Backbone transmission network and data network multi-domain association risk analysis early warning method and system

Abstract

The invention belongs to the field of power distribution network fault risk early warning, and discloses a backbone transmission network and data network multi-domain association risk analysis early warning method and system, and the method comprises the steps: building a risk factor set of a backbone transmission network and a data network, and a double-layer experience risk association relationship; carrying out statistics on historical faults and cost losses of the backbone transmission network and the data network, carrying out descending sort to form a multi-domain fault guiding sequence, and carrying out segmented multi-domain association risk fitting; obtaining a fault risk sequence of the data network and the transmission network based on the multi-domain association risk fitting result, the operation state monitoring result and the distribution of risk factors continuously input into the transmission network and the data network; and performing multi-domain feature driven fault risk prediction by adopting a fault risk prediction network, and performing cross early warning based on a prediction result. The comprehensiveness and accuracy of multi-domain risk analysis are effectively improved by considering the influence of space-time correlation of the risk and faults of different degrees on risk analysis.

Description

Technical Field

[0001] The present invention relates to the field of distribution network fault risk warning, and specifically to a multi-domain correlation risk analysis and warning method and system for backbone transmission networks and data networks. Background Art

[0002] With the acceleration of the informatization of power systems, the power backbone transmission network and the power data network play a vital role in ensuring power supply and system stability. The power backbone transmission network handles long-distance, high-speed data transmission and serves as a key hub connecting major power equipment and monitoring systems. The power data network, on the other hand, is responsible for storing, exchanging, and processing power data, providing real-time data support for power dispatching, monitoring, and decision-making. The stability and security of both directly impact the normal operation of the power system. Failures can lead to large-scale power outages, loss of critical data, or system crashes, resulting in significant economic losses and social impact. Therefore, ensuring the safe and stable operation of the power backbone transmission network and the power data network, and identifying and predicting potential failure risks in advance, are crucial.

[0003] Traditional risk early warning systems for power backbone transmission networks and power data networks typically employ static rules or empirical analysis methods based on historical data. While these methods can monitor certain known risk factors to a certain extent, they often have many shortcomings when dealing with complex and changing power environments. For example, traditional methods struggle to account for the spatiotemporal coupling of fault occurrences, lack a comprehensive analysis of the synergistic impact of multiple domains, and are slow to respond to sudden risks, making it impossible to adjust early warning strategies in a timely manner. Furthermore, traditional risk analysis and early warning methods for backbone transmission networks and data networks fail to consider the spatiotemporal correlation of risks and the impact of varying fault severity on risk analysis, resulting in incomplete and inaccurate risk analysis. Risk prediction and early warning only consider a single fault information source and lack cross-correlation warnings, leading to poor prediction accuracy and untimely and inaccurate early warnings. Summary of the Invention

[0004] The purpose of the present invention is to provide a backbone transmission network and data network multi-domain correlation risk analysis and early warning method and system to solve the problems of poor prediction accuracy and insufficient timely and accurate early warning in the existing technology.

[0005] To achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for analyzing and warning multi-domain correlation risks of a backbone transmission network and a data network, comprising: Obtain historical fault record data, establish a risk factor set for the backbone transmission network and data network, and establish a two-layer empirical risk association relationship based on the time interval and location of fault occurrence in the risk factor set; Based on the two-layer empirical risk correlation relationship, the historical failures and cost losses of the backbone transmission network and data network are counted and arranged in descending order to form a multi-domain fault-oriented sequence, and a segmented multi-domain correlation risk fitting is performed; Based on the multi-domain correlation risk fitting results, operation status monitoring results, and the risk factor distribution of the continuous input transmission network and data network, the fault risk sequence of the data network and transmission network is obtained; Based on the fault risk sequences of data networks and transmission networks, a fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction, and cross-warning is performed based on the prediction results.

[0006] Optionally, the acquiring of historical fault record data and establishing a risk factor set for the backbone transmission network and the data network includes: Risk factors are defined based on the dimensions of equipment failure rate, bandwidth utilization, and network topology complexity. The equipment failure rate risk factor for the backbone transmission network is expressed as: (1) Where, is the equipment failure rate risk factor of the backbone transmission network, The equipment failure rate of the backbone transmission network recorded in historical data, is the adjustment factor for equipment failure rate risk; The bandwidth utilization risk factor is expressed as (2) Where, is the bandwidth utilization risk factor of the backbone transmission network, The bandwidth utilization of the backbone transmission network recorded for historical data, is the adjustment factor for bandwidth utilization risk; The network topology complexity risk factor is expressed as (3) Where, is the network topology complexity risk factor of the backbone transmission network, is the network topology complexity index of the backbone transmission network recorded in historical data, is the adjustment factor of network topology complexity risk; The risk factor sets for backbone transmission network and data network constructed based on historical data are respectively expressed as (4) (5) Where, is a risk factor set for the backbone transmission network based on historical data. It is a set of risk factors for the data network built based on historical data. 、 and They are the equipment failure rate risk factor, bandwidth utilization risk factor, and network topology complexity risk factor of the data network, and the calculation method is the same as that of the backbone transmission network; In the backbone transmission network, the introduction of empirical data to the equipment failure rate risk factor Rating ,but The risk factor adjusted after empirical data scoring is expressed as (6) Combining historical data and empirical data, the backbone transmission network equipment failure rate risk factor is expressed as (7) Where, The failure rate risk factor of backbone transmission network equipment is obtained by combining historical data and empirical data. and are the weights of historical data and empirical data respectively; The risk factor sets of the backbone transmission network and data network finally established are expressed as (8) (9) Where, and are the risk factor sets for the backbone transmission network and data network obtained by combining historical data and empirical data; and They are respectively the backbone transmission network bandwidth utilization risk factor and the network topology complexity risk factor obtained by combining historical data and empirical data; , and They are the data network equipment failure rate risk factor, bandwidth utilization risk factor and network topology complexity risk factor obtained by combining historical data and empirical data.

