Particle swarm algorithm optimization method and system for predicting cable temperature field distribution

Abstract

The invention discloses a particle swarm optimization method and system for predicting cable temperature field distribution, and relates to the technical field of power system prediction.The method comprises the steps that a buried cable simulation model is built, and temperature field distribution data serve as a data set to be input into a neural network; optimizing the weight and threshold parameters of the neural network by using a particle swarm algorithm; and an error between the predicted temperature field and an actual measurement value is optimized by using a least square method. According to the method, the simulation model is constructed, the influence of multiple factors such as cable current-carrying capacity, environment temperature and soil thermal characteristics is comprehensively considered, and high-precision temperature field data is generated. The particle swarm optimization and the neural network are combined to adapt to the prediction problem of the buried cable temperature field with high dimension and strong coupling characteristics, final fitting is carried out in combination with the least square method, and the distribution characteristics of the buried cable temperature field are accurately captured.

Description

Technical Field

[0001] The invention relates to the technical field of power system prediction, and in particular to a particle swarm algorithm optimization method and system for predicting cable temperature field distribution. Background Art

[0002] With the acceleration of urbanization and the continuous improvement of urban infrastructure, direct buried cables, as an important way of urban power transmission, are crucial to ensuring the safety of urban power supply. During the operation of direct buried cables, due to the influence of various factors such as current flow, ambient temperature changes, and soil thermal properties, heat will accumulate inside the cable, resulting in uneven temperature field distribution. This uneven distribution of the temperature field may cause accelerated aging of cable insulation materials and even cause cable failures, affecting the reliability and safety of power supply. Predicting the temperature field distribution of direct buried cables is of great significance for realizing real-time monitoring of cables, preventing failures, and extending their service life. However, existing prediction technologies mostly rely on empirical formulas or simplified models, which cannot accurately reflect the complex underground environment and cable operation status, resulting in large errors in the prediction results and obvious deficiencies in processing large-scale data and real-time dynamic prediction. Summary of the invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the technical problem solved by the present invention is: how to accurately reflect the complex underground environment and cable operation status, reduce errors, process large-scale data and make real-time dynamic predictions.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a particle swarm algorithm optimization method for predicting cable temperature field distribution, comprising: building a direct buried cable simulation model, inputting temperature field distribution data as a data set into a neural network; using a particle swarm algorithm to optimize the weights and threshold parameters of the neural network; and using the least squares method to optimize the error between the predicted temperature field and the actual measured value.

[0006] As a preferred solution of the particle swarm algorithm optimization method for predicting cable temperature field distribution described in the present invention, the direct buried cable simulation model includes determining the number of input layer nodes of the neural network, the input variables are air thermal convection coefficient, cable current carrying capacity, soil temperature, air temperature, cable arrangement interval, determining the number of output layer nodes of the neural network, the output variable is temperature distribution, and determining the number of hidden layer nodes according to an empirical formula, which is expressed as follows:

[0007]

[0008] Among them, n His the number of hidden layer nodes, n1 is the number of input nodes, n2 is the number of output nodes, and n is an integer between 1 and 10.

[0009] As a preferred solution of the particle swarm algorithm optimization method for predicting cable temperature field distribution described in the present invention, the optimization of the particle swarm algorithm includes creating a group composed of multiple particles, each particle represents a set of potential solutions of the neural network, and the number of particles is n p , each particle is represented as a set of weights and thresholds P = {w, b}, where w is the weight matrix and b is the threshold vector;

[0010] For each particle p i Perform forward propagation and calculate the predicted output

[0011] And use the mean square error as the fitness function F;

[0012] Update the best position. For each particle, if F(p i )>F(p best,i ), then update the individual best position: p best,i =p i ; Find the particle with the best fitness value from all particles and update the global best position: g best =argmax(F(p best，i ));

[0013] Update the particle's velocity v i and position p i ;

[0014] Iterative optimization, repeat the steps forward to update the particle velocity v i and position p i , until the fitness value no longer increases significantly, select the global best position g best As the optimal weights and thresholds of the neural network.

[0015] As a preferred solution of the particle swarm algorithm optimization method for predicting cable temperature field distribution described in the present invention, the mean square error is expressed as:

[0016]

[0017] Update the particle's velocity v i and position p i It is expressed as:

[0018] v i (t+1)=w·v i (t)+c1·r 1i ·(p best，i -p i(t))+c2·r 2i ·(g best -p i (t))p i (t+1)=p i (t)+v i (t+1)

[0019] Where Y is the actual output, is the predicted output, n is the number of samples, w is the inertia weight, c1 and c2 are the personal and social acceleration coefficients, r 1i and r 2i are independent random numbers in the range [0,1].

