CN114861542A - Method, device and equipment for evaluating loss of direct current transmission project and storage medium - Google Patents

Method, device and equipment for evaluating loss of direct current transmission project and storage medium Download PDF

Info

Publication number
CN114861542A
CN114861542A CN202210527974.XA CN202210527974A CN114861542A CN 114861542 A CN114861542 A CN 114861542A CN 202210527974 A CN202210527974 A CN 202210527974A CN 114861542 A CN114861542 A CN 114861542A
Authority
CN
China
Prior art keywords
loss
engineering
direct current
training
current transmission
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210527974.XA
Other languages
Chinese (zh)
Inventor
黄之笛
彭光强
何竞松
武霁阳
毛炽祖
陈礼昕
国建宝
杨光源
夏谷林
王海军
谢惠藩
李士杰
杨育丰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Maintenance and Test Center of Extra High Voltage Power Transmission Co
Original Assignee
Maintenance and Test Center of Extra High Voltage Power Transmission Co
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Maintenance and Test Center of Extra High Voltage Power Transmission Co filed Critical Maintenance and Test Center of Extra High Voltage Power Transmission Co
Priority to CN202210527974.XA priority Critical patent/CN114861542A/en
Publication of CN114861542A publication Critical patent/CN114861542A/en
Priority to PCT/CN2022/131882 priority patent/WO2023221426A1/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/04—Power grid distribution networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/06—Power analysis or power optimisation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The application relates to the field of power loss calculation, and provides a method and a device for evaluating loss of a direct current transmission project, computer equipment, a storage medium and a computer program product, which can effectively improve the evaluation efficiency of loss of the direct current transmission project. The method comprises the following steps: acquiring engineering working condition parameters and engineering line information of a direct current transmission project to be evaluated; inputting engineering working condition parameters and engineering line information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples; and obtaining the loss information of the direct current transmission project output by the loss evaluation model, and determining the loss evaluation result of the direct current transmission project according to the loss information.

Description

Method, device and equipment for evaluating loss of direct current transmission project and storage medium
Technical Field
The present application relates to the field of power loss calculation, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for estimating loss in a dc power transmission project.
Background
With the continuous development of power grid technology, direct current transmission has the characteristics of low loss and high stability compared with alternating current transmission, and is widely applied to long-distance transmission engineering. However, as the voltage level and the capacity of direct current transmission are improved, the loss problem in the transmission process is more prominent.
In the related art, the loss of the direct current engineering power transmission engineering can be predicted through a simulation modeling method and an analysis algorithm. The method comprises the following steps of establishing a complete direct current engineering system model by a simulation modeling method, and obtaining losses of different positions in the engineering by means of a real-time simulation curve; and the analytic algorithm can quickly obtain the integral loss of the direct current engineering through a connection and disconnection formula algorithm.
However, the simulation modeling method has a long period for building a direct current engineering system model, often needs to perform iterative computation for many times, and a single iteration needs to consume a long time, and in practice, the direct current transmission engineering has complex operation condition changes, and if the loss under various operation conditions needs to be simulated, a large amount of time needs to be consumed. Although the analysis algorithm can quickly obtain the overall loss of the direct current transmission project, a certain error often exists between the result and the actual situation. Therefore, the related technology is difficult to consider both the calculation time and the calculation accuracy when evaluating the loss of the direct current transmission project, and the overall evaluation efficiency is low.
Disclosure of Invention
In view of the above, it is necessary to provide a method, an apparatus, a computer device, a computer readable storage medium, and a computer program product for estimating loss in dc power transmission engineering.
In a first aspect, the application provides a loss assessment method for a direct current transmission project. The method comprises the following steps:
acquiring engineering working condition parameters and engineering line information of a direct current transmission project to be evaluated;
inputting the engineering working condition parameters and the engineering line information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
and obtaining the loss information of the direct current transmission project output by the loss evaluation model, and determining the loss evaluation result of the direct current transmission project according to the loss information.
