault prediction result is further improved.

[0086] In one embodiment, as shown in FIG. 2, the current acquisition data of the terminals of the meter box are obtained, including: Figure 3

[0087] ​S302, collect real-time data of the terminal of the electric meter box at each time within a preset time length.

[0088] The terminal collects real-time data of the terminal of the electric meter box at each time within a preset time length through a sensor. The current collected data includes at least one real-time data. For example, the terminal collects temperature real-time data of the terminal of the electric meter box at each time within a preset time length through a temperature sensor. The terminal collects voltage real-time data of the terminal of the electric meter box at each time within a preset time length through a voltage sensor. The terminal collects current real-time data of the terminal of the electric meter box at each time within a preset time length through a current sensor.

[0089] S304, obtain the first linear change rate between the real-time data at each time within the preset time length relative to the starting time within the preset time length, and take the first linear change rate of the terminal of the electric meter box at the ending time within the preset time length relative to the starting time as the target change rate.

[0090] The real-time data at the starting time is the real-time data collected at the starting time among the real-time data at each time within the preset time length. The ending time is the ending time among the real-time data at each time within the preset time length.

[0091] The terminal obtains the first linear change rate between the real-time data at each time within the preset time length relative to the real-time data at the starting time within the preset time length. The first linear change rate is used to represent the change degree of the real-time data at each time within the preset time length relative to the starting time. The linear change rate calculation formula is:

[0092]

[0093] Wherein, The first linear change rate is represented by t0, which represents the starting time within the preset time length, and t i The real-time data at each time within the preset time length is represented by t The real-time data at the starting time within the preset time length is represented by t0. The real-time data at each time within the preset time length is represented by t

[0094] The terminal takes the first linear change rate of the terminal of the electric meter box at the ending time within the preset time length relative to the starting time as the target change rate, which can be represented as:

[0095]

[0096] Wherein, The target change rate is represented by t n The ending time within the preset time length is represented by t The real-time data at the ending time within the preset time length is represented by t

[0097] S306, if the absolute value of the difference between any first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter, the real-time data corresponding to the first linear change rate is updated to preset data; based on the updated real-time data, the current acquisition data is determined.

[0098] The terminal obtains the difference between each first linear change rate and the target change rate. If the absolute value of the difference between any first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter, the real-time data corresponding to the first linear change rate is updated to preset data. In some embodiments, the preset data can be data in the historical acquisition data that meets the linear change rate requirement.

[0099] The terminal determines the current acquisition data based on the updated real-time data. Specifically, the terminal collectively uses the updated real-time data and the un-updated data in the real-time data at each time point in the preset time period as the current acquisition data.

[0100] In this embodiment, by acquiring the real-time data of the terminal of the electric meter box at each time point in the preset time period, the first linear change rate corresponding to each time point is obtained, the real-time data when the absolute value of the difference between the first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter is updated to preset data, and the current acquisition data is determined based on the updated real-time data. This can ensure that any data in the current acquisition data meets the linear change rate requirement, i.e., the absolute value of the difference between the first linear change rate corresponding to any data in the current acquisition data and the target change rate is not greater than the initial value of the linear change rate threshold parameter. Since the aging of the terminal of the electric meter box is not a sudden process, the change of the current acquisition data is usually linear, therefore, this method of determining the current acquisition data can eliminate data that does not meet the linear change rate requirement, avoid the influence of abnormal data collected by the sensor on the fault prediction result, and is beneficial to improving the accuracy of the fault prediction result.

[0101] In one embodiment, as shown in FIG. 4, based on the initial prediction result, the initial value of the elastic network mixing parameter and the initial value of the linear change rate threshold parameter are adjusted to obtain an adjusted fault prediction function, including: Figure 4

[0102] S402, based on the initial prediction result and the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter is adjusted to obtain a first fault prediction function.

[0103] ​In order to improve the accuracy and precision of the fault prediction result, the initial value of the elastic network mixing parameter can also be adjusted, so as to obtain the first fault prediction function. In some embodiments, the terminal adjusts the initial value of the elastic network mixing parameter according to the initial prediction result and the initial value of the linear change rate threshold parameter by obtaining the linear change rate of the initial prediction result relative to the current collected data. The terminal inputs the adjusted initial value of the elastic network mixing parameter into the initial fault prediction function, and obtains the first fault prediction function. The first fault function is used for fault prediction of the meter box.

[0104] S404, according to the first fault prediction function and the current collected data, a first prediction result of the next moment of the cutoff time is predicted, and the initial value of the linear change rate threshold parameter is adjusted based on the first prediction result, to obtain an adjusted fault prediction function.

[0105] In order to further improve the accuracy and precision of the fault prediction result, the initial value of the linear change rate threshold parameter can also be adjusted, to obtain an adjusted fault prediction function.

[0106] The terminal inputs the current collected data into the first fault prediction function, and obtains a first prediction result. The first prediction result is the predicted data of the next moment of the cutoff time within a preset time length. The terminal adjusts the initial value of the linear change rate threshold parameter based on the first prediction result, to obtain an adjusted fault prediction function. In some embodiments, the terminal performs multiple predictions based on the first prediction result and the first fault prediction function, adjusts the initial value of the linear change rate threshold parameter based on the linear change rate of the results of the multiple predictions, and obtains an adjusted initial value of the linear change rate threshold parameter. The adjusted initial value of the linear change rate threshold parameter is input into the first fault prediction function, to obtain an adjusted fault prediction function.

