Fuse trigger warning method, server, medium and equipment

By predicting the number of information requests and feedback times by predicting the prediction model, the circuit breaker risk is identified in advance, and the existing circuit breaker mechanism lacks buffering time is solved, the stability and reliability of the system are improved, and the user experience is improved.

CN117785628BActive Publication Date: 2024-11-22SHENZHEN DIDATRAVEL TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202311656441.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-04
Publication Date
2024-11-22
Estimated Expiration
2043-12-04

AI Technical Summary

Technical Problem

The existing circuit breaker mechanism lacks buffering time to take early response measures when judging sudden failures or high concurrency conditions, resulting in poor system stability and reliability and poor user experience.

Method used

Through the actual number of unfeedbacks requests and the number of information requests/feedbacks in pre-designed cycles, data is encoded using prediction models (including forget gates, input gates, output gates, and unit status) to predict the number of unfeedbacks in subsequent count cycles. , if the threshold is exceeded, a circuit breaker risk is indicated and early warning is provided to take preventive measures.

Benefits of technology

Provide buffering time before the fuse occurs, improving the stability and reliability of the system and improving the user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117785628B_ABST
    Figure CN117785628B_ABST
Patent Text Reader

Abstract

The present invention discloses a fuse trigger warning method, server, medium and device. First, the current actual number of non-feedback times and the preset number of first information requests and the first information feedback times within the latest P counting cycles are counted. Then, the first information request times within the P counting cycles are encoded in chronological order and input into a convergent prediction model to predict the second information request times in the subsequent Q counting cycles. Similarly, the second information feedback times in the subsequent Q counting cycles are predicted. Based on the actual number of non-feedback times, the second information request times and the second information feedback times, the preset predicted number of non-feedback times in the latest Q subsequent counting cycles are calculated. Finally, if it is found that any predicted number of non-feedback times exceeds the preset non-feedback threshold, it will be prompted that there is a risk of triggering a fuse, so that there is buffer time to take preventive measures before the fuse occurs to ensure the stability and reliability of the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field, and in particular to a fuse triggering warning method, server, medium and equipment. Background Art

[0002] When requesting information, a circuit breaker is a communication mechanism used to protect the system from high concurrent request pressure. When the system is waiting for feedback from a large number of requests, if the processing capacity cannot meet the demand, it may cause the system to crash or become unavailable. To prevent this from happening, a circuit breaker mechanism can be used to limit the number of requests processed.

[0003] The existing fuse mechanism is triggered immediately when it is determined that there is a sudden failure or high concurrency. Its purpose is to protect the system from overload and ensure the stability and reliability of the system. However, due to the lack of buffer time for taking countermeasures in advance, the stability and reliability of the system will be poor, which will result in a poor user experience. Summary of the invention

[0004] Based on this, it is necessary to provide a fuse trigger warning method, server, medium and equipment to solve the problem that the system stability and reliability will be poor due to the lack of buffer time for taking countermeasures in advance, which will cause a poor user experience.

[0005] A fuse triggering warning method, the method comprising:

[0006] Count the actual number of unanswered requests for information from third parties;

[0007] Counting the number of first information requests and the number of first information feedbacks within the preset most recent P previous counting periods;

[0008] Encode the first information request times in P previous counting cycles in time sequence and input them into the converged prediction model in sequence to predict the second information request times in the preset next Q next counting cycles;

[0009] Encode the first information feedback times in P previous counting cycles in time sequence and input them into the converged prediction model in sequence to predict the second information feedback times in the preset next Q next counting cycles;

[0010] Calculate a preset predicted number of non-feedback times in the most recent Q subsequent counting cycles based on the actual number of non-feedback times, the second information request number, and the second information feedback number;

[0011] If the predicted number of non-feedback times in any subsequent counting cycle is greater than the preset non-feedback threshold, it indicates that there is a risk of triggering a circuit breaker in the future.

[0012] In one embodiment, the prediction model includes a plurality of loop structures, each loop structure including a forget gate, an input gate, an output gate, and a unit state;

[0013] The cell state connects the cell state at the previous moment with the cell state at the next moment for transmitting long-term memory;

[0014] The forget gate is used to control the amount of data of the unit state at the previous moment retained to the unit state at the current moment;

[0015] The input gate is used to control the amount of data of the unit state saved to the current moment in the data of the current input loop structure; wherein the data of the current input loop structure is the code corresponding to the first information request number or the code corresponding to the first information feedback number;

[0016] The output gate is used to control the amount of data outputted by the unit state at the current moment to transfer the short-term memory at the current moment; wherein the short-term memory indicates the second information request number or the second information feedback number.

