Wireless network out-of-service early warning and device and computer equipment

Abstract

The invention discloses a wireless network out-of-service early warning method and device and computer equipment. The method comprises the steps that a pre-trained rainfall prediction model is adopted to predict predicted rainfall in a to-be-early-warned time period; acquiring the weather forecast rainfall in the to-be-early-warned time period, and adjusting the predicted rainfall and the weather forecast rainfall according to a predetermined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, the predetermined adjustment factor is determined according to the accuracy of the predicted rainfall and the weather forecast rainfall; and determining a wireless network out-of-service probability in the time period to be pre-warned according to the adjusted predicted rainfall and the adjusted weather forecast rainfall, and generating pre-warning information according to the wireless network out-of-service probability in the time period to be pre-warned.

Description

Technical Field

[0001] This application relates to the field of wireless communication technologies, and in particular, to a method, apparatus, and computer device for early warning of wireless network service outage. Background Art

[0002] At present, wireless communication technologies are ubiquitous in all aspects of life. However, natural disasters, especially heavy rain brought by typhoons and continuous heavy rain during the rainy season, are extremely likely to cause urban waterlogging and power outages, which in turn lead to the outage of wireless network devices, resulting in communication interruption in the disaster area and having a great impact on local production and life. The wireless network service outage alarm is real-time alarm data. When the alarm is reported, the device is already in the outage state, with a certain lag in time. Existing technologies lack accurate early warning means for network service outages in disasters. Usually, it is estimated based on information such as weather forecasts and emergency response levels, with insufficient accuracy, a relatively large regional granularity, and differences in drainage and natural disaster resistance capabilities in different regions, resulting in inaccurate estimation situations. Therefore, it is difficult for emergency communication command and dispatch departments to take targeted preventive measures and advance dispatching and deployment of emergency resources based on existing early warning means, greatly affecting the efficiency of disaster relief work. There is a very close connection between the wireless network service outage early warning method and rainfall prediction. In the existing field of rainfall prediction technologies, the prediction methods include empirical formulas, statistical models (such as autoregressive models), and physical-based numerical weather prediction models. These methods are usually simplified relationships based on long-term observations, and their prediction capabilities are limited by the applicable range and accuracy of the empirical formulas. Summary of the Invention

[0003] Embodiments of this application provide a method, apparatus, and computer device for early warning of wireless network service outage, so as to at least solve the technical problem in related technologies that the accuracy of early warning is reduced due to the low accuracy of the prediction of the wireless network service outage probability.

[0004] According to one aspect of the embodiments of this application, a method for early warning of wireless network service outage is provided, including: predicting the predicted rainfall during the period to be warned by using a pre-trained rainfall prediction model; obtaining the weather forecast rainfall during the period to be warned, and respectively adjusting the predicted rainfall and the weather forecast rainfall according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, where the pre-determined adjustment factor is determined according to the accuracies of the predicted rainfall and the weather forecast rainfall; determining the wireless network service outage probability during the period to be warned according to the adjusted predicted rainfall and the adjusted weather forecast rainfall, and generating an early warning message according to the wireless network service outage probability during the period to be warned.

[0005] Optionally, the rainfall prediction model is determined in the following manner, including: obtaining historical rainfall data; using the rainfall data for each hour within a preset period in the historical rainfall data as the input of the rainfall prediction model to train the rainfall prediction model until the loss function value of the rainfall prediction model converges. Wherein, the rainfall prediction model includes: an input layer, two serially-connected long short-term memory layers, and an output layer. The loss function of the rainfall prediction model is a mean square error loss function, and the loss function of the rainfall prediction model is used to represent the difference between the predicted rainfall and the actual rainfall.

[0006] Optionally, the method further includes: extracting multiple time periods from the historical rainfall data that are the same as the predicted rainfall; respectively obtaining the wireless network outage probabilities for the multiple time periods, and determining the maximum value, minimum value, and average value of the wireless network outage probabilities for the multiple time periods; determining the wireless network outage probability corresponding to the predicted rainfall according to the maximum value, minimum value, and average value of the wireless network outage probabilities for the multiple time periods.

[0007] Optionally, determining the wireless network outage probability corresponding to the predicted rainfall according to the maximum value, minimum value, and average value of the wireless network outage probabilities for the multiple time periods includes: determining a first weight corresponding to the maximum value, a second weight corresponding to the minimum value, and a third weight corresponding to the average value according to the historical rainfall data; determining the weighted sum of the maximum value, the minimum value, and the average value as the wireless network outage probability corresponding to the predicted rainfall.

