Cable cross-layer optimization design method based on radius time sequence circulation network

Abstract

According to a cable cross-layer optimization design method based on a radius time sequence circulation network, a cross-layer structure and design of a cable generally cover a plurality of material and structure layers from a core to an outer layer, and the layer is subjected to stress and abrasion of different types and strengths in practical application. Therefore, a brand new cable design and optimization thought is explored in the application of the invention. According to the algorithm, the direction from the circle center to the radius of the outermost layer of the section of the cable is regarded as a time sequence axis, materials and design parameters on different radiuses are regarded as variable states on different time points along the axis, and a recurrent neural network is used for analysis and optimization. The method aims to realize cross-layer fine management and intelligent optimization design of the cable, so that the design of the flexible cable is optimized to the greatest extent.

Description

Technical Field

[0001] The invention relates to the field of metal packaging, and in particular to a cable cross-layer optimization design method based on a radius time series cyclic network. Background Art

[0002] Metal packaging is a packaging container made of metal materials and is widely used in various fields such as food, beverages, medicines, and cosmetics. Metal packaging is usually made of materials such as iron, tin, and aluminum. However, metal packaging also has some limitations and considerations. Since metal packaging is a sealed device, it is difficult to check whether the items in the package are damaged. Traditional metal packaging inspection mainly relies on manual visual inspection and some auxiliary means, such as manual touch, knocking sound, etc. However, this method has great defects. On the one hand, it relies on manual operation, and the human eye may not be able to accurately identify or miss small defects or hidden defects. In addition, factors such as fatigue and lack of concentration will also affect the accuracy and consistency of detection. Therefore, there is an urgent need for a solution that can not only inspect without destroying the packaging, but also ensure the accuracy of the detection. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes a cable cross-layer optimization design method based on a radius time series recurrent network, that is, a cable cross-layer optimization design method based on a radius time series recurrent network. The method comprehensively considers the key steps such as the current carrying capacity of the cable, the construction of the material model, the preprocessing of the cable data set, and the analysis and design based on the time series recurrent neural network (RNN), so as to realize the fine management and intelligent optimization design of the cable cross-layer structure.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] The cable cross-layer optimization design method based on radius time series recurrent network includes the following steps:

[0006] 1) Calculation of cable current carrying capacity:

[0007] The current carrying capacity of a cable takes into account factors such as conductor diameter, current carrying coefficient, resistance, and shielding layer coefficient, and is one of the important indicators for cable design and selection. Generally, the current carrying capacity of a cable is affected by factors such as the cross-sectional area of ​​the conductor, heat dissipation conditions, and ambient temperature. The cable carrying capacity determines the maximum current load that the cable can withstand under normal working conditions, which is directly related to the reliability and safety of the cable in practical applications, and also provides a guarantee for the stable operation of the power transmission system.

[0008] 2) Cable material model construction:

[0009] The construction of cable material model is a crucial link in the cable manufacturing process, which includes the material selection decisions of key components such as conductors, insulators, and sheaths. The conductivity of the conductor material, the insulation properties of the insulator material, and the environmental resistance of the sheath material will be considered in the cable construction to ensure that the cable has excellent electrical performance and transmission stability in practical applications.

[0010] 3) Cable data preprocessing:

[0011] Preprocessing the cable data set ensures the accuracy of analysis and optimization. The preprocessing process includes data cleaning (eliminating missing values), normalization and standardization, which can ensure the consistency and reliability of the data. At the same time, feature selection and full dimensionality reduction can eliminate redundant information and improve the efficiency of model training. The introduction of cross-validation technology also helps to enhance the generalization ability of the evaluation model and provide high-quality and reliable data sets for subsequent time series network material performance analysis and optimization design.

[0012] 4) Analysis and design based on time series recurrent network (RNN)

[0013] The analysis and design based on time series recurrent network is an innovative method that regards the characteristics of cable materials as time-varying sequence information and constructs a recurrent neural network (RNN) model to achieve fine management and intelligent optimization design of cable cross-layer structures. This method comprehensively considers the dynamic changes of cables under different stress and wear conditions, provides a new perspective for the selection and optimization of cable materials, and also provides strong support for the stable operation of power transmission systems.