[0007] Optionally, establishing a two-layer empirical risk association relationship based on the time interval and location of the fault occurrence in the risk factor set includes: According to the time interval and location of the faults, a two-layer empirical risk association model is established. The first layer is the time association model. The time correlation between faults decreases as the time interval increases, which is expressed as (10) Where, For failure and failure The temporal relationship between and is the adjustment factor to control the effect of time interval on the correlation relationship. For failure and failure the time interval between The second layer is the spatial correlation model. The spatial correlation between faults decreases as the distance increases, which can be expressed as (11) Where, For failure and failure The spatial correlation between and is the adjustment factor to control the effect of spatial distance on the association relationship. For failure and failure The spatial distance between them.

[0008] Optionally, based on the two-layer empirical risk association relationship, historical failures and cost losses of the backbone transmission network and the data network are counted and arranged in descending order to form a multi-domain fault-oriented sequence, including: The frequency of each fault, the scope of the fault, the difficulty of recovery and the cost loss recorded in the historical data of the backbone transmission network and data network are counted and comprehensively scored. , the comprehensive score is expressed as (12) Where, Failure of the backbone transmission network The comprehensive rating of Failure of the backbone transmission network Frequency of occurrence, Failure of the backbone transmission network The scope of the fault caused indicates the degree of impact of the fault on the network. Failure of the backbone transmission network The recovery difficulty indicates the difficulty of repairing the fault. Failure of the backbone transmission network Cost loss, which represents the economic loss caused by the failure; , , and is the weight coefficient; After calculating the comprehensive score of each fault, all faults are sorted from high to low according to the comprehensive score, and the multi-domain fault-oriented sequence is obtained, which is expressed as (13) (14) Where, and These are the multi-domain fault-oriented sequences for the backbone transmission network and the data network respectively; The first in the multi-domain fault-oriented sequence of the backbone transmission network elements, The first in the multi-domain fault-oriented sequence of the backbone transmission network The comprehensive score of the fault corresponding to each element, The first in the multi-domain fault-oriented sequence of the backbone transmission network The comprehensive score of each element corresponding to the fault; The first in the multi-domain fault-oriented sequence of the data network elements, The first in the multi-domain fault-oriented sequence of the data network The comprehensive score of the fault corresponding to each element, The first in the multi-domain fault-oriented sequence of the data network The comprehensive score of each element corresponding to the fault; In the descending multi-domain fault-oriented sequence, different subsequences are formed by cutting according to the comprehensive score threshold, which is expressed as (15) (16) Where, This is the upper part of the multi-domain fault-oriented sequence of the backbone transmission network. This is the lower part of the multi-domain fault-oriented sequence of the backbone transmission network. This is the upper part of the multi-domain fault-oriented sequence for data networks. This is the next section of the multi-domain fault-oriented sequence for data networks. is the comprehensive score threshold.

[0009] Optionally, performing segmented multi-domain correlation risk fitting includes: According to the multi-domain guidance sequence, the least square method is used to perform segmented multi-domain correlation risk fitting, and the weight of the upper segment is greater than the weight of the lower segment, that is, , ; When the least squares method is used for fitting, the goal of fitting is to minimize the weighted error, which can be expressed as (17) (18) Where, is the actual observed value, The values predicted by the fitted model.

[0010] Optionally, obtaining the fault risk sequence of the data network and the transmission network based on the multi-domain correlation risk fitting results, the operation status monitoring results, and the risk factor distribution of the continuous input transmission network and the data network includes: Based on the operation status monitoring results, the risk factor distribution of the backbone transmission network and data network is obtained continuously. The risk factor distribution of backbone transmission network and data network are expressed as and , the fault risk sequence of the backbone transmission network and the data network is generated through the fault risk sequence generation network, which is expressed as (19) (20) Where, Generate a network for failure risk sequences, and Separate moments Failure risk sequence of backbone transmission network and data network, For the moment Failures in the backbone transmission network For the moment A data network failure occurred.

