Alarm method and alarm system of distribution box and distribution box

Abstract

The invention provides an alarm method and an alarm system for a distribution box and the distribution box. The alarm method comprises the following steps: acquiring acquisition data of acquisition equipment; the acquisition data comprises an acquisition equipment identifier, a real-time data value and a corresponding timestamp; inputting the acquisition equipment identifier and the timestamp into a pre-trained threshold value model, and outputting a current threshold value; judging whether the real-time data value exceeds a current threshold value or not; if so, determining that the distribution box has a fault, and generating first alarm information; if not, inputting the acquisition equipment identifier and the real-time data value into a pre-trained data prediction model, and outputting a fault prediction result; and if the fault prediction result comprises the fault level, generating second alarm information based on the fault level. In the mode, by adopting the alarm method based on the threshold model and the data prediction model, the operation data of the acquisition equipment can be acquired in real time, and the data are analyzed and predicted, so that the fault judgment standard is dynamically adjusted, and the potential fault risk is identified in advance.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution boxes, and in particular to an alarm method, an alarm system and a distribution box for a distribution box. Background Art

[0002] With the development of modern power systems, distribution boxes, as core equipment for power distribution and management, are widely used in industrial, commercial and household scenarios, and the requirements for their safety and intelligence levels are increasing. Existing distribution box alarm methods mostly rely on simple threshold judgments or manual inspections to judge faults, but these methods have problems such as slow response and insufficient accuracy, making it difficult to identify potential fault risks in real time and accurately. Summary of the invention

[0003] In view of this, the object of the present invention is to provide an alarm method, an alarm system and a distribution box for a distribution box, which can dynamically adjust the fault judgment criteria of the distribution box and identify potential fault risks in advance, thereby improving the accuracy and timeliness of distribution box fault detection.

[0004] In the first aspect, an embodiment of the present invention provides an alarm method for a distribution box, which is applied to a control device of the distribution box; the distribution box also includes at least one acquisition device; the acquisition device is connected to the control device; it includes: acquiring acquisition data of the acquisition device; the acquisition data includes an acquisition device identifier, a real-time data value and its corresponding timestamp; inputting the acquisition device identifier and timestamp into a pre-trained threshold model, and outputting the current threshold; the threshold model is used to output the threshold corresponding to the acquisition data; judging whether the real-time data value exceeds the current threshold; if it exceeds the current threshold, determining that there is a fault in the distribution box, and generating a first alarm message; if it does not exceed the current threshold, inputting the acquisition device identifier and the real-time data value into a pre-trained data prediction model, and outputting a fault prediction result; if the fault prediction result includes a fault level, generating a second alarm message based on the fault level.

[0005] Furthermore, the threshold model is trained in the following manner: obtaining the historical timestamps, historical data values ​​and device abnormality records corresponding to each acquisition device identifier; extracting the time features of the historical timestamps and the data features in the historical data values, associating and saving the time features, data features and device abnormality records to obtain the historical association relationship; constructing a first training data set based on the acquisition device identifier, historical timestamp, historical data value, abnormal record and historical association relationship; training the initial regression model based on the first training data set until the initial regression model can determine the threshold of each acquisition device in different time periods to obtain a trained threshold model; the threshold model can output the current threshold based on the input real-time data value.

[0006] Furthermore, the data prediction model is trained in the following manner: obtaining the historical timestamps, historical data values ​​and equipment abnormality records corresponding to each collection device identifier; extracting the time features of the historical timestamps and the data features in the historical data values, associating and saving the time features, data features and equipment abnormality records to obtain the historical association relationship; constructing a second training data set based on the collection device identifier, historical timestamp, historical data value, abnormal record and historical association relationship; training the initial deep learning model based on the second training data set until the initial deep learning model can obtain the change trend of each collection device in different time periods, and obtain a trained data prediction model; the data training model can predict the next collection data corresponding to the collection device in the next time interval based on the input real-time data value, and output the fault prediction result based on the next collection data.

[0007] Furthermore, the collection device includes at least one of the following: a current collection device, a voltage collection device, a water immersion sensor, a temperature sensor, a humidity sensor and an electric energy meter; if the fault prediction result includes a fault level, after the step of generating a second alarm information based on the fault level, the method also includes: setting the collection device whose fault prediction result is a fault level as a fault collection device; determining the weight value corresponding to each fault collection device based on a preset correspondence between the collection device and the weight; calculating the risk value of the distribution box based on the weight value, the fault level corresponding to the fault collection device and a preset risk level calculation method; determining the risk level corresponding to the risk value based on the preset risk data and the risk level correspondence; the risk levels include low risk, medium risk and high risk.

[0008] Furthermore, the distribution box also includes a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; the circuit breaker is a main circuit breaker; after the step of determining the risk level corresponding to the risk value based on the correspondence between the preset risk data and the risk level, the method also includes: when the risk level is low risk, sending a second alarm information to the display device so that the display device displays the second alarm information; when the risk level is medium risk, generating a disconnect signal, and sending the disconnect signal to the main circuit breaker so that the main circuit breaker performs a disconnect operation; when the risk level is high risk, generating a disconnect signal and an alarm signal, sending the disconnect signal to the main circuit breaker, and sending the alarm signal to the alarm device so that the main circuit breaker performs a disconnect operation, so that the alarm device alarms.

[0009] Furthermore, the distribution box also includes a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; the circuit breaker includes a main circuit breaker and at least one branch circuit breaker connected to the main circuit breaker; each branch circuit breaker is connected to an external load to form a branch circuit; the input end of the main circuit breaker is connected to the power supply; the output end of the main circuit breaker is connected to the input end of the branch circuit breaker; when the acquisition device is a branch current acquisition device and / or a branch voltage acquisition device arranged on the branch circuit, the risk level corresponding to the preset risk data is closed. After the step of determining the risk level corresponding to the risk value, the method also includes: when the risk level is low risk, sending the second alarm information to the display device so that the display device displays the second alarm information; when the risk level is medium risk, generating a disconnect signal, and sending the disconnect signal to the branch circuit breaker corresponding to the fault collection device, so that the branch circuit breaker performs a disconnect operation; when the risk level is high risk, generating a disconnect signal and an alarm signal, sending the disconnect signal to the main circuit breaker and the branch circuit breaker, and sending the alarm signal to the alarm device, so that the main circuit breaker and the branch circuit breaker perform a disconnect operation, so that the alarm device alarms.