[0014] As a further improvement of the present invention, the calculation formula of the cable current carrying capacity in step 1) is expressed as:

[0015] Among them, the calculation formula for cable current carrying capacity is:

[0016]

[0017] Among them, s is the cross-sectional area of ​​the conductor (in square millimeters), t is the maximum allowable temperature rise (in degrees Celsius), θ is the ambient temperature of the conductor (in degrees Celsius), k is a coefficient related to the conductor material and heat dissipation conditions, D is the cable conductor diameter (unit: meter), λ is the cable shielding layer coefficient. The introduction of λ takes into account the shielding layer's carrying capacity for the cable, and R is the cable resistance (unit: ohm / meter).

[0018] As a further improvement of the present invention, the cable material model construction formula in step 2) is expressed as:

[0019] Among them, the cable material model construction formula is:

[0020]

[0021] Among them, E is Young's modulus, γ is the critical crack surface energy, V is Poisson's ratio, C is the experimental constant of the relevant material, ΔK is the stress intensity factor at the crack front, and n is the loading parameter of the relevant experimental conditions.

[0022] As a further improvement of the present invention, the cable data set preprocessing formula in step 3) is expressed as:

[0023] Among them, the cable data set preprocessing formula is:

[0024] S a =[f(C,I,S,P,F,M)-f min (C,I,S,P,F,M)] / [f max (C,I,S,P,F,M)-f min (C,I,S,P,F,M)] (3)

[0025] Among them, C is the conductor material, I is the insulation material, S is the shielding material, P is the sheath material, F is the fire-resistant material, M is the mechanical strength, f is a unified dimension function, and f min is the smallest dimension function, f max is the largest dimension function. Here S a is an average function, which acts as a combination of cable material characteristics.

[0026] As a further improvement of the present invention, the recurrent neural network (RNN) analysis formula in step 4) is expressed as:

[0027] Among them, the recurrent neural network (RNN) analysis formula is:

[0028] y t =σW hy (σ(W ih X t +b ih +W hh h t-1 +b hh )) (4)

[0029] Among them, W ih , W hh and W hy are all weight matrices, b ih , b hh and b hy are all bias terms, σ is the activation function, X t is the input vector value at time step t, h t-1 is the hidden state at time step t-1, y t is the output vector value at time step t.

[0030] This application has the following benefits:

[0031] The cable cross-layer optimization design method based on the radius time series cyclic network provided in this application can improve the current carrying capacity of the cable, and can accurately calculate the current carrying capacity of the cable by considering factors such as conductor diameter, current carrying coefficient, resistance, and shielding layer coefficient. This will help optimize the cable design so that it can withstand a larger current load under normal working conditions, thereby improving the overall performance of the cable.

[0032] The cable cross-layer optimization design method based on the radius time series cyclic network provided in this application can optimize the selection of cable materials. By constructing a cable material model, it can fully consider the material's conductivity, insulation properties, environmental resistance and other properties in the material selection decision-making process of key components such as conductors, insulators, and sheaths, thereby ensuring that the cable has excellent electrical performance and transmission stability in practical applications.

[0033] The cable cross-layer optimization design method based on the radius time series recurrent network provided in this application can improve the accuracy of data analysis and optimization, and ensure the consistency and reliability of the data by preprocessing the cable data set, including cleaning, normalization, and standardization. At the same time, the redundant information in the data is eliminated by using processing methods such as feature selection and full dimensionality reduction, thereby improving the efficiency of model training. The introduction of cross-validation technology also helps to enhance the generalization ability of the evaluation model.