[0107] In this embodiment, the initial value of the elastic network mixing parameter is adjusted according to the initial prediction result and the initial value of the linear change rate threshold parameter, to obtain the first fault prediction function; according to the first fault prediction function and the current collected data, a first prediction result of the next moment of the cutoff time is predicted, and the initial value of the linear change rate threshold parameter is adjusted based on the first prediction result, to obtain an adjusted fault prediction function. This adjustment of the initial value of the elastic network mixing parameter based on the initial prediction result and the initial value of the linear change rate threshold parameter is conducive to improving the accuracy and precision of the fault prediction result; and the adjustment of the initial value of the linear change rate threshold parameter based on the prediction result of the first fault prediction function is conducive to further improving the accuracy and precision of the fault prediction result.

[0108] In one embodiment, as Figure 5As shown, based on the initial prediction result and the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter is adjusted to obtain a first fault prediction function, including:

[0109] S502, obtaining a second linear change rate between the initial prediction result and real-time data at the starting moment within the preset time length.

[0110] Wherein, the terminal brings the initial prediction result and the real-time data at the starting moment within the preset time length into the linear change rate calculation formula, and calculates the result as the second linear change rate. The second linear change rate is used to represent the change degree of the initial prediction result and the real-time data at the starting moment within the preset time length.

[0111] S504, determining the absolute value of the difference between the second linear change rate and the target change rate as a change rate difference value.

[0112] Wherein, the real-time data of the terminal at the starting moment within the preset time length is subtracted from the target change rate, and the absolute value of the difference obtained is taken as the change rate difference value.

[0113] S506, if the change rate difference value is greater than the initial value of the linear change rate threshold parameter, adjusting the initial value of the elastic network mixing parameter based on the change rate difference value and the initial value of the linear change rate threshold parameter to obtain the first fault prediction function.

[0114] Wherein, the terminal compares the change rate difference value with the initial value of the linear change rate threshold parameter, and if the change rate difference value is greater than the initial value of the linear change rate threshold parameter, adjusts the initial value of the elastic network mixing parameter based on the change rate difference value and the initial value of the linear change rate threshold parameter to obtain the first fault prediction function. Exemplarily, the initial value of the elastic network mixing parameter should satisfy the following conditions:

[0115]

[0116] Wherein, ζ represents the initial value of the elastic network mixing parameter, represents the change rate difference value, and a max represents the initial value of the linear change rate threshold parameter. The change rate difference value greater than the initial value of the linear change rate threshold parameter represents that the prediction accuracy of the initial prediction result is not high, and the initial value of the elastic network mixing parameter can be increased to obtain the first fault prediction function, so as to improve the accuracy of the fault prediction result.

[0117] In some other embodiments, if the change rate difference value is not greater than the initial value of the linear change rate threshold parameter, returning to the step of determining the initial prediction result according to the current collected data and the initial fault prediction function, and continuing to predict. That is, the prediction accuracy and precision of the initial fault prediction function are high, and the parameter adjustment of the initial fault prediction function is not needed.

[0118] In this embodiment, by obtaining the change rate difference between the second linear change rate and the target change rate between the initial prediction result and the real-time data at the starting moment within the preset time length, and adjusting the initial value of the elastic network mixing parameter in the case where the change rate difference is greater than the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter is adjusted to obtain the first fault prediction function when the prediction accuracy of the initial prediction function is not high, which is beneficial to improve the accuracy and accuracy of the fault prediction result.

[0119] In one embodiment, as shown in Figure 6 Based on the first prediction result, the initial value of the linear change rate threshold parameter is adjusted to obtain an adjusted fault prediction function, including:

[0120] S602, based on the current acquisition data and the first prediction result, determining the first prediction sample data, the first prediction sample data being used for predicting the data of the next moment of the moment corresponding to the first prediction result.

[0121] The terminal determines the first prediction sample data based on the current acquisition data and the first prediction result, and the first prediction sample data is used for predicting the data of the next moment of the moment corresponding to the first prediction result. Illustratively, the corresponding acquisition moments of the current acquisition data are the first moment to the fifth moment, the moment corresponding to the first prediction result is the sixth moment, the first prediction sample data is the data of the second moment to the sixth moment, and the first prediction sample data is used for predicting the data of the seventh moment.

[0122] S604, based on the first prediction sample data and the first fault prediction function, continuing to perform the prediction process until the prediction number reaches the first preset number, obtaining the first target prediction result of the first prediction number.

[0123] The terminal continues to perform the prediction process based on the first prediction sample data and the first fault prediction function until the prediction number reaches the first preset number, and obtains the first target prediction result of the first prediction number. Illustratively, the terminal inputs the first prediction sample data into the first fault prediction function, and obtains the result as the prediction data of the seventh moment. The data from the second moment to the seventh moment is taken as the second sample prediction data and input into the first fault prediction function, and the obtained result is taken as the prediction data of the eighth moment. According to such a prediction method, the first target prediction result of the first prediction number is obtained until the prediction number reaches the first preset number. The first target prediction result is the prediction data obtained each time.