[0010] Furthermore, the distribution box also includes an automatic reclosing switch, which is respectively connected to the control device and the circuit breaker; after the circuit breaker performs the disconnection operation, the method also includes: after a preset time interval, the automatic reclosing switch sends a closing signal to the circuit breaker, so that the circuit breaker performs a closing operation according to the closing signal.

[0011] In the second aspect, an embodiment of the present invention provides an alarm system for a distribution box, which includes a control device, a network relay and at least one collection device; the network relay and the collection device are respectively connected to the control device; the alarm system includes: a data collection module, which is used to obtain the collection data of the collection device; the collection data includes a collection device identifier, a real-time data value and its corresponding timestamp; a fault judgment module, which is used to input the collection device identifier and timestamp into a pre-trained threshold model and output the current threshold; the threshold model is used to output the threshold corresponding to the collection data; the fault judgment module is also used to judge whether the real-time data value exceeds the current threshold; the fault judgment module is also used to determine that there is a fault in the distribution box if the current threshold is exceeded, and generate a first alarm message; the fault judgment module is also used to input the collection device identifier and the real-time data value into a pre-trained data prediction model if the current threshold is not exceeded, and output a fault prediction result; if the fault prediction result includes a fault level, a second alarm message is generated based on the fault level.

[0012] In the third aspect, an embodiment of the present invention provides a distribution box, comprising a distribution box shell, and a control device, a collection device, a power supply, a circuit breaker and an automatic reclosing switch arranged inside the distribution box shell; it also includes the above-mentioned alarm system of the distribution box, and the alarm system of the distribution box is arranged in the control device; the collection device, the power supply, the circuit breaker and the automatic reclosing switch are respectively connected to the control device; the automatic reclosing switch is connected to the circuit breaker; the power supply is used to power the distribution box.

[0013] Furthermore, the distribution box also includes a network relay; it also includes a display device, an alarm device, a power indicator light and a button arranged on the outer shell of the distribution box; the collection device includes at least one of the following: a current collection device, a voltage collection device, a water immersion sensor, a temperature sensor, a humidity sensor and an electric energy meter; the network relay is connected to the control device; the control device is connected to the host computer of the peripheral device through the network relay.

[0014] The embodiment of the present invention provides an alarm method, an alarm system and a distribution box for a distribution box, wherein the distribution box also includes at least one collection device; the network relay and the collection device are respectively connected to the control device; the method includes: obtaining the collection data of the collection device; the collection data includes the collection device identifier, the real-time data value and its corresponding timestamp; the collection device identifier and the timestamp are input into a pre-trained threshold model, and the current threshold is output; the threshold model is used to output the threshold corresponding to the collection data; judging whether the real-time data value exceeds the current threshold; if the current threshold is exceeded, it is determined that there is a fault in the distribution box, and a first alarm message is generated; if the current threshold is not exceeded, the collection device identifier and the real-time data value are input into a pre-trained data prediction model, and the fault prediction result is output; if the fault prediction result includes a fault level, a second alarm message is generated based on the fault level. In this method, by adopting an alarm method based on a threshold model and a data prediction model, the operation data of the collection device can be obtained in real time, and the data can be analyzed and predicted, so as to dynamically adjust the fault judgment standard and identify the potential fault risk in advance. In addition, the accuracy and timeliness of the fault detection of the distribution box are improved, the equipment damage and downtime caused by the fault are reduced, and the stability and safety of the power system are guaranteed.

[0015] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0016] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flow chart of an alarm method for a distribution box provided in an embodiment of the present invention;

[0019] Figure 2 A flow chart of a threshold model training method provided by an embodiment of the present invention;

[0020] Figure 3 A flow chart of a data prediction model training method provided by an embodiment of the present invention;

[0021] Figure 4 A flow chart of a method for calculating risk level provided by an embodiment of the present invention;

[0022] Figure 5 A flow chart of a risk level based alarm method provided by an embodiment of the present invention;

[0023] Figure 6 A flow chart of another risk level based alarm method provided by an embodiment of the present invention;

[0024] Figure 7 A schematic diagram of an alarm system for a distribution box provided in an embodiment of the present invention;

[0025] Figure 8 A schematic diagram of a distribution box provided in an embodiment of the present invention.

[0026] Icons: 1-data acquisition module; 2-fault judgment module; 11-electrical box housing; 12-control device; 13-acquisition device; 14-power supply; 15-circuit breaker; 16-automatic reclosing switch; 17-alarm system of distribution box. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] To facilitate understanding of this embodiment, the embodiment of the present invention is described in detail below.

[0029] Embodiment 1:

[0030] Figure 1 A flow chart of an alarm method for a distribution box provided in an embodiment of the present invention.

[0031] The alarm method of the distribution box is applied to the control device of the distribution box; the distribution box also includes at least one collection device; the collection device is connected to the control device.

[0032] Here, the control device may be a control chip.

[0033] The data acquisition device includes at least one of the following: a current data acquisition device, a voltage data acquisition device, a water immersion sensor, a temperature sensor, a humidity sensor and an electric energy meter.

[0034] The current acquisition device is used to monitor the change of current in the distribution box, and can be a Hall sensor or a current transformer, etc.

[0035] The voltage acquisition device is used to monitor the voltage changes in the distribution box, and can be a voltage sensor or a voltage transformer, etc.

[0036] The water sensor is used to monitor whether there is water accumulation inside or around the distribution box to prevent short circuit or equipment damage caused by moisture.

[0037] The temperature sensor is used to monitor the temperature inside the distribution box and can be a thermocouple or RTD (resistance temperature detector) sensor, etc.

[0038] Humidity sensors are used to monitor the relative humidity inside the distribution box to prevent electrical failures caused by excessive humidity.

[0039] Energy meters are used to monitor energy consumption in distribution boxes, usually by measuring current, voltage and power factor to calculate power usage.

[0040] Reference Figure 1 , the alarm method of the distribution box includes:

[0041] Step S101, acquiring the collected data of the collection device; the collected data includes a collection device identifier, a real-time data value and its corresponding timestamp.

[0042] Here, the data acquisition equipment in the distribution box is used to monitor the current, voltage, temperature, humidity and other related data in real time. The signals collected by each data acquisition device through the sensor are converted into digital data, and are accompanied by a device identifier (each data acquisition device corresponds to a unique identifier) ​​and a timestamp when the data is collected (recording the specific time of data collection).

