Digital cable parameter identification method, device and equipment and storage medium

Abstract

The invention discloses a parameter identification method, device and equipment for a digital cable and a storage medium, and the method comprises the steps: obtaining a to-be-identified parameter vector and the actual measurement temperature of each layer of material of the digital cable, the to-be-identified parameter vector being the heat conductivity coefficient of each layer of material of the digital cable; the to-be-identified parameter vector is input into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector, the adaptive particle swarm algorithm performs iterative optimization on the to-be-identified parameter vector based on a target function, and the target function is obtained through calculation based on the deviation between the predicted temperature and the actually measured temperature of each layer of material of the digital cable; under the condition that the target function value corresponding to the candidate optimal parameter vector meets the preset convergence condition, the candidate optimal parameter vector is determined as a target optimal parameter vector, and the target optimal parameter vector is the identification result of the to-be-identified parameter corresponding to the to-be-identified parameter vector. Therefore, the real-time identification requirement of the digital cable on the heat conductivity coefficient parameter under the complex dynamic working condition can be effectively met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power systems and information physics fusion, and particularly relates to a parameter identification method and device of a digital cable, equipment and a storage medium. BACKGROUND

[0002] As the core carrier of power transmission in the power system, the operation state of the digital cable is directly related to the safety and stability of the entire power system. In the operation process of the digital cable, key state quantities such as cable core temperature and insulation aging state are significantly affected by parameters such as the thermal conductivity of each layer of material. Therefore, accurately identifying these core parameters is the basis for realizing real-time evaluation of the carrying capacity of the digital cable and early warning of overload risk.

[0003] However, the traditional method mainly uses the least square method and its derivative method to fit the thermal conductivity of each layer of material of the digital cable by minimizing the sum of squares of errors. However, the linear assumption cannot reflect the nonlinear characteristics of the thermal conductivity of the digital cable changing with temperature, especially under high temperature working conditions, the error is obviously amplified; the weighted least square method relies on the empirical weight matrix, and it is difficult to adapt to diversified and complex scenes; the iterative least square method can partially improve the fitting effect, but it has problems such as slow convergence speed and large calculation amount, which cannot meet the real-time requirements of online monitoring. Therefore, the existing parameter identification methods generally have weak anti-interference ability, lack self-adaptive correction mechanism, and cannot balance efficiency and accuracy, and cannot meet the real-time identification requirements of the thermal conductivity parameters of the digital cable under complex dynamic working conditions. SUMMARY

[0004] The embodiments of the present application provide a parameter identification method, device, equipment and storage medium of a digital cable, which can effectively meet the real-time identification requirements of the thermal conductivity parameters of the digital cable under complex dynamic working conditions.

[0005] In a first aspect, the embodiments of the present application provide a parameter identification method of a digital cable, which comprises:

[0006] obtaining a to-be-identified parameter vector and a measured temperature of each layer of material of the digital cable, the to-be-identified parameter vector being the thermal conductivity of each layer of material of the digital cable;

[0007] inputting the to-be-identified parameter vector into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector, wherein the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on a target function, the target function being calculated based on the deviation between a predicted temperature and the measured temperature of each layer of material of the digital cable, and the predicted temperature being output by an eigenvalue orthogonal decomposition reduced-order digital twin model;

[0008] In a case where the target function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition, the candidate optimal parameter vector is determined as a target optimal parameter vector, and the target optimal parameter vector is an identification result of the to-be-identified parameter corresponding to the to-be-identified parameter vector.

[0009] An implementable embodiment of the method comprises:

[0010] A value range of the to-be-identified parameter is set, and a search space of the to-be-identified parameter is determined, the to-be-identified parameter including thermal conductivities of a core, a semiconductor shielding layer, a cross-linked polyethylene insulation layer, a metal shielding layer and an outer sheath of the digital cable.

[0011] The to-be-identified parameter vector is randomly generated based on the search space.

[0012] An implementable embodiment of the method comprises:

[0013] A plurality of temperature sensors are arranged on the layers of the digital cable to obtain temperature data of a plurality of monitoring points.

[0014] The measured temperature of the layers of the digital cable is obtained based on the temperature data of the plurality of monitoring points.

[0015] An implementable embodiment of the method comprises:

[0016] The adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on a target function to obtain a target function value corresponding to an updated parameter vector of each iteration.

[0017] The target function value corresponding to the updated parameter vector of each iteration is screened to obtain an optimal target function value.

[0018] The updated parameter vector corresponding to the optimal target function value is determined as a candidate optimal parameter vector.

[0019] An implementable embodiment of the method comprises:

[0020] The adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on an objective function to obtain an updated parameter vector of each iteration round;

[0021] The updated parameter vector of each iteration round is substituted into the objective function for calculation to obtain an objective function value corresponding to the updated parameter vector of each iteration round.