[0124] S606, respectively, obtain a second linear change rate between each first target prediction result and real-time data of the starting time in the first prediction sample data; average the second linear change rates corresponding to each first target prediction result to obtain an average change rate.

[0125] In the preset number of times, the terminal obtains a second linear change rate between each first target prediction result and real-time data of the starting time in the first prediction sample data. That is, each first target prediction result and real-time data of the starting time in the first prediction sample data are brought into the linear change rate calculation formula to obtain the second linear change rate. The terminal averages the second linear change rates corresponding to each first target prediction result to obtain an average change rate. Exemplarily, the calculation formula of the average change rate is as follows:

[0126]

[0127] In the preset number of times, the terminal obtains a second linear change rate between each first target prediction result and real-time data of the starting time in the first prediction sample data. That is, each first target prediction result and real-time data of the starting time in the first prediction sample data are brought into the linear change rate calculation formula to obtain the second linear change rate. The terminal averages the second linear change rates corresponding to each first target prediction result to obtain an average change rate. Exemplarily, the calculation formula of the average change rate is as follows: The average change rate is represented by P, t k The prediction time in the preset number of times is represented by k, The first target prediction result in the preset number of times is represented by k, and t'0 represents the starting time in the first prediction sample data, The real-time data of the starting time in the first prediction sample data.

[0128] S608, based on the average change rate and the target change rate, adjusting the initial value of the linear change rate threshold parameter to obtain an adjusted fault prediction function.

[0129] In the preset number of times, the terminal obtains a second linear change rate between each first target prediction result and real-time data of the starting time in the first prediction sample data. That is, each first target prediction result and real-time data of the starting time in the first prediction sample data are brought into the linear change rate calculation formula to obtain the second linear change rate. The terminal averages the second linear change rates corresponding to each first target prediction result to obtain an average change rate. Exemplarily, the calculation formula of the average change rate is as follows:

[0130]

[0131] wherein, γ represents an adjustment coefficient. The terminal brings the average change rate and the target change rate into the linear threshold adjustment formula to obtain an initial value of the adjusted linear change rate threshold parameter. The terminal brings the initial value of the adjusted linear change rate threshold parameter into the first fault prediction function to obtain the adjusted fault prediction function.

[0132] In this embodiment, the prediction process is continued by using the first prediction sample data determined based on the current collected data and the first prediction result and the first fault prediction function, and the first target prediction result of the first prediction quantity is obtained. The initial value of the linear change rate threshold parameter is adjusted based on the average change rate and the target change rate determined based on the second linear change rate corresponding to each first target prediction result, and the adjusted fault prediction function is obtained. This adjustment of the initial value of the linear change rate threshold parameter based on the average change rate corresponding to the prediction result of the first fault prediction function is beneficial to improve the accuracy and precision of the fault prediction result.

[0133] In one embodiment, as shown in Figure 7 the fault prediction result of the electric meter box is determined based on the adjusted fault prediction function, including:

[0134] S702, determining a second prediction result according to the adjusted fault prediction function and the current collected data.

[0135] wherein, the terminal inputs the current collected data into the adjusted fault prediction function, and the obtained result is taken as the second prediction result. The second prediction result is used to represent the prediction data of the next moment of the current collected data. The second prediction result predicted based on the current collected data and the second prediction result has higher accuracy and precision than the initial prediction result.

[0136] S704, determining second prediction sample data based on the current collected data and the second prediction result.

[0137] wherein, the terminal determines the second prediction sample data based on the current collected data and the second prediction result. Exemplarily, the corresponding collection time of the current collected data is the first time to the fifth time, the time corresponding to the second prediction result is the sixth time, the second prediction sample data is the data of the second time to the sixth time, and the second prediction sample data is used to predict the data of the seventh time.

[0138] S706, continuing to perform the prediction process based on the second prediction sample data and the adjusted fault prediction function until the prediction number reaches a second preset number, and obtaining a prediction result of a second prediction quantity.

[0139] The terminal continues to perform the prediction process based on the second prediction sample data and the adjusted fault prediction function until the number of predictions reaches a second preset number, to obtain a second number of prediction results.

[0140] In S708, it is determined whether the second number of prediction results satisfy a preset fault condition. If all the second number of prediction results satisfy the preset fault condition, it is determined that the fault prediction result of the meter box is a meter box fault.

[0141] In S708, the terminal determines whether the second number of prediction results satisfy a preset fault condition. Illustratively, the preset fault condition can be that the prediction result is greater than a preset threshold. If each of the second number of prediction results is greater than the preset threshold, it is determined that the fault prediction result of the meter box is a meter box fault.

[0142] In this embodiment, the second prediction result is predicted by using the adjusted fault prediction function and the current collected data, the second prediction sample data is determined, the prediction process is continued based on the second prediction sample data and the adjusted fault prediction function, and when the second number of prediction results all satisfy the preset fault condition, it is determined that the fault prediction result of the meter box is a meter box fault. This can ensure the accuracy of the meter box fault result, avoid the deviation of the one-time fault result from leading to a judgment error, and be conducive to improving the accuracy of the fault prediction result.

