Multi-span optical communication system power spectrum prediction method and system based on grey box model

Abstract

The invention provides a multi-span optical communication system power spectrum prediction method and system based on a grey box model, and the method comprises the steps: S1, carrying out the measurement to obtain a plurality of groups of spectrum data, and solving a model parameter of a power spectrum through a mode of solving an equation; s2, constructing a digital twinborn model of the power spectrum of the multi-span optical communication system by using the solved model parameters; and S3, predicting the power spectrum by using the constructed digital twinborn model, and outputting a prediction result. According to the power spectrum modeling method based on the Er-doped optical fiber amplifier, power spectrum modeling can be carried out under the condition that power spectrum information is lost in an actual system by adopting a method of simultaneously solving an equation through the physical law of the optical fiber and the Er-doped optical fiber amplifier; the problems that in power spectrum modeling, a data driving model needs a large amount of data, and physical model parameters are difficult to obtain are solved, the power spectrum model with physical interpretability can be obtained under the condition of small data amount, and the method has good application prospects.

Description

Technical Field

[0001] The present invention relates to the field of information and communication engineering technology, and in particular to a method and system for predicting power spectrum of a multi-span optical communication system based on a grey box model. Background Art

[0002] With the rapid development of information technology, optical communication systems, as the core infrastructure for data transmission, are facing the need to continuously increase transmission capacity and distance. In this context, accurately predicting the power spectrum of optical communication systems is of vital importance to ensure signal quality, optimize system design and maintenance.

[0003] Traditional power spectrum prediction methods can be divided into two categories: physical model-based methods and purely data-driven modeling methods.

[0004] The physical model-based approach relies on the accurate physical properties of key components such as optical fiber and erbium-doped fiber amplifier (EDFA) to establish a theoretical model. Although this model provides a clear physical explanation, its application is restricted by complex computational requirements and limited ability to respond to nonlinear effects and environmental changes, especially when the power spectrum information in the actual system is incomplete, it is difficult to achieve high-precision predictions.

[0005] The data-driven approach uses a large amount of measured data to train machine learning or other statistical models to capture the dynamic behavior of the system. This method can adapt to complex environmental changes and automatically adjust model parameters, but in the absence of sufficient data, its generalization ability and accuracy will be affected, and the generated models often lack physical interpretability.

[0006] A method for predicting power spectrum using cascaded learning is proposed in Z. Wang, Y. K. Huang, S. Han, T. Wang, D. Kilper and T. Chen, "Multi-Span Optical Power Spectrum Prediction using ML-based EDFA Models and Cascaded Learning," 2024 Optical Fiber Communications Conference and Exhibition (OFC), San Diego, CA, USA, 2024, pp. 1-3. However, this method requires pre-training of the model in advance and requires a large amount of measurement data.

[0007] The present invention proposes a power spectrum prediction method and system for a multi-span optical communication system based on a gray box model, aiming to solve the model parameters by solving equations simultaneously, so that power spectrum modeling can still be performed in the absence of power spectrum information, and a physically interpretable model can still be obtained with a small amount of data. Summary of the invention

[0008] In view of the defects in the prior art, the object of the present invention is to provide a method and system for predicting the power spectrum of a multi-span optical communication system based on a gray box model.

[0009] According to the present invention, a method for predicting power spectrum of a multi-span optical communication system based on a gray box model comprises:

[0010] Step S1: Measure and obtain multiple sets of spectrum data, and obtain the model parameters of the power spectrum by solving equations;

[0011] Step S2: constructing a digital twin model of the power spectrum of a multi-span optical communication system using the obtained model parameters;

[0012] Step S3: Use the constructed digital twin model to predict the power spectrum and output the prediction results.

