Operation and maintenance IoT maintenance method and system for intelligent optical transmission supporting equipment

Through the optical nerve model processing time period operation parameters and operation and maintenance data, the problem of low operation and maintenance efficiency of intelligent optical transmission supporting equipment is solved, efficient fault location and data transmission are achieved, and the reliability of the equipment is ensured.

CN118678255BActive Publication Date: 2024-11-12HANGZHOU AOKE PHOTOELECTRIC EQUIP CO LTD +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411161317.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-12
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

The IoT operation and maintenance technology of existing intelligent optical transmission supporting equipment is inefficient in fault handling and data transmission, and cannot meet the needs of high timeliness and high responsiveness.

Method used

Optical neural model is used for operation and maintenance of the Internet of Things. By obtaining the operating parameters and operation and maintenance data during the period, setting the input values ​​and converting and weight adjustments, adjusting the model in real time to realize operation and maintenance, predicting the parameters and data for the next period, and improving operation and maintenance efficiency.

Benefits of technology

The processing speed of the optical nerve model improves the operation and maintenance efficiency of the Internet of Things, ensures the data transmission reliability and fault positioning accuracy of the intelligent optical transmission supporting equipment, and meets the needs of efficient operation and maintenance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118678255B_ABST
    Figure CN118678255B_ABST
Patent Text Reader

Abstract

The present invention discloses an operation and maintenance Internet of Things maintenance method and system for intelligent optical transmission supporting equipment, belonging to the fields of optical network technology and operation and maintenance technology. The present invention realizes operation and maintenance through an optical neural model, fully utilizing the processing speed of the optical neural model, thereby further improving the operation and maintenance efficiency of the Internet of Things, thereby meeting the data transmission requirements of the intelligent optical transmission supporting equipment and ensuring reliability. Through time period operation parameters, the operation and maintenance of the Internet of Things is realized, while ensuring that the fault can be accurately located, and network maintenance is realized, thereby further meeting the data transmission requirements of the intelligent optical transmission supporting equipment and ensuring reliability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of optical network technology and operation and maintenance technology, and in particular to an operation and maintenance Internet of Things maintenance method and system for intelligent optical transmission supporting equipment. Background Art

[0002] As the number of Internet users increases, network construction and infrastructure capacity expansion will continue to increase, and optical transmission products will be used more and more.

[0003] Optical Transport Network (OTN) is a transmission network that organizes the network at the optical layer based on wavelength division multiplexing technology. It is the next-generation backbone transmission network. Through a new generation of digital transmission system and optical transmission system specified by a series of ITU-T standards such as G.872, G.709, and G.798, OTN network will solve the problems of traditional WDM network, such as poor wavelength / sub-wavelength service scheduling capability, weak networking capability, and weak protection capability.

[0004] The operation and maintenance method of the intelligent optical transmission supporting equipment in the prior art is to configure the Internet of Things for the intelligent optical transmission supporting equipment to achieve its remote maintenance and operation and maintenance.

[0005] Those skilled in the art have discovered that the above-mentioned method also has operation and maintenance problems of the Internet of Things. Since the fault handling and fault discovery of intelligent optical transmission supporting equipment have high timeliness requirements, and the data volume is determined, the response speed and transmission speed of the Internet of Things are required to be high. The existing Internet of Things operation and maintenance technology is inefficient in terms of network structure, response fault and transmission fault discovery and processing, and thus does not meet the requirements of intelligent optical transmission supporting equipment. Summary of the invention

[0006] In order to solve the problems of the prior art, the embodiment of the present invention provides an operation and maintenance Internet of Things maintenance method and system for intelligent optical transmission supporting equipment, including:

[0007] On the one hand, a maintenance method for operation and maintenance of an intelligent optical transmission supporting device is provided, the method comprising:

[0008] Obtaining time period operation parameters of the operation and maintenance Internet of Things, wherein the time period operation parameters at least include idle time period parameters and busy time period parameters;

[0009] Respectively obtaining first operation and maintenance data corresponding to the idle period parameter and second operation and maintenance data corresponding to the busy period parameter;

[0010] Setting the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data as input values ​​of the optical neural model;

[0011] According to the conversion model, the input value of the optical neural model is converted; the optical neural model is an operation and maintenance Internet of Things maintenance model;

[0012] respectively setting weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data;

[0013] According to the adjustment model, adjusting the input value of the optical neural model in real time;

[0014] The input value of the adjusted optical neural model is input into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things.