[0022] In an implementable embodiment, the candidate optimal parameter vector is determined as the target optimal parameter vector in the case where the objective function value corresponding to the candidate optimal parameter vector meets a preset convergence condition, which comprises:

[0023] The adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on an objective function to obtain an updated parameter vector of each iteration round;

[0024] In an implementable embodiment, the method further comprises:

[0025] The adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on an objective function to obtain an updated parameter vector of each iteration round;

[0026] In an implementable embodiment, the candidate optimal parameter vector is determined as the target optimal parameter vector in the case where the objective function value corresponding to the candidate optimal parameter vector meets a preset convergence condition, which comprises:

[0027] In a second aspect, the embodiments of the present application provide a parameter identification device of a digital cable, which comprises:

[0028] A vector acquisition module is configured to acquire a to-be-identified parameter vector and measured temperatures of materials of each layer of the digital cable, the to-be-identified parameter vector being a thermal conductivity coefficient of the materials of each layer of the digital cable;

[0029] A vector generation module is configured to input the to-be-identified parameter vector into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector, wherein the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on an objective function, the objective function being calculated based on a deviation between a predicted temperature and the measured temperature of the materials of each layer of the digital cable, the predicted temperature being output by a digital twin model subjected to intrinsic orthogonal decomposition and order reduction;

[0030] a target vector determination module, configured to determine the candidate optimal parameter vector as a target optimal parameter vector in a case where a target function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition, the target optimal parameter vector being a recognition result of the to-be-recognized parameter corresponding to the to-be-recognized parameter vector.

[0031] In a third aspect, an electronic device is provided, and the device includes: a processor, a memory, and a system bus;

[0032] The processor and the memory are connected through the system bus;

[0033] The memory is configured to store a program, the program including instructions that, when executed by the processor, cause the processor to perform any implementation step of the parameter recognition method of the digital cable.

[0034] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium is configured to store a computer program, the computer program being executed by a terminal device to implement any implementation step of the parameter recognition method of the digital cable.

[0035] As can be seen from the above technical solutions, the embodiments of the present application have the following advantages:

[0036] In the embodiments of the present application, first, a to-be-recognized parameter vector and measured temperatures of materials of each layer of the digital cable are obtained, wherein the to-be-recognized parameter vector is a thermal conductivity coefficient of the materials of each layer of the digital cable. Then, the to-be-recognized parameter vector is input into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector, wherein the adaptive particle swarm algorithm iteratively optimizes the to-be-recognized parameter vector based on a target function, the target function is calculated based on a deviation between a predicted temperature and the measured temperature of the materials of each layer of the digital cable, and the predicted temperature is output by an eigenvalue orthogonal decomposition reduced-order digital twin model. Finally, in a case where a target function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition, the candidate optimal parameter vector is determined as a target optimal parameter vector, and the target optimal parameter vector is a recognition result of the to-be-recognized parameter corresponding to the to-be-recognized parameter vector.

[0037] It can be seen that the scheme combines the adaptive particle swarm algorithm with the eigen-orthogonal decomposition reduced-order digital twin model, takes the deviation between the predicted temperature of each layer material of the digital cable and the measured temperature as the core to construct the objective function, iteratively optimizes the to-be-identified parameter vector composed of the thermal conductivity coefficient, and determines the optimal parameter vector based on the preset convergence condition. The eigen-orthogonal decomposition reduced-order digital twin model in the scheme can reflect the nonlinear characteristics of the thermal conductivity coefficient changing with temperature, breaking through the linear assumption limitation to reduce the identification error under high temperature working conditions; the adaptive particle swarm algorithm does not need to rely on empirical weight, and can independently iterate and optimize the to-be-identified parameter vector to adapt to complex scenes, and the algorithm has fast convergence speed and small calculation amount, and in combination with the preset convergence condition, the real-time performance of parameter identification can be guaranteed, and finally the problems of weak anti-interference ability, lack of adaptive correction mechanism and difficulty in balancing efficiency and accuracy of the existing method are effectively made up, and the real-time identification demand of the thermal conductivity coefficient parameter of the digital cable under complex dynamic working conditions is effectively met. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 A flowchart of a parameter identification method of a digital cable provided by an embodiment of the present application;

[0039] Figure 2 A structural schematic diagram of a parameter identification device of a digital cable provided by an embodiment of the present application. DETAILED DESCRIPTION

[0040] As described above, the digital cable is the core carrier of power transmission in the power system, and its operating state is directly related to the safety and stability of the entire power system. In the operation process of the digital cable, the core state quantities such as cable core temperature and insulation aging state are significantly affected by parameters such as thermal conductivity coefficient of each layer material. Therefore, accurately identifying these core parameters is the basis for realizing real-time evaluation of the carrying capacity of the digital cable and early warning of overload risk.

[0041] Traditional methods mostly use least squares method and its derivative methods to fit the thermal conductivity and other core parameters of each layer material of digital cable by minimizing the sum of squares of errors. Specifically, the traditional least squares method optimizes parameters by minimizing the sum of squares of errors between observed values and model predicted values, which has been used in the simplified model of digital cable temperature field based on thermoelectric analogy theory, such as estimating parameters by linearly fitting the relationship between core temperature and load current, ambient temperature, but this method ignores the nonlinear temperature characteristics of the thermal conductivity of cross-linked polyethylene insulation material, and large identification errors are easily produced in high temperature area. In order to suppress the influence of noise, the weighted least squares method introduces a weight matrix to give different weights to each layer of the digital cable, but the empirical setting of the weight matrix is difficult to adapt to the parameter variation characteristics in complex environments, such as uneven soil thermal resistance, which will lead to deviation in parameter identification. Although the iterative least squares method linearizes the nonlinear model by iteration, the strong nonlinear relationship between current and temperature in the digital cable temperature field leads to slow convergence speed, which cannot meet the real-time demand of online monitoring of parameters.