[0013] Preferably, the step S1 comprises:

[0014] Step S1.1: Based on the output power spectrum S of the erbium-doped fiber amplifier out The change in ΔS out Get the model parameters dG:

[0015] dG=ΔS out ;

[0016] Step S1.2: Input the total power P to the optical fiber in，fiber Two different groups of optical fiber input power spectra S in and S′ in , calculate the attenuation spectrum Loss and Loss′ of the fiber with self-excited Raman scattering effect;

[0017] Step S1.3: Calculate model parameter k ISRS :

[0018] var[(G EDFA (k ISRS , m)-G' EDFA (k ISRS , m')) / dG]=0

[0019] Among them, k ISRS The physical meaning is 10lge·C r L eff / δ,C ris the normalized Raman gain spectrum slope, L eff is the effective length, δ is the insertion loss, and m is the value of the known P in，EDFA Next, the shift of the ISRS gain spectrum is obtained;

[0020] Step S1.4: Calculate the gain spectrum G of the erbium-doped fiber amplifier EDFA and G' EDFA :

[0021] G EDFA =S out -(S in -Loss(k ISRS ,m))

[0022] Step S1.5: Calculate model parameter G 0 :

[0023] G 0 =mean(G EDFA (k ISRS , m), G' EDFA (k ISRS , m'));

[0024] Among them, dG and G 0 Represents the wavelength dependence of the gain spectrum of an erbium-doped fiber amplifier.

[0025] Preferably, the calculation process of step S1.2 includes:

[0026] Loss(k ISRS , m) = αL-G ISRS (k ISRS , m)

[0027] G ISRS (k ISRS , m) = k ISRS P in，fiber λ+m

[0028]

[0029] Where P in,fiber is the total fiber power, α is the fiber attenuation spectrum, L is the fiber length, and λ is the wavelength vector; P in,EDFA is the total input power of the erbium-doped fiber amplifier.

[0030] Preferably, the calculation process of the digital twin model includes:

[0031] G ISRS (k ISRS , m) = k ISRS P in，fiber λ+m

[0032]

[0033] G EDFA (x) = dG x + G 0

[0034]

[0035] Where x represents the input power dependence of the erbium-doped fiber amplifier gain spectrum, S in，EDFA is the input power spectrum of the erbium-doped fiber amplifier, G tot is the total gain of the erbium-doped fiber amplifier.

[0036] Preferably, step S3 includes the following sub-steps:

[0037] Step S3.1: S in and P in，EDFA Substitute the following formula to obtain the variable m:

[0038]

[0039] Step S3.2: Calculate Loss:

[0040] Loss(k ISRS , m) = αL-G ISRS (k ISRS , m):

[0041] according to

[0042] S in，EDFA =S in -Loss

[0043] Calculate S in，EDFA ;

[0044] Step S3.3: G tot and S in，EDFA Substitute the following formula to find the variable x:

[0045]

[0046] Calculate G using the following formula EDFA :

[0047] G EDFA (x) = dG x + G 0 ;

[0048] Step S3.4: Use

[0049] S out =S in，EDFA +G EDFA

[0050] Calculate the output S of the modelout .

[0051] A power spectrum prediction system for a multi-span optical communication system based on a gray box model is provided according to the present invention, comprising:

[0052] Module M1: Measure multiple sets of spectrum data and obtain the model parameters of the power spectrum by solving equations;

[0053] Module M2: Use the obtained model parameters to construct a digital twin model of the power spectrum of a multi-span optical communication system;

[0054] Module M3: Use the constructed digital twin model to predict the power spectrum and output the prediction results.