[0017] In one embodiment, the calculation formula for predicting the number of times of no feedback is expressed as:

[0018]

[0019] In the above formula, N q indicates the predicted number of times no feedback is given in the qth subsequent counting cycle, N0 indicates the actual number of times no feedback is given, Indicates the number of second information requests in the i-th subsequent counting cycle, Indicates the number of second information feedbacks in the i-th subsequent counting cycle.

[0020] In one embodiment, the method further comprises:

[0021] When q>1, the smoothing value S(q) is calculated and expressed as:

[0022] S(q)=α×N q +(1-α)×N q-1

[0023] In the above formula, S(q) represents the smoothed value of the qth subsequent counting period, and α represents the smoothing coefficient, which is between 0 and 1;

[0024] If all calculated smoothed values ​​show an upward trend as time increases, it is determined that the risk of triggering a circuit breaker is increasing. If all calculated smoothed values ​​show a downward trend as time increases, it is determined that the risk of triggering a circuit breaker is decreasing.

[0025] A server, comprising:

[0026] A first counter is used to count the actual number of unanswered information requests to third parties at present;

[0027] A second counter, used to count the number of first information requests within the preset most recent P previous counting periods;

[0028] A third counter, used to count the number of first information feedbacks within the preset most recent P previous counting cycles;

[0029] A controller, used for encoding the first information request times within P previous counting cycles in chronological order and inputting them into a convergent prediction model in sequence, so as to predict the second information request times within the preset next Q subsequent counting cycles; encoding the first information feedback times within P previous counting cycles in chronological order and inputting them into a convergent prediction model in sequence, so as to predict the second information feedback times within the preset next Q subsequent counting cycles; calculating the predicted non-feedback times within the preset most recent Q subsequent counting cycles based on the actual non-feedback times, the second information request times and the second information feedback times; if there is any predicted non-feedback times within the subsequent counting cycle that is greater than a preset non-feedback threshold, it is indicated that there will be a risk of triggering a fuse in the future.

[0030] In one embodiment, the prediction model includes a plurality of loop structures, each loop structure including a forget gate, an input gate, an output gate, and a unit state;

[0031] The cell state connects the cell state at the previous moment with the cell state at the next moment for transmitting long-term memory;

[0032] The forget gate is used to control the amount of data of the unit state at the previous moment retained to the unit state at the current moment;

[0033] The input gate is used to control the amount of data of the unit state saved to the current moment in the data of the current input loop structure; wherein the data of the current input loop structure is the code corresponding to the first information request number or the code corresponding to the first information feedback number;

[0034] The output gate is used to control the amount of data outputted by the unit state at the current moment to transfer the short-term memory at the current moment; wherein the short-term memory indicates the second information request number or the second information feedback number.

[0035] In one embodiment, the calculation formula for predicting the number of times of no feedback is expressed as:

[0036]

[0037] In the above formula, N q indicates the predicted number of times no feedback is given in the qth subsequent counting cycle, N0 indicates the actual number of times no feedback is given, Indicates the number of second information requests in the i-th subsequent counting cycle, Indicates the number of second information feedbacks in the i-th subsequent counting cycle.

[0038] In one embodiment, the controller is further configured to:

[0039] When q>1, the smoothing value S(q) is calculated and expressed as:

[0040] S(q)=α×N q +(1-α)×N q-1

[0041] In the above formula, S(q) represents the smoothed value of the qth subsequent counting period, and α represents the smoothing coefficient, which is between 0 and 1;

[0042] If all calculated smoothed values ​​show an upward trend as time increases, it is determined that the risk of triggering a circuit breaker is increasing. If all calculated smoothed values ​​show a downward trend as time increases, it is determined that the risk of triggering a circuit breaker is decreasing.

[0043] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the steps of the above-mentioned fuse triggering warning method.

[0044] A fuse trigger warning device comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned fuse trigger warning method.