[0042] Therefore, the existing parameter identification methods generally have weak anti-interference ability, lack adaptive correction mechanism, and are difficult to balance efficiency and accuracy, and overall cannot meet the real-time parameter identification demand of digital cable in complex dynamic working conditions.

[0043] Based on this, in order to solve the above problems, the embodiment of the present application provides a parameter identification method, device and equipment of digital cable and storage medium, first, the measured temperature of each layer material of digital cable and the to-be-identified parameter vector are obtained, wherein the to-be-identified parameter vector is the thermal conductivity of each layer material of digital cable. Then, the to-be-identified parameter vector is input into the preset adaptive particle swarm algorithm to obtain the candidate optimal parameter vector, wherein the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on the objective function, the objective function is calculated based on the deviation between the predicted temperature and the measured temperature of each layer material of digital cable, and the predicted temperature is output by the digital twin model of intrinsic orthogonal decomposition reduction. Finally, in the case that the objective function value corresponding to the candidate optimal parameter vector meets the preset convergence condition, the candidate optimal parameter vector is determined as the target optimal parameter vector, and the target optimal parameter vector is the identification result of the to-be-identified parameter corresponding to the to-be-identified parameter vector.

[0044] It can be seen that the scheme combines the adaptive particle swarm algorithm with the eigen-orthogonal decomposition reduced-order digital twin model, takes the deviation between the predicted temperature and the measured temperature of each layer material of the digital cable as the core to construct the objective function, iteratively optimizes the to-be-identified parameter vector composed of the thermal conductivity coefficients, and determines the optimal parameter vector based on the preset convergence condition. The eigen-orthogonal decomposition reduced-order digital twin model in the scheme can reflect the nonlinear characteristics of the thermal conductivity coefficients changing with temperature, breaking through the linear assumption limitation to reduce the identification error under high temperature working conditions. The adaptive particle swarm algorithm does not need to rely on empirical weights, and can iteratively optimize the to-be-identified parameter vector to adapt to complex scenarios. The algorithm has fast convergence speed and small calculation amount, and in combination with the preset convergence condition, the real-time performance of parameter identification can be guaranteed. Finally, the problems of weak anti-interference ability, lack of adaptive correction mechanism and difficulty in balancing efficiency and accuracy of the existing methods are effectively solved, and the real-time identification demand of the thermal conductivity parameters of the digital cable under complex dynamic working conditions is effectively met.

[0045] It should be noted that the parameter identification method of the digital cable according to the embodiments of the present application can not be limited to the execution subject. For example, the parameter identification method of the digital cable according to the embodiments of the present application can be applied to information processing devices such as servers or terminal devices. The server can be a stand-alone server, a cluster server or a cloud server. The terminal device can be a smart phone, a computer, a personal digital assistant (PDA), a tablet computer or the like.

[0046] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0047] Figure 1 A flowchart of a parameter identification method of a digital cable according to an embodiment of the present application is provided. As shown in FIG. 1, the method can include steps S101-S103. Figure 1

[0048] S101: Obtain the to-be-identified parameter vector and the measured temperature of each layer material of the digital cable. The to-be-identified parameter vector is the thermal conductivity coefficient of each layer material of the digital cable.

[0049] ​In the embodiment of the present application, first, the thermal conductivities of the cable core, the semiconductor shielding layer, the cross-linked polyethylene insulation layer, the metal shielding layer and the outer sheath of the digital cable are taken as the to-be-identified parameters. Then, according to the physical characteristics and the experience value interval of the thermal conductivities of the materials of the layers of the digital cable, the to-be-identified parameters are set with corresponding value ranges. A search space of the to-be-identified parameters is constructed based on the set value ranges, for limiting the value range of the to-be-identified parameters. On this basis, a to-be-identified parameter vector is randomly generated in the search space, and the to-be-identified parameter vector can be expressed as , wherein represents the thermal conductivity of the mth layer of material. The above process provides reasonable physical constraints and operable parameter ranges for subsequent parameter identification.

[0050] In addition, the temperature data of multiple monitoring points is obtained through the multiple temperature sensors arranged on the materials of the layers of the digital cable, and the measured temperatures of the materials of the layers of the digital cable are obtained based on the temperature data of the multiple monitoring points. It should be noted that the multiple temperature sensors arranged on the materials of the layers of the digital cable in the embodiment of the present application can be adjusted according to actual conditions, for example, 15 temperature sensors can be arranged on the materials of the layers of the digital cable to obtain the temperature data of 15 monitoring points.