[0011] Optionally, the fault risk sequence based on the data network and the transmission network uses a fault risk prediction network to perform multi-domain feature-driven fault risk prediction, and based on the prediction results, performs cross-warning, including: A fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction. The multi-domain features include the risk fluctuation slope, maximum value, minimum value, variance, number of occurrences of large risk values, frequency of occurrence of large risk values, and frequency change rate of the backbone transmission network and data network. The details are as follows: For the backbone transmission network, the risk fluctuation slope is The maximum risk is expressed as ; The minimum value is expressed as ; Variance is expressed as ,in is the total time, is the mean; the number of occurrences of large risk values is expressed as ,in is the large risk value threshold, is a counting function; the frequency of occurrence of large risk values is expressed as The frequency change rate is expressed as ; For the data network, the risk volatility slope is The maximum risk is expressed as ; The minimum value is expressed as ; Variance is expressed as ;in is the mean; the number of occurrences of large risk values is expressed as The frequency of occurrence of large risk values is expressed as The frequency change rate is expressed as ; The failure risk prediction of backbone transmission network and data network is expressed as (twenty one) (twenty two) Where, predicting networks for failure risk; Based on the prediction results, cross-warning is carried out, including two warning methods: 1) If the risk value obtained based on the self-fault risk sequence prediction is greater than the threshold, an early warning should be issued, which is expressed as (twenty three) Where, To indicate the time An indicator variable indicating whether the backbone transmission network issues a risk warning. If the value is 1, a warning is issued; if the value is 0, no warning is issued. It is an indicator function, and the value in the brackets is 1 when it holds true, and 0 when it does not hold true; is the risk value threshold; Similarly, for the data network, whether to issue an early warning is indicated by (twenty four) 2) Based on the fault risk prediction result of another domain, a warning is issued when the risk value obtained by the multi-domain correlation risk fitting model is greater than the threshold, which is expressed as (25) (26) Where, Represents piecewise multi-domain correlation risk fitting.

[0012] In a second aspect, the present invention provides a backbone transmission network and data network multi-domain correlation risk analysis and early warning system, comprising: The data acquisition module is used to obtain historical fault record data, establish a risk factor set for the backbone transmission network and the data network, and establish a two-layer empirical risk association relationship based on the time interval and location of the fault occurrence in the risk factor set; The risk fitting module is used to collect statistics on historical failures and cost losses of the backbone transmission network and data network based on the two-layer empirical risk correlation relationship, and arrange them in descending order to form a multi-domain fault-oriented sequence, and perform segmented multi-domain correlation risk fitting; The fault risk sequence acquisition module is used to obtain the fault risk sequences of the data network and transmission network based on the multi-domain correlation risk fitting results, the operation status monitoring results, and the risk factor distribution of the continuous input transmission network and data network; The early warning module is used to predict the fault risk sequences of data networks and transmission networks. It uses a fault risk prediction network to perform multi-domain feature-driven fault risk prediction and provides cross-warning based on the prediction results.

[0013] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the backbone transmission network and data network multi-domain correlation risk analysis and early warning method are implemented.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the backbone transmission network and data network multi-domain correlation risk analysis and early warning method.

[0015] Compared with the prior art, the present invention has the following technical effects: Based on historical data and expert experience, this paper establishes a set of risk factors for the backbone transmission network and the data network, and establishes a two-layer empirical risk correlation relationship based on the time interval and location of the failure. Secondly, the frequency of historical failures in the two networks, the scope of the failures caused, the difficulty of recovery, and the cost loss are counted and arranged in descending order to form a multi-domain fault-oriented sequence, and the sequence is cut. Finally, based on the multi-domain guided sequence, a segmented multi-domain correlation risk fitting is performed. By considering the spatiotemporal correlation of risks and the impact of different degrees of failure on risk analysis, the comprehensiveness and accuracy of multi-domain risk analysis are effectively improved.

[0016] Based on the results of multi-domain correlation risk analysis and operational status monitoring, this method continuously inputs the distribution of risk factors for the transmission and data networks to obtain fault risk sequences for the data and transmission networks. Secondly, a fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction. Then, based on the prediction results, cross-prediction warnings are issued. Finally, based on the prediction results and actual conditions, the prediction model is updated at a small scale and the multi-domain correlation risk fitting model is updated at a large time scale, effectively improving the accuracy and timeliness of risk prediction and warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0018] The present invention is further described below with reference to the accompanying drawings: Example 1, please refer to Figure 1 The present invention provides a backbone transmission network and data network multi-domain correlation risk analysis and early warning method, comprising: Obtain historical fault record data, establish a risk factor set for the backbone transmission network and data network, and establish a two-layer empirical risk association relationship based on the time interval and location of fault occurrence in the risk factor set; Based on the two-layer empirical risk correlation relationship, the historical failures and cost losses of the backbone transmission network and data network are counted and arranged in descending order to form a multi-domain fault-oriented sequence, and a segmented multi-domain correlation risk fitting is performed; Based on the multi-domain correlation risk fitting results, operation status monitoring results, and the risk factor distribution of the continuous input transmission network and data network, the fault risk sequence of the data network and transmission network is obtained; Based on the fault risk sequences of data networks and transmission networks, a fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction, and cross-warning is performed based on the prediction results.

