Optical network health degree assessment method and system based on digital twinning

Abstract

The invention discloses an optical network health degree assessment method and system based on digital twinning, and relates to the technical field of communication. The method comprises the following steps: receiving an optical network health degree prediction request sent from a network management side; obtaining the real-time state and historical data of each element of the current optical network from the equipment side according to the prediction request; according to the historical data and the real-time state of the optical network, constructing an optical network health degree evaluation and prediction model based on a digital twin technology; calculating the health degree of the optical network based on the model, wherein the health degree of the optical network is used for representing the performance reduction or deviation degree after the operation state of the current optical network is compared with the expected stable operation state; and the optical network health degree result is fed back to the network management side. According to the invention, a reasonable and effective solution is provided for evaluating the health degree of the optical network, and the deficiency in the field of evaluating the health degree of the optical network is made up; and the digital twinning technology can be applied to the health degree evaluation of the optical network, so that the actual application requirements are met.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and more specifically, to a method and system for evaluating the health of an optical network based on digital twin. Background Art

[0002] The development of the digital economy requires the support of information technology and communication technology, and the demand for digital transformation in the industry also promotes the evolution of the traditional communication network architecture towards a "cloud-network integration" architecture. It is difficult for the communication network architecture constructed by the traditional "physical device plus professional network management" mode to meet the digital transformation needs of various industries such as "resources on demand, flexible control, and secure and reliable". Moreover, with the continuous development of the informatization process of the whole society, the demands for network latency, capacity, bandwidth, etc. are also showing a rapid growth trend, and the traditional network is facing pressures such as large-scale and high-capacity data exchange and processing.

[0003] As the underlying foundation, the evaluation of the health of the optical network directly affects the communication network architecture. In the optical network, although the use of erbium-doped fiber amplifiers can compensate for some losses caused by long-distance fiber transmission, it introduces amplifier spontaneous emission noise. Affected by the environment, manufacturing process, and signal frequency, the optical fiber inevitably brings chromatic dispersion and polarization mode dispersion. These impairments can be measured by some parameters, such as the parameter optical signal-to-noise ratio (OSNR) and bit error rate (BER). When channel impairments occur, the BER will increase, and the optical power at the receiving end will decrease, resulting in a decrease in OSNR. Therefore, it is of great significance to evaluate the health of the optical network.

[0004] Digital twin originated from CPS (Cyber-Physical Systems, information-physical systems). CPS is a multi-dimensional complex system integrating the physical environment, information network, and computing. Through the deep integration and collaboration of control, communication, and computer technologies, it completes adaptive control, comprehensive perception, and information service supply. It consists of multiple elements, including physical devices and digital components, forming a complex network. However, CPS belongs to the field of scientific research, and digital twin is an important technology that brings the CPS scientific field into the industrial field.

[0005] However, there is currently no reasonable and effective technical solution for evaluating the health of the optical network in the industry, and there is no specific application that combines digital twin technology into the field of optical network health evaluation, which cannot meet the actual application requirements. Summary of the Invention

[0006] The object of the present invention is to provide a method and system for evaluating the health of an optical network based on digital twin, which not only provides a reasonable and effective solution for evaluating the health of an optical network, making up for the lack in the field of optical network health evaluation; but also can apply digital twin technology to the evaluation of optical network health to meet the actual application requirements.

[0007] To achieve the above object, in a first aspect, an embodiment of the present invention provides a method for evaluating the health of an optical network based on digital twin, the method comprising:

[0008] Receiving an optical network health prediction request sent from the network management side;

[0009] According to the optical network health prediction request, obtaining the real-time status and historical data of each element of the current optical network from the device side;

[0010] Based on the optical network historical data and real-time status, constructing an optical network health evaluation and prediction model based on digital twin technology;

[0011] Calculating the optical network health based on the optical network health evaluation and prediction model, where the optical network health is used to represent the degree of performance degradation or deviation after comparing the current operating state of the current optical network with the expected stable operating state; and feeding back the optical network health result to the network management side.

[0012] Combined with the first aspect, in an implementation manner, constructing an optical network health evaluation and prediction model based on digital twin technology includes:

[0013] Based on digital twin technology, from the information dimension, life cycle, and system hierarchy dimension, completing the data modeling at the overall level of the optical network health evaluation and prediction model;

[0014] Based on digital twin technology, through parallel hierarchical abstraction of the main path and standby protection path of the optical network, completing the hierarchical structure modeling of the optical network prediction model.

[0015] Combined with the first aspect, in an implementation manner, the method further includes:

[0016] In the stage of constructing the optical network health evaluation and prediction model, through hierarchical modeling of the optical network health, defining the optical network health as the probability that the performance difference between the current operating state of all subsystems included in the optical network at any time and the expected state performance is not less than the threshold w, and its calculation formula is:

[0017]

[0018] In the formula, R s (t) represents the optical network health; P l,i(t) = Pr{G l (t) = g l,i}, which represents the probability distribution of the i-state of any subsystem l at any time t, and g l,i represents the performance level when the subsystem l is in the i-state.