[0015] Optionally, converting the input value of the optical neural model into the input value of the optical neural model according to the conversion model includes: acquiring image parameters corresponding to the optical neural model;

[0016] Calculating, according to the image parameters, a convergence function and a number of input values ​​corresponding to the optical neural model composed of the optical neural model;

[0017] Setting the conversion model according to the convergence function and the number of input values;

[0018] The input value of the optical neural model is converted according to the conversion model.

[0019] Optionally, the respectively setting of weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data includes:

[0020] Set up the initial weight adjustment model;

[0021] setting an expected value according to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data;

[0022] According to the expected value, the initial weight adjustment model is adjusted.

[0023] Optionally, adjusting the input value of the optical neural model in real time according to the adjustment model includes:

[0024] Acquire environmental noise;

[0025] Correcting the adjustment model in real time according to the environmental noise;

[0026] According to the corrected adjustment model, the input value of the optical neural model is adjusted in real time.

[0027] Optionally, adjusting the input value of the optical neural model in real time according to the corrected adjustment model includes:

[0028] Acquiring the output value of the optical neural model in real time;

[0029] According to the output value, setting a feedback function;

[0030] Based on the feedback function, the input value of the optical neural model is adjusted in real time according to the corrected adjustment model. Optionally, before the input value of the adjusted optical neural model is input into the optical neural model to implement the maintenance of the operation and maintenance Internet of Things, the method further includes:

[0031] Set the convergence function;

[0032] The optical neural model is trained according to the convergence function.

[0033] Optionally, inputting the adjusted input value of the optical neural model into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things includes:

[0034] inputting the adjusted input value of the optical neural model into the optical neural model;

[0035] According to the optical neural model, predict the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data;

[0036] The operation and maintenance Internet of Things is maintained according to the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data.

[0037] Optionally, predicting the next idle period parameter and the next busy period parameter according to the optical neural model includes:

[0038] Get historical idle period parameters and historical busy period parameters;

[0039] Based on the time period operation parameters, filter out an operation parameter input data set from the historical idle time period parameters and the historical busy time period parameters;

[0040] According to the operating parameter input data set, based on the optical neural model, the next idle period parameter and the busy period parameter are predicted.

[0041] Optionally, predicting the next first operation and maintenance data and the next second operation and maintenance data according to the optical neural model includes:

[0042] Obtain historical operation and maintenance data;

[0043] Based on the time period operation parameters, an operation and maintenance data input data set is filtered out from the operation and maintenance data; and based on the optical neural model, the next first operation and maintenance data and the second operation and maintenance data are predicted according to the operation and maintenance data input data set.

[0044] On the other hand, an operation and maintenance IoT maintenance system for intelligent optical transmission supporting equipment is also provided, the system comprising a plurality of intelligent optical transmission supporting equipment, a plurality of IoT components and a plurality of processing devices, wherein the plurality of processing devices are used to:

[0045] Obtaining time period operation parameters of the operation and maintenance Internet of Things, wherein the time period operation parameters at least include idle time period parameters and busy time period parameters;

[0046] Respectively obtaining first operation and maintenance data corresponding to the idle period parameter and second operation and maintenance data corresponding to the busy period parameter;

[0047] Setting the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data as input values ​​of the optical neural model;

[0048] According to the conversion model, the input value of the optical neural model is converted into the input value of the optical neural model; the optical neural model is an operation and maintenance Internet of Things maintenance model;

[0049] respectively setting weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data;

[0050] According to the adjustment model, adjusting the input value of the optical neural model in real time;

[0051] The input value of the adjusted optical neural model is input into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things.

[0052] The present invention has at least the following beneficial effects:

[0053] 1. Operation and maintenance are realized through the optical neural model, and the processing speed of the optical neural model is fully utilized, thereby further improving the operation and maintenance efficiency of the Internet of Things, thereby meeting the data transmission requirements of intelligent optical transmission supporting equipment and ensuring reliability.