[0043] The control chip can continuously receive the collected data sent by the collection device, and can also periodically obtain the collected data sent by the collection device according to a preset collection cycle.

[0044] When the acquisition device is a current acquisition device, the acquisition device identifier is the identifier corresponding to the current acquisition device, the real-time data value is the collected current value (unit: ampere A), and the timestamp is the time point of current data acquisition.

[0045] When the acquisition device is a voltage acquisition device, the acquisition device identifier is the identifier corresponding to the voltage acquisition device, the real-time data value is the collected voltage value (unit: volt V), and the timestamp is the time point of voltage data acquisition.

[0046] When the acquisition device is a water immersion sensor, the acquisition device identifier is the identifier corresponding to the water immersion sensor, the real-time data value is the detection value of the water immersion sensor, which usually indicates the water level height or the presence or absence of water immersion (such as "1" or "0"), and the timestamp is the time when the water immersion occurs.

[0047] When the acquisition device is a temperature sensor, the acquisition device identifier is the identifier corresponding to the temperature sensor, the real-time data value is the collected temperature value (unit: Celsius), and the timestamp is the time when the temperature data is collected.

[0048] When the acquisition device is a humidity sensor, the acquisition device identifier is the identifier corresponding to the humidity sensor, the real-time data value is the collected humidity value (unit: %RH, indicating relative humidity), and the timestamp is the time when the humidity data is collected.

[0049] When the collection device is an electric energy meter, the collection device identifier is the identifier corresponding to the electric energy meter, the real-time data value is the collected electric energy consumption value (unit: kilowatt hour kWh or megawatt hour MWh), and the timestamp is the time when the electric energy data is collected.

[0050] Step S102, input the collection device identifier and timestamp into a pre-trained threshold model, and output the current threshold; the threshold model is used to output the threshold corresponding to the collection data.

[0051] Here, the identifier of the collection device and the corresponding timestamp are used as input to the trained threshold model. The threshold model can predict the normal operation threshold of each device in different time periods based on historical data. The current thresholds corresponding to different time periods can be the same or different.

[0052] In one embodiment, referring to Figure 2 , the threshold model in step S102 is trained in the following way:

[0053] Step S201, obtaining the historical timestamp, historical data value and device abnormality record corresponding to each collection device identifier.

[0054] Here, historical data is collected from each collection device in the power distribution system (such as current collection device, voltage collection device, etc.). These data include:

[0055] The historical timestamp is used to record the exact time of each data collection, ensuring that the data is associated with time and facilitating the analysis of timing.

[0056] Historical data values ​​refer to the actual data values ​​measured by each acquisition device in the past, such as current, voltage, temperature, etc. The historical data values ​​can reflect the operating status of the device in different time periods.

[0057] The device abnormality record is used to record the failures or abnormal conditions (such as overload, overheating, short circuit, etc.) that occurred in the past for each device. The device abnormality record helps to analyze the failure mode of the device under specific conditions.

[0058] Step S202, extracting the time features of the historical timestamp and the data features in the historical data value, and associating and saving the time features, the data features and the equipment abnormality records to obtain a historical association relationship.

[0059] Here, the collected historical data is subjected to feature extraction and association, including:

[0060] Time feature extraction: For example, extracting information such as "hour", "week", and "month" from historical timestamps. These time features may be related to the operating status of the device. For example, the device may show higher load or temperature during certain time periods (such as peak hours).

[0061] Data feature extraction: Extract meaningful features from historical data values, such as maximum, minimum, average, volatility, etc. These data features can help build models and identify the normal fluctuation range of data.

[0062] Abnormal record association: associate historical data values ​​with device abnormal records to form an association between data and abnormal events. For example, when the current exceeds a certain threshold, the corresponding abnormal event is recorded, such as excessive current or overload.

[0063] Step S203, constructing a first training data set based on the collection device identifier, historical timestamp, historical data value, abnormal record and historical association relationship.

[0064] Here, a complete training data set is organized according to each acquisition device identifier. Each data item should include:

[0065] Collect device identifiers to identify the device from which the data originates.

[0066] Historical timestamps and historical data values ​​are used to describe the operating status of the device at each point in time.

[0067] Equipment abnormality record, used to record abnormal conditions of equipment within a specific time.

[0068] Historical association relationships include the associations between time features, data features, and abnormal records.

[0069] This information constitutes a training data set for training subsequent regression models. Each data item should contain data in multiple dimensions to ensure that the model can learn the operating mode and possible failure conditions of the equipment.

[0070] Step S204, training the initial regression model based on the first training data set until the initial regression model can determine the threshold of each acquisition device in different time periods to obtain a trained threshold model; the threshold model can output the current threshold according to the input real-time data value.

[0071] Here, the first training data set is input into the regression model for training. The goal of the regression model is to learn how to predict the current threshold based on the historical data of the device (including time features, data features, and abnormal records). The specific steps include:

[0072] The model is trained using common regression methods (such as linear regression, decision tree regression, support vector regression, etc.), with the goal of enabling the model to output the threshold value of the device in different time periods.

[0073] Through repeated training and parameter adjustment, we ensure that the regression model can accurately predict the thresholds of different time periods and different devices. For example, the thresholds of current acquisition devices may be different during working hours and non-working hours.

[0074] After the training is completed, the model effect is verified through cross-validation and other methods to ensure the accuracy of the model.

[0075] Finally, a trained threshold model is obtained, and the trained threshold model can output the corresponding current threshold according to the input real-time data (such as real-time current, voltage, etc.).

[0076] In practical applications, when the control device of the distribution box receives real-time data from the acquisition device, the threshold model will output the corresponding threshold value based on the real-time data.

[0077] Step S103, determining whether the real-time data value exceeds the current threshold.

[0078] Here, the data value collected in real time is compared with the current threshold value calculated according to the threshold model.

[0079] Step S104: If the current threshold is exceeded, it is determined that a fault exists in the distribution box, and a first alarm message is generated.

[0080] Here, if the real-time data value exceeds the current threshold, it means that a device or circuit in the distribution box may have failed. The distribution box is determined to be faulty, and the first alarm information is generated.

[0081] The distribution box also includes a network relay. The control device is connected to the network relay via a communication interface, and the control device is connected to a host computer of the peripheral device via the network relay.

[0082] After the first alarm information is generated, the control device sends the first alarm information to the host computer.

[0083] The first alarm information usually includes the fault type, device identifier, specific data value and fault occurrence time, etc.