[0034] The cable cross-layer optimization design method based on the radius time series recurrent network provided in this application realizes fine management and intelligent optimization design. The analysis and design method based on the time series recurrent network (RNN) regards the characteristics of the cable material as sequence information that changes over time, and the dynamic changes of the cable under different stress and wear conditions are considered in the model construction. This makes the cable design more refined, and at the same time has intelligent optimization capabilities, which improves the reliability and performance of the cable. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Attached Figure 1 According to the overall flow chart of the cable cross-layer optimization design method based on the radius time series cyclic network provided by the present application;

[0036] Attached Figure 2 Schematic diagram of a method for calculating current carrying capacity according to a cable cross-layer optimization design method based on a radius time series cyclic network provided in the present application;

[0037] Attached Figure 3 Schematic diagram of cable material model construction according to the cable cross-layer optimization design method based on radius time series cyclic network provided by this application;

[0038] Attached Figure 4 Schematic diagram of recurrent neural network (RNN) for cable cross-layer optimization design method based on radius time series recurrent network provided in the present application. DETAILED DESCRIPTION

[0039] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0040] like Figure 1 Shown is the overall flow chart of the cable cross-layer optimization design method based on the radius time series recurrent network.

[0041] Step S1: Calculation of cable current carrying capacity.

[0042] like Figure 2 Shown is a schematic diagram of the current carrying capacity calculation method of the cable cross-layer optimization design method based on the radius time series cyclic network.

[0043] Correctly evaluating the current carrying capacity of cables is one of the key factors to ensure the stable operation of power transmission systems.

[0044] When designing and selecting cables, we need to consider many factors, including conductor diameter, current carrying coefficient, resistance and shielding layer coefficient. The conductor diameter determines the speed at which resistance and heat are generated when current passes through the conductor, which directly affects the cable's carrying capacity. The current carrying coefficient takes into account the load of the cable under actual working conditions, including instantaneous overload and short circuit conditions, to ensure the stability of the cable in emergency situations. In addition, the resistance of the cable is also one of the important factors affecting its current carrying capacity. The smaller the resistance, the less heat the cable generates when transmitting current, and the corresponding carrying capacity is higher. The shielding layer coefficient affects the cable's resistance to external interference and electromagnetic radiation, which is of great significance for ensuring the normal operation of the cable. In addition, ambient temperature and heat dissipation conditions are also important factors that determine the cable's current carrying capacity. High temperature environment will increase the temperature rise of the cable conductor and affect its carrying capacity. Therefore, when designing and selecting cables, it is necessary to consider the temperature conditions of the surrounding environment and the conditions of the heat dissipation facilities.

[0045] The calculation of the cable carrying capacity determines the maximum current load that the cable can withstand under normal working conditions, which will directly affect the reliability and safety of the cable in practical applications and provide a guarantee for the stable operation of the power transmission system. Figure 2 shown.

[0046] Among them, the calculation formula for cable current carrying capacity is:

[0047]

[0048] Among them, S is the cross-sectional area of ​​the conductor (in square millimeters), t is the maximum allowable temperature rise (in degrees Celsius), θ is the ambient temperature of the conductor (in degrees Celsius), k is a coefficient related to the conductor material and heat dissipation conditions, D is the cable conductor diameter (unit: meter), λ is the cable shielding layer coefficient. The introduction of λ takes into account the shielding layer's carrying capacity for the cable, and R is the cable resistance (unit: ohm / meter).

[0049] Step S2: Cable material model construction.

[0050] like Figure 3 Shown is a schematic diagram of cable material model construction based on the radius time series cyclic network for cable cross-layer optimization design method.

[0051] Cable material model construction is one of the core links in cable manufacturing, which directly affects the performance and reliability of the cable. At this stage, the materials of key components such as conductors, insulators and sheaths need to be carefully selected to ensure that the cable performs well in actual applications and has excellent electrical performance and stability.

[0052] The selection of conductor materials is crucial. The conductivity of the conductor directly determines the conduction efficiency of the cable. Conductors with high conductivity can reduce energy loss during current transmission and improve the efficiency of the cable. In addition, we also need to consider the mechanical strength of the conductor material to ensure that the cable will not be deformed or damaged during installation and use. The selection of insulator materials is also very important. The insulator needs to have good insulation properties to prevent current leakage or breakdown between conductors, thereby ensuring the safe and stable operation of the cable. In addition, the insulator also needs to have a certain degree of heat resistance to cope with the heat that may be generated under high load conditions. The selection of sheath materials should also take environmental factors into consideration. The sheath needs to have good environmental resistance and be able to resist the effects of corrosion and wear on the cable by the external environment to ensure that the cable can operate stably in various complex working environments.