[0054] 2. Through the idle period parameters, the operation and maintenance of the Internet of Things is realized, ensuring that the fault can be accurately located and the network maintenance can be realized, thereby further meeting the data transmission requirements of the intelligent optical transmission supporting equipment and ensuring reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0056] Figure 1 A schematic diagram of a maintenance method for operation and maintenance of an intelligent optical transmission supporting equipment provided by an embodiment of the present invention;

[0057] Figure 2 A schematic diagram of a maintenance method for operation and maintenance of an intelligent optical transmission supporting equipment provided by an embodiment of the present invention;

[0058] Figure 3 A schematic diagram of a maintenance method for operation and maintenance of an intelligent optical transmission supporting equipment provided by an embodiment of the present invention;

[0059] Figure 4 A schematic diagram of a flow chart of an operation and maintenance IoT maintenance method for intelligent optical transmission supporting equipment provided by an embodiment of the present invention;

[0060] Figure 5 A schematic diagram of an operation and maintenance Internet of Things maintenance system for intelligent optical transmission supporting equipment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0062] Reference Figure 1 As shown, a maintenance method for operation and maintenance of an intelligent optical transmission supporting device is provided, and the method includes:

[0063] 101. Obtaining time period operation parameters of the operation and maintenance Internet of Things, where the time period operation parameters at least include idle time period parameters and busy time period parameters;

[0064] Specifically, the idle period parameters described in the embodiment of the present invention mainly include:

[0065] In practical applications, the idle period parameters include: the number of available IoT nodes, data transmission delay, and data transmission volume;

[0066] In the embodiment of the present invention, the idle period and the busy period are defined as follows:

[0067] Set the statistical period. During the statistical period:

[0068] When the data transmission volume is less than or equal to the threshold A, it is an idle period;

[0069] When the data transmission volume is greater than A, it is a busy period;

[0070] The operation and maintenance Internet of Things described in the embodiment of the present invention, during use, corresponds to the above-mentioned idle period parameters.

[0071] 102. Obtain first operation and maintenance data corresponding to the idle period parameter and second operation and maintenance data corresponding to the busy period parameter respectively;

[0072] 103. Setting the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data as input values ​​of the optical neural model;

[0073] 104. According to the conversion model, the input value of the optical neural model is converted; the optical neural model is an operation and maintenance Internet of Things maintenance model;

[0074] 105. Set weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data respectively;

[0075] 106. According to the adjustment model, adjust the input value of the optical neural model in real time;

[0076] 107. Input the adjusted input value of the optical neural model into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things.

[0077] Optional, see Figure 2 As shown, step 104 converts the input value of the optical neural model into the input value of the optical neural model according to the conversion model, including:

[0078] 201. Obtain image parameters corresponding to the optical neural model;

[0079] 202. Calculate, according to the image parameters, a convergence function and a number of input values ​​corresponding to the optical neural model composed of the optical neural model;

[0080] 203. Setting a conversion model according to a convergence function and the number of input values;

[0081] 204. According to the conversion model, the input value of the optical neural model is converted.

[0082] Specifically, the above steps can be specifically as follows:

[0083] A neural network is preset; the neural network is based on optical devices, that is, a physical model of a neural network composed of multiple optical devices, and multiple optical devices are respectively set as convolutional layers and fully connected layers; the connections between the optical devices are set as weights; (the connections between the optical devices can be set manually);

[0084] The neural network model is set as an optical neural model; wherein the optical neural model includes a convolutional layer, a fully connected layer, and an activation function layer, which are the convolutional layers, fully connected layers, and activation function layers of the optical neural model, and the convolutional layers, fully connected layers, and activation function layers of the optical neural model, which are the convolutional layers, fully connected layers, and activation function layers of the neural network, and the difference is that the nodes in the neural network are neural network operators, and the nodes in the optical neural network are optical neural operators, which are actually optical devices. It should be noted that the convergence function of the optical neural model is the same as that of the neural network model;

[0085] Prune the convolutional layers and fully connected layers of the optical neural model to obtain a pruned model;

[0086] The pruned model is quantized, and the bit rate of the weight value of the pruned model is reduced to obtain the target model.

[0087] Optionally, the neural network model is set to an optical neural model, including:

[0088] Converting neural network model operators into optical neural model operators;

[0089] Assign the weights of the neural model operators to the corresponding optical neural model operators to obtain the optical neural model;

[0090] Optionally, the convolutional layers and fully connected layers of the optical neural model are pruned to obtain a pruned model, including:

[0091] The convolution layer is pruned according to the first preset threshold and the weight value of the convolution layer to obtain a pruned convolution layer.