[0019] Combined with the first aspect, in an implementation manner, calculating the optical network health based on the optical network health assessment and prediction model includes:

[0020] Defining the degradation probability λ l,(i,j) (t) of the optical network subsystem, which is used to represent the probability that any subsystem l degrades from state i to state j; and the original calculation formula of λ l,(i,j) (t) is:

[0021]

[0022] where t represents the usage time of the subsystem, β represents the basic feedback information of the device, and α represents the deviation between the digital twin simulation quantity and the actual quantity;

[0023] According to λ l,(i,j) (t), calculating the state distribution probability P l,i (t) of any subsystem l at any time t;

[0024] According to P l,i (t), calculating the optical network health R s (t).

[0025] Combined with the first aspect, in an implementation manner, the optical network state is divided into three stages: the early failure period, the accidental failure period, and the attenuation failure period; and for different failure periods, the calculation formula of the degradation probability λ l,(i,j) (t) of the optical network subsystem, as well as the calculation methods of α and β are all different.

[0026] Combined with the first aspect, in an implementation manner, in the early failure period, the calculation formula of the degradation probability λ l,(i,j) (t) of the optical network subsystem remains the original calculation formula; and α and β are calculated using the following formulas:

[0027] α = 1 - |simulation quantity - actual quantity| / simulation quantity;

[0028] β = |configured value - feedback value| / critical value.

[0029] Combined with the first aspect, in an implementation manner, in the accidental failure period, the calculation formula of the degradation probability λ l,(i,j) (t) of the optical network subsystem is advanced to:

[0030]

[0031] Moreover, α and β in the accidental failure period are calculated in the following manner:

[0032] First, according to the current formula for λ l,(i,j) (t), the mean and variance of λ l,(i,j) (t) are obtained as follows:

[0033] E(λ l,(i,j) (t)) = α / β

[0034] Var(λ l,(i,j) (t)) = α / β 2 ;

[0035] Then, the mean and variance of λ l,(i,j) (t) are converted through the change of the collectable parameters of the device;

[0036] Finally, by solving the equations of the converted mean and variance, α and β in the accidental failure period are obtained.

[0037] Combined with the first aspect, in one implementation, during the attenuation failure period, the formula for the degradation probability λ l,(i,j) (t) of the optical network subsystem is advanced as follows:

[0038]

[0039] where F represents the number of device failure data in the subsystem, s represents the individuals that exceed the usage threshold but do not fail to work, and T γ represents the time when the individuals that exceed the usage threshold do not fail to work;

[0040] Moreover, α and β in the attenuation failure period are calculated in the following manner:

[0041] By solving the limit formula of λ l,(i,j) (t), α and β in the attenuation failure period are calculated; the formula is as follows:

[0042]

[0043] Combined with the first aspect, in one implementation, the method further includes: judging the critical points between the early failure period, the accidental failure period, and the attenuation failure period; when the change in the optical network health at the later moment exceeds twice the change in the optical network health at the previous moment, it is determined that the optical network health reaches the critical point of the current failure period.

[0044] In the second aspect, an embodiment of the present invention further provides a digital - twin - based optical network health assessment system for implementing the method in the embodiment of the first aspect. The system includes:

[0045] A request receiving unit, which is configured to receive a prediction request for the optical network health degree sent from the network management side;

[0046] A data acquisition unit, which is configured to obtain the real-time status and historical data of each element of the current optical network from the device side according to the prediction request for the optical network health degree;

[0047] A model construction unit, which is configured to construct an optical network health degree evaluation and prediction model based on the digital twin technology according to the optical network historical data and real-time status;

[0048] A prediction calculation unit, which is configured to calculate the optical network health degree based on the optical network health degree evaluation and prediction model, where the optical network health degree is used to represent the degree of performance degradation or deviation after comparing the operating state of the current optical network with the expected stable operating state; and feedback the optical network health degree result to the network management side.

[0049] The beneficial effects brought by the technical solution provided by the embodiment of the present application include:

[0050] This embodiment will construct an optical network health degree evaluation and prediction model based on the digital twin technology. The goal of the optical network health degree evaluation and prediction model is to scientifically and efficiently summarize and statistically analyze the real-time status of each element of the network from the overall network level according to the prediction request for the optical network health degree, obtain the optical network historical performance data within a period of time, and based on the optical network health degree evaluation and prediction model, obtain the optical network prediction performance data, and then calculate the optical network health degree. By using the evaluation system of this embodiment, not only a reasonable and effective solution is proposed for the optical network health degree evaluation, making up for the lack in the field of optical network health degree evaluation; but also the digital twin technology can be applied to the optical network health degree evaluation to meet the actual application requirements. Description of the Drawings

[0051] Figure 1 It is a schematic flowchart of an embodiment of the optical network health degree evaluation method based on the digital twin of the present application;

[0052] Figure 2 It is a schematic diagram of modeling from the information dimension, life cycle, and system hierarchy dimension in the embodiment of the present application;