[0084] Step S105, if the current threshold is not exceeded, the acquisition device identifier and the real-time data value are input into a pre-trained data prediction model, and a fault prediction result is output; if the fault prediction result includes a fault level, a second alarm information is generated based on the fault level.

[0085] Here, if the real-time data does not exceed the current threshold, the identifier of the collection device and the real-time data value are input into a pre-trained data prediction model. The data prediction model can predict the future operation trend of the equipment based on historical data and determine whether there is a potential failure risk.

[0086] By analyzing the correlation between historical data and real-time data, it is possible to identify trend changes, predict possible equipment failures or abnormalities in the future, warn of potential failures in advance, and classify failures.

[0087] Depending on the prediction results, each fault is assigned a fault level. For example, a low-level fault may indicate a minor abnormality in the equipment status, while a high-level fault may indicate that the equipment is about to fail seriously or shut down.

[0088] The second alarm information will indicate different processing priorities and response measures according to the fault level. It may include the fault level, impact range, recommended emergency measures, etc.

[0089] After the second alarm information is generated, the control device sends the second alarm information to the host computer.

[0090] In one embodiment, referring to Figure 3 The data prediction model in step S105 is trained in the following way:

[0091] Step S301, obtaining the historical timestamp, historical data value and device abnormality record corresponding to each collection device identifier.

[0092] Here, similar to the threshold model training process, historical data of each acquisition device (such as current, voltage, water immersion sensor, etc.) in the distribution box is collected.

[0093] The historical timestamp is used to identify the time of each data collection, ensuring that the data is sorted in chronological order, making it easier to analyze the time series characteristics of the device.

[0094] The historical data value represents the specific measurement value of each acquisition device in the historical time, such as current, voltage, temperature, etc. The historical data value reflects the operating status of the acquisition device in the past period of time.

[0095] Device abnormality records are used to record abnormal events that occur on each device, such as device overload, overheating, etc.

[0096] Step S302, extracting the time features of the historical timestamp and the data features in the historical data value, and associating and saving the time features, the data features and the equipment abnormality records to obtain a historical association relationship.

[0097] Here, time feature extraction is to extract time features from historical timestamps, such as hours, days, weeks, months, etc. Time features help identify the periodic behavior of equipment, for example, the operation mode of equipment during peak load is different from that during off-peak periods.

[0098] Data feature extraction is to extract data features from historical data values, such as mean, standard deviation, maximum value, minimum value, volatility, etc. Data feature extraction helps to understand the fluctuation range and trend of data, and can help the model learn the difference between normal and abnormal behavior.

[0099] Abnormal record association is to associate the abnormal records of each device with the corresponding time features and data features to construct time series data of device abnormalities. For example, when the current exceeds a certain threshold, an overload event is recorded.

[0100] Step S303, constructing a second training data set based on the collection device identifier, historical timestamp, historical data value, abnormal record and historical association relationship.

[0101] Here, the extracted time features, data features and device abnormality records are classified according to the acquisition device identifier and organized into a complete second training data set.

[0102] The constructed second training data set provides input data for the subsequent deep learning model. Each data item in the second training data set should contain time series data and abnormal information of the equipment to ensure that the model can learn the long-term operation trend of the equipment.

[0103] Step S304: Based on the second training data set, the initial deep learning model is trained until the initial deep learning model can obtain the change trend of each acquisition device in different time periods, and obtain a trained data prediction model; the data training model can predict the next acquisition data corresponding to the acquisition device in the next time interval according to the input real-time data value, and output the fault prediction result according to the next acquisition data.

[0104] Here, the initial deep learning model can be a neural network model, etc.

[0105] The initial deep learning model is trained using the second training data set so that the initial deep learning model can output the change trend in different time periods. For example, the change trend of current may be affected by factors such as load, seasonal changes, or equipment aging.

[0106] The trained data prediction model can predict the measured values ​​of the equipment in future time intervals (such as future current, voltage and other data), and can also identify abnormal patterns of the equipment from historical data and predict possible failures.

[0107] Through hyperparameter adjustment and cross-validation during the training process, the accuracy of the deep learning model is optimized to ensure that the model can accurately predict the change trend of the equipment in real applications.

[0108] When real-time collected data is obtained, it is used as input to the trained data prediction model.

[0109] The data prediction model predicts the next data value of the device in the next time interval based on the input real-time data (such as current current, voltage, etc.) and the device's historical data. For example, based on the current current data, predict the current change trend in the next hour.

[0110] Based on the next data value predicted by the data prediction model, we can further analyze whether it exceeds the normal working range of the device, and then infer whether the device may fail. For example, if the predicted current value is far beyond the normal range of the device, it can be determined that the device is at risk of failure.

[0111] In one embodiment, referring to Figure 4 In step S105, if the fault prediction result includes a fault level, after the step of generating second alarm information based on the fault level, the method further includes:

[0112] Step S401, setting the collection device with the fault prediction result of the fault level as the fault collection device.

[0113] Here, if the fault prediction result output by the data prediction model shows the fault level of a certain acquisition device, the device is marked as a faulty acquisition device. The fault level is usually obtained through model prediction and indicates the severity or urgency of the device fault. For example, the fault level may be divided into minor fault, medium fault, and severe fault.

[0114] When the fault level of a certain acquisition device is identified, the acquisition device will be marked or set as a faulty acquisition device.

[0115] Step S402: determining a weight value corresponding to each faulty collection device based on a preset correspondence relationship between the collection device and the weight.

[0116] Here, each data collection device has a preset weight value based on its type, fault level, and impact on the overall operation of the distribution box. These weight values ​​are set through experience or historical data to quantify the impact of device failure on the overall safety of the distribution box.

[0117] According to the type of acquisition equipment (such as current acquisition equipment, voltage acquisition equipment, etc.) and the importance of the acquisition equipment in the distribution box, a weight value related to the equipment type and fault level is determined. For example, the current acquisition equipment may have a higher weight because it has a greater impact on the load and stability of the distribution box.

[0118] The size of the weight value is related to the potential risk of device failure. A device with a higher failure level (such as a serious failure) may have a higher weight, while a device with a minor failure may have a lower weight.

[0119] Step S403, based on the weight value, the fault level corresponding to the fault collection device and the preset risk level calculation method, the risk value of the distribution box is calculated.