[0051] S102: input the to-be-identified parameter vector into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector, wherein the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on a target function, the target function is calculated based on the deviation between the predicted temperature and the measured temperature of the materials of the layers of the digital cable, and the predicted temperature is output by the proper orthogonal decomposition reduced-order digital twin model.

[0052] In the embodiment of the present application, the predicted temperature of each monitoring point of the digital cable can be quickly calculated by the proper orthogonal decomposition reduced-order digital twin model, thereby greatly reducing the calculation amount. Specifically, the basis vector and the corresponding coefficient extracted based on the proper orthogonal decomposition (POD) are used to reconstruct the temperature field, thereby simplifying the original three-dimensional heat conduction problem into scalar calculation, and compared with the traditional finite element calculation method, the time consumption of online calculation is significantly reduced.

[0053] When the to-be-identified parameter vector , the sampling time t of the sensor and the cable core current I of the sensor are input, the predicted temperature of the mth monitoring point can be expressed as:

[0054] ;

[0055] wherein,​ This is a counter for the order of POD basis vectors. In this embodiment, the first three POD basis vectors extracted offline are used. Represents the parameter vector to be identified Sensor sampling time and the cable core current of the sensor The correlation coefficients reflect the weight of each basis vector under the current conditions; Indicates the first The first-order POD basis vectors are in the first order. The components at each monitoring point reflect the main spatial distribution of the temperature field, and its spatial distribution characteristics remain unchanged under different times and currents.

[0056] In the specific implementation, The acquisition is divided into two stages. In the offline stage, the entire dynamic process of the overlay digital cable from transient to steady state under multiple current and time step conditions is simulated using finite element method to extract the POD basis vector and construct the parameter vector to be identified. - Sensor core current -Sensor sampling time -coefficient The mapping table is used, while in the online phase, the coefficients are quickly determined by looking up the mapping table or by linear interpolation based on the sensor's sampling time t and the incoming cable core current I. This enables efficient reconstruction of the temperature field of digital cables.

[0057] Therefore, it can be seen that the digital twin model constructed by reducing the order through intrinsic orthogonal decomposition can obtain the POD coefficients by looking up the mapping table in the online stage after establishing the mapping table offline, and reconstruct the temperature field by combining it with the POD basis vectors, thereby efficiently obtaining the predicted temperature of each monitoring point of the digital cable.

[0058] Next, an objective function is calculated based on the deviation between the predicted and measured temperatures of each layer of the digital cable material, which is used to evaluate the fit of the parameter vector to be identified. In this embodiment, the weighted root mean square error is used as the objective function, specifically expressed as follows:

[0059] ;

[0060] in, Based on the parameter vector to be identified The Predicted temperature at each monitoring point Indicates the first The measured temperature at each monitoring point. The weight matrix... Let represent the weight coefficient of the m-th layer. The weight coefficient of each layer is dynamically updated with each iteration, and its update formula is:

[0061] ;

[0062] Among them, represents the weight coefficient of the m-th layer at the j-th iteration, represents the average temperature error of the m-th layer at the j-th iteration, and its calculation formula is:

[0063] ;

[0064] Among them, represents the number of monitoring points contained in the m-th layer, is the predicted temperature of the -th monitoring point, is the measured temperature of the -th monitoring point.

[0065] Subsequently, the parameter vector to be identified is input as a particle into the preset adaptive particle swarm algorithm. Initially, 50 particles are set, and the initial velocity is set to 0. Among them, the first 30 particles are evenly randomly distributed within the search space of the parameter to be identified to cover all reasonable value ranges of the parameter to be identified; the last 20 particles are generated near the sensitive interval of the thermal conductivity corresponding to the thinnest insulating layer of the digital cable to improve the initial search efficiency by using prior knowledge. The position of each particle corresponds to a parameter vector to be identified, which is used to explore the optimal thermal conductivity combination within the search space.

[0066] It should be noted that in the adaptive particle algorithm, its inertia weight adopts a three-stage dynamic adjustment to optimize the search strategy. In this embodiment of the application, it is illustrated with the iteration number j being 100 times. represents the i-th particle and the inertia weight at the j-th iteration. Specifically, in the early stage of iteration, that is, 0 < j ≤ 30, the inertia weight can be:

[0067] ;

[0068] Among them, in the early stage of iteration, the inertia weight value of the adaptive particle algorithm is relatively large to ensure that the particles have strong global search ability.

[0069] In the middle stage of iteration, that is, 30 < j ≤ 70, at this time the inertia weight can be:

[0070] ;

[0071] Among them, represents the objective function value of the i-th particle at the j-th iteration, represents the j-th iteration, and the average objective function value of all particles. The average objective function value can be expressed as:

[0072] ;

[0073] In the middle of the iteration, the particle swarm roughly locks in the optimal candidate area. High-quality particles increase their weights to strengthen local development, ordinary particles slightly reduce their weights to encourage exploration, and low-quality particles maintain their weights to continue exploring other areas.