[0053] Figure 3 It is a schematic diagram of the parallel hierarchical abstraction for the main path and the standby protection path of the optical network in the embodiment of the present application;

[0054] Figure 4 It is a schematic diagram of the change process of the subsystem state over time in an example;

[0055] Figure 5 It is a schematic diagram of a subsystem in an example;

[0056] Figure 6 Schematic diagram of the EDFA temperature change curve in an example;

[0057] Figure 7 Schematic diagram of the EDFA bias current change curve in an example;

[0058] Figure 8 Schematic diagram of the probability curve of the optical network health state distribution in an example;

[0059] Figure 9 Schematic diagram of the change curve of the optical network health degree from T0 - T1 in an example;

[0060] Figure 10 Schematic diagram of the change curve of the optical network health degree from T1 - T2 in an example;

[0061] Figure 11 Schematic diagram of the functional modules of an embodiment of the optical network health degree evaluation system based on digital twin in this application. Detailed implementation manners

[0062] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the technical solutions of the present invention will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0063] It should be noted that: the examples to be introduced next are only some specific examples, and do not limit that the embodiments of the present invention must be the following specific steps, numerical values, conditions, data, sequences, etc. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope protected by this application.

[0064] Embodiment 1

[0065] See Figure 1 As shown, this embodiment provides an optical network health degree evaluation method based on digital twin, and the method includes the following steps:

[0066] A. Receive the optical network health degree prediction request sent from the network management side;

[0067] B. According to the optical network health degree prediction request, obtain the real - time status and historical data of each element of the current optical network from the device side;

[0068] C. Based on the optical network historical data and real - time status, construct an optical network health degree evaluation and prediction model based on digital twin technology;

[0069] D. Calculate the optical network health based on the optical network health assessment and prediction model. The optical network health is used to represent the degree of performance degradation or deviation after comparing the current operating state of the optical network with the expected stable operating state; and feedback the optical network health result to the network management side.

[0070] It can be understood that in the embodiments of the present invention, an optical network health assessment and prediction model will be constructed based on digital twin technology. The goal of this optical network health assessment and prediction model is to scientifically and efficiently summarize and statistically analyze the real-time status of each network element from the overall network level according to the optical network health prediction request, obtain the historical performance data of the optical network over a period of time, and based on the optical network health assessment and prediction model, obtain the predicted performance data of the optical network, and then calculate the optical network health. Among them, the optical network health is used to represent the degree of performance degradation or deviation after comparing the current basic state of the optical network with the expected stable operating state. Using the evaluation method of this embodiment not only provides a reasonable and effective solution for the optical network health assessment, making up for the lack in the field of optical network health assessment; but also can apply digital twin technology to the optical network health assessment to meet the actual application requirements. And in actual applications, the evaluation object of the evaluation method of this embodiment can be the health assessment of devices and optical fiber cable infrastructure, or the end-to-end health assessment of the optical network, including but not limited to transmission quality, network reliability, resource availability, etc.

[0071] Further, as an optional implementation manner, in step C of this embodiment, constructing an optical network health assessment and prediction model based on digital twin technology includes:

[0072] C1. Based on digital twin technology, complete the data modeling at the overall level of the optical network health assessment and prediction model from the information dimension, life cycle, and system hierarchy dimension, as Figure 2 shown.

[0073] Among them, the information dimension includes performance information, experimental information, and expert information. The difference between performance information and experimental information is as follows: The information collected from the live network is performance information, and the information collected in the laboratory is experimental information. Among them, performance information includes network topology and device configuration, optical fiber parameters, amplifier parameters, single-wave and noise channel power, and service configuration parameters, etc. Expert information includes specification parameters, expert rules, constraint conditions, standards, and specifications. Specification parameters refer to the specification parameters of devices, components, etc., which describe the physical information of devices and components, including the physical dimensions of components, panel port information, etc. For example, the length, width, and height information, attenuation coefficient, attenuation spectrum, and gain spectrum factory data of an EDFA single disk, and the upper and lower voice port position information displayed on the panel of a WSS. Expert rules refer to a series of rules formulated to enable the system to operate normally. Generally, there are different empirical rules for the same system in different objective environments. For example, in a wavelength division multiplexing system, the main optical channel stipulates the maximum and minimum output power of each subcarrier and the maximum power difference. With the passage of the system's service life and the difference in the geographical location of use, differential expert rules will be adapted to the specificity of the network at that time. Constraint conditions refer to the artificial restrictions made to enable the service to be successfully launched and not interrupt existing services before the service is launched or during service switching. For example, restricting the EDFA gain adjustment range, etc. Standards / specifications refer to international and domestic industry standards related to the basic model, etc. Generally, there are the same standards and specifications for a certain same system in different objective environments. For example, in the technical requirements for an N×400Gbit / s optical wavelength division multiplexing (WDM) system, it is stipulated that the worst sensitivity of the receiver is -14dBm.

[0074] The life cycle corresponds to before and after the service is launched in the optical network.

[0075] The system level dimension includes the service level, link level, and device level.