In one embodiment, the method further comprises the following steps:
acquiring a training sample set, wherein the training sample set comprises training working condition parameters and training line information of a plurality of different first direct current transmission engineering samples, and an engineering loss label of each first direct current transmission engineering sample;
inputting training working condition parameters and training circuit information of a plurality of first direct current power transmission engineering samples in the training sample set into a neural network model to be trained, and outputting predicted loss information of each first direct current power transmission engineering sample by the neural network model based on the training working condition parameters of the first direct current power transmission engineering sample;
and adjusting the model parameters of the neural network model according to the predicted loss information and the engineering loss label until the training end condition is met, so as to obtain a trained loss evaluation model.
In one embodiment, the adjusting, according to the predicted loss information and the engineering loss label, model parameters of the neural network model until a training end condition is satisfied to obtain a trained loss evaluation model includes:
determining a loss value according to the predicted loss information of the plurality of first direct current transmission engineering samples and the engineering loss label of each first direct current transmission engineering sample;
and adjusting the model parameters of the neural network model according to the loss value and a back propagation algorithm until the training end condition is met, so as to obtain a trained loss evaluation model.
In one embodiment, the adjusting the model parameters of the neural network model to be trained according to the loss value and the back propagation algorithm until a training end condition is satisfied to obtain a trained loss evaluation model includes:
adjusting model parameters of the neural network model according to the loss value and a back propagation algorithm to obtain an adjusted neural network model;
obtaining a test sample set; the test sample set comprises training working condition parameters and training line information of a plurality of second direct current transmission engineering samples and an engineering loss label of each second direct current transmission engineering sample;
inputting training working condition parameters and training circuit information of a plurality of second direct current power transmission engineering samples in the test sample set into the neural network model, and outputting predicted loss information of each second direct current power transmission engineering sample by the neural network model based on the training working condition parameters and the training circuit information of the second direct current power transmission engineering sample;
determining a training error according to the predicted loss information of the plurality of second direct current transmission engineering samples and the engineering loss labels of the second direct current transmission engineering samples;
and when the training error is larger than a preset error threshold value, returning to the step of obtaining the training sample set, and adjusting the parameters of the neural network model again until the training error is not larger than the preset error threshold value to obtain the trained loss evaluation model.
In one embodiment, the method further comprises the following steps:
acquiring training working condition parameters and training circuit information of a plurality of direct current transmission engineering samples; the plurality of direct current transmission engineering samples are direct current transmission engineering samples with different training working condition parameters and/or line information;
determining loss information of each direct current power transmission engineering sample according to training working condition parameters, line information and a preset loss algorithm of each direct current power transmission engineering sample, and taking the loss information of each direct current power transmission engineering sample as an engineering loss label of the direct current power transmission engineering sample;
and acquiring a training sample set and a test sample set according to the plurality of direct current transmission engineering samples and the engineering loss labels thereof.
In one embodiment, the obtaining of the engineering condition parameters of the dc power transmission engineering to be evaluated includes:
acquiring power information, the number of full half-bridge sub-modules and fundamental frequency of a converter station in the direct current transmission project to be evaluated;
and determining engineering working condition parameters of the direct current transmission engineering according to the power information, the number of full half-bridge sub-modules and the fundamental frequency.
In a second aspect, the application further provides a device for evaluating the loss of the direct current transmission engineering. The device comprises:
the input information acquisition module is used for acquiring engineering working condition parameters and engineering line information of the direct current transmission engineering to be evaluated;
the input module is used for inputting the engineering working condition parameters and the engineering line information into a pre-trained loss evaluation model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
and the loss evaluation result acquisition module is used for acquiring the loss information of the direct current transmission project output by the loss evaluation model and determining the loss evaluation result of the direct current transmission project according to the loss information.
In a third aspect, the present application also provides a computer device comprising a memory storing a computer program and a processor implementing the steps of the method as described in any one of the above when the processor executes the computer program.
In a fourth aspect, the present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the method according to any one of the above.
In a fifth aspect, the present application also provides a computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method as described in any of the above.