[0074] In the later stage of the iteration, that is, 70 < j ≤ 100, the inertia weight can be:

[0075] ;

[0076] where is the chaotic function at the j-th iteration, and its initial value, is 0.3, and it can be specifically expressed as:

[0077] ;

[0078] In the later stage of the iteration, the particle swarm approaches the global optimal solution, the inertia weight is small, and the chaotic function can avoid falling into the local optimum.

[0079] Furthermore, the particle velocity and its position are iteratively updated according to the following formula:

[0080] ;

[0081] ;

[0082] where represents the velocity of the i-th particle at the j-th iteration; represents the position of the i-th particle at the j-th iteration; represents the historical optimal position of the i-th particle at the j-th iteration, that is, the individual optimal solution; represents the global optimal position of the particle swarm at the j-th iteration, that is, the swarm optimal solution; and are the introduced random numbers, represents the individual learning factor at the j-th iteration, represents the global learning factor at the j-th iteration. At the same time, + = 3 remains unchanged, ensuring a reasonable exploration and development ratio of the particles during the search process. It is specifically expressed as:

[0083] ;

[0084] ;

[0085] It should be noted that due to the introduction of the random numbers and and and a value range of 0 to 1, which increases the diversity of particle search. To ensure that the particles are within the physically reasonable search space, when the updated particle position exceeds the lower boundary, a mirror reflection correction strategy is adopted:

[0086] ;

[0087] and when the updated particle position exceeds the upper boundary, there is:

[0088] ;

[0089] wherein, is the lower boundary corresponding to the search space, is the upper boundary corresponding to the search space.

[0090] In addition, for each particle after each iteration, an evaluation is made based on the fitness function, and the greater the value, the better the combination of the obtained to-be-identified parameters, and the expression is:

[0091] ;

[0092] wherein, is a small positive number introduced to avoid computational anomalies caused by the denominator being 0 when the fitness function takes a value of 0. The fitness function is the basis for updating the individual optimal solution and the global optimal solution in the particle swarm iteration.

[0093] Therefore, through the above mechanism, the adaptive particle swarm algorithm can effectively explore within the search space and dynamically update the particle position according to the target function, i.e., through the adaptive particle swarm algorithm, the to-be-identified parameter vector is iteratively optimized based on the target function, and the updated parameter vector after each iteration is obtained; the updated parameter vector after each iteration is substituted into the target function for calculation to obtain the target function value corresponding to the updated parameter vector after each iteration; the target function value corresponding to the updated parameter vector after each iteration is selected to obtain the optimal target function value; the updated parameter vector corresponding to the optimal target function value is determined as the candidate optimal parameter vector. That is, the global optimal position obtained after the final iteration converges is the candidate optimal parameter vector obtained by the adaptive particle swarm algorithm.

[0094] S103: In the case where the target function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition, the candidate optimal parameter vector is determined as the target optimal parameter vector, and the target optimal parameter vector is the identification result of the to-be-identified parameter corresponding to the to-be-identified parameter vector.

[0095] In the embodiment of the present application, if the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on the target function more than a preset maximum iteration number of times, and the variation of the target function value corresponding to the candidate optimal parameter vector is less than or equal to a preset temperature deviation threshold in a preset number of continuous iterations, the candidate optimal parameter vector is determined as the target optimal parameter vector. For example, if the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector more than 100 times, and the variation of the target function value corresponding to the candidate optimal parameter vector is less than or equal to 0.05 ℃ in 5 continuous iterations, the candidate optimal parameter vector is determined as the target optimal parameter vector.

[0096] Specifically, the expression of the variation of the target function value corresponding to the candidate optimal parameter vector can be:

[0097]

[0098] indicates the target function value corresponding to the optimal parameter vector at the jth iteration, indicates the target function value corresponding to the optimal parameter vector at the (j-5)th iteration.

[0099] If the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on the target function more than a preset maximum iteration number of times, and the variation of the target function value corresponding to the candidate optimal parameter vector is greater than a preset temperature deviation threshold in a preset number of continuous iterations, that is, if the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector more than 100 times, and the variation of the target function value corresponding to the candidate optimal parameter vector is greater than 0.05 ℃ in 5 continuous iterations, a secondary optimization strategy needs to be triggered for the to-be-identified parameter vector to obtain a second candidate optimal parameter vector. Specifically, 10% of the particles in the current particle swarm are reset as historical suboptimal solutions, and the remaining particles remain unchanged in the current position. The to-be-identified parameter vector is iteratively optimized again according to the iteration rule of the adaptive particle swarm algorithm for 100 times, and the global optimal position obtained after iteration and convergence is the second candidate optimal parameter vector. If the target function value corresponding to the second candidate optimal parameter vector satisfies a preset convergence condition, the second candidate optimal parameter vector is determined as the target optimal parameter vector.

[0100] It should be noted that during the whole process of the first iteration optimization, the parameter vector updated after each iteration and the target function value corresponding thereto are recorded. After the parameter vector corresponding to the global optimal target function value is removed, the remaining target function values are sorted in ascending order, and the parameter vectors updated after iteration corresponding to the first 5 target function values after sorting are selected, that is, the historical suboptimal solutions.