[0008] Optionally, the method further includes: respectively obtaining the groups to which the predicted rainfall and the weather forecast rainfall belong, wherein different groups correspond to different rainfall levels, and the rainfall difference between each group and the corresponding adjacent group is a preset rainfall; determining an adjustment factor for the predicted rainfall based on the mean absolute percentage error between all predicted values and actual observed values within the first group, wherein the first group contains multiple observation points, each observation point includes a model predicted value and a corresponding actual observed value, and the group to which the predicted rainfall belongs is the first group; determining an adjustment factor for the weather forecast rainfall based on the mean absolute percentage error between all predicted values and actual observed values within the second group, wherein the second group contains multiple observation points, each observation point includes a weather forecast predicted value and a corresponding actual observed value, and the group to which the weather forecast rainfall belongs is the second group.

[0009] Optionally, adjust the predicted rainfall and the weather forecast rainfall respectively according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, including: determining the product of the adjustment factor of the predicted rainfall and the predicted rainfall as the adjusted predicted rainfall; determining the product of the adjustment factor of the weather forecast rainfall and the weather forecast rainfall as the adjusted weather forecast rainfall.

[0010] Optionally, determine the wireless network disconnection probability during the warning period according to the adjusted predicted rainfall and the adjusted weather forecast rainfall, including: respectively determining the first wireless network disconnection probability and the second wireless network disconnection probability during the warning period according to the adjusted predicted rainfall and the adjusted weather forecast rainfall; determining the weight of the first wireless network disconnection probability and the weight of the second wireless network disconnection probability according to the adjustment factor of the predicted rainfall and the adjustment factor of the weather forecast rainfall; determining the weighted sum of the first wireless network disconnection probability and the second wireless network disconnection probability as the network disconnection probability during the warning period.

[0011] According to another aspect of the embodiments of the present application, there is also provided a wireless network disconnection warning device, including: a prediction module, configured to predict the predicted rainfall during the warning period by using a pre-trained rainfall prediction model; an acquisition module, configured to acquire the weather forecast rainfall during the warning period, and adjust the predicted rainfall and the weather forecast rainfall respectively according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, where the pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall and the weather forecast rainfall; a warning module, configured to determine the wireless network disconnection probability during the warning period according to the adjusted predicted rainfall and the adjusted weather forecast rainfall, and generate a warning message according to the wireless network disconnection probability during the warning period.

[0012] According to still another aspect of the embodiments of the present application, there is also provided a computer device, including: a memory and a processor, where the memory is used to store program instructions; the processor is connected to the memory and is configured to execute the above-mentioned wireless network disconnection warning.

[0013] According to yet another aspect of the embodiments of the present application, there is also provided a non-volatile storage medium, which includes a stored computer program, where the device where the non-volatile storage medium is located executes the above-mentioned wireless network disconnection warning by running the computer program.

[0014] According to another aspect of the embodiments of the present application, a computer program product is further provided, including computer instructions, which implement the above-mentioned wireless network service outage warning when executed by a processor.

[0015] In the embodiments of the present application, a pre-trained rainfall prediction model is used to predict the predicted rainfall during the period to be warned; the weather forecast rainfall during the period to be warned is obtained, and the predicted rainfall and the weather forecast rainfall are respectively adjusted according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, and the pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall and the weather forecast rainfall; the wireless network service outage probability during the period to be warned is determined according to the adjusted predicted rainfall and the adjusted weather forecast rainfall, and a warning message is generated according to the wireless network service outage probability during the period to be warned, so as to achieve the purpose of predicting rainfall using an artificial intelligence model, adjusting the predicted rainfall using an adjustment factor, and finally predicting the wireless network service outage probability using the adjusted predicted rainfall and the weather forecast rainfall, thereby achieving the technical effect of improving the prediction accuracy rate of the wireless network service outage probability, and further solving the technical problem in the related art that the warning accuracy is reduced due to the low prediction accuracy rate of the wireless network service outage probability. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments and descriptions thereof of the present application are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0017] Figure 1 is a hardware structure block diagram of a computer terminal for implementing wireless network service outage warning according to an embodiment of the present application;

[0018] Figure 2 is a flowchart of a wireless network service outage warning according to an embodiment of the present application;

[0019] Figure 3 is a flowchart of training a rainfall prediction model according to an embodiment of the present application;

[0020] Figure 4 is a flowchart of a method for determining a wireless network service outage according to an embodiment of the present application;

[0021] Figure 5 is a flowchart of another wireless network service outage warning according to an embodiment of the present application;

[0022] Figure 6 is a structural diagram of a wireless network service outage warning device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] In order to enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0025] The information collected in the embodiments of this application is information and data authorized by the user or fully authorized by all parties. Moreover, for the processing of relevant data such as collection, storage, use, processing, transmission, provision, disclosure, and application, all comply with the relevant laws, regulations and standards of the relevant regions, necessary confidentiality measures are taken, it does not violate public order and good customs, and a corresponding operation entry is provided for the user to choose to authorize or reject the automated decision-making result; if the user chooses to reject, the expert decision-making process will be entered.