[0143] In one embodiment, based on a plurality of sets of historical collected data and a target function containing a linear change rate threshold parameter and an elastic network mixing parameter, an initial value of the linear change rate threshold parameter and an initial value of the elastic network mixing parameter are determined, including: for each set of historical collected data in the plurality of sets of historical collected data, inputting the current set of historical collected data into the target function to determine a target fault prediction function corresponding to the current set of historical collected data; based on the target fault prediction function corresponding to each historical sample set, obtaining a minimum value in the function values of each target fault prediction function; taking the value of the linear change rate threshold parameter corresponding to the minimum value as the initial value of the linear change rate threshold parameter, and taking the value of the elastic network mixing parameter corresponding to the minimum value as the initial value of the elastic network mixing parameter.

[0144] For each set of historical collected data in the plurality of sets of historical collected data, the terminal inputs the current set of historical collected data into the target function to determine a target fault prediction function corresponding to the current set of historical collected data. Based on the target fault prediction function corresponding to each historical sample set, a minimum value in the function values of each target fault prediction function is obtained. The value of the linear change rate threshold parameter corresponding to the minimum value is taken as the initial value of the linear change rate threshold parameter, and the value of the elastic network mixing parameter corresponding to the minimum value is taken as the initial value of the elastic network mixing parameter. Illustratively, the target function can be represented by the following formula:

[0145]

[0146] wherein, N represents the number of each set of historical acquisition data, the data corresponding to the time point as the prediction value, and the other data as the sample data; β0represents the average value of the Euclidean distance between the sample data and the prediction value; β represents the Euclidean distance between the sample data and the prediction value; x i represents the acquisition time corresponding to each historical acquisition data; y i represents the historical acquisition data, that is, any one of the terminal temperature, the voltage flowing through the terminal or the current flowing through the terminal; λ represents a complex coefficient, λ can be obtained through data experiments according to historical experimental experience, and λ controls the degree of penalty, wherein 0 represents no penalty, and ∞ represents complete penalty; α max represents a linear change rate threshold parameter; represents a complex parameter adjustment term based on the fault prediction result of the meter box, when the prediction result does not meet the linear change rate, is greater than 1, the penalty degree needs to be improved, and the influence of each historical acquisition data on the prediction result needs to be improved, when the prediction result meets the linear threshold, is less than 1, the penalty degree needs to be reduced, and the interpretability of the objective function needs to be increased; ζ (0≤ζ≤1) represents a elastic network mixing parameter, which controls the degree to which the objective function is ridge regression or Lasso (Least absolute shrinkage and selection operator) regression, ζ equal to 0 represents that the objective function is complete ridge regression, and ζ equal to 1 represents that the objective function is complete Lasso regression; represents a ridge regression term; ||β|| l1 represents a Lasso term. The ridge regression term is the square root of the sum of squares of each sample data corresponding to β, and the Lasso term is the sum of absolute values of each sample data corresponding to β.

[0147] In some embodiments, the historical acquisition data of each data type corresponds to an objective function, and the objective functions corresponding to each data type are the same. The terminal inputs the historical acquisition data of each data type into the respective corresponding objective function, and can train the respective corresponding fault prediction function of each data type. When the results predicted by each data type respectively corresponding fault prediction function meet the fault condition, the terminal determines that the fault prediction result of the meter box is the meter box fault, which can further improve the accuracy of the meter box fault prediction result.

[0148] In this embodiment, by using multiple sets of historical data and an objective function that includes a linear rate of change threshold parameter and an elastic network hybrid parameter, the value of the linear rate of change threshold parameter corresponding to the minimum value is used as the initial value of the linear rate of change threshold parameter, and the value of the elastic network hybrid parameter corresponding to the minimum value is used as the initial value of the elastic network hybrid parameter, in order to determine the initial fault prediction function, which is beneficial to improving the accuracy and precision of the fault prediction results.

[0149] In one embodiment, to illustrate the meter box fault prediction method and its effects in this solution in detail, the following is a detailed description of a specific embodiment:

[0150] The terminal acquires multiple sets of historical data from the meter box terminals. The types of historical data include at least one of the following: voltage flowing through the meter box terminals, current flowing through the meter box terminals, or temperature of the meter box terminals. In some embodiments, the multiple sets of historical data record the voltage flowing through the meter box terminals five minutes before the hour during normal operation of the meter box within the previous week. Current flowing through the terminals of the meter box Temperature of meter box terminals Historical data can be specifically represented as:

[0151]

[0152] Where m represents the historical data collected from the m-th meter box terminal, and t represents the collection time. to These represent historical data collected from Monday to Sunday. Taking Monday's historical data as an example, It is expressed as follows:

[0153]

[0154] In some embodiments, the terminal also acquires multiple sets of fault acquisition data for the meter box terminals. The fault acquisition data includes the voltage flowing through the meter box terminals per minute for the five minutes preceding the fault occurrence within a six-month historical period. Current flowing through the terminals of the meter box and the temperature of the meter box terminals That is, when a fault is determined to have occurred, data from the 5 minutes prior to the fault is extracted from the currently collected data and used as the fault collection data.