[0092] Based on the above optical neural model, the image parameters of the input values ​​are set, that is, the input is an image, where the pixels in the image are the input values, and the number of pixels in the image is the number of input values; the specific values ​​of the idle period parameters, the first operation and maintenance data, the busy period parameters, and the second operation and maintenance data are set as the pixels in the image; since the idle period parameters and the first operation and maintenance data are continuous multiple data in the idle period, the pixel values ​​in the image also change with time, and the same is true for the busy period. If the number of input values ​​is different from the values ​​of the idle period parameters, the first operation and maintenance data, the busy period parameters, and the second operation and maintenance data, in actual applications, the number of idle period parameters, the first operation and maintenance data, the busy period parameters, and the second operation and maintenance data is often greater than the number of input values, then for any one of the idle period parameters, the first operation and maintenance data, the busy period parameters, and the second operation and maintenance data, perform the following operations:

[0093] Set the conversion model;

[0094] According to the conversion model, the input value of the optical neural model is converted.

[0095] Taking the idle period parameters as an example, the specific parameters are:

[0096] If the difference between two consecutive values ​​in the idle period parameter is less than m, the average value is taken; finally, the number of converted values ​​is obtained. If the number of values ​​is less than or equal to the number of input values, the conversion is completed. If the number of values ​​is greater than the number of input values, the execution continues:

[0097] If the difference between three consecutive values ​​in the idle period parameter is less than m, take the average value; finally, get the number of converted values. If the number of values ​​is less than or equal to the number of input values, the conversion is completed. If the number of values ​​is greater than the number of input values, continue to perform the processing between the three consecutive values ​​until the number of values ​​is less than or equal to the number of input values.

[0098] Optionally, step 105 of respectively setting weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data includes:

[0099] Set up the initial weight adjustment model;

[0100] Setting an expected value according to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data;

[0101] According to the expected value, adjust the initial weights to adjust the model.

[0102] Optional, see Figure 3 As shown, step 106 adjusts the input value of the optical neural model in real time according to the adjustment model, including:

[0103] 301. Obtain environmental noise;

[0104] 302. Correct the adjustment model in real time according to the environmental noise;

[0105] 303. According to the corrected adjustment model, the input value of the optical neural model is adjusted in real time.

[0106] The process can be specifically as follows:

[0107] Get descriptive data of environmental noise.

[0108] The description data of the environmental noise is added to the training data set of the optical neural model, and the optical neural model is optimized in combination with gradient backpropagation until the training end conditions are met.

[0109] Due to the particularity of the optical neural model, noise enhancement of the optical neural model is achieved. Noise information refers to the interference to the correct classification of the optical neural model. If the noise is too large, the correct classification cannot be made. The noise information of this embodiment includes hardware error noise and / or imaging error noise. Hardware error noise refers to the discrepancy between the actual value of the spatial geometric relationship of the hardware and the original theoretical design value. Imaging error noise refers to the noise caused by external environmental factors during the imaging process. During the model training process, one of these two types of noise can be selected, or they can be included at the same time.

[0110] After the noise information is introduced during the training process, the model training is based on the model under the noise for gradient backpropagation optimization. Each training can introduce one or more noises, and then repeat the training tens of thousands of times.

[0111] Optionally, step 303 of adjusting the input value of the optical neural model in real time according to the corrected adjustment model includes:

[0112] Obtain the output value of the optical neural model in real time;

[0113] A feedback function is set according to the output value; based on the feedback function, the input value of the optical neural model is adjusted in real time according to the corrected adjustment model.

[0114] Optionally, before step 107 inputs the adjusted input value of the optical neural model into the optical neural model to implement the maintenance of the operation and maintenance of the Internet of Things, the method further includes

[0115] Set the convergence function;

[0116] According to the convergence function, the optical neural model is trained.

[0117] Optional, see Figure 4 As shown, step 107 inputs the adjusted input value of the optical neural model into the optical neural model to implement the maintenance of the operation and maintenance of the Internet of Things, including:

[0118] 401. Inputting the adjusted input value of the optical neural model into the optical neural model;

[0119] 402. Predict the next idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data according to the optical neural model;

[0120] 403. Maintain the operation and maintenance Internet of Things according to the next idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data.