According to the direct-current transmission project loss evaluation method, the direct-current transmission project loss evaluation device, the computer equipment, the storage medium and the computer program product, the project working condition parameters and the project line information of the direct-current transmission project to be evaluated are obtained, and the project working condition parameters and the project line information are input into a pre-trained loss evaluation model, wherein the loss evaluation model is used for determining the loss information of the direct-current transmission project based on the project working condition parameters and the project line information, and is obtained by training the neural network model based on the training working condition parameters, the training line information and the project loss labels of a plurality of different direct-current transmission project samples; and the terminal can further obtain the loss information of the direct current transmission project output by the loss evaluation model, and determine the loss evaluation result of the direct current transmission project according to the loss information. According to the method, the neural network model is trained by using the plurality of direct current transmission engineering samples to obtain the loss assessment model, the loss assessment model can rapidly learn the working condition parameters of different direct current transmission engineering and the mapping relation between the engineering lines and the direct current transmission engineering loss, so that the parameters do not need to be modified manually under the condition of changing the working condition and the lines, the accuracy of loss information can be ensured while the loss information is rapidly acquired, and the assessment efficiency of the direct current transmission engineering loss is effectively improved.
Drawings
Fig. 1 is a schematic flow chart of a loss estimation method in a dc transmission project according to an embodiment;
FIG. 2 is a diagram illustrating a neural network model according to one embodiment;
FIG. 3 is a schematic flow chart illustrating testing a neural network model in one embodiment;
fig. 4 is a block diagram of a loss evaluation apparatus in an embodiment of a dc power transmission project;
FIG. 5 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In order to facilitate understanding of the embodiments of the present application, a method for evaluating loss in dc transmission engineering in the related art is described first.
With the continuous development of power grid technology, direct current transmission has the characteristics of low loss and high stability compared with alternating current transmission, and is widely applied to long-distance transmission engineering. In a direct current transmission project, a flexible direct current transmission system gradually becomes a mainstream by virtue of excellent output characteristics, low switching loss and other advantages, and the problem of converter loss of a converter station in the direct current transmission project is more prominent along with the improvement of the voltage grade and the capacity of the flexible direct current transmission.
In the related technology, the loss of the direct current engineering power transmission engineering can be predicted through a simulation modeling method and an analysis algorithm. The simulation modeling method mainly comprises the steps of building a complete direct current engineering system model on a simulation platform (such as PSCAD (power system computer aided design) and MATLAB (matrix laboratory)/Simulink, obtaining losses at different positions in a project by means of a real-time simulation curve, wherein the accuracy of a loss result is in direct proportion to the simulation precision of a simulation program, more accurate results can be generated by more comprehensive and accurate simulation, but because the period for building the simulation system is too long, iterative computation needs to be carried out for many times, and long time is consumed for a single time, the simulation method is only suitable for small project simulation. Moreover, the actual direct current transmission project usually has complex operation condition changes, a simulation modeling method needs to consume a large amount of time if all conditions under different working conditions are to be simulated, the requirement on computer hardware configuration is high, different operation lines can exist in the direct current transmission project, different parameters need to be changed for different lines through an analysis algorithm, and the calculation process is complex. Therefore, if a simulation modeling method is used for simulating and calculating the loss of a large direct current transmission project, a large amount of time is consumed, the requirement on hardware is high, and if multiple working conditions and the loss of different lines are required in a project period, parameters need to be changed and simulation steps need to be repeated, so that the time consumption is extremely long.
The analytical algorithm is mainly used for calculating the loss of the converter valve in the direct current project, the final loss can be obtained by calculating the loss of each part of the converter valve and finally adding the loss, and the total loss of the direct current transmission project can also be obtained by calculating the loss of each project part in the direct current transmission project and then adding the loss. Under the fixed power, the overall loss of the direct current engineering can be quickly obtained through an analytical formula algorithm, but the problems of sampling of pulse waveforms, nonlinear impedance and the like are not considered due to the fact that average bridge arm current is adopted for input of the analytical algorithm, and a calculation result has certain errors. Therefore, although the overall time consumption of the analysis algorithm is slightly shorter than that of the simulation modeling method, the algorithm needs to be adjusted under the conditions of different working conditions and different lines, the calculation process is complex, data needs to be repeatedly changed if different powers and line losses are needed, and the method is complex and prone to human errors.
Therefore, in the related technology, the calculation time and the calculation accuracy are difficult to be considered when the loss of the direct current transmission project is evaluated, and only one group of calculation examples can be calculated once. Based on the above, the application provides a loss assessment method for direct current transmission engineering.