[0045] The present invention provides a fuse trigger warning method, a server, a medium and a device. First, the current actual number of non-feedbacks and the number of first information requests and the number of first information feedbacks in the latest P counting cycles are counted. Then, the number of first information requests in the P counting cycles is encoded in chronological order and input into a convergent prediction model to predict the number of second information requests in the subsequent Q counting cycles. Similarly, the number of first information feedbacks in the P counting cycles is encoded and input into another prediction model to predict the number of second information feedbacks in the subsequent Q counting cycles. Based on the actual number of non-feedbacks, the number of second information requests and the number of second information feedbacks, the preset number of predicted non-feedbacks in the latest Q subsequent counting cycles is calculated. Finally, if it is found that any predicted number of non-feedbacks in the subsequent counting cycle exceeds the preset non-feedback threshold, it will be prompted that there is a risk of triggering a fuse, so that there is a buffer time to take preventive measures before the fuse occurs to ensure the stability and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] in:

[0048] Figure 1 It is a flowchart of the fuse triggering warning method;

[0049] Figure 2 It is a schematic diagram of the structure of the prediction model;

[0050] Figure 3 This is a schematic diagram of the server structure;

[0051] Figure 4 This is the structural block diagram of the fuse triggering warning device. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0053] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0054] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] like Figure 1 As shown, Figure 1: is a flow chart of a fuse triggering warning method in an embodiment. The steps provided by the fuse triggering warning method in this embodiment include:

[0056] S101, counting the actual number of unanswered information requests to third parties.

[0057] Specifically, the terminal will send an information request to the system end, and the information request is used to call the third-party target service, that is, the third-party service. Then our system end forwards the call request to the third-party service. If the third-party service returns an error message to our system end (for example, the third-party service is unavailable or the feedback has timed out), our system end controls the counter to increase the actual number of times no feedback is given by 1; our system end returns the service call failure information to the terminal. On the contrary, if the third-party service returns feedback information to our system end, our system end controls the counter to reduce the actual number of times no feedback is given by 1; our system end returns feedback information to the terminal. In this step, we count the current actual number of times no feedback is given.

[0058] Exemplarily, the system end is an online hotel reservation system, which cooperates with a third-party hotel supplier named "HotelSupplier".

[0059] In case of an error (e.g., the third-party service is unavailable or the feedback timeout): The terminal sends a message containing a request to query the room price to the system. The system forwards this request to "HotelSupplier". For some reason (e.g., the "HotelSupplier" server fails), it fails to respond within the expected time. The system receives an error message after the timeout, and the actual number of unanswered requests is increased by 1. The system returns the service call failure information to the terminal to let the user know that there is a problem.

[0060] Normal process: The terminal sends a message containing a request to query the room price to the system. The system forwards this request to "HotelSupplier". "HotelSupplier" successfully processes the request and returns the room price information to the system. The system receives the feedback information from "HotelSupplier", reduces the actual number of unanswered requests by 1, and then returns the room price information to the terminal.

[0061] S102: Count the number of first information requests and the number of first information feedbacks within the preset most recent P previous counting periods.

[0062] Among them, P and the duration of each counting cycle are preset values ​​and can be set according to actual usage requirements.

[0063] The number of first information requests in the most recent P previous counting cycles can be expressed as:

[0064]

[0065] In the above formula, Indicates the number of first information requests within the previous P-1 counting cycles; Indicates the number of first information requests in the current counting period;

[0066] Similarly, the number of first information feedbacks in the most recent P previous counting cycles can be expressed as:

[0067]

[0068] In the above formula, Indicates the number of first information feedbacks within the previous P-1 counting cycles; Indicates the number of first information feedbacks in the current counting cycle.

[0069] S103, encoding the first information request times in P previous counting cycles in chronological order and inputting them into the converged prediction model in sequence to predict the second information request times in the preset next Q next counting cycles.

[0070] LSTM models usually expect input data in the format of (samples, time_steps, features), where samples represents the number of samples in the dataset. time_steps represents the number of time steps in each sample. features represents the number of features at each time step. The prepared data needs to be encoded in this format first, and then input into the converged prediction model in sequence.

[0071] In a specific embodiment, Figure 2 As shown, the prediction model includes multiple loop structures, each of which includes a forget gate, an input gate, an output gate, and a unit state.