[0120] Here, the risk value of the distribution box is calculated based on the fault level, weight value and preset risk level calculation method of the equipment. The specific steps are as follows:

[0121] The fault level of the data acquisition device and its corresponding weight value will affect the calculation of the risk value. For example, if the fault level of a data acquisition device is medium and the corresponding weight value is 0.7, then its contribution to the overall risk of the distribution box is the weight multiplied by the impact coefficient of its fault level.

[0122] According to the preset algorithm or calculation rules, the fault information of each data collection device is converted into a risk value. For example, the algorithm may combine the risks of all faulty data collection devices through weighted average, weighted sum, etc. to obtain the overall risk value of the distribution box.

[0123] The risk value (fault level × weight value) of each fault collection device will be summarized into an overall distribution box risk value, reflecting the overall risk level of the distribution box.

[0124] Step S404, based on the correspondence between the preset risk data and the risk level, determine the risk level corresponding to the risk value; the risk level includes low risk, medium risk and high risk.

[0125] Here, after the risk value of the distribution box is calculated, the calculated risk value is mapped to a clear risk level according to a preset risk level correspondence.

[0126] The preset risk data and risk level correspondence is used to map the risk value to the risk level (low risk, medium risk and high risk). For example: low risk (such as: risk value is 0-20), medium risk (such as: risk value is 21-50), high risk (such as: risk value is above 51).

[0127] By comparing the calculated risk value with the preset rules, the risk value of the distribution box is mapped to the corresponding risk level. For example, if the risk value of the distribution box is 45, it is assessed as medium risk according to the corresponding relationship.

[0128] A specific risk level is output based on the risk value of the distribution box for decision support and alarm generation.

[0129] In one embodiment, the distribution box also includes a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; and the circuit breaker is a main circuit breaker.

[0130] Here, the power supply provides power support for all equipment in the distribution box (including circuit breakers, display devices, alarm devices, etc.).

[0131] A circuit breaker is a circuit protection device that shuts off a circuit when needed. When an electrical fault occurs, the circuit breaker prevents conditions such as overloads or short circuits from causing fire or damaging equipment.

[0132] The display device is used to display the status and alarm information of the distribution box, usually an LED display screen or other types of display devices, used to present the alarm content.

[0133] The alarm device is used to sound an alarm, usually an audible alarm or other alarm device, to notify maintenance personnel or operators of abnormal conditions of the distribution box.

[0134] The main circuit breaker is a key circuit breaker in the distribution box, responsible for the switch control of the entire distribution box circuit. The main circuit breaker can perform a disconnection operation when there is a risk in the distribution box system, cut off the power supply, and prevent further electrical accidents.

[0135] Reference Figure 5 After step S404, the method further includes:

[0136] Step S501: When the risk level is low risk, the second warning information is sent to the display device so that the display device displays the second warning information.

[0137] Here, low risk means that the system determines that the abnormal situation of the equipment has little impact on the overall safety of the distribution box. It usually only prompts that a fault exists, but will not cause significant harm to equipment or personnel.

[0138] At this time, the second alarm information is sent to the display device so that the display device can visually display the alarm information. The display device may display information such as "low risk" or "warning: minor abnormality" to remind the operator or maintenance personnel to pay attention to the status, but no emergency measures need to be taken immediately.

[0139] Step S502: when the risk level is medium risk, a disconnection signal is generated and sent to the main circuit breaker so that the main circuit breaker performs a disconnection operation.

[0140] Here, medium risk means that the fault or abnormal situation in the distribution box may pose a certain degree of threat to the safety of equipment or personnel, but has not yet reached a serious emergency. In order to prevent the risk from further escalating, a disconnection signal is generated to instruct the main circuit breaker to perform a disconnection operation.

[0141] The main circuit breaker disconnects the power supply, thus reducing the potential harm to equipment or personnel caused by the fault.

[0142] Step S503, when the risk level is high risk, a disconnection signal and an alarm signal are generated, the disconnection signal is sent to the main circuit breaker, and the alarm signal is sent to the alarm device, so that the main circuit breaker performs a disconnection operation, so that the alarm device alarms.

[0143] Here, high risk means that a failure or abnormality in the distribution box may pose a serious threat to equipment, personnel safety, or even the entire distribution system, which usually requires immediate emergency measures.

[0144] The control device generates a disconnect signal, which is sent to the main circuit breaker, instructing it to immediately cut off the power supply, ensuring that the power supply to the distribution box is interrupted, thereby reducing hazards to equipment or personnel.

[0145] In addition to disconnecting the power supply, the control device will also generate an alarm signal, which will be sent to the alarm device, usually in the form of a beep or other type of emergency alarm tone, prompting personnel to respond immediately and perform necessary processing operations. The alarm reminder of the alarm device can help maintenance personnel to promptly detect and respond to high-risk conditions in the distribution box.

[0146] In one embodiment, the distribution box also includes a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; the circuit breaker includes a main circuit breaker and at least one branch circuit breaker connected to the main circuit breaker; each branch circuit breaker is connected to an external load to form a branch circuit; the input end of the main circuit breaker is connected to the power supply; the output end of the main circuit breaker is connected to the input end of the branch circuit breaker.

[0147] Here, the circuit breaker includes a main circuit breaker and at least one branch circuit breaker. The main circuit breaker is used to control the power supply of the entire distribution box, while the branch circuit breaker controls each load circuit connected thereto.

[0148] Branch circuit breakers are connected to the main circuit breaker, and each branch circuit breaker controls one or more peripheral load circuits. For example, a branch circuit breaker may control lights, sockets, and other devices in a distribution box.

[0149] Branch current acquisition devices and branch voltage acquisition devices: These acquisition devices are used to monitor the current and voltage in each branch circuit. For example, the branch current acquisition device monitors the current connected to the branch circuit breaker, while the branch voltage acquisition device monitors the voltage in the circuit.

[0150] The collected data includes information such as the identifier of the branch current collection device or the branch voltage collection device, the real-time collected data value (current or voltage), and the corresponding timestamp.

[0151] When the acquisition device is a branch current acquisition device and / or a branch voltage acquisition device installed on a branch circuit, refer to Figure 6 After step S404, the method further includes:

[0152] Step S601: When the risk level is low risk, the second warning information is sent to the display device so that the display device displays the second warning information.

[0153] Here, when the risk level is assessed as low risk, it means that there is a certain abnormality in the equipment, but the abnormality is unlikely to pose a serious threat to the equipment or personnel. At this time, the distribution box will not take power-off measures, but only send a second alarm message through the display device to alert the operator or maintenance personnel.