[0076] Exemplarily, taking the constructed optical network health assessment and prediction model as an example to realize the end-to-end transmission quality assessment of the optical network, the performance information that needs to be collected includes:

[0077] a) Network topology and device configuration, etc.;

[0078] b) Optical fiber, including type and parameters, length, attenuation spectrum, etc.;

[0079] c) Amplifier, including amplifier type and parameters, calibrated full-wave gain spectrum, noise coefficient spectrum, etc.;

[0080] d) Actual operating parameters of the amplifier: input and output power, actual gain, attenuation;

[0081] e) Single-wave and noise channel power, etc.;

[0082] f) Service configuration parameters, including service path, corresponding OTU wavelength, modulation format, baud rate, sensitivity, OSNR tolerance, and the transmit power and bit error rate during actual operation, etc.

[0083] After that, using the digital twin system based on the collected data, simulation modeling can be carried out in a knowledge-driven manner or a data-driven manner based on machine learning to obtain an optical network health assessment and prediction model, and digital twin simulation quantities (i.e., optical network prediction performance data) can be obtained based on this model. Corresponding to this example, the transmission quality assessment parameters of simulation / prediction can be obtained, including:

[0084] a) Single-wave output power of each channel for each node (including amplification nodes)

[0085] b) OSNR of each OMS section on the service path

[0086] c) End-to-end service OSNR

[0087] d) Nonlinear cost

[0088] e) Service BER

[0089] f) Delay

[0090] g) GSNR (Gain of Signal to Noise Radio)

[0091] C2. Based on digital twin technology, through parallel hierarchical abstraction of the main path and standby protection path of the optical network, hierarchical structural modeling of the optical network prediction model is completed, as Figure 3 shown.

[0092] Furthermore, as an optional implementation manner, in this embodiment, during the stage of constructing the optical network health assessment and prediction model, hierarchical modeling of the optical network health will also be carried out, so as to define the optical network health in this embodiment as the probability that the difference between the current operating state performance and the expected state performance of all subsystems included in the optical network at any moment is not less than the requirement of the threshold w, that is:

[0093]

[0094] In the formula, R s (t) represents the optical network health; P l,i (t) = Pr{G l (t) = g l,i} represents the state distribution probability of any subsystem l at any moment t, g l,i represents the performance level when the subsystem l is in the i state; the threshold w is a manually set value, usually set according to experience.

[0095] Compared with the prior art, the present embodiment optimizes the existing data processing flow. In the existing data processing, data is collected from the original network management side at fixed times, and then the data within this time period is collectively processed. Taking 15 minutes of collection time as an example, this means that the optical network health information cannot be evaluated within 15 minutes and can only be evaluated after 15 minutes, failing to achieve real-time effects. In the present embodiment, however, the probability distribution of the system state is calculated for information at any moment, and the optical network health can be calculated and evaluated in a timely manner at any moment, enabling the system to more sensitively find the change trend of the optical network health, thereby preventing failures in a timely and effective manner.

[0096] Specifically, by hierarchically modeling the optical network health, the optical network health is defined as the probability that the difference between the current operating state performance and the expected state performance of all subsystems included in the optical network at any moment is not less than the threshold w. The specific process is as follows:

[0097] (1) According to the primary and backup paths of the optical network, the optical system is divided into M sub subsystems, where for any subsystem l, it has performance states, which can be represented by the set g l :

[0098]

[0099] Among them, g l,1 represents the performance level when subsystem l is in state 1, g l,2 represents the performance level when subsystem l is in state 2, represents the performance level when subsystem l is in state ; state 1 represents the complete failure state, and state represents the complete working state.

[0100] (2) For the state distribution probability of any subsystem l at any moment t, it can be represented by the set P l (t):

[0101]

[0102] P l,1 (t) represents the distribution probability that subsystem l is in state 1 at any moment t, P l,2 (t) represents the distribution probability that subsystem l is in state 2 at any moment t, represents the distribution probability that subsystem l is in state at any moment t;

[0103] Among them, P l,i (t) = Pr{G l (t) = gl,i}, and satisfy

[0104] For example, taking Figure 4 as an example, within t0, subsystem l is in state, between t0 and t1, subsystem l is in state, between t1 and t2, subsystem l is in g l,2 state, between t2 and t3, subsystem l is in state.

[0105] (3) According to the above process, the optical network health R s (t) can be defined as the probability that the current operating state of the optical network including all subsystems at any time is not lower than the required probability of the performance difference between the expected state w, that is:

[0106]

[0107] For example, as Figure 5 shown, taking the simplest subsystem OTS structure in the optical system as an example, this subsystem includes two EDFAs and a section of optical fiber, which are divided into a main path and a standby path. The service path will only select one of the EDFAs plus a section of optical fiber for transmission. For example, select EDFA-1 plus FIBER-1, or EDFA-2 plus FIBER-1.