[0055] Preferably, the module M1 comprises:

[0056] Module M1.1: Based on the output power spectrum S of the erbium-doped fiber amplifier out The change in ΔS out Get the model parameters dG:

[0057] dG=ΔS out ;

[0058] Module M1.2: Total optical fiber input power P in，fiber Two different groups of optical fiber input power spectra S in and S′ in , calculate the attenuation spectrum Loss and Loss′ of the fiber with self-excited Raman scattering effect;

[0059] Module M1.3: Calculation of model parameter k ISRS :

[0060] var[(G EDFA (k ISRS , m)-G' EDFA (k ISRS , m')) / dG]=0

[0061] Among them, k ISRS The physical meaning is 10lge·C r L eff / δ,C r is the normalized Raman gain spectrum slope, L eff is the effective length, δ is the insertion loss, and m is the value of the known P in，EDFA Next, the shift of the ISRS gain spectrum is obtained;

[0062] Module M1.4: Calculate the gain spectrum G of the erbium-doped fiber amplifier EDFA and G' EDFA :

[0063] GEDFA =S out -(S in -Loss(k ISRS ,m))

[0064] Module M1.5: Calculation of model parameters G 0 :

[0065] G 0 =mean(G EDFA (k ISRS , m), G' EDFA (k ISRS , m'));

[0066] Among them, dG and G 0 Represents the wavelength dependence of the gain spectrum of an erbium-doped fiber amplifier.

[0067] Preferably, the calculation process of the module M1.2 includes:

[0068] Loss(k ISRS , m) = αL-G ISRS (k ISRS , m)

[0069] G ISRS (k ISRS , m) = k ISRS P in，fiber λ+m

[0070]

[0071] Where P in,fiber is the total fiber power, α is the fiber attenuation spectrum, L is the fiber length, and λ is the wavelength vector; P in,EDFA is the total input power of the erbium-doped fiber amplifier.

[0072] Preferably, the calculation process of the digital twin model includes:

[0073] G ISRS (k ISRS , m) = k ISRS P in，fiber λ+m

[0074]

[0075] G EDFA (x) = dG x + G 0

[0076]

[0077] Where x represents the input power dependence of the erbium-doped fiber amplifier gain spectrum, S in，EDFAis the input power spectrum of the erbium-doped fiber amplifier, G tot is the total gain of the erbium-doped fiber amplifier.

[0078] Preferably, the module M3 includes the following submodules:

[0079] Module M3.1: S in and P in，EDFA Substitute the following formula to obtain the variable m:

[0080]

[0081] Module M3.2: Calculate Loss:

[0082] Loss(k ISRS , m) = αL-G ISRS (k ISRS , m);

[0083] according to

[0084] S in，EDFA =S in -Loss

[0085] Calculate S in，EDFA ;

[0086] Module M3.3: G tot and S in，EDFA Substitute the following formula to find the variable x:

[0087]

[0088] Calculate G using the following formula EDFA :

[0089] G EDFA (x) = dG x + G 0 ;

[0090] Module M3.4: Use

[0091] S out =S in，EDFA +G EDFA

[0092] Calculate the output S of the model out .

[0093] Compared with the prior art, the present invention has the following beneficial effects:

[0094] 1. The present invention adopts a method of solving equations by combining the physical laws of optical fiber and erbium-doped fiber amplifier, so as to perform power spectrum modeling in the absence of power spectrum information in an actual system, and has good practicality.

[0095] 2. The present invention solves the problem that the data-driven model in power spectrum modeling requires a large amount of data and the physical model parameters are difficult to obtain by adopting a method that combines physical laws with data-driven. It can obtain a power spectrum model with physical interpretability with a small amount of data and has good application prospects.

[0096] Other beneficial effects of the present invention will be explained in the specific implementation manner through the introduction of specific technical features and technical solutions. Through the introduction of these technical features and technical solutions, those skilled in the art should be able to understand the beneficial technical effects brought about by the technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0098] Figure 1 The figure is a flow chart of the method of the present invention.

[0099] Figure 2 It is a schematic diagram of the calculation process of the present invention. DETAILED DESCRIPTION

[0100] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0101] Reference Figure 1 and Figure 2 As shown, a method for predicting power spectrum of a multi-span optical communication system based on a gray box model comprises:

[0102] Step 1: Measure and obtain multiple sets of spectrum data, and obtain the model parameters of the power spectrum by solving equations;

[0103] Step 2: Use the obtained model parameters to construct a digital twin model of the power spectrum of the multi-span optical communication system;

[0104] Step 3: Use the constructed digital twin model to predict the power spectrum.