In an embodiment, as shown in fig. 1, a method for evaluating loss in dc transmission engineering is provided, and this embodiment is illustrated by applying the method to a terminal, it is to be understood that the method may also be applied to a server, and may also be applied to a system including a terminal and a server, and is implemented by interaction between the terminal and the server. In this embodiment, the method includes the steps of:
s101, acquiring engineering working condition parameters and engineering line information of the direct current transmission engineering to be evaluated.
As an example, the direct current transmission project to be evaluated may be any flexible direct current project, and the project working condition parameter may be a working parameter of each project component in the operation process of the direct current transmission project; the engineering line information may be a transmission line corresponding to a dc transmission engineering, for example, the dc transmission engineering to be evaluated may be for an a-B line, or may also be another dc transmission engineering for a B-C line.
In specific implementation, a user can input the engineering working condition parameters and the engineering line information of the direct current transmission project to be evaluated at the terminal, and the terminal can acquire the engineering working condition parameters and the engineering line information of the direct current transmission project to be evaluated in response to the detected trigger operation.
S102, inputting engineering working condition parameters and engineering circuit information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training the neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct current transmission engineering samples.
As an example, the loss information of the dc power transmission project may be a specific loss value, such as the power loss of the whole dc power transmission project, or may include the loss of each project component in the dc power transmission project, and may include, for example, the loss of at least one of the following project components: valve bank loss, converter transformer loss, direct current reactor loss, alternating current filter loss, direct current filter loss, valve cooling system loss, bridge arm reactor loss, converter transformer loss and transmission line loss. Alternatively, the loss information may be a loss interval, for example, the loss of the dc transmission project is between a% and b%.
In practical application, training condition parameters, training circuit information and engineering loss labels of a plurality of different direct current transmission engineering samples can be adopted to train a neural network model in advance, and a trained loss evaluation model is obtained. And then after the engineering working condition parameters and the engineering line information are obtained, the engineering working condition parameters and the engineering line information can be input into a pre-trained loss assessment model, and the loss information of the direct current transmission engineering is determined by the model based on the input engineering working condition parameters and the engineering line information.
Specifically, in the training process, because a large number of direct current transmission engineering samples are used for training the neural network model, the model can learn the combination of all engineering working condition parameters and engineering lines of the direct current transmission engineering during training, determine one or more loss characteristics which affect the loss of the direct current transmission engineering in the engineering working condition parameters and the engineering lines under different conditions, and endow corresponding weight to the one or more loss characteristics. After training is finished, the trained neural network model, namely the loss evaluation model can acquire the engineering working condition parameters and the mapping relation between the engineering line information and the direct-current transmission engineering loss information, so that under the condition of changing the working condition and the line, the parameters do not need to be modified manually, and accurate loss information is acquired quickly.
And S103, obtaining the loss information of the direct current transmission project output by the loss evaluation model, and determining the loss evaluation result of the direct current transmission project according to the loss information.
In practical application, after the loss evaluation model inputs the loss information of the direct current transmission project, the terminal can obtain the loss information and determine the loss evaluation result of the direct current transmission project according to the loss information. Specifically, the terminal may directly use the loss information of the dc transmission project as the loss evaluation result of the dc transmission project, or may perform a qualitative or quantitative evaluation on the loss condition of the dc transmission project according to the loss information, such as an evaluation level or an evaluation score, and further generate an evaluation report in combination with the evaluation and output the evaluation report as the loss evaluation result.
In this embodiment, engineering condition parameters and engineering line information of a dc power transmission engineering to be evaluated may be obtained, and the engineering condition parameters and the engineering line information are input to a loss evaluation model trained in advance, where the loss evaluation model is used to determine loss information of the dc power transmission engineering based on the engineering condition parameters and the engineering line information, and the loss evaluation model is obtained by training a neural network model based on training condition parameters, training line information, and engineering loss labels of a plurality of different dc power transmission engineering samples; and the terminal can further obtain the loss information of the direct current transmission project output by the loss evaluation model, and determine the loss evaluation result of the direct current transmission project according to the loss information. According to the method, the neural network model is trained by using the plurality of direct current transmission engineering samples to obtain the loss assessment model, the loss assessment model can rapidly learn the working condition parameters of different direct current transmission engineering and the mapping relation between the engineering lines and the direct current transmission engineering loss, so that the parameters do not need to be modified manually under the condition of changing the working condition and the lines, the accuracy of loss information can be ensured while the loss information is rapidly acquired, and the assessment efficiency of the direct current transmission engineering loss is effectively improved.