[0019] The present invention first establishes a risk factor set for the backbone transmission network and the data network based on historical data and expert experience, and establishes a two-layer empirical risk association relationship based on the time interval and location of the fault. Then, the frequency of historical faults in the two networks and the fault range, recovery difficulty and cost loss are counted and arranged in descending order to form a multi-domain fault-oriented sequence, and the sequence is cut. Finally, based on the multi-domain oriented sequence, a segmented multi-domain correlation risk fitting is performed. By considering the spatiotemporal correlation of risks and the impact of different degrees of faults on risk analysis, the comprehensiveness and accuracy of the multi-domain risk analysis are effectively improved. Secondly, based on the results of the multi-domain correlation risk analysis and the results of the operation status monitoring, the risk factor distribution of the transmission network and the data network is continuously input to obtain the fault risk sequence of the data network and the transmission network. Then, a fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction. Based on the prediction results, a cross-warning is performed. Finally, based on the prediction results and the actual situation, the prediction model is updated at a small scale and the multi-domain correlation risk fitting model is updated at a large time scale, effectively improving the accuracy and timeliness of risk prediction and warning.

[0020] In Example 2, the present invention provides a method for analyzing and warning multi-domain correlation risks of a backbone transmission network and a data network, specifically comprising: Step 1: Multi-domain correlation risk analysis method based on risk spatiotemporal coupling and differentiated fault impacts First, based on historical data and expert experience, a risk factor set for the backbone transmission network and data network was established. A two-layer empirical risk correlation was then established based on the time interval and location of fault occurrences. Second, the frequency of historical faults in both networks, the resulting fault scope, recovery difficulty, and cost were statistically analyzed and sorted in descending order to form a multi-domain fault-oriented sequence, which was then segmented. Finally, a segmented multi-domain correlation risk fitting was performed based on the multi-domain-oriented sequence.

[0021] (1) Constructing a risk factor set The present invention converts the fault record data in the historical data into risk factors.

[0022] The present invention defines risk factors from the dimensions of equipment failure rate, bandwidth utilization and network topology complexity. Taking the backbone transmission network as an example, the equipment failure rate risk factor is expressed as (1) Where, is the equipment failure rate risk factor of the backbone transmission network, The equipment failure rate of the backbone transmission network recorded in historical data, is the adjustment factor for equipment failure rate risk.

[0023] The higher the bandwidth utilization of a network device, the more likely it is to fail. The bandwidth utilization risk factor is expressed as (2) Where, is the bandwidth utilization risk factor of the backbone transmission network, The bandwidth utilization of the backbone transmission network recorded for historical data, is the adjustment factor for bandwidth utilization risk.

[0024] The network topology complexity has a great impact on fault propagation and recovery. The network topology complexity risk factor is expressed as (3) Where, is the network topology complexity risk factor of the backbone transmission network, is the network topology complexity index of the backbone transmission network recorded in historical data, It is the adjustment factor of network topology complexity risk.

[0025] Therefore, the risk factor sets of the backbone transmission network and data network constructed based on historical data are expressed as (4) (5) Where, is a risk factor set for the backbone transmission network based on historical data. It is a set of risk factors for the data network built based on historical data. 、 and They are the equipment failure rate risk factor, bandwidth utilization risk factor and network topology complexity risk factor of the data network, and the calculation method is similar to that of the backbone transmission network.

[0026] Expert experience influences the weight of risk factors by evaluating the impact of each risk factor on the overall risk. Assume that expert experience scores each factor in the range of [0,1], which ranges from unimportant to very important. Taking the backbone transmission network as an example, for each risk factor in the risk factor set of the backbone transmission network constructed based on historical data, the risk factor of equipment failure rate is used as the weight of risk factor. For example, assuming that the expert experience scores it as ,but The risk factor adjusted after expert experience scoring is expressed as (6) Combining historical data and expert experience, the backbone transmission network equipment failure rate risk factor is expressed as (7) Where, The failure rate risk factor of backbone transmission network equipment is obtained by combining historical data and expert experience. and are the weights of historical data and expert experience respectively.

[0027] The risk factor sets of the backbone transmission network and data network finally established are expressed as (8) (9) Where, and are risk factor sets for backbone transmission networks and data networks obtained by combining historical data and expert experience; and They are respectively the backbone transmission network bandwidth utilization risk factor and the network topology complexity risk factor obtained by combining historical data and expert experience; , and They are the data network equipment failure rate risk factor, bandwidth utilization risk factor and network topology complexity risk factor obtained by combining historical data and expert experience.

[0028] (2) Constructing a two-layer empirical risk correlation relationship According to the time interval and location of the faults, a two-layer empirical risk association model is established. The first layer is the time association model. The time correlation between faults decreases as the time interval increases, which is expressed as (10) Where, For failure and failure The temporal relationship between and is the adjustment factor to control the effect of time interval on the correlation relationship. For failure and failure The time interval between.

[0029] The second layer is the spatial correlation model. The spatial correlation between faults decreases as the distance increases, which can be expressed as (11) Where, For failure and failure The spatial correlation between and is the adjustment factor to control the effect of spatial distance on the association relationship. For failure and failure The spatial distance between them.