[0154] The display device will display low-risk warnings, such as "minor abnormality" and "low risk", so that operators can understand the current equipment status and stay concerned.

[0155] Step S602: When the risk level is medium risk, a disconnection signal is generated and sent to a branch circuit breaker corresponding to the fault collection device, so that the branch circuit breaker performs a disconnection operation.

[0156] Here, when a medium risk is assessed, it means that the equipment failure or abnormality may cause a certain impact, but has not yet reached a serious level. In order to prevent the fault from escalating further, a disconnection signal is generated and sent to the corresponding branch circuit breaker.

[0157] The branch circuit breaker performs a power-off operation according to the disconnection signal received, cutting off the current supply to the branch circuit. This helps to isolate the fault source and prevent the impact from spreading to other parts.

[0158] Step S603, when the risk level is high risk, a disconnection signal and an alarm signal are generated, the disconnection signal is sent to the main circuit breaker and the branch circuit breaker, and the alarm signal is sent to the alarm device, so that the main circuit breaker and the branch circuit breaker perform a disconnection operation, so that the alarm device alarms.

[0159] Here, when assessed as high risk, it means there is a potential risk that poses a significant threat to equipment or personnel safety, and emergency measures must be taken immediately.

[0160] The control device generates a disconnect signal and sends it to the main circuit breaker and branch circuit breaker respectively, instructing them to disconnect the power supply and cut off all power in the distribution box.

[0161] The main circuit breaker cuts off the power supply to the distribution box according to the disconnection signal to prevent further fault propagation.

[0162] The branch circuit breaker performs a disconnect operation to cut off the power supply to the external load circuit connected to it, thereby preventing the fault from affecting other circuits or equipment.

[0163] At the same time, an alarm signal is generated and sent to an alarm device, such as a sound alarm or other forms of emergency prompts, to remind maintenance personnel to immediately handle the fault and take further measures.

[0164] In one embodiment, the distribution box further comprises an automatic reclosing switch, which is connected to the control device and the circuit breaker respectively. After the circuit breaker performs the disconnection operation, the method further comprises:

[0165] After a preset time interval, the automatic reclosing sends a closing signal to the circuit breaker, so that the circuit breaker performs a closing operation according to the closing signal.

[0166] Here, after the circuit breaker performs the disconnection operation, it waits for a certain period of time to confirm whether the fault is caused by a temporary current overload, instantaneous short circuit, etc. If the fault is temporary, the distribution box can reconnect the power supply to restore the power supply.

[0167] This time interval is preset by the control device according to the specific scenario or the working characteristics of the device, for example, it may be a few seconds to a few minutes.

[0168] After the preset time interval, the auto recloser sends a closing signal to the circuit breaker, instructing it to close. The closing signal means reconnecting the circuit breaker to the power supply and restoring the current supply.

[0169] When the circuit breaker receives a closing signal, it performs a closing operation, that is, reclosing the circuit of the circuit breaker so that power is restored to the original circuit.

[0170] For the main circuit breaker, closing operation means restoring power supply to the entire distribution box.

[0171] For branch circuit breakers, closing operation means reconnecting the faulty circuit and restoring power to the connected equipment or external loads.

[0172] After the closing operation is completed, the control device of the distribution box may continue to monitor the real-time data of related acquisition devices (such as current acquisition devices, voltage acquisition devices, etc.) to determine whether there are new abnormalities. If the system detects an abnormality, it may perform the disconnection operation again, or remind the operator to intervene manually through the alarm device.

[0173] The embodiment of the present invention provides an alarm method for a distribution box, which is applied to a control device of a distribution box; the distribution box also includes at least one acquisition device; the network relay and the acquisition device are respectively connected to the control device; the method includes: obtaining the acquisition data of the acquisition device; the acquisition data includes an acquisition device identifier, a real-time data value and its corresponding timestamp; the acquisition device identifier and the timestamp are input into a pre-trained threshold model, and the current threshold is output; the threshold model is used to output the threshold corresponding to the acquisition data; judging whether the real-time data value exceeds the current threshold; if the current threshold is exceeded, it is determined that there is a fault in the distribution box, and a first alarm message is generated; if the current threshold is not exceeded, the acquisition device identifier and the real-time data value are input into a pre-trained data prediction model, and a fault prediction result is output; if the fault prediction result includes a fault level, a second alarm message is generated based on the fault level. In this method, by adopting an alarm method based on a threshold model and a data prediction model, the operation data of the acquisition device can be obtained in real time, and the data can be analyzed and predicted, so as to dynamically adjust the fault judgment standard and identify potential fault risks in advance. In addition, the accuracy and timeliness of fault detection in the distribution box are improved, the equipment damage and downtime caused by the fault are reduced, and the stability and safety of the power system are guaranteed.

[0174] Embodiment 2:

[0175] Figure 7 A schematic diagram of an alarm system for a distribution box provided in an embodiment of the present invention.

[0176] Reference Figure 7 , the alarm system of the distribution box includes:

[0177] The data acquisition module 1 is used to obtain the acquisition data of the acquisition device; the acquisition data includes the acquisition device identifier, the real-time data value and its corresponding timestamp.

[0178] Fault judgment module 2 is used to input the collection device identifier and timestamp into a pre-trained threshold model and output the current threshold; the threshold model is used to output the threshold corresponding to the collection data.

[0179] The fault judgment module 2 is also used to judge whether the real-time data value exceeds the current threshold.

[0180] The fault judgment module 2 is also used to determine that there is a fault in the distribution box and generate a first alarm message if the current threshold is exceeded.

[0181] The fault judgment module 2 is also used to input the collection device identifier and the real-time data value into a pre-trained data prediction model if the current threshold is not exceeded, and output a fault prediction result; if the fault prediction result includes a fault level, a second alarm information is generated based on the fault level.

[0182] In one embodiment, the fault judgment module 2 is further used to:

[0183] Get the historical timestamp, historical data value and device exception record corresponding to each collection device identifier.

[0184] The time features of the historical timestamps and the data features in the historical data values ​​are extracted, and the time features, data features and equipment abnormality records are associated and saved to obtain the historical association relationship.

[0185] A first training data set is constructed based on a collection device identifier, a historical timestamp, a historical data value, an abnormal record, and a historical association relationship.

[0186] The initial regression model is trained based on the first training data set until the initial regression model can determine the threshold of each acquisition device in different time periods to obtain a trained threshold model; the threshold model can output the current threshold according to the input real-time data value.