[0121] Optionally, in step 402, predicting the next idle period parameter and the next busy period parameter according to the optical neural model includes:

[0122] Get historical idle period parameters and historical busy period parameters;

[0123] Based on the time period operation parameters, the operation parameter input data set is filtered out from the historical idle time period parameters and the historical busy time period parameters;

[0124] According to the operating parameter input data set, the next idle period parameters and busy period parameters are predicted based on the optical neural model.

[0125] Optionally, step 402 predicts the next first operation and maintenance data and the second operation and maintenance data according to the optical neural model, including: obtaining historical operation and maintenance data;

[0126] Based on the time period operation parameters, filter out the operation and maintenance data input data set from the operation and maintenance data;

[0127] According to the operation and maintenance data input data set, based on the optical neural model, the next first operation and maintenance data and the second operation and maintenance data are predicted.

[0128] The above process can be specifically described as follows:

[0129] Set the basis of the optical neural model to the BP optical neural model:

[0130] The BP optical neural model was optimized using genetic algorithm and the GA-BP optical neural model was constructed.

[0131] First, individual encoding is performed. Each individual contains all the weight and threshold information of the optical neural model. The population size is set and the population is initialized. The fitness value of each individual in the initial population is calculated through the selected fitness function, and the next generation population is obtained by selection, crossover and mutation operations.

[0132] Calculate the fitness value of each individual again, and iterate continuously to finally get the optimal individual. Decode the optimal individual to get the weight and threshold, assign them to the BP optical neural model, calculate the error through forward propagation, calculate the gradient through back propagation, update the weight and threshold along the negative gradient direction until the target error is reached, terminate the training, and save the optical neural model.

[0133] Select the population size.

[0134] The inverse of the mean square error function MSE of the BP optical neural model is taken as the fitness function of the genetic algorithm.

[0135] This embodiment adopts the roulette wheel method for selection operation, that is, the probability of chromosome individual j (j=1, 2, ..., N) being selected is proportional to the fitness value. First, the fitness value of each individual is calculated, and then the probability Pj of each individual being inherited to the next generation is calculated;

[0136] The real number crossover method is used for crossover operation. Two individuals are selected from the population. One or more crossover points are selected according to the crossover probability to exchange some genes of the individuals to generate new individuals. The uth chromosome Ru and the vth chromosome Rv are crossed at the lth position. The method is as follows:

[0137] Randomly change the genes at certain positions of the individual to generate new individuals, and the mutation probability is set to 0.05;

[0138] A new population is obtained by cycling the above genetic operations such as selection, crossover and mutation. When the fitness value reaches the target value or the number of iterations reaches the set value, the cycle ends and the optimal individual is generated. The maximum number of iterations is set to 50.

[0139] The optimal individual is decoded, and the obtained optimal optical neural model weights and thresholds are assigned to the BP optical neural model.

[0140] The operation parameter input data set and the operation and maintenance data input data set are respectively input into the BP optical neural model according to the number of input values, and finally the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data are obtained.

[0141] In addition, the operation and maintenance strategy described in the embodiment of the present invention includes a technician determining whether to adopt an on-site maintenance or remote maintenance solution after obtaining the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data.

[0142] Reference Figure 5 As shown, an operation and maintenance IoT maintenance system for intelligent optical transmission supporting equipment is also provided, the system includes multiple intelligent optical transmission supporting equipment, multiple IoT components and multiple processing devices, the multiple IoT components are configured on the intelligent optical transmission supporting equipment, wherein the multiple processing devices are used to:

[0143] Obtaining time period operation parameters of the operation and maintenance Internet of Things, the time period operation parameters at least including idle time period parameters and busy time period parameters; respectively obtaining first operation and maintenance data corresponding to the idle time period parameters, and second operation and maintenance data corresponding to the busy time period parameters;

[0144] Setting the idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data as input values ​​of the optical neural model;

[0145] According to the conversion model, the input value of the optical neural model is converted into the input value of the optical neural model; the optical neural model is the maintenance model of the operation and maintenance Internet of Things;

[0146] respectively setting weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data;

[0147] According to the adjustment model, the input value of the optical neural model is adjusted in real time;

[0148] The input value of the adjusted optical neural model is input into the optical neural model to realize the maintenance of the operation and maintenance of the Internet of Things.