[0030] (3) Constructing a multi-domain fault-oriented sequence The frequency of each fault recorded in the historical data of the backbone transmission network and data network, the scope of the fault, the difficulty of recovery and the cost loss are counted and a comprehensive score is given. For example, the comprehensive score is expressed as (12) Where, Failure of the backbone transmission network The comprehensive rating of Failure of the backbone transmission network Frequency of occurrence, Failure of the backbone transmission network The scope of the fault caused indicates the degree of impact of the fault on the network. Failure of the backbone transmission network The recovery difficulty indicates the difficulty of repairing the fault. Failure of the backbone transmission network Cost loss refers to the economic loss caused by the failure. , , and is the weight coefficient.

[0031] After calculating the comprehensive score of each fault, all faults are sorted from high to low according to the comprehensive score, and the multi-domain fault-oriented sequence is obtained, which is expressed as (13) (14) Where, and These are the multi-domain fault-oriented sequences for the backbone transmission network and the data network respectively. The first in the multi-domain fault-oriented sequence of the backbone transmission network elements, The first in the multi-domain fault-oriented sequence of the backbone transmission network The comprehensive score of the fault corresponding to each element, The first in the multi-domain fault-oriented sequence of the backbone transmission network The comprehensive score of each element corresponds to the fault. The first in the multi-domain fault-oriented sequence of the data network elements, The first in the multi-domain fault-oriented sequence of the data network The comprehensive score of the fault corresponding to each element, The first in the multi-domain fault-oriented sequence of the data network The comprehensive score of each element corresponds to the fault.

[0032] Since the fitting accuracy of different elements in the sequence corresponding to faults is different when the multi-domain fault-oriented sequence is used to fit the multi-domain correlation risk, the multi-domain fault-oriented sequence arranged in descending order is cut according to the comprehensive score threshold to form different subsequences, which are expressed as (15) (16) Where, This is the upper part of the multi-domain fault-oriented sequence of the backbone transmission network. This is the lower part of the multi-domain fault-oriented sequence of the backbone transmission network. This is the upper part of the multi-domain fault-oriented sequence for data networks. This is the next section of the multi-domain fault-oriented sequence for data networks. is the comprehensive score threshold.

[0033] (4) Multi-domain correlation risk fitting According to the multi-domain guidance sequence, the least square method is used to perform segmented multi-domain correlation risk fitting. For different segments of the guidance sequence, the fitting accuracy requirements are different. For example, the upper segment of the multi-domain fault guidance sequence of the backbone transmission network above Since the comprehensive score of the upper fault is higher, the impact of the same fitting error is greater than that of the lower fault. is higher, so when fitting, the weight of the upper segment is greater than the weight of the lower segment, that is, , so as to achieve a smaller error in the upper segment fitting. Similarly, the upper segment of the multi-domain fault-oriented sequence of the data network , the impact of the same fitting error is compared with the lower High, so when fitting, it should be ensured .

[0034] When the least squares method is used for fitting, the goal of fitting is to minimize the weighted error, which can be expressed as (17) (18) Where, is the actual observed value, The values predicted by the fitted model.

[0035] Step 2: Fault risk prediction and cross-warning method based on multi-domain feature drive First, based on the results of multi-domain correlation risk analysis and operational status monitoring, the risk factor distribution of the transmission and data networks is continuously input to obtain fault risk sequences for the data and transmission networks. Second, a fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction. Then, based on the prediction results, cross-prediction warnings are issued. Finally, based on the prediction results and actual conditions, the prediction model is updated on a small scale and the multi-domain correlation risk fitting model is updated on a large time scale.

[0036] (1) Failure risk sequence Based on the operation status monitoring results, the risk factor distribution of the backbone transmission network and data network is obtained continuously. The risk factor distribution of backbone transmission network and data network are expressed as and , the fault risk sequence of the backbone transmission network and the data network is generated through the fault risk sequence generation network, which is expressed as (19) (20) Where, Generate a network for failure risk sequences, and Separate moments Failure risk sequence of backbone transmission network and data network, For the moment Failures in the backbone transmission network For the moment A data network failure occurred.

[0037] (2) Failure risk prediction The present invention uses a fault risk prediction network to perform multi-domain feature-driven fault risk prediction. The multi-domain features include the risk fluctuation slope, maximum value, minimum value, variance, number of occurrences of large risk values, frequency of occurrence of large risk values, and frequency change rate of the backbone transmission network and the data network. The details are as follows: For the backbone transmission network, the risk fluctuation slope is The maximum risk is expressed as ; The minimum value is expressed as ; Variance is expressed as ,in is the total time, is the mean; the number of occurrences of large risk values is expressed as ,in is the large risk value threshold, is a counting function; the frequency of occurrence of large risk values is expressed as The frequency change rate is expressed as .

[0038] For the data network, the risk volatility slope is The maximum risk is expressed as ; The minimum value is expressed as ; Variance is expressed as ;in is the mean; the number of occurrences of large risk values is expressed as The frequency of occurrence of large risk values is expressed as The frequency change rate is expressed as .

[0039] Therefore, the failure risk prediction of the backbone transmission network and data network is expressed as (twenty one) (twenty two) Where, Predicting networks for failure risk.

[0040] (3) Cross-warning Based on the prediction results, cross-warning is carried out, including two warning methods: 1) If the risk value obtained based on the self-fault risk sequence prediction is greater than the threshold, an early warning should be issued, which is expressed as (twenty three) Where, To indicate the time An indicator variable indicating whether the backbone transmission network issues a risk warning. If the value is 1, a warning is issued; if the value is 0, no warning is issued. It is an indicator function, and the value in the brackets is 1 when it holds true, and 0 when it does not hold true; is the risk value threshold.