[0020] As a preferred solution of the particle swarm algorithm optimization method for predicting cable temperature field distribution described in the present invention, wherein: the least squares optimization includes defining an error function, and selecting mean square error MSE as the error function;

[0021] Pass the input data forward through the neural network and calculate the output of each layer until the final predicted output y^ is obtained;

[0022] Use the chain rule to calculate the gradient of the error function MSE with respect to each weight W and threshold b;

[0023] Use gradient descent to update the network weights and thresholds;

[0024] Iterative optimization repeats the forward propagation step to update the network weights and thresholds until the error meets the set range.

[0025] As a preferred solution of the particle swarm algorithm optimization method for predicting cable temperature field distribution described in the present invention, the mean square error MSE is expressed as:

[0026]

[0027] Where N is the number of samples, y i is the actual output of the ith sample, is the output predicted by the model.

[0028] As a preferred solution of the particle swarm algorithm optimization method for predicting cable temperature field distribution described in the present invention, the gradient calculated by the chain rule is a vector whose components are the partial derivatives of the loss function with respect to each parameter:

[0029]

[0030] The weights and thresholds of the network updated by the gradient descent method are expressed as:

[0031]

[0032] Here, η is the learning rate, a hyperparameter that determines the step size of the parameter update in each iteration.

[0033] In the second aspect, another object of the present invention is to provide a remote data transmission prediction system for cable thermal degradation gas, comprising: a temperature field simulation and data generation module, a neural network prediction module and an error correction and optimization module; the temperature field simulation and data generation module is used to generate temperature field distribution data of direct buried cables, and provide multi-dimensional feature data as training input of the neural network; the neural network prediction module is used to adjust the weights and thresholds of the neural network using particle swarm optimization; the error correction and optimization module is used to correct the error between the neural network prediction value and the actual measurement value by least squares method.

[0034] In a third aspect, a computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the particle swarm algorithm optimization method for predicting the cable temperature field distribution as described above are implemented.

[0035] In a fourth aspect, a computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the particle swarm algorithm optimization method for predicting cable temperature field distribution as described above.

[0036] Beneficial effects of the present invention: The particle swarm algorithm optimization method and system for predicting the cable temperature field distribution provided by the present invention generates high-precision temperature field data by constructing a simulation model, comprehensively considering the influence of multiple factors such as cable current carrying capacity, ambient temperature, and soil thermal characteristics. The particle swarm algorithm is combined with a neural network to adapt to the prediction problem of the direct-buried cable temperature field, which has high dimensionality and strong coupling characteristics, and the least squares method is combined for final fitting to accurately capture the distribution characteristics of the direct-buried cable temperature field. By predicting the cable temperature field distribution and potential abnormal points in real time, the risk of insulation aging or failure can be prevented in advance, which helps to extend the service life of the cable and ensure the stability and safety of urban power transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0038] Figure 1 An overall flow chart of a particle swarm algorithm optimization method for predicting cable temperature field distribution provided by one embodiment of the present invention;

[0039] Figure 2 A prediction algorithm model structure diagram of a particle swarm algorithm optimization method for predicting cable temperature field distribution provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0042] Example 1

[0043] Reference Figure 1-Figure 2 , as an embodiment of the present invention, provides a particle swarm algorithm optimization method for predicting cable temperature field distribution, comprising:

[0044] S1: Build a direct buried cable simulation model and input the temperature field distribution data into the neural network as a data set;

[0045] Furthermore, COMSOL simulation is used to build a soil buried cable model to obtain temperature field distribution simulation data. COMSOL is an advanced numerical simulation software based on the finite element method that can simulate complex physical phenomena, including fluid flow, heat conduction, structural mechanics, electromagnetic analysis and other physical fields.

[0046] It should be noted that when building the cable geometry model, a single-core cable is selected as the simulation object. The single-core cable structure consists of a conductor layer, an insulation layer, a semiconductor layer, a metal sheath layer and an outer sheath. The material library provided by COMSOL is used to define the material properties of each layer, and the air thermal convection coefficient, cable current carrying capacity, soil temperature, air temperature, and cable arrangement spacing are set.