[0155] Based on multiple sets of historical data and an objective function that includes a linear rate of change threshold parameter and elastic network hybrid parameters, the initial values ​​of the linear rate of change threshold parameter and the elastic network hybrid parameters are determined. Based on these initial values ​​and the objective function, an initial fault prediction function is determined. For example, the objective function can be expressed by the following formula:

[0156]

[0157] wherein N represents the number of each set of historical acquisition data, the data corresponding to the time point as the prediction value, and the other data as the sample data; β0represents the average of the Euclidean distance between the sample data and the prediction value; β represents the Euclidean distance between the sample data and the prediction value; x i represents the acquisition time corresponding to each historical acquisition data; y i represents the historical acquisition data, that is, any one of the terminal temperature, the voltage flowing through the terminal, or the current flowing through the terminal; λ represents a complex coefficient, which can be obtained through data experiments and historical experimental experience, and λ controls the degree of penalty, wherein 0 represents no penalty, and ∞ represents complete penalty; α max represents a linear change rate threshold parameter; represents a complex parameter adjustment term based on the fault prediction result of the electric meter box, when the prediction result does not meet the linear change rate, is greater than 1, the penalty degree needs to be improved, and the influence of each historical acquisition data on the prediction result needs to be improved, when the prediction result meets the linear threshold, is less than 1, the penalty degree needs to be reduced, and the interpretability of the objective function needs to be increased; ζ (0≤ζ≤1) represents a elastic network mixing parameter, which controls to what extent the objective function is ridge regression or Lasso (Least absolute shrinkage and selection operator) regression, ζ equal to 0 represents that the objective function is complete ridge regression, and ζ equal to 1 represents that the objective function is complete Lasso regression; represents a ridge regression term; ||β|| l1 represents a Lasso term. The ridge regression term is the square root of the sum of squares of each sample data corresponding to β, and the Lasso term is the sum of absolute values of each sample data corresponding to β.

[0158] In some embodiments, the historical acquisition data of each data type corresponds to an objective function, and the objective functions corresponding to each data type are the same. The terminal inputs the historical acquisition data of each data type into the respective corresponding objective function, and can train the respective corresponding fault prediction function of each data type. When the results predicted by each data type respectively corresponding fault prediction function meet the fault condition, the terminal determines that the fault prediction result of the electric meter box is the electric meter box fault, which can further improve the accuracy of the electric meter box fault prediction result.

[0159] As Figure 8 shown is a general flowchart of an electric meter box fault prediction method. The current acquisition data of the electric meter box terminal is obtained. As Figure 9The acquisition step of the current acquisition data is shown. The terminal acquires the real-time data of the terminal of the electric meter box at each time within the preset time length. Exemplarily, the real-time data of the terminal acquisition at each time within the preset time length is to record the voltage of each minute in the past 5 minutes current terminal temperature The real-time data at each time within the preset time length is acquired, the first linear change rate between the real-time data relative to the starting time within the preset time length is calculated, and the first linear change rate of the ending time within the preset time length relative to the starting time is taken as the target change rate. If the absolute value of the difference between any first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter, the real-time data corresponding to the corresponding first linear change rate is updated as the preset data, and the current acquisition data is determined based on the updated real-time data. The linear change rate calculation formula is:

[0160]

[0161] wherein, the first linear change rate, t0 represents the starting time within the preset time length, t i each time within the preset time length, the real-time data of the starting time within the preset time length, the real-time data of each time within the preset time length.

[0162] The terminal brings the real-time data at each time within the preset time length and the real-time data of the starting time within the preset time length into the linear change rate calculation formula to obtain the first linear change rate.

[0163] The terminal takes the first linear change rate of the ending time within the preset time length relative to the starting time as the target change rate, which can be represented as:

[0164]

[0165] wherein, the target change rate, t n the ending time within the preset time length, The real-time data represents a cut-off time within a preset time length. When the absolute value of the difference between the first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter, it indicates that the deviation of the real-time data collected by the terminal at the corresponding time within the preset time length is too large. The terminal filters the data that meets the linear change rate requirement from the historical collected data as preset data, replaces the real-time data with too large deviation with the preset data, and thus determines the current collected data. Since the aging of the terminal of the electric meter box is not a sudden process, the change of the current collected data is usually linear, and thus this method of determining the current collected data can eliminate the data that does not meet the linear change rate requirement, avoid the influence of abnormal data collected by the sensor on the fault prediction result, and is beneficial to improving the accuracy of the fault prediction result.

[0166] The terminal determines an initial prediction result according to the current collected data and the initial fault prediction function. Based on the initial prediction result, the initial value of the elastic network mixing parameter and the linear change rate threshold parameter are adjusted to obtain an adjusted fault prediction function.

[0167] The initial value of the elastic network mixing parameter is adjusted based on the initial prediction result and the initial value of the linear change rate threshold parameter to obtain a first fault prediction function. Specifically, the terminal obtains a second linear change rate between the initial prediction result and the real-time data at the starting time within the preset time length. The absolute value of the difference between the second linear change rate and the target change rate is determined as a change rate difference value, and if the change rate difference value is greater than the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter is adjusted based on the change rate difference value and the initial value of the linear change rate threshold parameter to obtain the first fault prediction function. Exemplarily, the initial value of the elastic network mixing parameter should satisfy the following condition:

[0168]

[0169] wherein ζ represents the initial value of the elastic network mixing parameter, The change rate difference value, α max The initial value of the linear change rate threshold parameter. The change rate difference value greater than the initial value of the linear change rate threshold parameter indicates that the prediction accuracy of the initial prediction result is not high, and the initial value of the elastic network mixing parameter can be increased to obtain the first fault prediction function, so as to improve the accuracy of the fault prediction result.