[0041] Similarly, for the data network, whether to issue an early warning can be expressed as (twenty four) 2) If the risk value obtained by multi-domain correlation risk fitting model based on the fault risk prediction result of another domain is greater than the threshold, an early warning should also be issued, which is expressed as (25) (26) Where, Represents a piecewise multi-domain correlation risk fitting method.

[0042] (4) Model update Based on the prediction results and actual conditions, the prediction model is updated on a small scale, and the multi-domain correlation risk fitting model is updated on a large time scale.

[0043] At a small scale, for the fault risk prediction model, the gradient descent method is used to update the model parameters, which is expressed as (27) Where, is the updated parameter, is the parameter before updating, is the learning rate, which controls the step size of parameter updates, is the loss function, expressed as (28) Where, is the actual risk value of the backbone transmission network, is the actual risk value of the data network.

[0044] On a large time scale, the multi-domain correlation risk fitting model is updated. Specifically, historical data is updated according to the operation status monitoring results. The risk factor set of the backbone transmission network and the data network is re-established according to formulas (1)-(9). The historical fault frequency of the backbone transmission network and the data network and the resulting fault range, recovery difficulty and cost loss are re-calculated. The multi-domain fault guidance sequence is established according to formulas (12)-(16) and the sequence is cut. According to the latest multi-domain fault guidance sequence segmentation and the actual fault situation, the risk factor set of the backbone transmission network and the data network is re-established according to formulas (17)-(18). and to update.

[0045] This embodiment effectively improves the accuracy of risk analysis of multi-domain faults in the backbone transmission network and data network, takes into account the spatiotemporal coupling effect of risks and the differentiated impacts of different faults, and thus can more accurately predict and assess potential risks in the network.

[0046] By combining multi-domain features with real-time monitoring data, this embodiment can more accurately predict network failures and issue cross-warnings in advance, reducing the impact of failures on the entire network system. By dynamically updating prediction and risk analysis models, this embodiment greatly improves the accuracy and timeliness of failure risk prediction and warnings.

[0047] First, based on historical data and expert experience, the present invention establishes a set of risk factors for the backbone transmission network and the data network. Furthermore, a two-layer empirical risk correlation is established based on the time interval and location of the failure. Secondly, the frequency of historical failures in the two networks, the scope of the failures caused, the difficulty of recovery, and the cost loss are statistically analyzed and sorted in descending order to form a multi-domain fault-oriented sequence, which is then segmented. Finally, based on the multi-domain-oriented sequence, a segmented multi-domain correlation risk fitting is performed. By considering the spatiotemporal correlation of risks and the impact of different levels of failure on risk analysis, the comprehensiveness and accuracy of multi-domain risk analysis are effectively improved.

[0048] Based on the results of multi-domain correlation risk analysis and operational status monitoring, the risk factor distribution of the transmission and data networks is continuously input to obtain fault risk sequences for the data and transmission networks. Secondly, a fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction. Then, based on the prediction results, cross-prediction warnings are issued. Finally, based on the prediction results and actual conditions, the prediction model is updated on a small scale and the multi-domain correlation risk fitting model is updated on a large time scale, effectively improving the accuracy and timeliness of risk prediction and warning.

[0049] In yet another embodiment of the present invention, a backbone transmission network and data network multi-domain correlation risk analysis and early warning system is provided, which can be used to implement the above-mentioned backbone transmission network and data network multi-domain correlation risk analysis and early warning method. Specifically, the system includes: The data acquisition module is used to obtain historical fault record data, establish a risk factor set for the backbone transmission network and the data network, and establish a two-layer empirical risk association relationship based on the time interval and location of the fault occurrence in the risk factor set; The risk fitting module is used to collect statistics on historical failures and cost losses of the backbone transmission network and data network based on the two-layer empirical risk correlation relationship, and arrange them in descending order to form a multi-domain fault-oriented sequence, and perform segmented multi-domain correlation risk fitting; The fault risk sequence acquisition module is used to obtain the fault risk sequences of the data network and transmission network based on the multi-domain correlation risk fitting results, the operation status monitoring results, and the risk factor distribution of the continuous input transmission network and data network; The early warning module is used to predict the fault risk sequences of data networks and transmission networks. It uses a fault risk prediction network to perform multi-domain feature-driven fault risk prediction and provides cross-warning based on the prediction results.

[0050] The module division in the embodiments of the present invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in various embodiments of the present invention may be integrated into a single processor, exist physically as separate modules, or two or more modules may be integrated into a single module. The integrated modules may be implemented in either hardware or software functional modules.

[0051] In another embodiment of the present invention, a computer device is provided, comprising a processor and a memory, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the backbone transmission network and data network multi-domain correlation risk analysis and early warning method.

[0052] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device, used to store programs and data. It is understood that the computer-readable storage medium herein may include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space stores one or more instructions suitable for being loaded and executed by a processor. These instructions may be one or more computer programs (including program code). It should be noted that the computer-readable storage medium herein may be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device. The processor may load and execute the one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the multi-domain correlation risk analysis and early warning method for the backbone transmission network and data network described in the above-mentioned embodiment.