[0149] Optionally, according to the conversion model, converting the input value of the optical neural model into the input value of the optical neural model includes:

[0150] Obtain image parameters corresponding to the optical neural model;

[0151] According to the image parameters, a convergence function and the number of input values ​​corresponding to the optical neural model composed of the optical neural model are calculated;

[0152] Set the conversion model according to the convergence function and the number of input values;

[0153] According to the conversion model, the input value of the optical neural model is converted into the input value of the optical neural model.

[0154] Optionally, respectively setting weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data includes:

[0155] Setting an initial weight adjustment model; setting an expected value according to the idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data;

[0156] According to the expected value, adjust the initial weights to adjust the model.

[0157] Optionally, according to the adjustment model, adjusting the input value of the optical neural model in real time includes:

[0158] Acquire environmental noise;

[0159] According to the environmental noise, the adjustment model is corrected in real time;

[0160] According to the corrected adjustment model, the input value of the optical neural model is adjusted in real time.

[0161] Optionally, according to the corrected adjustment model, adjusting the input value of the optical neural model in real time includes:

[0162] Obtain the output value of the optical neural model in real time;

[0163] According to the output value, set the feedback function;

[0164] Based on the feedback function, the input value of the optical neural model is adjusted in real time according to the corrected adjustment model.

[0165] Optionally, before inputting the adjusted input value of the optical neural model into the optical neural model to implement the maintenance of the operation and maintenance of the Internet of Things, the method also includes

[0166] Set the convergence function;

[0167] According to the convergence function, the optical neural model is trained.

[0168] Optionally, the input value of the adjusted optical neural model is input into the optical neural model to achieve the maintenance of the operation and maintenance of the Internet of Things including:

[0169] Inputting the adjusted input value of the optical neural model into the optical neural model;

[0170] According to the optical neural model, the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data are predicted; according to the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data, the operation and maintenance Internet of Things is maintained.

[0171] Optionally, according to the optical neural model, predicting the next idle period parameter and the next busy period parameter includes:

[0172] Get historical idle period parameters and historical busy period parameters;

[0173] Based on the time period operation parameters, the operation parameter input data set is filtered out from the historical idle time period parameters and the historical busy time period parameters;

[0174] According to the operating parameter input data set, the next idle period parameters and busy period parameters are predicted based on the optical neural model.

[0175] Optionally, according to the optical neural model, predicting the next first operation and maintenance data and the second operation and maintenance data includes:

[0176] Obtain historical operation and maintenance data;

[0177] Based on the time period operation parameters, filter out the operation and maintenance data input data set from the operation and maintenance data;

[0178] According to the operation and maintenance data input data set, based on the optical neural model, the next first operation and maintenance data and the second operation and maintenance data are predicted.

[0179] The above specific embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0180] The technical features of the above embodiments may be arbitrarily combined (as long as there is no contradiction in the combination of these technical features). To make the description concise, not all possible combinations of the technical features in the above embodiments are described; these embodiments that are not explicitly written should also be considered to be within the scope of this specification.

[0181] The present invention is described in more detail through general description and specific embodiments. It should be noted that, without departing from the concept of the present invention, it is obvious that several variations and improvements can be made to these specific embodiments, which all belong to the protection scope of this application. Therefore, the protection scope of the patent of this application shall be subject to the attached claims.

[0182] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A maintenance method for the operation and maintenance of an intelligent optical transmission supporting equipment, characterized in that: The method comprises: Obtaining time period operation parameters of the operation and maintenance Internet of Things, wherein the time period operation parameters at least include idle time period parameters and busy time period parameters; Respectively obtaining first operation and maintenance data corresponding to the idle period parameter and second operation and maintenance data corresponding to the busy period parameter; Setting the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data as input values ​​of the optical neural model; According to the conversion model, the input value of the optical neural model is converted; the optical neural model is an operation and maintenance Internet of Things maintenance model; respectively setting weight adjustment models corresponding to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data; According to the adjustment model, adjusting the input value of the optical neural model in real time; Inputting the adjusted input value of the optical neural model into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things; The converting the input value of the optical neural model according to the conversion model includes: Obtain image parameters corresponding to the optical neural model; Calculating the convergence function and the number of input values ​​corresponding to the optical neural model according to the image parameters; Setting the conversion model according to the convergence function and the number of input values; According to the conversion model, converting the input value of the optical neural model; The step of inputting the adjusted input value of the optical neural model into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things includes: inputting the adjusted input value of the optical neural model into the optical neural model; According to the optical neural model, predict the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data; The operation and maintenance Internet of Things is maintained according to the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data.