[0101] ​​Therefore, the embodiments of this application employ a dual convergence criterion: only when the objective function value corresponding to the candidate optimal parameter vector satisfies the preset convergence condition is the candidate optimal parameter vector determined as the target optimal parameter vector. * The target optimal parameter vector * The identification result is the identification result of the parameter to be identified corresponding to the parameter vector to be identified.

[0102] Furthermore, the target optimal parameter vector finally identified in the embodiments of this application * It is also used for dynamic current-carrying capacity assessment of digital cables, and the current-carrying capacity assessment process can be derived in reverse based on the mapping table constructed above. Specifically, it is based on the target optimal parameter vector. * The sensor's core current I and basis coefficient are determined by the sensor's acquisition time t. The two-dimensional correspondence, and then through the above The formula is used to determine the sensor's position on the 1st... Measured temperature at monitoring point Replace the formula ,and Since it has already been determined in advance, it is based on the target optimal parameter vector. * The basis coefficients under the current operating conditions can be obtained by inversely solving the data collected by the sensor over time t. Furthermore, combining the aforementioned core current I and the base coefficient The two-dimensional correspondence is obtained, thus solving the problem inversely. The corresponding core current value I is the current dynamic current carrying capacity assessment value of the digital cable. .

[0103] It should be noted that in the current carrying capacity assessment value Compared with the actual value of the cable core current When the deviation reaches a preset threshold, that is, when the relative error between the two satisfies:

[0104] ;

[0105] Alternatively, the absolute error satisfies:

[0106] ;

[0107] If the two errors exceed the preset threshold, it indicates that the to-be-identified parameters of the materials of each layer of the digital cable change, and thus the full process of the parameter identification of the digital cable must be re-executed to update the to-be-identified parameters and ensure the accuracy of subsequent current-carrying capacity evaluation and insulation life prediction. Conversely, if the two errors do not exceed the preset threshold, it indicates that the prediction result of the current-carrying capacity is relatively accurate, and the target parameter vector identified previously is also relatively accurate, and a new round of parameter identification process does not need to be triggered.

[0108] In addition, the embodiment of the present application can calculate the remaining life of the digital cable based on the Arrhenius equation. Since the cross-linked polyethylene insulation layer of the digital cable is the most sensitive core layer to aging, the aging of the cross-linked polyethylene insulation layer directly determines the life and safety of the digital cable. Therefore, when calculating the remaining life L of the digital cable, the cross-linked polyethylene insulation layer is focused on, and the calculation is as follows:

[0109]

[0110] wherein Ea represents the activation energy, i.e., the minimum energy required for the aging reaction of the insulation material of the digital cable, the value of which is determined by the characteristics of the material itself. For the cross-linked polyethylene insulation layer of the digital cable, Ea can be taken as 0.9 eV; L0 is the predicted life of the digital cable under standard working conditions; T0 is the reference temperature, i.e., the standard ambient temperature without thermal aging acceleration effect; is the Boltzmann constant; T XLPE is based on the target parameter vector * The predicted temperature of the cross-linked polyethylene insulation layer.

[0111] To convert the identification results of the to-be-identified parameters, the temperature field distribution, and the remaining life prediction results into operable operation and maintenance decision basis, the present application further proposes to develop a digital cable visual operation and maintenance platform. The platform directly presents the full-dimensional temperature field distribution characteristics of the digital cable in the form of a three-dimensional cloud map, and simultaneously displays the core parameters such as the thermal conductivity coefficients of each layer identified in the foregoing, the current-carrying capacity evaluation value, the remaining life, etc. in the form of dynamic curves, clearly presenting the change trend of various parameters over time. Through this visual method, reference basis is provided for the operation and maintenance safety control, life warning, and maintenance strategy formulation of the digital cable.

[0112] ​Based on the related content of steps S101-S103, in the embodiment of the present application, first, the measured temperature of each layer material of the digital cable and the to-be-identified parameter vector are obtained, wherein the to-be-identified parameter vector is the thermal conductivity of each layer material of the digital cable. Then, the to-be-identified parameter vector is input into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector, wherein the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on an objective function, the objective function is calculated based on the deviation between the predicted temperature and the measured temperature of each layer material of the digital cable, and the predicted temperature is output by the proper orthogonal decomposition reduced-order digital twin model. Finally, in the case that the objective function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition, the candidate optimal parameter vector is determined as a target optimal parameter vector, and the target optimal parameter vector is the identification result of the to-be-identified parameter corresponding to the to-be-identified parameter vector. It can be seen that, by combining the adaptive particle swarm algorithm with the proper orthogonal decomposition reduced-order digital twin model, and taking the deviation between the predicted temperature and the measured temperature of each layer material of the digital cable as the core to construct the objective function, the to-be-identified parameter vector composed of the thermal conductivities is iteratively optimized, and the target optimal parameter vector is determined based on the preset convergence condition. Among them, the proper orthogonal decomposition reduced-order digital twin model in the present scheme can reflect the nonlinear characteristics of the thermal conductivity changing with temperature, breaking through the linear hypothesis limitation to reduce the identification error under high temperature working condition; the adaptive particle swarm algorithm does not need to rely on the experience weight, and can iteratively optimize the to-be-identified parameter vector to adapt to complex scenes, and the algorithm has fast convergence speed and small calculation amount, which can guarantee the real-time performance of parameter identification combined with the preset convergence condition, and finally effectively make up for the problems of weak anti-interference ability, lack of adaptive correction mechanism and difficulty in balancing efficiency and accuracy of the existing method, effectively meet the real-time identification demand of the thermal conductivity parameter of the digital cable under complex dynamic working condition.