[0105] Wherein, step 1 comprises the following steps:

[0106] In order to explain the principle simply, a cascade of a section of optical fiber and an erbium-doped fiber amplifier (EDFA) is used as an example. 0 , k ISRS , x and m, a total of 5 parameters need to be solved, among which dG, G0 and k ISRS are model parameters, x and m are variables that need to be solved for different input spectra, which will be explained in the following steps.

[0107] Step 1.1: Change the EDFA gain G tot , measure the EDFA output power spectrum S out The change in ΔS out , the model parameter dG can be calculated using formula (1).

[0108] dG=ΔS out ,#(1)

[0109] Step 1.2: Input the total power P to the fiber in，fiber Two different groups of optical fiber input power spectra S in and S′ in The attenuation spectrum Loss of the optical fiber considering the self-excited Raman scattering effect (ISRS) can be calculated using formulas (2)-(4). Loss′ (corresponding to S′ in ) is calculated in the same way as Loss.

[0110] Loss(k ISRS , m) = αL-G ISRS (k ISRS ,m),#(2)

[0111] G ISRS (k ISRS , m) = k ISRS P in，fiber λ+m,#(3)

[0112]

[0113] Where P in，fiber is the total fiber power, α is the fiber attenuation spectrum, L is the fiber length, and λ is the wavelength vector. m in formula (3) can be obtained using formula (4), where P in，EDFA is the total input power of EDFA. Next, the EDFA gain spectrum G is calculated by formula (5): EDFA . G′ EDFA (corresponding to S′ in ) is calculated in the same way.

[0114] G EDFA =S out -(S in -Loss(k ISRS ,m)),#(5)

[0115] Step 1.3: Model parameter k ISRS Formula (6) can be used to solve it.

[0116] var[(G EDFA (k ISRS , m)-G' EDFA (k ISRS , m')) / dG]=0, #(6)

[0117] ISRS can be expressed as:

[0118] Δρ(z)[dB]=10lge·P in，fiber C r L eff B, #(10)

[0119] In the formula, C r is the normalized Raman gain spectrum slope, L eff is the effective length, and B is the total spectrum width. Considering the unknown insertion loss δ before the span, the actual total fiber power is P in，fiber / δ.

[0120] The present invention focuses on P in，fiber Therefore, the other parts in formula (10) are simplified to k ISRS , that is, k ISRS The physical meaning is 10lge·C r L eff / δ. This is the physical meaning of formula (3). The physical meaning of m is that in the known P in，EDFA Under this condition, the offset of ISRS gain spectrum is obtained. dG and G in formula (8) 0 represents the wavelength dependency of the EDFA gain spectrum. x represents the input power dependency of the EDFA gain spectrum.

[0121] Step 1.4: Set k ISRS Substituting into formula (5) we can get G EDFA and G' EDFA Model parameter G 0 Formula (7) can be used to solve it.

[0122] G 0 =mean(G EDFA (k ISRS , m), G' EDFA (k ISRS ,m')),#(7)

[0123] Wherein, step 2 comprises the following steps:

[0124] Step 2.1: Substitute the model parameters dG, G obtained in step 1 0 and k ISRS Substituting into formulas (3), (4), (8), and (9) we can obtain the digital twin model of the power spectrum.

[0125] G EDFA (x) = dG x + G 0 ,#(8)

[0126]

[0127] Among them, S in，EDFA is the input power spectrum of the erbium-doped fiber amplifier, G tot is the total gain of the erbium-doped fiber amplifier.

[0128] Step 3 includes the following steps:

[0129] Step 3.1: S in and P in，EDFA Substitute into formula (4) to obtain the variable m.

[0130] Step 3.2: Use formula (2) to calculate Loss. S in，EDFA You can use S in，EDFA =S in -Loss is calculated.