[0026] In order to solve the problems existing in the related art, the embodiments of this application provide a wireless network service outage warning, and this method can run on Figure 1 the computer terminal shown below, and the following is an explanation of this computer terminal.

[0027] The wireless network service outage warning embodiments provided by the embodiments of this application can be executed on a mobile terminal, a computer terminal or a similar computing device. Figure 1 The following shows a hardware structure block diagram of a computer terminal for implementing wireless network service outage warning. As Figure 1As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ……, 102n in the figure) (the processor may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions connected through wired and / or wireless networks. In addition, it may further include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those of ordinary skill in the art can understand that Figure 1 the structure shown is only schematic and does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 may further include more or fewer components than Figure 1 shown in, or have a different configuration from Figure 1 that shown.

[0028] It should be noted that the above one or more processors and / or other data processing circuits are generally referred to as "data processing circuits" herein. The data processing circuit may be embodied in software, hardware, firmware, or any combination thereof, in whole or in part. In addition, the data processing circuit may be a single independent processing module, or be incorporated in whole or in part into any one of the other elements in the computer terminal 10. As involved in the embodiments of the present application, the data processing circuit is a kind of processor control (such as the selection of a variable resistance terminal path connected to an interface).

[0029] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the wireless network service outage warning in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizes the above-mentioned wireless network service outage warning. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely set relative to the processor, and these remote memories can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0030] The transmission module 106 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission module 106 can be a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0031] The display can be, for example, a touch-screen liquid crystal display (LCD), which enables a user to interact with the user interface of the computer terminal 10.

[0032] It should be noted here that in some alternative embodiments, the above Figure 1 illustrated computer terminal may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware elements and software elements. It should be pointed out that Figure 1 is only an example of a specific specific instance and is intended to illustrate the types of components that may exist in the above computer terminal.

[0033] Under the above operating environment, an embodiment of a wireless network service outage warning is provided in an embodiment of the present application. It should be noted that the steps illustrated in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is illustrated in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0034] Figure 2 is a flowchart of a wireless network service outage warning according to an embodiment of the present application. As Figure 2 shown, the method includes the following steps:

[0035] Step S202, predicting the predicted rainfall during the warning period using a pre-trained rainfall prediction model;

[0036] Step S204, obtaining the weather forecast rainfall during the warning period, and respectively adjusting the predicted rainfall and the weather forecast rainfall according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, where the pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall and the weather forecast rainfall;

[0037] Step S206: Determine the wireless network service outage probability during the to-be-forecast period based on the adjusted predicted rainfall and the adjusted weather forecast rainfall, and generate a warning message according to the wireless network service outage probability during the to-be-forecast period.

[0038] Through the above steps S202 to S206, the predicted rainfall during the to-be-forecast period is predicted by using a pre-trained rainfall prediction model; the weather forecast rainfall during the to-be-forecast period is obtained, and the predicted rainfall and the weather forecast rainfall are respectively adjusted according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, where the pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall and the weather forecast rainfall; the wireless network service outage probability during the to-be-forecast period is determined based on the adjusted predicted rainfall and the adjusted weather forecast rainfall, and a warning message is generated according to the wireless network service outage probability during the to-be-forecast period, thereby achieving the purpose of predicting rainfall by using an artificial intelligence model, adjusting the predicted rainfall by using an adjustment factor, and finally predicting the wireless network service outage probability by using the adjusted predicted rainfall and the weather forecast rainfall, thus realizing the technical effect of improving the prediction accuracy rate of the wireless network service outage probability, and further solving the technical problem in the related art that the warning accuracy is reduced due to the low prediction accuracy rate of the wireless network service outage probability. The following is a detailed description.

[0039] In some embodiments of the present application, the rainfall prediction model is determined in the following manner, including: obtaining historical rainfall data; using the rainfall data of each hour within a preset period in the historical rainfall data as the input of the rainfall prediction model to train the rainfall prediction model until the loss function value of the rainfall prediction model converges, where the rainfall prediction model includes: an input layer, two serially-connected long short-term memory layers, and an output layer, the loss function of the rainfall prediction model is a mean square error loss function, and the loss function of the rainfall prediction model is used to represent the difference between the predicted rainfall and the actual rainfall.