[0108] Assume that only the equivalent loss is used to represent the performance state. At this time, the performance states of the EDFA and the optical fiber are shown in Table 1 below:

[0109] Table 1. Performance states of each device (unit: dB)

[0110] Component Status 1 Status 2 Status 3 EDFA-1 2.5 1 0 EDFA-2 3 2.5 0 FIBER-1 / (Fiber break) 2 0

[0111] Assume that the optical network health of this subsystem at any time can represent the probability that the system state at this time is not higher than the expected state (state ) and the insertion loss is 1.5 dB. Then, the optical network health R of this subsystem at any time can be expressed as:

[0112]

[0113] Among them, P 3,3 (t) represents the probability that FIBER-1 is in state 3, P 1,3 (t) represents the probability that EDFA-1 is in state 3, P 1,2 (t) represents the probability that EDFA-1 is in state 2, P 2,3 (t) represents the probability that EDFA-2 is in state 3.

[0114] Further, as an optional implementation manner, in step D of this embodiment, calculating the optical network health degree based on the optical network health degree evaluation and prediction model includes:

[0115] D1. Define the degradation probability λ l,(i,j) (t) of the optical network subsystem. The degradation probability λ l,(i,j) (t) of the optical network subsystem is used to represent the probability that any subsystem l degrades from state i to state j; and the calculation formula of the degradation probability λ l,(i,j) (t) of the optical network subsystem is:

[0116]

[0117] where t represents the usage time of the subsystem, β represents the basic feedback information of the device, and α represents the deviation between the digital twin simulation quantity and the actual quantity. The digital twin simulation quantity is the simulation quantity (i.e., the optical network prediction performance data) obtained based on the optical network health degree evaluation and prediction model. For example, as described in the previous example, the digital twin simulation quantity may include: the single-wavelength output power of each channel of each node (including the amplification node), the OSNR of each OMS section on the service path, the end-to-end service OSNR, the nonlinear cost, the service BER, the delay, and the GSNR, etc.

[0118] In addition, it can be understood that the degradation trend of the optical network subsystem is closely related to the performance indicators of each optical device. Therefore, the present invention introduces the degradation probability λ l,(i,j) (t) of the optical network subsystem to represent the probability that any subsystem degrades from one state to another state, and defines it as Then the cumulative distribution function F of the subsystem failure can be expressed as:

[0119] D2. Calculate the state distribution probability P l,(i,j) (t) of any subsystem l at any time t according to the degradation probability λ l,i (t) of the optical network subsystem.

[0120] It can be understood that since the degradation probability λ l,(i,j) (t) of the optical network subsystem is used to represent the probability that any subsystem degrades from one state to another state, then, assuming that the probability that subsystem l changes from state 1 to state 2 is λ l,(1,2) (t), then P l,2 (t) = λ l,(1,2) (t)P l,1 (t). Therefore, the state distribution probability P l,(i,j) (t) of any subsystem l at any time t can be calculated according to the degradation probability λ l,i (t) of the optical network subsystem.

[0121] D3. Calculate the optical network health degree R l,i (t) according to P s (t).

[0122] Specifically, according to the calculated P l,i (t), and combined with the definition of the optical network health degree R s (t), that is the optical network health degree R s (t) can be obtained.

[0123] Furthermore, in order to accurately and effectively obtain the degradation probability of the optical network subsystem, thus ensuring the correct calculation of the optical network health degree, as an optional implementation manner, in this embodiment, the optical network state is divided into three stages: the early failure stage, the accidental failure stage, and the attenuation failure stage. Among them, the early failure stage corresponds to the period shortly after the services in the optical network are launched and the optical devices have been used for a short time. The entire optical network is in a stable operation period, and there are basically no failures at this time. However, as time goes by, the quality of the optical devices gradually deteriorates, and the optical network enters the accidental failure stage, where the optical devices occasionally have some failures, such as WSS frequency offset, fiber bending, etc.; as time further goes by, the optical network enters the attenuation failure stage, where there are more failures. Although the services can still be maintained at this time, the quality has dropped to the critical point.

[0124] Moreover, for different failure stages, the calculation formula of the degradation probability of the optical network subsystem, as well as the calculation methods of α and β are all different. The specific description is as follows:

[0125] 1. Early failure stage

[0126] In the early failure stage, the optical devices have been used for a short time, and the entire optical network is in a stable operation period, and there are basically no failures at this time. Therefore, for the early failure stage, the calculation formula of the degradation probability of the optical network subsystem can maintain the original calculation formula, that is:

[0127]

[0128] For the calculation methods of α and β, the following formulas can be used to directly calculate:

[0129] α = 1 - |simulation quantity - actual quantity| / simulation quantity;

[0130] β = |configured value - feedback value| / critical value.

[0131] It is understandable that during the early failure period, α reflects the deviation between the digital twin simulation quantity and the actual quantity. The greater the deviation between the simulation quantity and the actual quantity, the further the subsystem is from the ideal stable operating state, that is, the more severely the subsystem deteriorates. Similarly, during the early failure period, β reflects the basic feedback information of the device. Taking the EDFA gain configuration as an example, the degree of deviation between the configured value sent and the actual EDFA configuration reflects the degree of internal aging of the device. The greater the degree of internal aging of the device, the greater the degree of deviation. Among them, the critical value is the set value, usually set by experts. For example, the EDFA gain configuration critical value is 1 dBm.