[0047] Furthermore, the number of nodes in the input layer of the neural network is determined to be 5 according to the needs, and the input variables are air thermal convection coefficient, cable current carrying capacity, soil temperature, air temperature, and cable arrangement interval;

[0048] According to the needs, the number of nodes in the output layer of the neural network is determined to be 1, and the output variables are temperature distribution;

[0049] The number of hidden layer nodes is determined according to the empirical formula, which is as follows:

[0050]

[0051] Where n H is the number of hidden layer nodes, n1 is the number of input nodes, n2 is the number of output nodes, n is an integer between 1 and 10, so the number of hidden layer nodes is determined to be 8.

[0052] S2: Use particle swarm algorithm to optimize the weights and threshold parameters of the neural network;

[0053] Furthermore, after determining the number of nodes in each layer of the neural network, the particle swarm algorithm is used to train the neural network model to obtain the optimal weights and thresholds. The specific steps are as follows:

[0054] (1) Create a group of particles, each particle represents a set of potential solutions of the neural network, and the number of particles is n p , each particle is represented as a set of weights and thresholds P = {w, b}, where w is the weight matrix and b is the threshold vector;

[0055] (2) For each particle p i , perform forward propagation and calculate the predicted output

[0056] (3) Use mean square error as fitness function F

[0057]

[0058] Where Y is the actual output, is the predicted output, n is the number of samples;

[0059] (4) Update the best position. For each particle, if F(p i )>F(p best,i ), then update the individual best position: p best,i =p i ; Find the particle with the best fitness value from all particles and update the global best position: g best =argmax(F(p best,i ));

[0060] (5) Update the particle's velocity v i and position p i :

[0061] v i (t+1)=w·v i (t)+c1·r 1i ·(P best,i -pi (t))+c2·r 2i ·(g best -p i (t))p i (t+1)

[0062] =p i (t)+v i (t+1)

[0063] Where w is the inertia weight, c1 and c2 are the personal and social acceleration coefficients, and r 1i and r 2i are independent random numbers in the range [0,1].

[0064] (6) Iterative optimization, repeating steps (2) to (5) until the fitness value no longer increases significantly.

[0065] (7) Select the global optimal position g best As the optimal weights and thresholds of the neural network.

[0066] S3: Use the least squares method to optimize the error between the predicted temperature field and the actual measured value.

[0067] Furthermore, the least square method is used to minimize the error between the predicted temperature field and the actual measured value, further improving the accuracy of the prediction model.

[0068] (1) Define the error function and select the mean square error MSE as the error function:

[0069]

[0070] Where N is the number of samples, y i is the actual output of the ith sample, is the output predicted by the model.

[0071] (2) The input data is forward propagated through the neural network and the output of each layer is calculated until the final predicted output y^ is obtained.

[0072] (3) Use the chain rule to calculate the gradient of the error function MSE with respect to each weight W and threshold b. The gradient is a vector whose components are the partial derivatives of the loss function with respect to each parameter:

[0073]

[0074] (4) Use gradient descent to update the network weights and thresholds to reduce errors:

[0075]

[0076] Here, η is the learning rate, a hyperparameter that determines the step size of the parameter update in each iteration.

[0077] (5) Iterative optimization, repeating steps (2) to (4) until the error meets the set range.

[0078] Example 2

[0079] One embodiment of the present invention provides a particle swarm algorithm optimization system for predicting cable temperature field distribution, including: a temperature field simulation and data generation module, a neural network prediction module and an error correction and optimization module; the temperature field simulation and data generation module is used to generate temperature field distribution data of direct buried cables, and provide multi-dimensional feature data as training input of the neural network; the neural network prediction module is used to adjust the weights and thresholds of the neural network using particle swarm optimization; the error correction and optimization module is used to correct the error between the neural network prediction value and the actual measurement value through the least squares method.

[0080] Example 3

[0081] An embodiment of the present invention is different from the first two embodiments in that:

[0082] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0083] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0084] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0085] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0086] Example 4

[0087] An embodiment of the present invention provides a particle swarm algorithm optimization method for predicting cable temperature field distribution. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0088] A direct buried cable laying model was built, with a soil width of 40.5m, a soil height of 21m, a cable buried horizontally in the middle, and a buried depth of 1m. After meshing, setting the electromagnetic thermal coupling physical field, and setting the boundary conditions, the temperature field simulation results were calculated as shown in the following table:

[0089]

[0090] In the experiment, after optimization, the maximum error was reduced from 4.85℃ of the empirical formula to 0.85℃, and the average error was reduced from 2.13℃ without optimization to 0.31℃, reflecting the system's significant error optimization capability. Although the calculation time increased slightly, the system achieved a good balance between high-precision prediction and practical application. Through optimization and correction methods, the system's adaptability to complex environments was significantly improved, and it had higher prediction accuracy and reliability.