[0040] It should be noted that before training the rainfall prediction model, it is necessary to collect historical rainfall data first. The specific data acquisition process is as follows:

[0041] Step 1: Data collection. The collected original data is stored in a suitable database for easy management and access. The selected database is MySQL.

[0042] Step 2: Data cleaning. Remove any non-numerical items or data points that do not meet the experimental requirements in the records; identify and process outliers, and identify and replace or delete outliers by the IQR (interquartile range) method.

[0043] Step 3, Missing value handling: For the missing rainfall data, the time series interpolation method, the nearest neighbor interpolation method, is used to fill in the missing values. For the value of the service outage rate, if it is missing, the method based on the historical average is used to fill it in.

[0044] Step 4, Data transformation: Standardize and normalize the hourly rainfall and service outage rate data to improve the efficiency and stability of model training.

[0045] Step 5, Dataset division: Using the time series division method, divide the dataset composed of historical rainfall data into a training set (80%) and a test set (20%), ensuring that the test set contains the latest data.

[0046] Furthermore, the prediction model training steps are as Figure 3 shown, including:

[0047] Build the model structure. The rainfall prediction model includes: Input layer: Input Shape, which is used to receive the rainfall data of a preset period in the training set as input. For example, the rainfall data of each hour in the past 24 hours is used as input. This means that the shape of the input vector is (24, 1), where each time step represents the rainfall of one hour, 24 represents the length of the time series, that is, the number of consecutive points of data in time. In this example, it refers to the data of the past 24 hours, and 1 means that only one feature is considered at each time step: the rainfall of that hour.

[0048] LSTM layer: Units, which consists of a preset number of units. For example, it consists of 128 units and can capture the complex patterns of rainfall changes over time. To improve the performance of the model, two LSTM layers can be set, which are connected in series to form a deeper network structure to better learn the time series features of rainfall.

[0049] Output layer: Output Shape: (1), which outputs the predicted value of the cumulative rainfall for the period to be warned by a single neuron. For example, the predicted value of the cumulative rainfall in the next 24 hours.

[0050] The training process of the rainfall prediction model is as follows:

[0051] Determine the loss function respectively: Use the mean squared error (MSE) loss function, which measures the difference between the predicted value and the actual rainfall.

[0052] Optimizer: Use the Adam optimizer to optimize the parameters, and set the initial learning rate to 0.001, which can be adjusted according to the performance during training.

[0053] Batch size and number of training epochs: Select a batch size of 64 and set the number of training epochs to 50. Generally, the training will continue for multiple epochs until the value of the loss function converges or no longer decreases significantly.

[0054] After the training is completed, the model needs to be evaluated on a test data set that was not involved in the training to verify its prediction ability. We use the mean squared error (MSE) and mean absolute error (MAE) as evaluation metrics, which can help us understand the accuracy and stability of the model when predicting future rainfall. The evaluation results will indicate the performance of the model on new data, ensuring that it not only performs well on the training set but also has the generalization ability to provide accurate rainfall predictions in practical applications.

[0055] After the training iteration, the prediction model finally outputs the predicted value of the cumulative rainfall (P_{rain}) within the warning period, with the unit of millimeters (mm).

[0056] In some embodiments of the present application, the wireless network outage probability corresponding to the rainfall can be calculated in the following manner: Extract multiple time periods from the historical rainfall data that are the same as the predicted rainfall; respectively obtain the wireless network outage probabilities of the multiple time periods, and determine the maximum value, minimum value, and average value of the wireless network outage probabilities of the multiple time periods according to the wireless network outage probabilities of the multiple time periods; determine the wireless network outage probability corresponding to the predicted rainfall according to the maximum value, minimum value, and average value of the wireless network outage probabilities of the multiple time periods.

[0057] Specifically, compare the rainfall predicted by the model with historical rainfall events, and match multiple historical time periods with the same rainfall, such as: from March 10th to March 11th, from May 10th to May 11th, from July 1st to July 2nd, etc.

[0058] It can be understood that the above-mentioned same rainfall means that the rainfall difference does not exceed a preset threshold, then it is determined that the rainfall is the same.

[0059] For each found historical time period with the same rainfall, extract the corresponding outage rate data.

[0060] For the collected outage rate data, calculate the maximum value (Max), minimum value (Min), and average value (Mean) in multiple historical time periods.

[0061] Among them, determining the wireless network service outage probability corresponding to the predicted rainfall according to the maximum value, minimum value, and average value of the wireless network service outage probabilities in the multiple time periods includes: determining the first weight corresponding to the maximum value, the second weight corresponding to the minimum value, and the third weight corresponding to the average value according to the historical rainfall data; determining the weighted sum of the maximum value, the minimum value, and the average value as the wireless network service outage probability corresponding to the predicted rainfall.