[0072] Among them, the cell state connects the cell state of the previous moment with the cell state of the next moment, and has only a few linear interactions with other parts. The cell state is used to transmit long-term memory. The long-term memory passed in at the previous moment is represented by C t-1 , the long-term memory passed to the next moment is represented by C t .

[0073] The forget gate is used to control the amount of data that is retained from the cell state at the previous moment to the cell state at the current moment, that is, to control how to discard information, thereby achieving the forgetting function. The calculation formulas involved in the forget gate include:

[0074] f t =σ(W f [h t-1 , xt ]+b f )

[0075] In the above formula, W f is the weight matrix of the forget gate, b f is the constant of the forget gate, h t-1 is the output of the previous unit, x t is the input of this unit; the forget gate is the output h of the previous unit t-1 and the input x of this unit t is the input sigmoid function, C t-1 Each item in produces a value in [0, 1] to control the degree to which the unit state at the previous moment is forgotten.

[0076] This input gate is used to control the data x of the current input loop structure. t The amount of data in the cell state stored in (the code corresponding to the number of first information requests in this step) to the current moment. That is, it controls how to store information to achieve the memory function. The calculation formulas involved in the input gate include:

[0077] i t =σ(W i [h t-1 , x t ]+b i )

[0078] C′ t =tanh(W c [h t-1 , x t ]+b c )

[0079] In the above formula, W i is the weight matrix of the input gate, b i is the constant of the input gate, W c is the weight matrix of the tanh function, W c is the constant of the tanh function; the input gate and a tanh function cooperate to control what new information is added. The tanh function generates a new candidate vector C′ t , the input gate is C′ t Each entry in produces a value in [0, 1], controlling how much new information is added.

[0080] So far, we already have the output f of the forget gate t , used to control the degree to which the previous unit is forgotten, and also has the output i of the input gate t , which is used to control how much new information is added. Then the long-term memory is updated through the following formula:

[0081] Ct =f t C t-1 +i t C′ t

[0082] The output gate is used to control the cell state C at the current moment. t The amount of data output to convey the current moment's short-term memory h t (This step corresponds to the second information request number). That is, control the current unit state C t How much is filtered out. The calculation formulas involved in the output gate include:

[0083] o t =σ(W o [h t-1 , x t ]+b o )

[0084] h t =o t tanh(C t )

[0085] In the above formula, W o is the weight matrix of the output gate, b o is the constant of the output gate; first use the tanh function to convert the unit state C t Activate, the output gate generates a value in [0, 1] for each of them, controlling the cell state C t The degree of filtering.

[0086] Finally, we get the number of second information requests in the next Q counting cycles, which is recorded as:

[0087]

[0088] In the above formula, Indicates the number of second information requests in the next counting period; Indicates the number of second information requests in the next Q counting periods.

[0089] S104, encoding the first information feedback times in P previous counting cycles in chronological order and sequentially inputting them into the converged prediction model to predict the second information feedback times in the preset next Q next counting cycles.

[0090] In a specific embodiment, the prediction method of this step is basically the same as that of S103, except that the input data is the first information feedback times, and the output data is the second information feedback times in the next Q subsequent counting cycles, expressed as:

[0091]

[0092] In the above formula, Indicates the number of second information feedbacks in the next counting cycle; Indicates the number of second information feedbacks in the next Q counting cycles.

[0093] S105, calculating a preset predicted number of non-feedback times in the latest Q subsequent counting cycles based on the actual number of non-feedback times, the number of second information requests and the number of second information feedback times.

[0094] Optionally, the calculation formula for predicting the number of times no feedback is given is expressed as:

[0095]

[0096] In the above formula, N q indicates the predicted number of times no feedback is given in the qth subsequent counting cycle, N0 indicates the actual number of times no feedback is given, Indicates the number of second information requests in the i-th subsequent counting cycle, Indicates the number of second information feedbacks in the i-th subsequent counting cycle.

[0097] Finally, the predicted number of non-feedback times in the subsequent counting cycles can be obtained, which is expressed as:

[0098] [N1…N q …N Q ]

[0099] S106: If the predicted number of non-feedback times in any subsequent counting cycle is greater than a preset non-feedback threshold, it is indicated that there is a risk of triggering a fuse in the future.