[0170] In other embodiments, if the change rate difference value is not greater than the initial value of the linear change rate threshold parameter, the step of determining the initial prediction result according to the current collected data and the initial fault prediction function is returned, and the prediction continues. That is, the prediction accuracy and precision of the initial fault prediction function are high, and there is no need to adjust the parameters of the initial fault prediction function.

[0171] The first prediction result of the next moment of the cutoff time is predicted according to the first fault prediction function and the current acquisition data, and the initial value of the linear change rate threshold parameter is adjusted based on the first prediction result to obtain an adjusted fault prediction function. Specifically, the first prediction sample data is determined based on the current acquisition data and the first prediction result, and the first prediction sample data is used to predict the data of the next moment of the time corresponding to the first prediction result. The prediction process is continued based on the first prediction sample data and the first fault prediction function until the prediction number reaches a first preset number, and a first prediction number of first target prediction results is obtained. A second linear change rate between each first target prediction result and the real-time data of the starting time in the first prediction sample data is obtained, and the average of the second linear change rates corresponding to each first target prediction result is obtained to obtain the average change rate. The calculation formula of the average change rate is as follows:

[0172]

[0173] wherein, represents the average change rate, P represents the preset number, t k represents the prediction time of the kth time in the preset number, represents the first target prediction result of the kth time in the preset number, t'0 represents the starting time in the first prediction sample data, the real-time data of the starting time in the first prediction sample data.

[0174] The initial value of the linear change rate threshold parameter is adjusted based on the average change rate and the target change rate to obtain the adjusted fault prediction function. The linear threshold adjustment formula is as follows:

[0175]

[0176] wherein, γ represents an adjustment coefficient. The terminal brings the average change rate and the target change rate into the linear threshold adjustment formula to obtain the initial value of the adjusted linear change rate threshold parameter. The initial value of the adjusted linear change rate threshold parameter is brought into the first fault prediction function to obtain the adjusted fault prediction function.

[0177] The fault prediction result of the electric meter box is determined based on the adjusted fault prediction function. Specifically, the terminal determines a second prediction result according to the adjusted fault prediction function and the current acquisition data. The second prediction sample data is determined based on the current acquisition data and the second prediction result, and the prediction process is continued based on the second prediction sample data and the adjusted fault prediction function until the prediction number reaches a second preset number, and a second prediction number of prediction results is obtained. It is determined whether the second prediction number of prediction results meets the preset fault condition, and if they all meet, it is determined that the fault prediction result of the electric meter box is the electric meter box fault.

[0178] The aforementioned meter box fault prediction method acquires multiple sets of historical data from the meter box terminals. Based on this data and an objective function containing a linear rate of change threshold parameter and elastic network hybrid parameters, it determines the initial values ​​of the linear rate of change threshold parameter and the elastic network hybrid parameters. Then, based on these initial values ​​and the objective function, it determines an initial fault prediction function. This initial fault prediction function, determined from historical data, can be used for meter box fault prediction, improving the accuracy of the prediction results. Next, it acquires current data from the meter box terminals and, based on this data and the initial fault prediction function, determines the initial prediction result. Based on this result, it adjusts the initial values ​​of the elastic network hybrid parameters and the linear rate of change threshold parameter to obtain an adjusted fault prediction function. This adjusted function, obtained by adjusting the initial values ​​of the elastic network hybrid parameters and the linear rate of change threshold parameter based on the initial prediction result, further improves the accuracy of the fault prediction results.

[0179] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0180] Based on the same inventive concept, this application also provides a meter box fault prediction device for implementing the meter box fault prediction method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more meter box fault prediction device embodiments provided below can be found in the limitations of the meter box fault prediction method described above, and will not be repeated here.

[0181] In one embodiment, such as Figure 10 As shown, a meter box fault prediction device 100 is provided, including: a historical data acquisition module 110, an initial function determination module 120, a current data acquisition module 130, an adjustment function acquisition module 140, and a result determination module 150, wherein:

[0182] The historical data acquisition module 110 is configured to acquire a plurality of sets of historical acquisition data of the terminal of the meter box.

[0183] The initial function determination module 120 is configured to determine an initial value of the linear change rate threshold parameter and an initial value of the elastic network mixing parameter based on the plurality of sets of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic network mixing parameter, and determine an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the target function.

[0184] The current data acquisition module 130 is configured to acquire current acquisition data of the terminal of the meter box.

[0185] The adjusted function acquisition module 140 is configured to determine an initial prediction result according to the current acquisition data and the initial fault prediction function, and adjust the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function.

[0186] The result determination module 150 is configured to determine a fault prediction result of the meter box based on the adjusted fault prediction function.