[0053] Through scientific and reasonable material selection and model construction, it is possible to ensure that the cable can play the best performance in practical applications and provide reliable guarantee for the stable operation of the power transmission system, such as Figure 3 shown.

[0054] Among them, the cable material model construction formula is:

[0055]

[0056] Among them, E is Young's modulus, γ is the critical crack surface energy, V is Poisson's ratio, C is the experimental constant of the relevant material, ΔK is the stress intensity factor at the crack front, and n is the loading parameter of the relevant experimental conditions.

[0057] Step S3: cable dataset preprocessing.

[0058] Cable data set preprocessing is a key step to ensure the accuracy of analysis and optimization. At this stage, data consistency and reliability can be guaranteed by cleaning, normalizing and standardizing the data, thus laying a solid foundation for subsequent analysis and optimization. Cleaning data includes processing missing values, outliers, and duplicate data to ensure the integrity and accuracy of the data, avoid interference from invalid or misleading information during the analysis process, and ensure the credibility of the analysis results. Normalization and standardization can adjust the numerical range of different features to a similar range to avoid certain features from having too much impact on model training, while improving the convergence speed and performance of the model.

[0059] In addition, feature selection and sufficient dimensionality reduction are also important steps. By eliminating redundant information in the data set, the complexity of the model can be reduced, the training efficiency can be improved, and key features can be retained to ensure the performance of the model. Sufficient dimensionality reduction can also reduce storage and computing costs, which is particularly important when processing large-scale data.

[0060] Among them, the cable data set preprocessing formula is:

[0061] S a =[f(C,I,S,P,F,M)-f min (C,I,S,P,F,M)] / [f max (C,I,S,P,F,M)-f min (C,I,S,P,E,M)] (3)

[0062] Among them, C is the conductor material, I is the insulation material, S is the shielding material, P is the sheath material, F is the fire-resistant material, M is the mechanical strength, f is a unified dimension function, and f min is the smallest dimension function, f max is the largest dimension function. Here S a is an average function, which acts as a combination of cable material characteristics.

[0063] Step S4: Analysis and design based on time series neural network (RNN).

[0064] like Figure 4 Shown is a schematic diagram of a recurrent neural network (RNN) for a cable cross-layer optimization design method based on a radius time series recurrent network.

[0065] Analysis and design of RNN (recurrent neural network) time series networks, such as Figure 4As shown in the figure, this is an innovative method. It regards the characteristics of cable materials as sequence information that changes over time, so as to achieve fine management and intelligent optimization design of the cable cross-layer structure. This method comprehensively considers the dynamic changes of cables under different stress and wear conditions, and provides a new perspective for the selection and optimization of cable materials. By adopting the RNN network, we can effectively capture the trends and laws of cable material characteristics over time, thereby providing a more accurate basis for cable design. This method can not only be applied to material property analysis under static conditions, but also play a role in dynamic working environments, providing a comprehensive evaluation of the durability and stability of cable materials.

[0066] Among them, the recurrent neural network (RNN) analysis formula is:

[0067] y t =σW hy (σ(W ih X t +b ih +W hh h t-1 +b hh )) (4)

[0068] Among them, W ih , W hh and W hy are all weight matrices, b ih , b hh and b hy are all bias terms, σ is the activation function, X t is the input vector value at time step t, h t-1 is the hidden state at time step t-1, y t is the output vector value at time step t.

[0069] The above description is only a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent change made based on the technical essence of the present invention still falls within the scope of protection required by the present invention.