[0100] Furthermore, if it is determined that there is N q >threN, where threN is the preset no-feedback threshold, it can be indicated that there is a risk of triggering a fuse in q cycles.

[0101] The purpose of this step is to detect potential problems or signs of failure in advance, predict possible circuit breakers in the future, and take preventive measures to ensure the stability and reliability of the system. For example, remove a certain room type from the hotel in advance to avoid no feedback during subsequent reservations.

[0102] Furthermore, the present embodiment may also perform the following steps:

[0103] When q>1, the smoothing value S(q) is calculated and expressed as:

[0104] S(q)=α×N q +(1-α)×N q-1

[0105] In the above formula, S(q) represents the smoothed value of the qth subsequent counting period, and α represents the smoothing coefficient, which is between 0 and 1. This formula describes how to adjust the smoothed value of the previous period by the smoothing coefficient to calculate the smoothed value of the current period. The smoothing coefficient α controls the degree of influence of historical data on the smoothed value. The larger α is, the more obvious the influence of historical data is.

[0106] By observing the changing trend of all calculated smoothed values ​​over time, risk assessment can be performed. If all calculated smoothed values ​​show an upward trend over time, the risk of triggering a circuit breaker is determined to be increased. If all calculated smoothed values ​​show a downward trend over time, the risk of triggering a circuit breaker is determined to be reduced. This can help system administrators or operation and maintenance personnel make corresponding decisions and actions more timely from another perspective to ensure the stability and reliability of the system.

[0107] The above-mentioned fuse triggering warning method, first, counts the current actual number of non-feedbacks and the first information request number and the first information feedback number in the preset most recent P counting cycles. Then, the first information request number in the P counting cycles is encoded in chronological order and input into a convergent prediction model to predict the second information request number in the subsequent Q counting cycles. Similarly, the first information feedback number in the P counting cycles is encoded and input into another prediction model to predict the second information feedback number in the subsequent Q counting cycles. Based on the actual number of non-feedbacks, the second information request number and the second information feedback number, the preset most recent Q predicted non-feedback times in the subsequent counting cycles are calculated. Finally, if it is found that any predicted non-feedback number in the subsequent counting cycle exceeds the preset non-feedback threshold, it will be prompted that there is a risk of triggering a fuse, so that there is a buffer time to take preventive measures before the fuse occurs to ensure the stability and reliability of the system.

[0108] In one embodiment, Figure 3 As shown, a server is proposed, and the server includes:

[0109] A first counter 301 is used to count the actual number of times that a third-party information request has not been responded to at present;

[0110] The second counter 302 is used to count the number of first information requests within the preset most recent P previous counting periods;

[0111] The third counter 303 is used to count the number of first information feedbacks within the preset most recent P previous counting cycles;

[0112] The controller 304 is used to encode the first information request times in P previous counting cycles in chronological order and input them into the convergent prediction model in sequence to predict the second information request times in the preset next Q subsequent counting cycles; encode the first information feedback times in P previous counting cycles in chronological order and input them into the convergent prediction model in sequence to predict the second information feedback times in the preset next Q subsequent counting cycles; calculate the predicted non-feedback times in the preset most recent Q subsequent counting cycles based on the actual non-feedback times, the second information request times and the second information feedback times; if the predicted non-feedback times in any subsequent counting cycle are greater than the preset non-feedback threshold, it is indicated that there is a risk of triggering a fuse in the future.

[0113] In one embodiment, the prediction model includes multiple loop structures, each loop structure includes a forget gate, an input gate, an output gate, and a unit state; the unit state connects the unit state at the previous moment with the unit state at the next moment, and is used to transfer long-term memory; the forget gate is used to control the amount of data of the unit state at the previous moment retained to the unit state at the current moment; the input gate is used to control the amount of data of the current input loop structure that is saved to the unit state at the current moment; wherein the data of the current input loop structure is the code corresponding to the first number of information requests or the code corresponding to the first number of information feedbacks; the output gate is used to control the amount of data of the unit state output at the current moment to transfer the short-term memory at the current moment; wherein the short-term memory indicates the second number of information requests or the second number of information feedbacks.