[0053] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0054] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0055] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0056] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A multi-domain correlation risk analysis and early warning method for backbone transmission networks and data networks, characterized in that: include: Obtain historical fault record data, establish a risk factor set for the backbone transmission network and data network, and establish a two-layer empirical risk association relationship based on the time interval and location of fault occurrence in the risk factor set; Based on the two-layer empirical risk correlation relationship, the historical failures and cost losses of the backbone transmission network and data network are counted and arranged in descending order to form a multi-domain fault-oriented sequence, and a segmented multi-domain correlation risk fitting is performed; Based on the multi-domain correlation risk fitting results and pre-collected operation status monitoring results, the fault risk sequence of the data network and transmission network is obtained; Based on the fault risk sequences of data networks and transmission networks, a fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction, and cross-warning is performed based on the prediction results.

2. The backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to claim 1 is characterized in that: The acquisition of historical fault record data and establishment of a risk factor set for the backbone transmission network and the data network includes: Risk factors are defined based on the dimensions of equipment failure rate, bandwidth utilization, and network topology complexity. The equipment failure rate risk factor for the backbone transmission network is expressed as: (1) Where, is the equipment failure rate risk factor of the backbone transmission network, The equipment failure rate of the backbone transmission network recorded in historical data, is the adjustment factor for equipment failure rate risk; The bandwidth utilization risk factor is expressed as (2) Where, is the bandwidth utilization risk factor of the backbone transmission network, The bandwidth utilization of the backbone transmission network recorded for historical data, is the adjustment factor for bandwidth utilization risk; The network topology complexity risk factor is expressed as (3) Where, is the network topology complexity risk factor of the backbone transmission network, is the network topology complexity index of the backbone transmission network recorded in historical data, is the adjustment factor of network topology complexity risk; The risk factor sets for backbone transmission network and data network constructed based on historical data are respectively expressed as (4) (5) Where, is a risk factor set for the backbone transmission network based on historical data. It is a set of risk factors for the data network built based on historical data. 、 and They are the equipment failure rate risk factor, bandwidth utilization risk factor, and network topology complexity risk factor of the data network, and the calculation method is the same as that of the backbone transmission network; In the backbone transmission network, the introduction of empirical data to the equipment failure rate risk factor Rating ,but The risk factor adjusted after empirical data scoring is expressed as (6) Combining historical data and empirical data, the backbone transmission network equipment failure rate risk factor is expressed as (7) Where, The failure rate risk factor of backbone transmission network equipment is obtained by combining historical data and empirical data. and are the weights of historical data and empirical data respectively; The risk factor sets of the backbone transmission network and data network finally established are expressed as (8) (9) Where, and are the risk factor sets for the backbone transmission network and data network obtained by combining historical data and empirical data; and They are respectively the backbone transmission network bandwidth utilization risk factor and the network topology complexity risk factor obtained by combining historical data and empirical data; , and They are the data network equipment failure rate risk factor, bandwidth utilization risk factor and network topology complexity risk factor obtained by combining historical data and empirical data.

3. The backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to claim 1, characterized in that: The two-layer empirical risk association relationship is established based on the time interval and location of the fault occurrence in the risk factor set, including: According to the time interval and location of the faults, a two-layer empirical risk association model is established. The first layer is the time association model. The time correlation between faults decreases as the time interval increases, which is expressed as (10) Where, For failure and failure The temporal relationship between and is the adjustment factor to control the effect of time interval on the correlation relationship. For failure and failure the time interval between The second layer is the spatial correlation model. The spatial correlation between faults decreases as the distance increases, which can be expressed as (11) Where, For failure and failure The spatial correlation between and is the adjustment factor to control the effect of spatial distance on the association relationship. For failure and failure The spatial distance between them.

4. The backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to claim 1, characterized in that: Based on the two-layer empirical risk association, historical failures and cost losses of the backbone transmission network and the data network are counted and arranged in descending order to form a multi-domain fault-oriented sequence, including: The frequency of each fault, the scope of the fault, the difficulty of recovery and the cost loss recorded in the historical data of the backbone transmission network and data network are counted and comprehensively scored. , the comprehensive score is expressed as (12) Where, Failure of the backbone transmission network The comprehensive rating of Failure of the backbone transmission network Frequency of occurrence, Failure of the backbone transmission network The scope of the fault caused indicates the degree of impact of the fault on the network. Failure of the backbone transmission network The recovery difficulty indicates the difficulty of repairing the fault. Failure of the backbone transmission network Cost loss, which represents the economic loss caused by the failure; , , and is the weight coefficient; After calculating the comprehensive score of each fault, all faults are sorted from high to low according to the comprehensive score, and the multi-domain fault-oriented sequence is obtained, which is expressed as (13) (14) Where, and These are the multi-domain fault-oriented sequences for the backbone transmission network and the data network respectively; The first in the multi-domain fault-oriented sequence of the backbone transmission network elements, The first in the multi-domain fault-oriented sequence of the backbone transmission network The comprehensive score of the fault corresponding to each element, The first in the multi-domain fault-oriented sequence of the backbone transmission network The comprehensive score of each element corresponding to the fault; The first in the multi-domain fault-oriented sequence of the data network elements, The first in the multi-domain fault-oriented sequence of the data network The comprehensive score of the fault corresponding to each element, The first in the multi-domain fault-oriented sequence of the data network The comprehensive score of each element corresponding to the fault; In the descending multi-domain fault-oriented sequence, different subsequences are formed by cutting according to the comprehensive score threshold, which is expressed as (15) (16) Where, This is the upper part of the multi-domain fault-oriented sequence of the backbone transmission network. This is the lower part of the multi-domain fault-oriented sequence of the backbone transmission network. This is the upper part of the multi-domain fault-oriented sequence for data networks. This is the next section of the multi-domain fault-oriented sequence for data networks. is the comprehensive score threshold.