[0187] In one embodiment, the fault judgment module 2 is further used to:

[0188] Get the historical timestamp, historical data value and device exception record corresponding to each collection device identifier.

[0189] The time features of the historical timestamps and the data features in the historical data values ​​are extracted, and the time features, data features and equipment abnormality records are associated and saved to obtain the historical association relationship.

[0190] A second training data set is constructed based on the collection device identifier, historical timestamps, historical data values, abnormal records, and historical association relationships.

[0191] Based on the second training data set, the initial deep learning model is trained until the initial deep learning model can obtain the change trend of each acquisition device in different time periods, and obtain a trained data prediction model; the data training model can predict the next acquisition data corresponding to the acquisition device in the next time interval based on the input real-time data value, and output the fault prediction result based on the next acquisition data.

[0192] In one embodiment, the acquisition device includes at least one of the following: a current acquisition device, a voltage acquisition device, a water immersion sensor, a temperature sensor, a humidity sensor, and an electric energy meter. The fault judgment module 2 is also used to:

[0193] The collection device whose fault prediction result is a fault level is set as the fault collection device.

[0194] Based on the preset corresponding relationship between the collection devices and the weights, the weight value corresponding to each faulty collection device is determined.

[0195] Based on the weight value, the fault level corresponding to the fault collection device and the preset risk level calculation method, the risk value of the distribution box is calculated.

[0196] Based on the correspondence between preset risk data and risk levels, the risk level corresponding to the risk value is determined; the risk levels include low risk, medium risk and high risk.

[0197] In one embodiment, the distribution box further includes a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; the circuit breaker is a main circuit breaker. The fault judgment module 2 is also used to:

[0198] When the risk level is low risk, the second warning information is sent to the display device so that the display device displays the second warning information.

[0199] When the risk level is medium risk, a disconnection signal is generated and sent to the main circuit breaker so that the main circuit breaker performs a disconnection operation.

[0200] When the risk level is high risk, a disconnection signal and an alarm signal are generated, the disconnection signal is sent to the main circuit breaker, and the alarm signal is sent to the alarm device, so that the main circuit breaker performs a disconnection operation, so that the alarm device alarms.

[0201] In one embodiment, the distribution box also includes a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; the circuit breaker includes a main circuit breaker and at least one branch circuit breaker connected to the main circuit breaker; each branch circuit breaker is connected to an external load to form a branch circuit; the input end of the main circuit breaker is connected to the power supply; the output end of the main circuit breaker is connected to the input end of the branch circuit breaker.

[0202] The fault judgment module 2 is also used for:

[0203] When the risk level is low risk, the second warning information is sent to the display device so that the display device displays the second warning information.

[0204] When the risk level is medium risk, a disconnection signal is generated and sent to the branch circuit breaker corresponding to the fault collection device, so that the branch circuit breaker performs a disconnection operation.

[0205] When the risk level is high risk, a disconnection signal and an alarm signal are generated, and the disconnection signal is sent to the main circuit breaker and the branch circuit breaker, and the alarm signal is sent to the alarm device, so that the main circuit breaker and the branch circuit breaker perform a disconnection operation to cause the alarm device to alarm.

[0206] In one embodiment, the distribution box further includes an automatic reclosing switch, which is connected to the control device and the circuit breaker respectively. The fault judgment module 2 is also used for:

[0207] After a preset time interval, the automatic reclosing device sends a closing signal to the circuit breaker, so that the circuit breaker performs a closing operation according to the closing signal.

[0208] The embodiment of the present invention provides an alarm system for a distribution box, in which, by adopting an alarm method based on a threshold model and a data prediction model, the operation data of the acquisition device can be obtained in real time, and the data can be analyzed and predicted, so as to dynamically adjust the fault judgment standard and identify potential fault risks in advance. In addition, the accuracy and timeliness of distribution box fault detection are improved, equipment damage and downtime caused by faults are reduced, and the stability and safety of the power system are guaranteed.

[0209] Embodiment three:

[0210] Figure 8 A schematic diagram of a distribution box provided in an embodiment of the present invention.

[0211] Reference Figure 8 The distribution box includes a distribution box housing 11, and a control device 12, a collection device 13, a power supply 14, a circuit breaker 15 and an automatic reclosing switch 16 arranged inside the distribution box housing 11; it also includes an alarm system 17 of the distribution box, and the alarm system 17 of the distribution box is arranged in the control device 12.

[0212] The data acquisition device 13 , the power supply 14 , the circuit breaker 15 and the automatic reclosing switch 16 are respectively connected to the control device 12 ; the automatic reclosing switch 16 is connected to the circuit breaker 15 .

[0213] The power supply 14 is used to supply power to the distribution box.

[0214] In one embodiment, the distribution box also includes a network relay; it also includes a display device, an alarm device, a power indicator light and a button arranged on the outer shell of the distribution box; the collection device includes at least one of the following: a current collection device, a voltage collection device, a water immersion sensor, a temperature sensor, a humidity sensor and an electric energy meter.

[0215] The network relay is connected to the control device; the control device is connected to the host computer of the peripheral device through the network relay.

[0216] The embodiment of the present invention provides a distribution box, in which the operation data of the collection device can be obtained in real time, and the data can be analyzed and predicted, so as to dynamically adjust the fault judgment standard and identify potential fault risks in advance. The computer program product provided by the embodiment of the present invention includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be referred to the method embodiment, which will not be repeated here.

[0217] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0218] In addition, in the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0219] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.

[0220] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0221] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An alarm method for a distribution box, characterized in that: A control device applied to a distribution box; the distribution box also includes at least one acquisition device; the acquisition device is connected to the control device; the device includes: Acquire the collected data of the collection device; the collected data includes a collection device identifier, a real-time data value and a corresponding timestamp; Input the acquisition device identifier and the timestamp into a pre-trained threshold model, and output the current threshold; the threshold model is used to output the threshold corresponding to the acquisition data; Determining whether the real-time data value exceeds the current threshold; If the current threshold is exceeded, it is determined that the distribution box has a fault, and a first alarm message is generated; If the current threshold is not exceeded, the acquisition device identifier and the real-time data value are input into a pre-trained data prediction model, and a fault prediction result is output; if the fault prediction result includes a fault level, a second alarm information is generated based on the fault level.