[0131] Step 3.3: G tot and S in，EDFA Substitute into formula (9) to obtain the variable x. Use formula (8) to calculate G EDFA .

[0132] Step 3.4: S out You can use S out =S in，EDFA +G EDFA Perform calculations.

[0133] The present invention adopts a method of solving equations by simultaneously applying the physical laws of optical fiber and erbium-doped fiber amplifier, and can perform power spectrum modeling in the absence of power spectrum information in an actual system, and has good practicality; by adopting a method combining physical laws with data-driven, the problem that the data-driven model in power spectrum modeling requires a large amount of data and the physical model parameters are difficult to obtain is solved, and a power spectrum model with physical interpretability can be obtained with a small amount of data, and has good application prospects.

[0134] The present invention also provides a power spectrum prediction system for a multi-span optical communication system based on a gray box model. The power spectrum prediction system for a multi-span optical communication system based on a gray box model can be implemented by executing the process steps of the power spectrum prediction method for a multi-span optical communication system based on a gray box model, that is, those skilled in the art can understand the power spectrum prediction method for a multi-span optical communication system based on a gray box model as a preferred implementation of the power spectrum prediction system for a multi-span optical communication system based on a gray box model.

[0135] Specifically, a power spectrum prediction system for a multi-span optical communication system based on a gray box model includes:

[0136] Module M1: Measure multiple sets of spectrum data and obtain the model parameters of the power spectrum by solving equations;

[0137] Module M2: Use the obtained model parameters to construct a digital twin model of the power spectrum of a multi-span optical communication system;

[0138] Module M3: Use the constructed digital twin model to predict the power spectrum and output the prediction results.

[0139] The module M1 comprises:

[0140] Module M1.1: Based on the output power spectrum S of the erbium-doped fiber amplifier out The change in ΔS out Get the model parameters dG:

[0141] dG=ΔS out ;

[0142] Module M1.2: Total optical fiber input power P in，fiber Two different groups of optical fiber input power spectra S in and S′ in , calculate the attenuation spectrum Loss and Loss' of the fiber with self-excited Raman scattering effect;

[0143] Module M1.3: Calculation of model parameter k ISRS :

[0144] var[(G EDFA (k ISRS , m)-G′ EDFA (k ISRS , m')) / dG]=0

[0145] Among them, k ISRS The physical meaning is 10lge·C r L eff / δ,C r is the normalized Raman gain spectrum slope, L eff is the effective length, δ is the insertion loss, and m is the value of the known P in，EDFA Next, the shift of the ISRS gain spectrum is obtained;

[0146] Module M1.4: Calculate the gain spectrum G of the erbium-doped fiber amplifier EDFA and G' EDFA :

[0147] G EDFA =S out -(S in-Loss(k ISRS ,m))

[0148] Module M1.5: Calculation of model parameters G 0 :

[0149] G 0 =mean(G EDFA (k ISRS , m), G' EDFA (k ISRS , m'));

[0150] Among them, dG and G 0 Represents the wavelength dependence of the gain spectrum of an erbium-doped fiber amplifier.

[0151] The calculation process of the module M1.2 includes:

[0152] Loss(k ISRS , m) = αL-G ISRS (k ISRS , m)

[0153] G ISRS (k ISRS , m) = k ISRS P in，fiber λ+m

[0154]

[0155] Where P in,fiber is the total fiber power, α is the fiber attenuation spectrum, L is the fiber length, and λ is the wavelength vector; P in,EDFA is the total input power of the erbium-doped fiber amplifier.

[0156] The calculation process of the digital twin model includes:

[0157] G ISRS (k ISRS , m) = k ISRS P in，fiber λ+m

[0158]

[0159] G EDFA (x) = dG x + G 0

[0160]

[0161] Where x represents the input power dependence of the erbium-doped fiber amplifier gain spectrum, S in，EDFA is the input power spectrum of the erbium-doped fiber amplifier, G tot is the total gain of the erbium-doped fiber amplifier.