[0132] 2. Random Failure Period

[0133] During the random failure period, without human intervention, the state of the subsystem has been in a declining period. At this time, the calculation formula for the deterioration probability of the optical network subsystem can be further advanced to:

[0134]

[0135] At this time, the deterioration of the subsystem shows non - negativity and monotonicity, indicating that the deterioration process of the subsystem is irreversible. At this time, the mean and variance of the deterioration probability of the optical network subsystem are:

[0136] E(λ l,(i,j) (t)) = α / β

[0137] Var(λ l,(i,j) (t)) = α / β 2

[0138] The mean and variance of the deterioration probability of the optical network subsystem are related to the stability of each device within the subsystem, and the stability of each device is reflected by the change of the collectable parameters of the device. Taking the example that the subsystem only contains one EDFA, the collectable parameters of the EDFA include temperature and bias current. The temperature change curve of the EDFA is as Figure 6 shown, and its change function is F1(t). The bias current change curve of the EDFA is as Figure 7 shown, and its change function is F2(t). Therefore, the mean and variance of the deterioration trend of the subsystem can be converted through the change of the collectable parameters of the device. For example, through the temperature change function F1(t) and the bias current change function F2(t), we can get:

[0139] E(λ l,(i,j) (t)) = E(F1F2)

[0140] Var(λ l,(i,j) (t)) = Var(F1F2)

[0141] By further solving the mean and variance equations after conversion, α and β during the random failure period can be calculated.

[0142] 3. Attenuation Failure Period

[0143] In the attenuation failure period, when subsystem failures occur frequently, assume that the failure data of F devices inside the subsystem are observed as x1, x2, ……, x F , then the calculation formula for the degradation probability of the optical network subsystem can be:

[0144]

[0145] Furthermore, considering that during the frequent attenuation failure period of the subsystem, it is mostly because the product has been used for too long and has some degradation and aging characteristics. At this time, although the overall degradation and aging trend of the product is the same, there may be differences in the degradation and aging time points or patterns of different sample individuals. Especially for active devices such as EDFA, due to the differences in some sample individuals themselves and the manufacturing process, even if they reach the usage threshold of certain conditions, they may not necessarily fail. At this time, there is a problem of correlation and possibility between exceeding the usage threshold range and whether there is a failure, that is, only a certain proportion of products will have failures when exceeding the usage threshold. If a failure is reported when the threshold is exceeded, the failure indicator light will also be on, but the product can still continue to work and is not completely damaged. Therefore, considering the proportion of this part of the products, the calculation formula for the degradation probability of the optical network subsystem can be further advanced to:

[0146]

[0147] Among them, F represents the number of device failure data in the subsystem, s represents the individuals that do not stop working when exceeding the usage threshold, and T γ represents the time when the individuals that do not stop working among those exceeding the usage threshold. In actual application, the above parameters F, s, and T γ are all data obtained by correcting historical data considering that "only a certain proportion of products will have failures when exceeding the usage threshold, but this part of the products can still continue to work and are not completely damaged". And, since in the above calculation formula for the degradation probability of the optical network subsystem, except for α, β, other parameters (such as F, s, and T γ etc.) are known parameters, therefore, according to the known parameters, the limit formula for the degradation probability of the optical network subsystem can be solved, as follows:

[0148]

[0149] Thus, α and β can be calculated.

[0150] For example, taking a single EDFA within a week's statistics in the attenuation failure period as an example, the EDFA status statistical table of historical data can be as shown in Table 2 below:

[0151] Table 2

[0152]

[0153]

[0154] When considering that the probability that the device is still effective when the EDFA is under the condition of over-limit bias current is 0.6, Table 2 is corrected to obtain the corrected table of EDFA states, as shown in Table 3 below:

[0155] Table 3

[0156] Serial number Fault time (s) Fault cause Fault status Quantity 1 200 Overheat Fault 1 2 350 Bias current overlimit Fault 0.4 3 350 / 0.6 4 800 Overheat Fault 1 5 400 Bias current overlimit Fault 0.4 6 400 / 0.6 7 200 Overheat Fault 1

[0157] According to Table 3, the degradation probability of the optical network subsystem is solved as follows:

[0158]

[0159] It can be calculated that α = 3862.84 and β = 2.195. At this time, according to the calculated α and β, combined with the calculation formula of the degradation probability of the optical network subsystem, the degradation probability of this EDFA from the attenuation failure period to complete damage is λ = e -52.64 .

[0160] Furthermore, in order to better understand and evaluate the change trend of the optical network health, as an optional implementation manner, in the method of this embodiment, the critical points between the early failure period, the accidental failure period, and the attenuation failure period are also determined. That is, when the change in the optical network health at the later moment exceeds twice the change in the optical network health at the previous moment, it is determined that the optical network health reaches the critical point of the current failure period.