5. The backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to claim 4 is characterized in that: The step of performing segmented multi-domain correlation risk fitting includes: According to the multi-domain guidance sequence, the least square method is used to perform segmented multi-domain correlation risk fitting, and the weight of the upper segment is greater than the weight of the lower segment, that is, , ; When the least squares method is used for fitting, the goal of fitting is to minimize the weighted error, which can be expressed as (17) (18) Where, is the actual observed value, The values predicted by the fitted model.

6. The backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to claim 1, characterized in that: The method of obtaining the fault risk sequence of the data network and the transmission network based on the multi-domain correlation risk fitting results and the pre-collected operation status monitoring results includes: Based on the operation status monitoring results, the risk factor distribution of the backbone transmission network and data network is continuously input. The risk factor distribution of backbone transmission network and data network are expressed as and , the fault risk sequence of the backbone transmission network and the data network is generated through the fault risk sequence generation network, which is expressed as (19) (20) Where, Generate a network for failure risk sequences, and Separate moments Failure risk sequence of backbone transmission network and data network, For the moment Failures in the backbone transmission network For the moment A data network failure occurred.

7. The backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to claim 1 is characterized in that: The fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction based on the fault risk sequence of the data network and the transmission network, including: A fault risk prediction network is used to perform multi-domain feature-driven fault risk prediction. The multi-domain features include the risk fluctuation slope, maximum value, minimum value, variance, number of occurrences of large risk values, frequency of occurrence of large risk values, and frequency change rate of the backbone transmission network and data network. The details are as follows: For the backbone transmission network, the risk fluctuation slope is The maximum risk is expressed as ; The minimum value is expressed as ; Variance is expressed as ,in is the total time, is the mean; the number of occurrences of large risk values is expressed as ,in is the large risk value threshold, is a counting function; the frequency of occurrence of large risk values is expressed as The frequency change rate is expressed as ; For the data network, the risk volatility slope is The maximum risk is expressed as ; The minimum value is expressed as ; Variance is expressed as ;in is the mean; the number of occurrences of large risk values is expressed as The frequency of occurrence of large risk values is expressed as The frequency change rate is expressed as ; The failure risk prediction of backbone transmission network and data network is expressed as (21) (22) Where, Predicting networks for failure risk.

8. The backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to claim 7, characterized in that: The cross-warning is carried out based on the prediction results, including two warning methods: 1) If the risk value obtained based on the self-fault risk sequence prediction is greater than the threshold, an early warning should be issued, which is expressed as (23) Where, To indicate the time An indicator variable indicating whether the backbone transmission network issues a risk warning. If the value is 1, a warning is issued; if the value is 0, no warning is issued. It is an indicator function, and the value in the brackets is 1 when it holds true, and 0 when it does not hold true; is the risk value threshold; Similarly, for the data network, whether to issue an early warning is indicated by (24) 2) Based on the fault risk prediction result of another domain, a warning is issued when the risk value obtained by the multi-domain correlation risk fitting model is greater than the threshold, which is expressed as (25) (26) Where, Represents piecewise multi-domain correlation risk fitting.

9. The backbone transmission network and data network multi-domain correlation risk analysis and early warning system is characterized by: include: The data acquisition module is used to obtain historical fault record data, establish a risk factor set for the backbone transmission network and the data network, and establish a two-layer empirical risk association relationship based on the time interval and location of the fault occurrence in the risk factor set; The risk fitting module is used to collect statistics on historical failures and cost losses of the backbone transmission network and data network based on the two-layer empirical risk correlation relationship, and arrange them in descending order to form a multi-domain fault-oriented sequence, and perform segmented multi-domain correlation risk fitting; The fault risk sequence acquisition module is used to obtain the fault risk sequence of the data network and transmission network based on the multi-domain correlation risk fitting results and the pre-collected operation status monitoring results; The early warning module is used to predict the fault risk sequences of data networks and transmission networks. It uses a fault risk prediction network to perform multi-domain feature-driven fault risk prediction and provides cross-warning based on the prediction results.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the backbone transmission network and data network multi-domain correlation risk analysis and early warning method according to any one of claims 1 to 7 are implemented.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the backbone transmission network and data network multi-domain correlation risk analysis and early warning method as claimed in any one of claims 1 to 7 are implemented.

Images