[0187] The above-mentioned meter box fault prediction device, by acquiring a plurality of sets of historical acquisition data of the terminal of the meter box, determining an initial value of the linear change rate threshold parameter and an initial value of the elastic network mixing parameter based on the plurality of sets of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic network mixing parameter, and determining an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the target function. The initial fault prediction function determined based on the historical acquisition data can be used for fault prediction of the meter box, which is beneficial to improve the accuracy of the fault prediction result. The current acquisition data of the terminal of the meter box is acquired, and an initial prediction result is determined according to the current acquisition data and the initial fault prediction function. The initial value of the elastic network mixing parameter and the linear change rate threshold parameter is adjusted based on the initial prediction result to obtain an adjusted fault prediction function. The adjusted fault prediction function is obtained by adjusting the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result predicted by the initial fault prediction function. The accuracy of the fault prediction result is further improved based on the adjusted fault prediction function to determine the fault prediction result of the meter box.

[0188] In one embodiment, in terms of acquiring the current collection data of the terminal of the electric meter box, the current data acquisition module 130 is further configured to: collect real-time data of the terminal of the electric meter box at each time within a preset time period; acquire a first linear change rate between the real-time data at each time within the preset time period relative to the real-time data at a starting time within the preset time period, and take the first linear change rate of a terminal time relative to the starting time within the preset time period as a target change rate; if the absolute value of the difference between any first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter, update the real-time data corresponding to the corresponding first linear change rate to preset data; and determine the current collection data based on the updated real-time data.

[0189] In one embodiment, in terms of adjusting the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function, the adjustment function acquisition module 140 is further configured to: adjust the initial value of the elastic network mixing parameter based on the initial prediction result and the initial value of the linear change rate threshold parameter to obtain a first fault prediction function; predict a first prediction result of the next time of the terminal time according to the first fault prediction function and the current collection data, and adjust the initial value of the linear change rate threshold parameter based on the first prediction result to obtain the adjusted fault prediction function.

[0190] In one embodiment, in terms of adjusting the initial value of the elastic network mixing parameter based on the initial prediction result and the initial value of the linear change rate threshold parameter to obtain a first fault prediction function, the adjustment function acquisition module 140 is further configured to: acquire a second linear change rate between the initial prediction result and the real-time data at the starting time within the preset time period; determine the absolute value of the difference between the second linear change rate and the target change rate as a change rate difference; and if the change rate difference is greater than the initial value of the linear change rate threshold parameter, adjust the initial value of the elastic network mixing parameter based on the change rate difference and the initial value of the linear change rate threshold parameter to obtain the first fault prediction function.

[0191] In one of the embodiments, in the aspect of adjusting the initial value of the linear change rate threshold parameter based on the first prediction result to obtain the adjusted fault prediction function, the adjustment function obtaining module 140 is further configured to: determine first prediction sample data based on the current collected data and the first prediction result, the first prediction sample data being used to predict the data of the next moment of the moment corresponding to the first prediction result; continue to perform the prediction process based on the first prediction sample data and the first fault prediction function until the number of predictions reaches a first preset number to obtain a first prediction number of first target prediction results; obtain a second linear change rate between each first target prediction result and the real-time data of the starting moment in the first prediction sample data; average the second linear change rates corresponding to the first prediction results to obtain an average change rate; and adjust the initial value of the linear change rate threshold parameter based on the average change rate and the target change rate to obtain the adjusted fault prediction function.

[0192] In one of the embodiments, in the aspect of determining the fault prediction result of the meter box based on the adjusted fault prediction function, the result determining module 150 is configured to: determine a second prediction result according to the adjusted fault prediction function and the current collected data; determine second prediction sample data based on the current collected data and the second prediction result; continue to perform the prediction process based on the second prediction sample data and the adjusted fault prediction function until the number of predictions reaches a second preset number to obtain a second prediction number of prediction results; determine whether the second prediction number of prediction results meet a preset fault condition; and if all meet, determine that the fault prediction result of the meter box is the meter box fault.

[0193] The above-mentioned modules in the meter box fault prediction device can be realized by software, hardware and combinations thereof in whole or in part. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned modules.

[0194] In one of the embodiments, a computer device is provided, which can be a terminal, and the internal structure diagram thereof can be as shown in Figure 11As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be realized through WIFI, mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by the processor to realize an electric meter box fault prediction method.

[0195] Those skilled in the art can understand that, Figure 11 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0196] In one embodiment, a computer device is provided, including a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the steps in each of the above method embodiments.

[0197] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to realize the steps in each of the above method embodiments.

[0198] In one embodiment, a computer program product is provided, including a computer program, and the computer program is executed by a processor to realize the steps in each of the above method embodiments.

[0199] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0200] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0201] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0202] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method of predicting failure of an electrical meter box, the method comprising: The method comprises: acquiring a plurality of groups of historical acquisition data of a meter box terminal; determining an initial value of a linear change rate threshold parameter and an initial value of an elastic network mixing parameter based on the plurality of groups of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic network mixing parameter, and determining an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic network mixing parameter and the target function; acquiring current acquisition data of the meter box terminal; determining an initial prediction result according to the current acquisition data and the initial fault prediction function, adjusting the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function; determining a fault prediction result of the meter box based on the adjusted fault prediction function; the adjusting of the initial value of the elastic network mixing parameter and the linear change rate threshold parameter based on the initial prediction result to obtain the adjusted fault prediction function comprises: adjusting the initial value of the elastic network mixing parameter based on the initial prediction result and the initial value of the linear change rate threshold parameter to obtain a first fault prediction function; predicting a first prediction result of a next moment at a cutoff moment within a preset time period according to the first fault prediction function and the current acquisition data, and adjusting the initial value of the linear change rate threshold parameter based on the first prediction result to obtain the adjusted fault prediction function; the adjusting of the initial value of the linear change rate threshold parameter based on the first prediction result to obtain the adjusted fault prediction function comprises: performing multiple predictions based on the first prediction result and the first fault prediction function, adjusting the initial value of the linear change rate threshold parameter based on a linear change rate of results of the multiple predictions to obtain an adjusted initial value of the linear change rate threshold parameter, and bringing the adjusted initial value of the linear change rate threshold parameter into the first fault prediction function to obtain the adjusted fault prediction function.