[0062] Specifically, the calculation process of the wireless network service outage probability is as Figure 4 shown.

[0063] The specific formula for calculating the wireless network service outage probability is as follows:

[0064] Q = w1 * Max + w2 * Min + w3 * Mean;

[0065] In the formula, w1 represents the first weight, w2 represents the second weight, and w3 represents the third weight.

[0066] It should be noted that the calculation methods of the first weight, the second weight, and the third weight can be determined in the following way: construct a linear regression model, where the service outage rate is the dependent variable, and the maximum value, the minimum value, and the average value are the independent variables, in the form of: Q = w1 * Max + w2 * Min + w3 * Mean. Use the historical rainfall data to train the regression model. During the training process, the least squares method is adopted to find the optimal weights w1, w2, and w3 by minimizing the sum of the squares of the residuals.

[0067] In some embodiments of the present application, the adjustment factor can be determined in the following way, including: respectively obtaining the groups where the predicted rainfall and the weather forecast rainfall are located. Among them, different groups correspond to different rainfall levels, and the rainfall difference between each group and the adjacent group is a preset rainfall; determining the adjustment factor of the predicted rainfall based on the mean absolute percentage error between all the predicted values and the actual observed values within the first group. Among them, the first group contains multiple observation points, and each observation point contains a model predicted value and a corresponding actual observed value, and the group where the predicted rainfall is located is the first group; determining the adjustment factor of the weather forecast rainfall based on the mean absolute percentage error between all the predicted values and the actual observed values within the second group. Among them, the second group contains multiple observation points, and each observation point contains a weather forecast predicted value and a corresponding actual observed value, and the group where the weather forecast rainfall is located is the second group.

[0068] For example: Step 1, calculation of grouped accuracy: Divide the predicted rainfall value P of the model and the forecasted rainfall value F into groups (such as 0 - 5mm, 5 - 10mm, etc.) according to each preset rainfall amount, for example, 5mm, and divide them into groups: P_group (the first group) and F_group (the second group) to evaluate the accuracy of different rainfall levels.

[0069] Step 2, for each group, calculate their accuracy metrics respectively, that is, the Grouped Mean Absolute Percentage Error (Grouped MAPE):

[0070] Grouped MAPE_group = (100% / n_group) * ∑|(Pt - Ot) / Ot|;

[0071] Where Pt represents the predicted value within the group, Ot represents the actual observed value within the group, n_group represents the total number of observation points within the group, and t = 1…n_group.

[0072] It can be understood that each group contains multiple observation points, and the predicted rainfall of each observation point is within the rainfall range corresponding to the group.

[0073] Based on the above formula for calculating the Grouped Mean Absolute Percentage Error, determine the Grouped MAPE_model_group of the predicted rainfall and the Grouped MAPE_forecast_group of the forecasted rainfall respectively.

[0074] It can be understood that when calculating the Grouped Mean Absolute Percentage Error of the predicted rainfall, the data of the first group is used, and when calculating the Grouped Mean Absolute Percentage Error of the forecasted rainfall, the data of the second group is used.

[0075] The adjustment factor for the predicted rainfall is:

[0076] α_group = 1 - (Grouped MAPE_model_group / 100%);

[0077] The adjustment factor for the forecasted rainfall is:

[0078] β_group = 1 - (Grouped MAPE_forecast_group / 100%);

[0079] It should be noted that the predicted rainfall and the weather forecast rainfall are adjusted respectively according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall, including: determining the product of the adjustment factor of the predicted rainfall and the predicted rainfall as the adjusted predicted rainfall; determining the product of the adjustment factor of the weather forecast rainfall and the weather forecast rainfall as the adjusted weather forecast rainfall.

[0080] Specifically, the adjusted predicted rainfall is:

[0081] P'_group = α_group * P;

[0082] The adjusted weather forecast rainfall is:

[0083] F'_group = β_group * F.

[0084] In some embodiments of the present application, determining the wireless network outage probability during the period to be warned according to the adjusted predicted rainfall and the adjusted weather forecast rainfall includes: respectively determining the first wireless network outage probability and the second wireless network outage probability during the period to be warned according to the adjusted predicted rainfall and the adjusted weather forecast rainfall; determining the weight of the first wireless network outage probability and the weight of the second wireless network outage probability according to the adjustment factor of the predicted rainfall and the adjustment factor of the weather forecast rainfall; determining the weighted sum of the first wireless network outage probability and the second wireless network outage probability as the network outage probability during the period to be warned.