[0162] The module M3 includes the following submodules:

[0163] Module M3.1: S in and P in，EDFA Substitute the following formula to obtain the variable m:

[0164]

[0165] Module M3.2: Calculate Loss:

[0166] Loss(k ISRS , m) = αL-G ISRS (k ISRS , m);

[0167] according to

[0168] S in，EDFA =S in -Loss

[0169] Calculate S in，EDFA ;

[0170] Module M3.3: G tot and S in，EDFA Substitute the following formula to find the variable x:

[0171]

[0172] Calculate G using the following formula EDFA :

[0173] G EDFA (x) = dG x + G 0 ;

[0174] Module M3.4: Use

[0175] S out =S in，EDFA +G EDFA

[0176] Calculate the output S of the model out .

[0177] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0178] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for predicting power spectrum of a multi-span optical communication system based on a gray box model, characterized in that: include: Step S1: Measure and obtain multiple sets of spectrum data, and obtain the model parameters of the power spectrum by solving equations; Step S2: constructing a digital twin model of the power spectrum of a multi-span optical communication system using the obtained model parameters; Step S3: Use the constructed digital twin model to predict the power spectrum and output the prediction results.

2. The method for predicting power spectrum of a multi-span optical communication system based on a gray box model according to claim 1, characterized in that: The step S1 comprises: Step S1.1: Based on the output power spectrum S of the erbium-doped fiber amplifier out The change in ΔS out Get the model parameter dG: dG=ΔS out ; Step S1.2: Input the total power P to the optical fiber in，fiber Two different groups of optical fiber input power spectra S in and S′ in , calculate the attenuation spectrum Loss and Loss′ of the fiber with self-excited Raman scattering effect; Step S1.3: Calculate model parameter k ISRS : var[(G EDFA (k ISRS (m)-G' EDFA (k ISRS ,m')) / dG]=0, Among them, k ISRS The physical meaning is 10lge·C r L eff / δ,C r is the normalized Raman gain spectrum slope, L eff is the effective length, δ is the insertion loss, and m is the value of the known P in，EDFA Next, the shift of the ISRS gain spectrum is obtained; Step S1.4: Calculate the gain spectrum G of the erbium-doped fiber amplifier EDFA and G' EDFA : G EDFA =S out -(S in -Loss(k ISRS ,m)) Step S1.5: Calculate model parameter G0: G0=mean(G EDFA (k ISRS ,m),G’ EDFA (k ISRS ,m’)); Where dG and G0 represent the wavelength dependence of the erbium-doped fiber amplifier gain spectrum.

3. The method for predicting power spectrum of a multi-span optical communication system based on a gray box model according to claim 2, characterized in that: The calculation process of step S1.2 includes: Loss(k ISRS ,m)=αL-G ISRS (k ISRS ,m) G ISRS (k ISRS ,m)=k ISRS P in，fiber λ+m Where P in，fiber is the total fiber power, α is the fiber attenuation spectrum, L is the fiber length, and λ is the wavelength vector; P in，EDFA is the total input power of the erbium-doped fiber amplifier.

4. The method for predicting power spectrum of a multi-span optical communication system based on a gray box model according to claim 3, characterized in that: The calculation process of the digital twin model includes: G ISRS (k ISRS ,m)=k ISRS P in，fiber λ+m G EDFA (x)=dG·x+G0 Where x represents the input power dependence of the erbium-doped fiber amplifier gain spectrum, S in,EDFA is the input power spectrum of the erbium-doped fiber amplifier, G tot is the total gain of the erbium-doped fiber amplifier.