2. The method of claim 1, wherein, the acquiring of the current acquisition data of the meter box terminal comprises: acquiring real-time data of each moment within a preset time period of the meter box terminal; acquiring a first linear change rate between real-time data of each moment within the preset time period relative to real-time data of a starting moment within the preset time period, and taking a first linear change rate of a cutoff moment within the preset time period relative to the starting moment as a target change rate; if an absolute value of a difference between any first linear change rate and the target change rate is greater than the initial value of the linear change rate threshold parameter, updating real-time data corresponding to the corresponding first linear change rate to preset data; determining the current acquisition data based on the updated real-time data.

3. The method of claim 2, wherein, the adjusting of the initial value of the elastic network mixing parameter based on the initial prediction result and the initial value of the linear change rate threshold parameter to obtain the first fault prediction function comprises: acquiring a second linear change rate between the initial prediction result and real-time data of a starting moment within the preset time period; determine an absolute value of a difference between the second linear change rate and the target change rate as a change rate difference value; if the change rate difference value is greater than an initial value of the linear change rate threshold parameter, adjust the initial value of the elastic net mixing parameter based on the change rate difference value and the initial value of the linear change rate threshold parameter to obtain a first fault prediction function.

4. The method of claim 2, wherein, the adjusting the initial value of the linear change rate threshold parameter based on the first prediction result to obtain an adjusted fault prediction function comprises: based on the current acquisition data and the first prediction result, determine first prediction sample data, the first prediction sample data being used to predict data of a next time point of a time point corresponding to the first prediction result; continue to perform a prediction process based on the first prediction sample data and the first fault prediction function until a prediction number reaches a first preset number to obtain a first target prediction result of the first prediction number; respectively obtain a second linear change rate between each first target prediction result and real-time data of a starting time point in the first prediction sample data; average the second linear change rates corresponding to the respective first prediction results to obtain an average change rate; adjust the initial value of the linear change rate threshold parameter based on the average change rate and the target change rate to obtain an adjusted fault prediction function.

5. The method of claim 1, wherein, the determining the fault prediction result of the electric meter box based on the adjusted fault prediction function comprises: determine a second prediction result according to the adjusted fault prediction function and the current acquisition data; determine second prediction sample data based on the current acquisition data and the second prediction result; continue to perform a prediction process based on the second prediction sample data and the adjusted fault prediction function until a prediction number reaches a second preset number to obtain a prediction result of the second prediction number; determine whether the prediction result of the second prediction number meets a preset fault condition; if all meet, determine that the fault prediction result of the electric meter box is an electric meter box fault.

6. A meter box failure prediction apparatus characterized by comprising: the device comprises: a historical data acquisition module configured to acquire a plurality of groups of historical acquisition data of a terminal of an electric meter box; an initial function determination module configured to determine an initial value of a linear change rate threshold parameter and an initial value of an elastic net mixing parameter based on the plurality of groups of historical acquisition data and a target function containing the linear change rate threshold parameter and the elastic net mixing parameter, and determine an initial fault prediction function according to the initial value of the linear change rate threshold parameter, the initial value of the elastic net mixing parameter and the target function; a current data acquisition module configured to acquire current acquisition data of the terminal of the electric meter box; an adjusted function acquisition module configured to determine an initial prediction result according to the current acquisition data and the initial fault prediction function, and adjust the initial value of the elastic net mixing parameter and the initial value of the linear change rate threshold parameter based on the initial prediction result to obtain an adjusted fault prediction function; a result determination module configured to determine a fault prediction result of the electric meter box based on the adjusted fault prediction function. The adjustment function obtaining module is further configured to adjust the initial value of the elastic network hybrid parameter based on the initial prediction result and the initial value of the linear change rate threshold parameter to obtain a first fault prediction function; predict a first prediction result of a next moment at a cutoff moment within a preset time length according to the first fault prediction function and the current acquisition data, and adjust the initial value of the linear change rate threshold parameter based on the first prediction result to obtain an adjusted fault prediction function; The adjustment function obtaining module is further configured to perform multiple predictions based on the first prediction result and the first fault prediction function, adjust the initial value of the linear change rate threshold parameter based on a linear change rate of results of the multiple predictions to obtain an adjusted initial value of the linear change rate threshold parameter, and input the adjusted initial value of the linear change rate threshold parameter into the first fault prediction function to obtain an adjusted fault prediction function. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6 when the computer program is executed by the processor. The processor, when executing the computer program, implements the steps of the method of any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 5.

Images