[0085] The network outage probability during the period to be warned can be determined by the following formula:

[0086] FinalAlert_group = γ_group * Q'_group + (1 - γ_group) * W'_group;

[0087] In the formula, Q'_group and W'_group are the first wireless network outage probability and the second wireless network outage probability obtained based on the adjusted model predicted rainfall (P'_group) and the weather forecast rainfall (F'_group) respectively, and γ_group represents the weight of the first wireless network outage probability.

[0088] Among them, the calculation methods of the first wireless network outage probability and the second wireless network outage probability are as follows:

[0089] Q'_group = w1 * Max_group + w2 * Min_group + w3 * Mean_group;

[0090] W'_group = w1 * Max_forecast_group + w2 * Min_forecast_group + w3 * Mean_forecast_group;

[0091] Wherein, Max_group, Min_group and Mean_group respectively represent the maximum value, minimum value and average value of the historical network outage probability obtained based on the rainfall predicted by the adjusted model; Max_forecast_group, Min_forecast_group and Mean_forecast_group respectively represent the maximum value, minimum value and average value of the historical network outage probability obtained based on the rainfall of the adjusted weather forecast. The calculation method can be determined by the above wireless network outage probability calculation formula and will not be elaborated here.

[0092] γ_group can be determined by the following formula:

[0093] γ_group = α_group / (α_group + β_group).

[0094] As Figure 5 shown, the embodiment of the present application also provides another prediction method for the wireless network outage probability, including: collecting historical data of rainfall and outage rate in recent years, saving them to a database, and ensuring the quality and availability of the data through a series of strict preprocessing steps. Based on the classical long short-term memory network (LSTM), an accurate rainfall prediction model is constructed, which effectively captures the time series characteristics of rainfall data. After the model is trained and evaluated, the predicted value of the cumulative rainfall within the next 24 hours is output. Based on the predicted rainfall value of the model, a retrospective analysis of the historical outage rate is carried out to establish an outage rate prediction formula. By grouping the rainfall data of the prediction model and the weather forecast, the accuracy rate is evaluated, and an adjustment factor is introduced to correct the predicted rainfall. Based on the predicted rainfall of the adjusted model and the weather forecast rainfall, the outage rate is calculated respectively according to the outage rate prediction formula, and the two are weighted and synthesized to obtain the final predicted value of the outage rate.

[0095] Figure 6 is a wireless network outage warning device according to an embodiment of the present application. The device includes:

[0096] A prediction module 60, configured to predict the predicted rainfall within the period to be warned by using a pre-trained rainfall prediction model;

[0097] An acquisition module 62, configured to acquire the predicted rainfall amount within the to-be-forecast period, and adjust the predicted rainfall amount and the forecast rainfall amount respectively according to a pre-determined adjustment factor to obtain an adjusted predicted rainfall amount and an adjusted forecast rainfall amount, where the pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall amount and the forecast rainfall amount;

[0098] An early warning module 64, configured to determine the wireless network service outage probability within the to-be-forecast period according to the adjusted predicted rainfall amount and the adjusted forecast rainfall amount, and generate an early warning message according to the wireless network service outage probability within the to-be-forecast period.

[0099] Through the above-mentioned wireless network service outage early warning device, the predicted rainfall amount within the to-be-forecast period is predicted by using a pre-trained rainfall prediction model; the forecast rainfall amount within the to-be-forecast period is acquired, and the predicted rainfall amount and the forecast rainfall amount are respectively adjusted according to a pre-determined adjustment factor to obtain an adjusted predicted rainfall amount and an adjusted forecast rainfall amount, where the pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall amount and the forecast rainfall amount; the wireless network service outage probability within the to-be-forecast period is determined according to the adjusted predicted rainfall amount and the adjusted forecast rainfall amount, and an early warning message is generated according to the wireless network service outage probability within the to-be-forecast period, thereby achieving the purpose of predicting the rainfall amount by using an artificial intelligence model, adjusting the predicted rainfall amount by using an adjustment factor, and finally predicting the wireless network service outage probability by using the adjusted predicted rainfall amount and the forecast rainfall amount, thereby realizing the technical effect of improving the prediction accuracy rate of the wireless network service outage probability, and further solving the technical problem in the related art that the early warning accuracy is reduced due to the low prediction accuracy rate of the wireless network service outage probability.

[0100] The prediction module 60 includes: a model sub-module, configured to determine a rainfall prediction model in the following manner, including: acquiring historical rainfall data; using the rainfall data of each hour within a preset period in the historical rainfall data as the input of the rainfall prediction model to train the rainfall prediction model until the loss function value of the rainfall prediction model converges, where the rainfall prediction model includes: an input layer, two serially-connected long short-term memory layers, and an output layer, the loss function of the rainfall prediction model is a mean square error loss function, and the loss function of the rainfall prediction model is used to represent the difference between the predicted rainfall amount and the actual rainfall amount.