[0161] Specifically, the probability distribution curve of the optical network in the state at the moment when the service is just opened (time T0) satisfies the Gaussian distribution, and the specific curve diagram is as shown. As time goes by (T0, T1, T2...), the probability distribution curve of the state of the optical network health will gradually change, but generally remains the Gaussian distribution. For example, the schematic diagram of the forward and backward movement of the health change curve of an optical network is as Figure 8 shown. Among them, h1 represents the degree of change in the optical network health from time T0 to T1, and h2 represents the degree of change in the optical network health from time T1 to T2. Figure 9 , Figure 10 shown. Among them, h1 represents the degree of change in the optical network health from time T0 to T1, and h2 represents the degree of change in the optical network health from time T1 to T2.

[0162] It can be known that the distribution probability of the optical network health at time T0 is The distribution probability of the optical network health at time T1 is Then the change from T0 to T1 is:

[0163]

[0164] However, if only the changes in the two Gaussian units are analyzed at the data level, the correlation between the system feedback parameters before and after time is often ignored, and the actual situation of the optical network cannot be reflected. Therefore, the α (the deviation between the digital twin simulation quantity and the actual quantity) at the current moment is further introduced, and the probability distribution of the optical network health at time T0 is The probability distribution of the optical network health at time T1 is Then the change from T0 to T1 is:

[0165]

[0166] Through the above calculation method, the change h2 from T1 to T2 and the change h3 from T2 to T3 can also be calculated. Then, when the change in the optical network health at the latter moment exceeds twice the change in the optical network health at the previous moment, it can be determined that the optical network health reaches the critical point of the current fault period. That is, when h2 > 2h1, the optical network health from T1 to T2 drops from the early fault period to the accidental fault period; when h3 ≤ 2h2, the optical network health from T2 to T3 remains in the accidental fault period.

[0167] Embodiment 2

[0168] Participate Figure 11 As shown in , based on the same inventive concept, the embodiment of the present invention also provides a digital twin-based optical network health assessment system for implementing the method in the first embodiment. The system includes:

[0169] A request receiving unit, which is used to receive an optical network health prediction request sent from the network management side;

[0170] A data acquisition unit, which is used to obtain the real-time status and historical data of each element of the current optical network from the device side according to the optical network health prediction request;

[0171] A model construction unit, which is used to construct an optical network health assessment and prediction model based on digital twin technology according to the optical network historical data and real-time status;

[0172] A prediction calculation unit, which is used to calculate the optical network health based on the optical network health assessment and prediction model. The optical network health is used to represent the degree of performance degradation or deviation after comparing the current operating state of the optical network with the expected stable operating state; and feedback the optical network health result to the network management side.

[0173] As can be seen from the above, the model construction unit in this embodiment constructs an optical network health assessment and prediction model based on digital twin technology. The goal of this optical network health assessment and prediction model is to scientifically and efficiently summarize and statistically analyze the real-time status of each network element at the overall network level according to the optical network health prediction request, obtain the historical performance data of the optical network over a period of time, and based on the optical network health assessment and prediction model, obtain the predicted performance data of the optical network, and then calculate the optical network health. Using the evaluation system of this embodiment not only provides a reasonable and effective solution for the optical network health assessment, making up for the lack in the field of optical network health assessment, but also can apply digital twin technology to the optical network health assessment to meet the actual application requirements.

[0174] It should be noted that the various change methods and specific examples in the foregoing method embodiments are equally applicable to the system in this embodiment. Through the detailed description of the foregoing method, those skilled in the art can clearly know the implementation manner of the system in this embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here.

[0175] Note: The serial numbers of the embodiments of the present application above are only for description and do not represent the superiority or inferiority of the embodiments.

[0176] The terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusion. 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 further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or devices. The descriptions of the terms "first", "second", "third", etc. are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit that "first", "second", and "third" are different types.

[0177] In the description of the embodiments of the present application, terms such as "exemplary", "for example" or "for instance" are used to indicate examples, illustrations or explanations. Any embodiment or design solution described as "exemplary", "for example" or "for instance" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of terms such as "exemplary", "for example" or "for instance" is intended to present relevant concepts in a specific manner.

[0178] In the description of the embodiments of the present application, unless otherwise specified, " / " means "or". For example, A / B may mean A or B. The "and / or" in the text is merely a description of the relationship between associated objects, indicating that there can be three relationships. For example, A and / or B may mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present application, "a plurality of" means two or more than two.

[0179] In some of the processes described in the embodiments of the present application, there are a plurality of operations or steps that appear in a specific order. However, it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or may be executed in parallel. The serial numbers of the operations are only used to distinguish different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in order or in parallel, and these operations or steps may be combined.

[0180] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium as described above (such as ROM / RAM, magnetic disk, optical disc) and includes several instructions for causing a terminal device to execute the methods described in the various embodiments of the present application.

[0181] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be included in the patent protection scope of the present application in the same way.