2. The alarm method for a distribution box according to claim 1, characterized in that: The threshold model is trained in the following way: Obtaining the historical timestamp, historical data value and device abnormality record corresponding to each of the acquisition device identifiers; Extracting the time feature of the historical timestamp and the data feature in the historical data value, associating and saving the time feature, the data feature and the device abnormality record to obtain a historical association relationship; constructing a first training data set based on the collection device identifier, the historical timestamp, the historical data value, the abnormal record and the historical association relationship; The initial regression model is trained based on the first training data set until the initial regression model can determine the threshold of each acquisition device in different time periods to obtain the trained threshold model; the threshold model can output the current threshold according to the input real-time data value.

3. The alarm method for a distribution box according to claim 1, characterized in that: The data prediction model is trained in the following way: Obtaining the historical timestamp, historical data value and device abnormality record corresponding to each of the acquisition device identifiers; Extracting the time feature of the historical timestamp and the data feature in the historical data value, associating and saving the time feature, the data feature and the device abnormality record to obtain a historical association relationship; constructing a second training data set based on the collection device identifier, the historical timestamp, the historical data value, the abnormal record and the historical association relationship; Based on the second training data set, the initial deep learning model is trained until the initial deep learning model can obtain the change trend of each of the acquisition devices in different time periods, and obtain the trained data prediction model; the data training model can predict the next acquisition data corresponding to the acquisition device in the next time interval based on the input real-time data value, and output the fault prediction result based on the next acquisition data.

4. The alarm method for a distribution box according to claim 1, characterized in that: The acquisition device includes at least one of the following: a current acquisition device, a voltage acquisition device, a water immersion sensor, a temperature sensor, a humidity sensor and an electric energy meter; If the fault prediction result includes a fault level, after the step of generating second alarm information based on the fault level, the method further includes: Setting the collection device whose fault prediction result is the fault level as a fault collection device; Based on the preset correspondence between the collection devices and the weights, determining the weight value corresponding to each of the faulty collection devices; Based on the weight value, the fault level corresponding to the fault collection device and a preset risk level calculation method, a risk value of the distribution box is calculated; Based on the correspondence between preset risk data and risk levels, the risk level corresponding to the risk value is determined; the risk level includes low risk, medium risk and high risk.

5. The alarm method for a distribution box according to claim 4, characterized in that: The distribution box further comprises a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; the circuit breaker is a main circuit breaker; After the step of determining the risk level corresponding to the risk value based on the correspondence between the preset risk data and the risk level, the method further includes: When the risk level is the low risk, sending the second warning information to the display device so that the display device displays the second warning information; When the risk level is the medium risk, generating a disconnection signal, and sending the disconnection signal to the main circuit breaker, so that the main circuit breaker performs a disconnection operation; When the risk level is the high risk, the disconnection signal and the alarm signal are generated, the disconnection signal is sent to the main circuit breaker, and the alarm signal is sent to the alarm device, so that the main circuit breaker performs the disconnection operation, so that the alarm device alarms.

6. The alarm method for a distribution box according to claim 4, characterized in that: The distribution box also includes a power supply and a circuit breaker, a display device and an alarm device respectively connected to the power supply; the circuit breaker, the display device and the alarm device are respectively connected to the control device; the circuit breaker includes a main circuit breaker and at least one branch circuit breaker connected to the main circuit breaker; each branch circuit breaker is connected to an external load to form a branch circuit; the input end of the main circuit breaker is connected to the power supply; the output end of the main circuit breaker is connected to the input end of the branch circuit breaker; When the acquisition device is a branch current acquisition device and / or a branch voltage acquisition device provided on the branch circuit, after the step of determining the risk level corresponding to the risk value based on the correspondence between the preset risk data and the risk level, the method further includes: When the risk level is the low risk, sending the second warning information to the display device so that the display device displays the second warning information; When the risk level is the medium risk, a disconnection signal is generated, and the disconnection signal is sent to the branch circuit breaker corresponding to the fault collection device, so that the branch circuit breaker performs a disconnection operation; When the risk level is the high risk, the disconnection signal and the alarm signal are generated, the disconnection signal is sent to the main circuit breaker and the branch circuit breaker, and the alarm signal is sent to the alarm device so that the main circuit breaker and the branch circuit breaker perform the disconnection operation and the alarm device alarms.

7. The alarm method for a distribution box according to claim 5 or 6, characterized in that: The distribution box further includes an automatic reclosing switch, which is connected to the control device and the circuit breaker respectively; after the circuit breaker performs the disconnection operation, the method further includes: After a preset time interval, the automatic reclosing switch sends a closing signal to the circuit breaker, so that the circuit breaker performs a closing operation according to the closing signal.

8. An alarm system for a distribution box, characterized in that: The distribution box includes a control device, a network relay and at least one acquisition device; the network relay and the acquisition device are respectively connected to the control device; the alarm system includes: A data acquisition module, used to obtain the collected data of the collection device; the collected data includes a collection device identifier, a real-time data value and its corresponding timestamp; A fault judgment module is used to input the acquisition device identifier and the timestamp into a pre-trained threshold model and output a current threshold; the threshold model is used to output a threshold corresponding to the acquisition data; The fault judgment module is further used to judge whether the real-time data value exceeds the current threshold; The fault judgment module is further configured to determine that a fault exists in the distribution box and generate a first alarm message if the current threshold is exceeded; The fault judgment module is also used to input the acquisition device identifier and the real-time data value into a pre-trained data prediction model if the current threshold is not exceeded, and output a fault prediction result; if the fault prediction result includes a fault level, generate a second alarm information based on the fault level.

9. A distribution box, characterized in that: The invention comprises a distribution box housing, and a control device, a collection device, a power supply, a circuit breaker and an automatic reclosing switch arranged inside the distribution box housing; and further comprises an alarm system of the distribution box as claimed in claim 8, wherein the alarm system of the distribution box is arranged in the control device; The acquisition device, the power supply, the circuit breaker and the automatic reclosing switch are respectively connected to the control device; the automatic reclosing switch is connected to the circuit breaker; The power supply is used to supply power to the distribution box.

10. The distribution box according to claim 9, characterized in that: The distribution box also includes a network relay; a display device, an alarm device, a power indicator light and a button arranged on the housing of the distribution box; the acquisition device includes at least one of the following: a current acquisition device, a voltage acquisition device, a water immersion sensor, a temperature sensor, a humidity sensor and an electric energy meter; The network relay is connected to the control device; the control device is connected to the host computer of the peripheral device through the network relay.

Images