[0101] The obtaining module 62 includes: a probability sub-module for extracting multiple time periods identical to the predicted rainfall amount from the historical rainfall data; respectively obtaining the wireless network service outage probabilities for the multiple time periods, and determining the maximum value, minimum value, and average value of the wireless network service outage probabilities for the multiple time periods based on the wireless network service outage probabilities for the multiple time periods; and determining the wireless network service outage probability corresponding to the predicted rainfall amount based on the maximum value, minimum value, and average value of the wireless network service outage probabilities for the multiple time periods.

[0102] The probability sub-module includes: a probability unit for determining the wireless network service outage probability corresponding to the predicted rainfall amount based on the maximum value, minimum value, and average value of the wireless network service outage probabilities for the multiple time periods, including: determining a first weight corresponding to the maximum value, a second weight corresponding to the minimum value, and a third weight corresponding to the average value based on the historical rainfall data; and determining the weighted sum of the maximum value, the minimum value, and the average value as the wireless network service outage probability corresponding to the predicted rainfall amount.

[0103] The obtaining module 62 further includes: an adjustment factor sub-module for respectively obtaining the groups to which the predicted rainfall amount and the weather forecast rainfall amount belong, where different groups correspond to different rainfall levels, and the rainfall difference between each group and the corresponding adjacent group is a preset rainfall amount; determining the adjustment factor for the predicted rainfall amount based on the mean absolute percentage error between all the predicted values and the actual observed values within the first group, where the first group contains multiple observation points, each observation point includes a model predicted value and a corresponding actual observed value, and the group to which the predicted rainfall amount belongs is the first group; and determining the adjustment factor for the weather forecast rainfall amount based on the mean absolute percentage error between all the predicted values and the actual observed values within the second group, where the second group contains multiple observation points, each observation point includes a weather forecast predicted value and a corresponding actual observed value, and the group to which the weather forecast rainfall amount belongs is the second group.

[0104] The adjustment factor sub-module includes: an adjustment unit for respectively adjusting the predicted rainfall amount and the weather forecast rainfall amount according to the pre-determined adjustment factors to obtain the adjusted predicted rainfall amount and the adjusted weather forecast rainfall amount, including: determining the product of the adjustment factor of the predicted rainfall amount and the predicted rainfall amount as the adjusted predicted rainfall amount; and determining the product of the adjustment factor of the weather forecast rainfall amount and the weather forecast rainfall amount as the adjusted weather forecast rainfall amount.

[0105] The adjustment unit includes: a warning subunit, configured to determine the probability of wireless network service interruption during the to-be-warned period according to the adjusted predicted rainfall and the adjusted weather forecast rainfall, including: determining the first probability of wireless network service interruption and the second probability of wireless network service interruption during the to-be-warned period according to the adjusted predicted rainfall and the adjusted weather forecast rainfall respectively; determining the weight of the first probability of wireless network service interruption and the weight of the second probability of wireless network service interruption according to the adjustment factor of the predicted rainfall and the adjustment factor of the weather forecast rainfall; and determining the weighted sum of the first probability of wireless network service interruption and the second probability of wireless network service interruption as the probability of network service interruption during the to-be-warned period.

[0106] It should be noted that Figure 6 the wireless network service interruption warning device shown is used to execute Figure 2 the wireless network service interruption warning shown. Therefore, the relevant explanations in the above wireless network service interruption warning also apply to this wireless network service interruption warning device, and will not be elaborated here.

[0107] The embodiment of the present application further provides a computer device, including: a memory and a processor. The memory is used to store program instructions. The processor is connected to the memory and is used to execute the above wireless network service interruption warning.

[0108] The embodiment of the present application further provides a non-volatile storage medium, which includes a stored computer program. The device where the non-volatile storage medium is located executes the above wireless network service interruption warning by running the computer program.

[0109] The embodiment of the present application further provides a computer program product, including computer instructions, which implement the steps of the wireless network service interruption warning in the present application when executed by a processor.

[0110] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0111] In the above embodiments of the present application, the descriptions of the respective embodiments have their own focuses. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0112] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units can be a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of units or modules can be in electrical or other forms.

[0113] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0114] In addition, in each embodiment of the present application, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0115] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. And the aforementioned storage medium includes: USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks or optical discs and other various media that can store program codes.