5. The method for predicting power spectrum of a multi-span optical communication system based on a gray box model according to claim 4, characterized in that: The step S3 comprises the following sub-steps: Step S3.1: S in and P in，EDFA Substitute the following formula to obtain the variable m: Step S3.2: Calculate Loss: Loss(k ISRS ,m)=αL-G ISRS (k ISRS ,m); according to S in，EDFA =S in -Loss Calculate S in，EDFA ; Step S3.3: G tot and S in，EDFA Substitute the following formula to find the variable x: Calculate G using the following formula EDFA : G EDFA (x)=dG·x+G0; Step S3.4: Use S out =S in，EDFA +G EDFA Calculate the output S of the model out .

6. A power spectrum prediction system for a multi-span optical communication system based on a gray box model, characterized in that: include: Module M1: Measure multiple sets of spectrum data and obtain the model parameters of the power spectrum by solving equations; Module M2: Use the obtained model parameters to construct a digital twin model of the power spectrum of a multi-span optical communication system; Module M3: Use the constructed digital twin model to predict the power spectrum and output the prediction results.

7. The power spectrum prediction system for multi-span optical communication system based on gray box model according to claim 6, characterized in that: The module M1 comprises: Module M1.1: Based on the output power spectrum S of the erbium-doped fiber amplifier out The change in ΔS out Get the model parameter dG: dG=ΔS out ; Module M1.2: Total optical fiber input power P in，fiber Two different groups of optical fiber input power spectra S in and S′ in , calculate the attenuation spectrum Loss and Loss′ of the optical fiber with self-excited Raman scattering effect; Module M1.3: Calculation of model parameter k ISRS : var[(G EDFA (k ISRS (m)-G′ EDFA (k ISRS ,m')) / dG]=0 Among them, k ISRS The physical meaning is 10lge·C r L eff / δ,C r is the normalized Raman gain spectrum slope, L eff is the effective length, δ is the insertion loss, and m is the value of the known P in，EDFA Next, the shift of the ISRS gain spectrum is obtained; Module M1.4: Calculate the gain spectrum G of the erbium-doped fiber amplifier EDFA and G' EDFA : G EDFA =S out -(S in -Loss(k ISRS ,m)) Module M1.5: Calculation of model parameters G0: G0=mean(G EDFA (k ISRS ,m),G’ EDFA (k ISRS ,m’)); Where dG and G0 represent the wavelength dependence of the erbium-doped fiber amplifier gain spectrum.

8. The power spectrum prediction system for multi-span optical communication system based on gray box model according to claim 7, characterized in that: The calculation process of the module M1.2 includes: Loss(k ISRS ,m)=αL-G ISRS (k ISRS ,m) G ISRS (k ISRS ,m)=k ISRS P in，fiber λ+m Where P in，fiber is the total fiber power, α is the fiber attenuation spectrum, L is the fiber length, and λ is the wavelength vector; P in，EDFA is the total input power of the erbium-doped fiber amplifier.

9. The power spectrum prediction system for multi-span optical communication system based on gray box model according to claim 8, characterized in that: The calculation process of the digital twin model includes: G ISRS (k ISRS ,m)=k ISRS P in，fiber λ+m G EDFA (x)=dG·x+G0 Where x represents the input power dependence of the erbium-doped fiber amplifier gain spectrum, S in,EDFA is the input power spectrum of the erbium-doped fiber amplifier, G tot is the total gain of the erbium-doped fiber amplifier.

10. The power spectrum prediction system for multi-span optical communication system based on gray box model according to claim 9, characterized in that: The module M3 includes the following submodules: Module M3.1: S in and P in,EDFA Substitute the following formula to obtain the variable m: Module M3.2: Calculate Loss: Loss(k ISRS ,m)=αL-G ISRS (k ISRS ,m); according to S in,EDFA =S in -Loss Calculate S in,EDFA ; Module M3.3: G tot and S in,EDFA Substitute the following formula to find the variable x: Calculate G using the following formula EDFA : G EDFA (x)=dG·x+G0; Module M3.4: Use S out =S in,EDFA +G EDFA Calculate the output S of the model out .

Images