In an optional embodiment, the loss evaluation model may be packaged as an interface and embedded in a monitoring interface of the dc power transmission project, and when a user needs to evaluate the loss of the dc power transmission project under different working conditions or lines, engineering working condition parameters and engineering line information may be input, and the interface may be called, so that the loss information output by the loss evaluation model for the dc power transmission project may be quickly and conveniently obtained.
In an exemplary embodiment, the method may further include the steps of:
s201, a training sample set is obtained, wherein the training sample set comprises training working condition parameters and training line information of a plurality of different first direct current transmission engineering samples, and an engineering loss label of each first direct current transmission engineering sample.
The direct-current transmission engineering samples are used as direct-current transmission engineering of a sample training neural network model, and in order to distinguish direct-current transmission engineering samples in a training sample set from direct-current transmission engineering samples in a test sample set in the following, the direct-current transmission engineering samples in the training sample set are called as first direct-current transmission engineering samples.
In practical application, before acquiring engineering working condition parameters and engineering line information of a direct current transmission project to be evaluated, a terminal can train a neural network model. Specifically, a training sample set can be obtained, the training sample set comprises a large number of first direct current transmission engineering samples, the first direct current transmission engineering samples can be different, and therefore direct current transmission engineering under different working conditions and lines can be covered, meanwhile, the training sample set can be used for associating engineering loss labels for each first direct current transmission engineering, and the engineering loss labels record loss of the first direct current transmission engineering samples in the operation process.
Illustratively, the training condition parameters, training route information, and engineering loss labels of a plurality of first direct current transmission engineering samples in the training sample set may have the following structures as shown in table 1 below:
TABLE 1
Figure BDA0003645352210000091
The input data 1-1, 1-2 … 1-i are i pieces of input information of the first direct current transmission engineering sample, wherein the input information includes training condition parameters and training line information of the first direct current transmission engineering sample, such as converter station active power, reactive power, full half-bridge submodule number, fundamental frequency and the like; the output data is an engineering loss label.
S202, inputting training working condition parameters and training circuit information of a plurality of first direct current transmission engineering samples in a training sample set into a neural network model to be trained, and outputting predicted loss information of the first direct current transmission engineering samples by the neural network model based on the training working condition parameters of each first direct current transmission engineering sample.
And S203, adjusting model parameters of the neural network model according to the predicted loss information and the engineering loss label until the training end condition is met, and obtaining a trained loss evaluation model.
After the training sample set is obtained, training working condition parameters and training circuit information of a plurality of first direct current power transmission engineering samples in the training sample set can be input into a neural network model to be trained, the neural network model outputs predicted loss information of the first direct current power transmission engineering samples based on the training working condition parameters of each first direct current power transmission engineering sample, and model parameters of the neural network model are adjusted according to the predicted loss information and an engineering loss label until a training end condition is met, so that a trained loss evaluation model is obtained.
Specifically, the neural network model in this embodiment may be a Back-propagation (BP) neural network model, the BP neural network model is a multi-layer neural network, the structure of the BP neural network model includes an input layer, a hidden layer and an output layer, neurons in each layer are independent, neurons between layers are linked with each other through neural lines, the neural lines may be given different weights, in the process of training the neural network, training condition parameters and training line information of a first dc power transmission engineering sample input to the neural network model are propagated backwards through the input layer and the hidden layer, along with continuous change of an input signal, the neural network may continuously update weights and thresholds of the network model, and errors of a final result are continuously reduced through self-learning.
Fig. 2 shows a structure of a neural network model, in practical applications, a plurality of first dc power transmission engineering samples x1, x2, x3 … … xj may be input through an input layer of the neural network model, and data is input into a hidden layer in the model through forward propagation.