[0116] The above is only the preferred embodiment of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A method for early warning of wireless network service outage, characterized in that, Including: Predict the predicted rainfall during the warning period using a pre-trained rainfall prediction model; Obtain the weather forecast rainfall during the warning period, and adjust the predicted rainfall and the weather forecast rainfall respectively according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted weather forecast rainfall. The pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall and the weather forecast rainfall; Determine the wireless network service outage probability during the warning period according to the adjusted predicted rainfall and the adjusted weather forecast rainfall, and generate a warning message according to the wireless network service outage probability during the warning period.

2. The method according to claim 1, characterized in that The rainfall prediction model is determined by the following method, including: Obtain historical rainfall data; Use the rainfall data of each hour within a preset period in the historical rainfall data as the input of the rainfall prediction model to train the rainfall prediction model until the loss function value of the rainfall prediction model converges. Among them, the rainfall prediction model includes: an input layer, two cascaded long short-term memory layers and an output layer. The loss function of the rainfall prediction model is the mean square error loss function, and the loss function of the rainfall prediction model is used to represent the difference between the predicted rainfall and the actual rainfall.

3. The method according to claim 1, wherein The method further includes: Extract multiple periods from the historical rainfall data that are the same as the predicted rainfall; Obtain the wireless network service outage probabilities of the multiple periods respectively, and determine the maximum value, minimum value and average value of the wireless network service outage probabilities of the multiple periods according to the wireless network service outage probabilities of the multiple periods; Determine the wireless network service outage probability corresponding to the predicted rainfall according to the maximum value, minimum value and average value of the wireless network service outage probabilities of the multiple periods.

4. The method according to claim 3, characterized in that, Determine the wireless network service outage probability corresponding to the predicted rainfall according to the maximum value, minimum value and average value of the wireless network service outage probabilities of the multiple periods, including: Determine the first weight corresponding to the maximum value, the second weight corresponding to the minimum value and the third weight corresponding to the average value according to the historical rainfall data; Determine the weighted sum of the maximum value, the minimum value and the average value as the wireless network service outage probability corresponding to the predicted rainfall.

5. The method according to claim 1, characterized in that The method further includes: Obtain the groups to which the predicted rainfall and the weather forecast rainfall belong respectively. Among them, different groups correspond to different rainfall levels, and the rainfall difference between each group and the adjacent group is the preset rainfall; Determine the adjustment factor of the predicted rainfall based on the mean absolute percentage error between all predicted values and actual observed values within the first group. Among them, the first group contains multiple observation points, each observation point contains a model predicted value and a corresponding actual observed value, and the group to which the predicted rainfall belongs is the first group. Determine an adjustment factor for the forecast rainfall based on the mean absolute percentage error between all predicted values and actual observed values within a second group, where the second group contains multiple observation points, each observation point includes a weather forecast predicted value and a corresponding actual observed value, and the group to which the forecast rainfall belongs is the second group.

6. The method according to claim 5, wherein Adjust the predicted rainfall and the forecast rainfall respectively according to the pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted forecast rainfall, including: Determine the product of the adjustment factor of the predicted rainfall and the predicted rainfall as the adjusted predicted rainfall; Determine the product of the adjustment factor of the forecast rainfall and the forecast rainfall as the adjusted forecast rainfall.

7. The method according to claim 6, wherein Determine the wireless network outage probability during the warning period according to the adjusted predicted rainfall and the adjusted forecast rainfall, including: Determine the first wireless network outage probability and the second wireless network outage probability during the warning period respectively according to the adjusted predicted rainfall and the adjusted forecast rainfall; Determine the weight of the first wireless network outage probability and the weight of the second wireless network outage probability according to the adjustment factor of the predicted rainfall and the adjustment factor of the forecast rainfall; Determine the weighted sum of the first wireless network outage probability and the second wireless network outage probability as the network outage probability during the warning period.

8. A wireless network service outage warning device, characterized in that, Include: A prediction module, configured to predict the predicted rainfall during the warning period by using a pre-trained rainfall prediction model; An acquisition module, configured to acquire the forecast rainfall during the warning period, and adjust the predicted rainfall and the forecast rainfall respectively according to a pre-determined adjustment factor to obtain the adjusted predicted rainfall and the adjusted forecast rainfall, where the pre-determined adjustment factor is determined according to the accuracy rates of the predicted rainfall and the forecast rainfall; A warning module, configured to determine the wireless network outage probability during the warning period according to the adjusted predicted rainfall and the adjusted forecast rainfall, and generate a warning message according to the wireless network outage probability during the warning period.

9. A computer device, characterized in that, Include: A memory and a processor, where the memory is used to store program instructions; The processor, connected to the memory, is configured to execute the wireless network outage warning method according to any one of claims 1 to 7.

10. A computer program product, comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, the wireless network outage warning method according to any one of claims 1 to 7 is implemented.

Images