[0113] Further, Figure 2 A structural schematic diagram of a parameter identification device for a digital cable is provided in the embodiment of the present application. As shown in Figure 2 The parameter identification device 200 for a digital cable provided in the embodiment of the present application can include:

[0114] A vector acquisition module 201, configured to acquire a to-be-identified parameter vector and a measured temperature of each layer material of a digital cable, wherein the to-be-identified parameter vector is a thermal conductivity of each layer material of the digital cable;

[0115] The vector generation module 202 is configured to input the to-be-identified parameter vector into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector, wherein the adaptive particle swarm algorithm performs iterative optimization on the to-be-identified parameter vector based on a target function, and the target function is calculated based on a deviation between a predicted temperature of each layer of material of the digital cable and a measured temperature, and the predicted temperature is output by the proper orthogonal decomposition reduced-order digital twin model.

[0116] The target vector determination module 203 is configured to determine the candidate optimal parameter vector as a target optimal parameter vector in a case where a target function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition, and the target optimal parameter vector is an identification result of a to-be-identified parameter corresponding to the to-be-identified parameter vector.

[0117] Optionally, the vector acquisition module 201 is specifically configured to:

[0118] set a value range of the to-be-identified parameter, and determine a search space of the to-be-identified parameter, wherein the to-be-identified parameter includes thermal conductivities of a core, a semiconductor shielding layer, a cross-linked polyethylene insulation layer, a metal shielding layer and an outer sheath of the digital cable.

[0119] randomly generate the to-be-identified parameter vector based on the search space.

[0120] Optionally, the vector acquisition module 201 is specifically configured to:

[0121] obtain temperature data of multiple monitoring points based on multiple temperature sensors arranged on each layer of material of the digital cable.

[0122] obtain the measured temperature of each layer of material of the digital cable based on the temperature data of the multiple monitoring points.

[0123] Optionally, the vector generation module 202 can include:

[0124] The iterative optimization module is configured to perform iterative optimization on the to-be-identified parameter vector based on the target function by using the adaptive particle swarm algorithm, to obtain a target function value corresponding to the parameter vector updated in each iteration.

[0125] The numerical value screening module is configured to screen the target function value corresponding to the parameter vector updated in each iteration to obtain an optimal target function value.

[0126] The vector determination module is configured to determine the parameter vector updated in each iteration corresponding to the optimal target function value as the candidate optimal parameter vector.

[0127] Optionally, the iterative optimization module is specifically configured to:

[0128] The adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on the objective function to obtain an updated parameter vector of each iteration round;

[0129] The updated parameter vector of each iteration round is substituted into the objective function to obtain an objective function value corresponding to the updated parameter vector of each iteration round.

[0130] Optionally, the target vector determination module 203 is specifically configured to:

[0131] The adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on the objective function to obtain an updated parameter vector of each iteration round;

[0132] Optionally, the parameter identification apparatus for a digital cable 200 can include:

[0133] The quadratic optimization module is configured to, when the adaptive particle swarm algorithm iteratively optimizes the to-be-identified parameter vector based on the objective function to obtain an updated parameter vector of each iteration round, and when the variation of the objective function value corresponding to the candidate optimal parameter vector is greater than the preset temperature deviation threshold in a continuous preset iteration number, trigger a quadratic optimization strategy for the to-be-identified parameter vector to obtain a second candidate optimal parameter vector.

[0134] The vector determination module is configured to, when the objective function value corresponding to the second candidate optimal parameter vector satisfies a preset convergence condition, determine the second candidate optimal parameter vector as the target optimal parameter vector.

[0135] Further, the embodiments of the present application also provide an electronic device, including: a processor, a memory, a system bus;

[0136] The processor and the memory are connected through the system bus;

[0137] The memory is configured to store one or more programs, the one or more programs including instructions, the instructions causing the processor to execute any implementation step of the above-mentioned parameter identification method for a digital cable when executed by the processor.

[0138] Further, the embodiments of the present application also provide a computer readable storage medium, the computer readable storage medium is configured to store a computer program, the computer program is executed by a terminal device to implement any implementation step of the above-mentioned parameter identification method for a digital cable.

[0139] Those skilled in the art can clearly understand that all or part of the steps of the method in the above embodiments can be implemented by means of software and necessary universal hardware platforms based on the description of the above embodiments. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which can be stored in a storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) execute the method described in the various embodiments or some parts of the embodiments of the present application. It should be noted that the various embodiments in the present specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0140] For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant parts are referred to the method part.

[0141] It should also be noted that the relationship terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or device including the element.