[0114] In one embodiment, the calculation formula for predicting the number of times of no feedback is expressed as:

[0115]

[0116] In the above formula, N q indicates the predicted number of times no feedback is given in the qth subsequent counting cycle, N0 indicates the actual number of times no feedback is given, Indicates the number of second information requests in the i-th subsequent counting cycle, Indicates the number of second information feedbacks in the i-th subsequent counting cycle.

[0117] In one embodiment, the controller 304 is further configured to:

[0118] When q>1, the smoothing value S(q) is calculated and expressed as:

[0119] S(q)=α×N q +(1-α)×N q-1

[0120] In the above formula, S(q) represents the smoothed value of the qth subsequent counting period, and α represents the smoothing coefficient, which is between 0 and 1;

[0121] If all calculated smoothed values ​​show an upward trend as time increases, it is determined that the risk of triggering a circuit breaker is increasing. If all calculated smoothed values ​​show a downward trend as time increases, it is determined that the risk of triggering a circuit breaker is decreasing.

[0122] Figure 4 FIG. 2 shows an internal structure diagram of a fuse trigger warning device in an embodiment. Figure 4 As shown, the fuse trigger warning device includes a processor, a memory and a network interface connected via a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the fuse trigger warning device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the fuse trigger warning method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute the fuse trigger warning method. Those skilled in the art can understand that Figure 4 The structure shown in the figure is only a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the fuse trigger warning device to which the scheme of the present application is applied. The specific fuse trigger warning device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0123] A computer-readable storage medium stores a computer program, which implements the following steps when executed by a processor: counting the current actual number of non-feedback of information requests for third parties; counting the number of first information requests and the number of first information feedbacks in the latest P previous counting cycles; encoding the number of first information requests in the P previous counting cycles in chronological order and inputting them into a convergent prediction model in sequence to predict the number of second information requests in the next Q subsequent counting cycles; encoding the number of first information feedbacks in the P previous counting cycles in chronological order and inputting them into a convergent prediction model in sequence to predict the number of second information feedbacks in the next Q subsequent counting cycles; calculating the predicted number of non-feedback in the latest Q subsequent counting cycles based on the actual number of non-feedback, the number of second information requests and the number of second information feedback; if the predicted number of non-feedback in any subsequent counting cycle is greater than the preset non-feedback threshold, it is prompted that there will be a risk of triggering a fuse in the future.

[0124] A fuse trigger warning device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: counting the current actual number of non-feedback of information requests for third parties; counting the number of first information requests and the number of first information feedbacks in the latest P previous counting cycles; encoding the number of first information requests in the P previous counting cycles in chronological order and inputting them into a convergent prediction model in sequence to predict the number of second information requests in the next Q subsequent counting cycles in chronological order; encoding the number of first information feedbacks in the P previous counting cycles in chronological order and inputting them into a convergent prediction model in sequence to predict the number of second information feedbacks in the next Q subsequent counting cycles in chronological order; calculating the predicted number of non-feedback in the latest Q subsequent counting cycles in chronological order based on the actual number of non-feedback, the number of second information requests and the number of second information feedback; if the predicted number of non-feedback in any subsequent counting cycle is greater than the preset non-feedback threshold, it is prompted that there will be a risk of triggering a fuse in the future.

[0125] It should be noted that the above-mentioned fuse trigger warning method, server, device and computer-readable storage medium belong to a general inventive concept, and the contents in the embodiments of the fuse trigger warning method, server, device and computer-readable storage medium are applicable to each other.