Claims

1. A method for evaluating the health of an optical network based on digital twins, characterized in that: The method includes: Receiving an optical network health prediction request sent from a network management side; According to the optical network health prediction request, the real-time status and historical data of each element of the current optical network are obtained from the device side; Based on the historical data and real-time status of the optical network, an optical network health assessment and prediction model is built based on digital twin technology; The optical network health is calculated based on the optical network health evaluation and prediction model, and the optical network health is used to indicate the degree of performance degradation or deviation after comparing the current operating state of the optical network with the expected stable operating state; and the optical network health result is fed back to the network management side.

2. The optical network health assessment method based on digital twin according to claim 1, characterized in that: Construct an optical network health assessment and prediction model based on digital twin technology, including: Based on digital twin technology, complete the overall data modeling of optical network health assessment and prediction model from the information dimension, life cycle and system level dimension; Based on digital twin technology, the hierarchical structural modeling of the optical network prediction model is completed through the parallel hierarchical abstraction of the main path and backup protection path of the optical network.

3. The optical network health assessment method based on digital twin according to claim 1, characterized in that: The method further includes: In the stage of building the optical network health evaluation and prediction model, the optical network health is defined as the probability that the difference between the current operating state performance and the expected state performance of all subsystems of the optical network at any time is not less than the limit w requirement by performing hierarchical modeling on the optical network health. The calculation formula is: In the formula, R s (t) represents the health of the optical network; P l,i (t) = Pr{G l (t) = g l,i }, represents the probability of state i of any subsystem l at any time t, g l,i Indicates the performance level of subsystem l when it is in state i.

4. The optical network health assessment method based on digital twin according to claim 3 is characterized in that: Calculating the health of the optical network based on the optical network health evaluation and prediction model includes: Define the degradation probability λ of the optical network subsystem l,(i,j) (t), which is used to represent the probability of any subsystem l deteriorating from state i to state j; and λ l,(i,j) The original calculation formula of (t) is: Among them, t represents the use time of the subsystem, β represents the basic feedback information of the device, and α represents the deviation between the digital twin simulation quantity and the actual quantity; According to λ l,(i,j) (t), calculate the state distribution probability P of any subsystem l at any time t l,i (t); According to P l,i (t), calculate the optical network health R s (t).

5. The optical network health assessment method based on digital twin according to claim 4, characterized in that: The optical network status is divided into three stages: early failure period, accidental failure period and attenuation failure period; and for different failure periods, the degradation probability of the optical network subsystem λ l,(i,j) The calculation formula of (t) and the calculation methods of α and β are different.

6. The optical network health assessment method based on digital twin according to claim 5, characterized in that: In the early failure period, the optical network subsystem degradation probability λ l,(i,j) The calculation formula of (t) remains the original calculation formula; and α and β are calculated using the following formula: α=1-|simulation amount-actual amount| / simulation amount; β = |configuration value - feedback value| / critical value.

7. The optical network health assessment method based on digital twin according to claim 5, characterized in that: During the accidental failure period, the optical network subsystem degradation probability λ l,(i,j) The calculation formula of (t) is as follows: And the α and β of the accidental failure period are calculated in the following way: First, according to the current λ l,(i,j) (t) is calculated by l,(i,j) The mean and variance of (t) are: E(λ l,(i,j) (t))=α / β Var(λ l,(i,j) (t))=α / β 2 ; Then λ l,(i,j) The mean and variance of (t) are converted by the changes in the collectible parameters of the device; Finally, by solving the converted mean and variance equations, α and β of the accidental failure period are obtained.

8. The optical network health assessment method based on digital twin according to claim 5, characterized in that: During the attenuation fault period, the optical network subsystem degradation probability λ l,(i,j) The calculation formula of (t) is as follows: Among them, F represents the number of device failure data in the subsystem, s represents the individual that exceeds the usage threshold but does not fail to work, T γ It indicates that no time was lost from work among individuals who exceeded the threshold of use; And the α and β of the attenuation failure period are calculated in the following way: By l,(i,j) (t) Solve the limiting formula to calculate the α and β of the attenuation failure period; the formula is as follows:

9. The optical network health assessment method based on digital twin according to claim 5, characterized in that: The method further includes: The critical points between the early failure period, the accidental failure period and the attenuation failure period are judged; when the change in the health of the optical network at a later moment exceeds twice the change in the health of the optical network at a previous moment, it is determined that the health of the optical network has reached the critical point of the current failure period.

10. An optical network health assessment system based on digital twins that implements the method described in any one of claims 1 to 9, characterized in that: The system includes: A request receiving unit, which is used to receive an optical network health prediction request sent from a network management side; A data acquisition unit, which is used to obtain the real-time status and historical data of various elements of the current optical network from the device side according to the optical network health prediction request; A model building unit, which is used to build an optical network health assessment and prediction model based on digital twin technology according to the historical data and real-time status of the optical network; A prediction calculation unit, which is used to calculate the health of the optical network based on the optical network health evaluation and prediction model, wherein the optical network health is used to indicate the degree of performance degradation or deviation after comparing the current operating state of the optical network with the expected stable operating state; and feed back the optical network health result to the network management side.

Images