Illustratively, the ith node (also called neuron) f of the hidden layer i Input1 i Comprises the following steps:
Figure BDA0003645352210000101
wherein, w ij For input layer ith node, input x j Assigned weight, θ i Is the threshold for that node. Hidden layer ith node f i Output o of i Comprises the following steps:
o i ＝f(input1 i )
input2 of kth node of output layer k Comprises the following steps:
Figure BDA0003645352210000102
wherein w ki For the k-th node of the output layer to input o i Assigned weight, a k Is the threshold for that node.
Output o of kth node of output layer k Comprises the following steps:
Figure BDA0003645352210000103
wherein for computing the node output o i And for calculating the output layer output o k Function of (2)
Figure BDA0003645352210000104
May be an activation function.
In the embodiment, by obtaining a training sample set, training working condition parameters and training circuit information of a plurality of first direct current transmission engineering samples in the training sample set are input into a neural network model to be trained, the neural network model outputs predicted loss information of each first direct current transmission engineering sample based on the training working condition parameters of the first direct current transmission engineering sample, further, according to the predicted loss information and the engineering loss label, model parameters of the neural network model are adjusted until the training end condition is met, a trained loss evaluation model is obtained, the neural network model can learn the loss of the direct current transmission engineering under various working conditions and lines, further fitting the mapping relation among the working condition parameters, the line information and the direct current transmission engineering loss, therefore, the loss of the direct current transmission project under different working conditions or lines can be quickly obtained subsequently under the condition that parameters are not required to be modified.
In an exemplary embodiment, the step S203 of adjusting model parameters of the neural network model according to the predicted loss information and the engineering loss label until a training end condition is satisfied to obtain a trained loss evaluation model, which may include the following steps:
determining a loss value according to the predicted loss information of the plurality of first direct current transmission engineering samples and the engineering loss label of each first direct current transmission engineering sample; and adjusting model parameters of the neural network model according to the loss value and a back propagation algorithm until a training end condition is met, and obtaining a trained loss evaluation model.
In practical application, after the predicted loss information of the plurality of first direct current transmission engineering samples is obtained, for each first direct current transmission engineering sample, the predicted loss information of the first direct current transmission engineering sample may be compared with the engineering loss label of the first direct current transmission engineering sample to determine a difference between the predicted loss information and the engineering loss label, and then the loss value of the neural network model may be determined according to the difference between the plurality of first direct current transmission engineering samples and the engineering loss label thereof.
After obtaining the loss value, the model parameters of the neural network model can be adjusted according to the loss value and a back propagation algorithm, in other words, the model parameters of the neural network model can be adjusted correspondingly by backward propagation of the loss value (also called error) and an engineering loss label (expected value), in the process of adjusting the model parameters, the weight and the threshold in the model parameters are adjusted by calculating the error between the output (namely predicted loss information) and the actual (engineering loss label) of all neurons in an output layer, in the engineering, the error can be utilized to ensure that the neural network determines how to correct the weight and the threshold of each layer of neurons in the model, and then different first direct current transmission engineering samples are used for repeated training until the number of iterations or the error between the predicted loss information and the engineering label of the first direct current transmission engineering samples is lower than a preset threshold, it may be determined that training is complete.
In this embodiment, a loss value may be determined according to predicted loss information of a plurality of first dc transmission engineering samples and an engineering loss label of each first dc transmission engineering sample, a model parameter of a neural network model is adjusted according to the loss value and a back propagation algorithm until a training end condition is satisfied, a trained loss evaluation model is obtained, a back propagation algorithm (i.e., error back propagation) training is performed, a nonlinear function is fitted, and in a dc transmission engineering (e.g., a flexible dc transmission engineering), influences of changes of a working condition and a line on a dc transmission engineering loss are also nonlinear, so that a nonlinear relationship between an engineering working condition, a line and a loss can be fitted according to the loss value and in combination with the back propagation algorithm, and rapid estimation of losses of the dc transmission engineering under different working conditions or lines is realized.
In one embodiment, adjusting the model parameters of the neural network model according to the loss values and the back propagation algorithm until the training end condition is satisfied to obtain a trained loss assessment model, may include the following steps:
s301, adjusting model parameters of the neural network model according to the loss value and the back propagation algorithm to obtain the adjusted neural network model.