[0126] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0127] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0128] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A fuse triggering warning method, characterized in that: The method comprises: Count the actual number of unanswered requests for third-party information; Counting the number of first information requests and first information feedback within the latest P preset counting periods; Encode the first number of information requests in P preceding counting periods in chronological order and sequentially input them into a converged prediction model to predict the second number of information requests in the next Q subsequent counting periods; Encode the first information feedback times in P preceding counting cycles in chronological order and input them into the converged prediction model in sequence to predict the second information feedback times in the next Q subsequent counting cycles; Calculating a preset predicted number of non-feedback times in the most recent Q subsequent counting cycles based on the actual number of non-feedback times, the second information request number, and the second information feedback number; If the number of predicted non-feedbacks in any subsequent counting cycle exceeds the preset non-feedback threshold, it indicates that there is a risk of triggering a circuit breaker in the future; The prediction model includes multiple loop structures, each loop structure includes a forget gate, an input gate, an output gate, and a unit state; the unit state connects the unit state at the previous moment and the unit state at the next moment for transferring long-term memory; the forget gate is used to control the amount of data of the unit state at the previous moment retained to the unit state at the current moment; the input gate is used to control the amount of data of the current input loop structure that is saved to the unit state at the current moment; wherein the data of the current input loop structure is a code corresponding to the first number of information requests or a code corresponding to the first number of information feedbacks; the output gate is used to control the amount of data of the unit state at the current moment that is output to transfer the short-term memory at the current moment; wherein the short-term memory indicates the second number of information requests or the second number of information feedbacks; The calculation formula for predicting the number of times no feedback is given is expressed as: In the above formula, N q indicates the predicted number of times no feedback is given in the qth subsequent counting cycle, N0 indicates the actual number of times no feedback is given, Indicates the number of second information requests in the i-th subsequent counting cycle, Indicates the number of second information feedbacks in the i-th subsequent counting cycle.

2. The method according to claim 1, characterized in that The method further comprises: When q>1, the smoothing value S(q) is calculated and expressed as: S(q)=α×N q +(1-α)×N q-1 In the above formula, S(q) represents the smoothed value of the qth subsequent counting period, and α represents the smoothing coefficient, which is between 0 and 1; If all calculated smoothed values ​​show an upward trend as time increases, it is determined that the risk of triggering a circuit breaker is increasing. If all calculated smoothed values ​​show a downward trend as time increases, it is determined that the risk of triggering a circuit breaker is decreasing.

3. A server, characterized in that: The server includes: A first counter is used to count the actual number of unanswered information requests to third parties; A second counter, used to count the number of first information requests within the latest P previous counting periods; A third counter is used to count the number of first information feedbacks within the latest P preset counting periods; A controller, configured to encode the first information request number within P preceding counting cycles in chronological order and sequentially input the number into a converged prediction model to predict a preset second information request number in the next Q succeeding counting cycles; encode the first information feedback number within P preceding counting cycles in chronological order and sequentially input the number into a converged prediction model to predict a preset second information feedback number in the next Q succeeding counting cycles; calculate a preset predicted number of no feedbacks for the most recent Q succeeding counting cycles based on the actual number of no feedbacks, the second number of information requests, and the second number of information feedbacks; and if the predicted number of no feedbacks for any succeeding counting cycle is greater than a preset no feedback threshold, indicate a risk of triggering a circuit breaker in the future; The prediction model includes multiple loop structures, each loop structure includes a forget gate, an input gate, an output gate, and a unit state; the unit state connects the unit state at the previous moment and the unit state at the next moment for transferring long-term memory; the forget gate is used to control the amount of data of the unit state at the previous moment retained to the unit state at the current moment; the input gate is used to control the amount of data of the current input loop structure that is saved to the unit state at the current moment; wherein the data of the current input loop structure is a code corresponding to the first number of information requests or a code corresponding to the first number of information feedbacks; the output gate is used to control the amount of data of the unit state at the current moment that is output to transfer the short-term memory at the current moment; wherein the short-term memory indicates the second number of information requests or the second number of information feedbacks; The calculation formula for predicting the number of times no feedback is given is expressed as: In the above formula, N q indicates the predicted number of times no feedback is given in the qth subsequent counting cycle, N0 indicates the actual number of times no feedback is given, Indicates the number of second information requests in the i-th subsequent counting cycle, Indicates the number of second information feedbacks in the i-th subsequent counting cycle.

4. The server according to claim 3, wherein: The controller is also used to: When q>1, the smoothing value S(q) is calculated and expressed as: S(q)=α×N q +(1-α)×N q-1 In the above formula, S(q) represents the smoothed value of the qth subsequent counting period, and α represents the smoothing coefficient, which is between 0 and 1; If all calculated smoothed values ​​show an upward trend as time increases, it is determined that the risk of triggering a circuit breaker is increasing. If all calculated smoothed values ​​show a downward trend as time increases, it is determined that the risk of triggering a circuit breaker is decreasing.

5. A computer-readable storage medium, characterized in that A computer program is stored, and when the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 2.

6. A fuse trigger warning device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 2.

Citation Information

Cited By

  • Intelligent traffic scheduling method and system based on application perception and storage medium

    CN122933751A