Claims

1. A cable cross-layer optimization design method based on a radius time series recurrent network, characterized in that: The following steps are involved: 1) Calculation of cable current carrying capacity: The current carrying capacity of a cable takes into account factors such as conductor diameter, current carrying coefficient, resistance, and shielding layer coefficient, and is one of the important indicators for cable design and selection. Generally, the current carrying capacity of a cable is affected by factors such as the cross-sectional area of ​​the conductor, heat dissipation conditions, and ambient temperature. The cable carrying capacity determines the maximum current load that the cable can withstand under normal working conditions, which is directly related to the reliability and safety of the cable in practical applications, and also provides a guarantee for the stable operation of the power transmission system. 2) Cable material model construction: The construction of cable material model is a crucial link in the cable manufacturing process, which includes the material selection decisions of key components such as conductors, insulators, and sheaths. The conductivity of the conductor material, the insulation properties of the insulator material, and the environmental resistance of the sheath material will be considered in the cable construction to ensure that the cable has excellent electrical performance and transmission stability in practical applications. 3) Cable data preprocessing: Preprocessing the cable data set ensures the accuracy of analysis and optimization. The preprocessing process includes data cleaning (eliminating missing values), normalization and standardization, which can ensure the consistency and reliability of the data. At the same time, feature selection and full dimensionality reduction can eliminate redundant information and improve the efficiency of model training. The introduction of cross-validation technology also helps to enhance the generalization ability of the evaluation model and provide high-quality and reliable data sets for subsequent time series network material performance analysis and optimization design. 4) Analysis and design based on time series recurrent network (RNN) The analysis and design based on time series recurrent network is an innovative method that regards the characteristics of cable materials as time-varying sequence information and constructs a recurrent neural network (RNN) model to achieve fine management and intelligent optimization design of cable cross-layer structures. This method comprehensively considers the dynamic changes of cables under different stress and wear conditions, provides a new perspective for the selection and optimization of cable materials, and also provides strong support for the stable operation of power transmission systems.

2. The cable cross-layer optimization design method based on the radius time series cyclic network according to claim 1 is characterized in that: The calculation formula for the cable current carrying capacity in step 1) is: Among them, S is the cross-sectional area of ​​the conductor (in square millimeters), t is the maximum allowable temperature rise (in degrees Celsius), θ is the ambient temperature of the conductor (in degrees Celsius), k is a coefficient related to the conductor material and heat dissipation conditions, D is the cable conductor diameter (unit: meter), λ is the cable shielding layer coefficient. The introduction of λ takes into account the shielding layer's carrying capacity for the cable, and R is the cable resistance (unit: ohm / meter).

3. The cable cross-layer optimization design method based on radius time series cyclic network according to claim 1 is characterized in that: The cable material model construction formula in step 2) is: Among them, E is Young's modulus, γ is the critical crack surface energy, V is Poisson's ratio, C is the experimental constant of the relevant material, ΔK is the stress intensity factor at the crack front, and n is the loading parameter of the relevant experimental conditions.

4. The cable cross-layer optimization design method based on radius time series cyclic network according to claim 1 is characterized in that: The cable data set preprocessing formula in step 3) is: S a =[f(C,I,S,P,F,M)-f min (C,I,S,P,F,M)] / [f max (C,I,S,P,F,M)-f min (C,I,S,P,F,M)](3) Among them, C is the conductor material, I is the insulation material, S is the shielding material, P is the sheath material, F is the fire-resistant material, M is the mechanical strength, f is a unified dimension function, and f min is the smallest dimension function, f max is the largest dimension function. Here S a is an average function, which acts as a combination of cable material characteristics.

5. The cable cross-layer optimization design method based on radius time series cyclic network according to claim 1 is characterized in that: The recurrent neural network (RNN) analysis formula in step 4) is: y t =σW hy (σ(W ih X t +b ih +W hh h t-1 +b hh )) (4) Among them, W ih , W hh and W hy are all weight matrices, b ih , b hh and b hy are all bias terms, σ is the activation function, X t is the input vector value at time step t, h t-1 is the hidden state at time step t-1, y t is the output vector value at time step t.

Images