Claims

1. A particle swarm algorithm optimization method for predicting cable temperature field distribution, characterized in that: include: A direct buried cable simulation model was built and the temperature field distribution data was input into the neural network as a data set; The particle swarm algorithm is used to optimize the weights and threshold parameters of the neural network; The error between the predicted temperature field and the actual measured value is optimized using the least squares method.

2. The particle swarm algorithm optimization method for predicting cable temperature field distribution according to claim 1, characterized in that: The direct buried cable simulation model includes determining the number of input layer nodes of the neural network, the input variables are air thermal convection coefficient, cable current carrying capacity, soil temperature, air temperature, cable arrangement interval, determining the number of output layer nodes of the neural network, the output variable is temperature distribution, and determining the number of hidden layer nodes according to an empirical formula, which is expressed as follows: Among them, n H is the number of hidden layer nodes, n1 is the number of input nodes, n2 is the number of output nodes, and n is an integer between 1 and 10.

3. The particle swarm algorithm optimization method for predicting cable temperature field distribution according to claim 2, characterized in that: The optimization of the particle swarm algorithm involves creating a group consisting of multiple particles, each particle represents a set of potential solutions of the neural network, and the number of particles is n p , each particle is represented as a set of weights and thresholds P = {w, b}, where w is the weight matrix and b is the threshold vector; For each particle p i Perform forward propagation and calculate the predicted output And use the mean square error as the fitness function F; Update the best position. For each particle, if F(p i )>F(p best,i ), then update the individual best position: p best,i =p i ; Find the particle with the best fitness value from all particles and update the global best position: g best =argmax(F(p best,i )); Update the particle's velocity v i and position p i ; Iterative optimization, repeat the steps forward to update the particle velocity v i and position p i , until the fitness value no longer increases significantly, select the global best position g best As the optimal weights and thresholds of the neural network.

4. The particle swarm algorithm optimization method for predicting cable temperature field distribution according to claim 3, characterized in that: The mean square error is expressed as: Update the particle's velocity v i and position p i It is expressed as: v i (t+1)=w·v i (t)+c1·r 1i ·(p best,i -p i (t))+c2·r 2i ·(g best -p i (t))p i (t+1)=p i (t)+v i (t+1) Where Y is the actual output, is the predicted output, n is the number of samples, w is the inertia weight, c1 and c2 are the personal and social acceleration coefficients, r 1i and r 2i are independent random numbers in the range [0,1].

5. The particle swarm algorithm optimization method for predicting cable temperature field distribution according to claim 4, characterized in that: The least squares optimization includes defining an error function and selecting mean square error MSE as the error function; Pass the input data forward through the neural network and calculate the output of each layer until the final predicted output y^ is obtained; Use the chain rule to calculate the gradient of the error function MSE with respect to each weight W and threshold b; Use gradient descent to update the network weights and thresholds; Iterative optimization repeats the forward propagation step to update the network weights and thresholds until the error meets the set range.

6. The particle swarm algorithm optimization method for predicting cable temperature field distribution according to claim 5, characterized in that: The mean square error MSE is expressed as: Where N is the number of samples, y i is the actual output of the ith sample, is the output predicted by the model.

7. The particle swarm algorithm optimization method for predicting cable temperature field distribution according to claim 6, characterized in that: The gradient calculated by the chain rule is a vector whose components are the partial derivatives of the loss function with respect to each parameter: The weights and thresholds of the network updated by the gradient descent method are expressed as: Here, η is the learning rate, a hyperparameter that determines the step size of the parameter update in each iteration.

8. A system using the particle swarm algorithm optimization method for predicting cable temperature field distribution as claimed in any one of claims 1 to 7, characterized in that: include: Temperature field simulation and data generation module, neural network prediction module and error correction and optimization module; The temperature field simulation and data generation module is used to generate temperature field distribution data of the direct buried cable and provide multi-dimensional feature data as training input of the neural network; The neural network prediction module is used to adjust the weights and thresholds of the neural network using particle swarm optimization; The error correction and optimization module is used to correct the error between the neural network prediction value and the actual measurement value through the least square method.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the particle swarm algorithm optimization method for predicting cable temperature field distribution according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the particle swarm algorithm optimization method for predicting cable temperature field distribution according to any one of claims 1 to 7 are implemented.

Images