[0142] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for parameter identification of a digital cable, characterized in that, The method includes: Obtain the vector of parameters to be identified and the measured temperature of each layer of material in the digital cable, wherein the vector of parameters to be identified is the thermal conductivity of each layer of material in the digital cable. The parameter vector to be identified is input into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector. The adaptive particle swarm algorithm iteratively optimizes the parameter vector to be identified based on an objective function. The objective function is calculated based on the deviation between the predicted temperature and the measured temperature of each layer of the digital cable material. The predicted temperature is output by a digital twin model with reduced order by intrinsic orthogonal decomposition. If the objective function value corresponding to the candidate optimal parameter vector satisfies the preset convergence condition, the candidate optimal parameter vector is determined as the target optimal parameter vector, and the target optimal parameter vector is the identification result of the parameter to be identified corresponding to the parameter to be identified.

2. The method according to claim 1, characterized in that, The process of obtaining the parameter vector to be identified and the measured temperatures of each layer of the digital cable material, wherein the parameter vector to be identified is the thermal conductivity of each layer of the digital cable material, includes: Set the value range of the parameter to be identified, and determine the search space of the parameter to be identified. The parameter to be identified includes the thermal conductivity of the cable core, semiconductor shielding layer, cross-linked polyethylene insulation layer, metal shielding layer and outer sheath of the digital cable. Based on the search space, the vector of parameters to be identified is randomly generated.

3. The method according to claim 1, characterized in that, The process of obtaining the parameter vector to be identified and the measured temperatures of each layer of the digital cable material, wherein the parameter vector to be identified is the thermal conductivity of each layer of the digital cable material, includes: Multiple temperature sensors are deployed on each layer of the digital cable to obtain temperature data at multiple monitoring points; Based on the temperature data from the multiple monitoring points, the measured temperatures of each layer of the digital cable material were obtained.

4. The method according to claim 1, characterized in that, The step of inputting the parameter vector to be identified into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector includes: The adaptive particle swarm optimization algorithm iteratively optimizes the parameter vector to be identified based on the objective function to obtain the objective function value corresponding to the parameter vector after each iteration. The objective function value corresponding to the parameter vector updated in each round of iteration is filtered to obtain the optimal objective function value; The iteratively updated parameter vector corresponding to the optimal objective function value is determined as the candidate optimal parameter vector.

5. The method according to claim 4, characterized in that, The adaptive particle swarm optimization algorithm iteratively optimizes the parameter vector to be identified based on an objective function to obtain the objective function value corresponding to the parameter vector after each iteration, including: The adaptive particle swarm optimization algorithm iteratively optimizes the parameter vector to be identified based on the objective function to obtain the parameter vector updated in each iteration. The parameter vector updated in each iteration is substituted into the objective function for calculation to obtain the objective function value corresponding to the parameter vector updated in each iteration.

6. The method according to claim 1, characterized in that, The step of determining the candidate optimal parameter vector as the target optimal parameter vector when the objective function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition includes: The adaptive particle swarm optimization algorithm iterates the target parameter vector based on the objective function a maximum number of times. If the change in the objective function value corresponding to the candidate optimal parameter vector is less than or equal to a preset temperature deviation threshold in a series of preset iterations, the candidate optimal parameter vector is determined as the target optimal parameter vector.

7. The method according to claim 1, characterized in that, The method further includes: The adaptive particle swarm optimization algorithm iterates the target parameter vector to be identified a maximum number of times based on the objective function. If the change in the objective function value corresponding to the candidate optimal parameter vector is greater than the preset temperature deviation threshold in the consecutive preset number of iterations, a secondary optimization strategy is triggered on the target parameter vector to obtain a second candidate optimal parameter vector. If the objective function value corresponding to the second candidate optimal parameter vector satisfies the preset convergence condition, the second candidate optimal parameter vector is determined as the target optimal parameter vector.

8. A parameter identification device for a digital cable, characterized in that, include: The vector acquisition module is used to acquire the parameter vector to be identified and the measured temperature of each layer of the digital cable material, wherein the parameter vector to be identified is the thermal conductivity of each layer of the digital cable material. The vector generation module is used to input the parameter vector to be identified into a preset adaptive particle swarm algorithm to obtain a candidate optimal parameter vector. The adaptive particle swarm algorithm iteratively optimizes the parameter vector to be identified based on an objective function. The objective function is calculated based on the deviation between the predicted temperature and the measured temperature of each layer of the digital cable material. The predicted temperature is output by a digital twin model with reduced order by intrinsic orthogonal decomposition. The target vector determination module is used to determine the candidate optimal parameter vector as the target optimal parameter vector when the objective function value corresponding to the candidate optimal parameter vector satisfies a preset convergence condition. The target optimal parameter vector is the identification result of the parameter to be identified corresponding to the parameter to be identified.

9. An electronic device, characterized in that, The device includes: a processor, a memory, and a system bus; The processor and the memory are connected via the system bus; The memory is used to store a program, the program including instructions that, when executed by the processor, cause the processor to perform the steps of the parameter identification method for a digital cable according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which, when executed by a terminal device, implements the steps of the parameter identification method for digital cables according to any one of claims 1 to 7.

Images