2. The method according to claim 1, characterized in that The weight adjustment models respectively set to correspond to the idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data include: Set up the initial weight adjustment model; setting an expected value according to the idle period parameter, the first operation and maintenance data, the busy period parameter, and the second operation and maintenance data; According to the expected value, the initial weight adjustment model is adjusted.

3. The method according to claim 2, characterized in that The step of adjusting the input value of the optical neural model in real time according to the adjustment model includes: Acquire environmental noise; Correcting the adjustment model in real time according to the environmental noise; According to the corrected adjustment model, the input value of the optical neural model is adjusted in real time.

4. The method according to claim 3, characterized in that The step of adjusting the input value of the optical neural model in real time according to the corrected adjustment model includes: Acquiring the output value of the optical neural model in real time; According to the output value, setting a feedback function; Based on the feedback function, the input value of the optical neural model is adjusted in real time according to the corrected adjustment model.

5. The method according to claim 4, characterized in that Before inputting the adjusted input value of the optical neural model into the optical neural model to implement maintenance of the operation and maintenance Internet of Things, the method further includes: Set the convergence function; The optical neural model is trained according to the convergence function.

6. The method according to claim 1, characterized in that According to the optical neural model, predicting the next idle period parameter and the next busy period parameter includes: Get historical idle period parameters and historical busy period parameters; Based on the time period operation parameters, filter out an operation parameter input data set from the historical idle time period parameters and the historical busy time period parameters; According to the operating parameter input data set, based on the optical neural model, the next idle period parameter and the busy period parameter are predicted.

7. The method according to claim 1, characterized in that Predicting the next first operation and maintenance data and the next second operation and maintenance data according to the optical neural model includes: Obtain historical operation and maintenance data; Based on the time period operation parameters, filter out an operation and maintenance data input data set from the operation and maintenance data; According to the operation and maintenance data input data set, based on the optical neural model, the next first operation and maintenance data and the second operation and maintenance data are predicted.

8. An operation and maintenance IoT system for intelligent optical transmission supporting equipment, characterized in that: The system includes a plurality of intelligent optical transmission supporting devices, a plurality of Internet of Things components and a plurality of processing devices, wherein the plurality of processing devices are used to: Obtaining time period operation parameters of the operation and maintenance Internet of Things, wherein the time period operation parameters at least include idle time period parameters and busy time period parameters; Respectively obtaining first operation and maintenance data corresponding to the idle period parameter and second operation and maintenance data corresponding to the busy period parameter; Setting the period operation parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data as input values ​​of the optical neural model; According to the conversion model, the input value of the optical neural model is converted into the input value of the optical neural model; the optical neural model is an operation and maintenance Internet of Things maintenance model; respectively setting weight adjustment models corresponding to the time period operation parameters, the first operation and maintenance data, the busy period parameters, and the second operation and maintenance data; According to the adjustment model, adjusting the input value of the optical neural model in real time; Inputting the adjusted input value of the optical neural model into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things; The converting the input value of the optical neural model according to the conversion model includes: Obtain image parameters corresponding to the optical neural model; Calculating the convergence function and the number of input values ​​corresponding to the optical neural model according to the image parameters; Setting the conversion model according to the convergence function and the number of input values; According to the conversion model, converting the input value of the optical neural model; The step of inputting the adjusted input value of the optical neural model into the optical neural model to achieve maintenance of the operation and maintenance Internet of Things includes: inputting the adjusted input value of the optical neural model into the optical neural model; According to the optical neural model, predict the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data; The operation and maintenance Internet of Things is maintained according to the next idle period parameter, the first operation and maintenance data, the busy period parameter and the second operation and maintenance data.

Citation Information

Cited By

  • IoT equipment maintenance method and system for edge computing big data processing

    CN119232580B