In specific implementation, model parameters of the neural network model can be adjusted according to the loss value and the back propagation algorithm, and when the iteration times meet the preset times or the error between the predicted loss information of the first direct current transmission engineering sample and the engineering label is lower than a preset threshold value, the current neural network model can be used as the adjusted neural network model.
S302, obtaining a test sample set; the test sample set comprises training working condition parameters and training line information of a plurality of second direct current transmission engineering samples and an engineering loss label of each second direct current transmission engineering sample.
And the direct current transmission project samples in the test sample set are called second direct current transmission project samples.
In practical application, a test sample set can be constructed in advance, and the test sample set can be used for testing the trained neural network model and verifying whether the training of the current neural network model is completed. After one or more training sessions are completed using the data in the training sample set, the trained neural network model may be tested for accuracy using the data in the test sample set.
Illustratively, the training condition parameters, training route information, and engineering loss labels of a plurality of first direct current transmission engineering samples in the training sample set may have the following structures as shown in table 2 below:
TABLE 2
Figure BDA0003645352210000121
The input data m +1-1, m +1-2 … m +1-i are i pieces of input information of the first direct current transmission engineering sample, wherein training working condition parameters and training line information of the second direct current transmission engineering sample are included, such as converter station active power, reactive power, the number of full half-bridge submodules, fundamental frequency and the like; and outputting the data as an engineering loss label of the second direct current transmission engineering sample.
And S303, inputting the training working condition parameters and the training circuit information of a plurality of second direct current power transmission engineering samples in the test sample set into a neural network model, and outputting the predicted loss information of each second direct current power transmission engineering sample by the neural network model based on the training working condition parameters and the training circuit information of each second direct current power transmission engineering sample.
And S304, determining training errors according to the predicted loss information of the plurality of second direct current transmission engineering samples and the engineering loss labels of the second direct current transmission engineering samples.
After the test sample set is obtained, training working condition parameters and training circuit information of a plurality of second direct current power transmission engineering samples in the test sample set can be input into a neural network model to be trained, the neural network model outputs predicted loss information of each second direct current power transmission engineering sample based on the training working condition parameters of the second direct current power transmission engineering sample, and then training errors can be determined according to the predicted loss information and the engineering loss labels of the second direct current power transmission engineering samples. For example, when determining the training error, the training errors of the plurality of second direct current transmission engineering samples may be summed to obtain a training error sum D sum And summing the training errors D sum Average value D of mean As the final determined training error.
S305, when the training error is larger than the preset error threshold, returning to the step of obtaining the training sample set, and adjusting the parameters of the neural network model again until the training error is not larger than the preset error threshold, so as to obtain the trained loss evaluation model.
As an example, the error threshold may be set according to different dc transmission engineering, for example, may be set to 1%.
After the training error is obtained, whether the training error is larger than a preset error threshold value or not can be judged, when the training error is larger than the preset error threshold value, the fact that the accuracy of the currently trained neural network model is low is indicated, the step of obtaining the training sample set can be returned again, the data of the training sample set and the data of the testing sample set are updated again, the steps are repeated, and the neural network model continues to be trained. And until the adjusted neural network model is tested by using a new test sample set and the obtained training error is not greater than the error threshold value, the model error of the current neural network model is within an acceptable range, and the current neural network model can be used as a trained loss evaluation model.
For example, as shown in FIG. 3, a set of test samples and a set of training samples may be obtained. Aiming at a training sample set, a first direct current transmission engineering sample in the training sample set can be input into a neural network model after model parameter initialization, training is started, after iteration is carried out for multiple times, the current neural network model is used as an adjusted neural network model, a second direct current transmission engineering sample in a test sample set is input into the neural network model, the accuracy of the neural network model is verified, whether the model is close to an engineering loss label or not can be judged according to a model output result, if yes, a loss evaluation model can be obtained, and if not, the neural network model continues to be trained.
After the training is determined to be completed, the training data may be further encapsulated into an application module, so that a user subsequently invokes the application module to obtain loss information of the dc transmission project to be evaluated, for example, the operating condition parameters and the line information of the dc transmission project are input into a trained neural network model (i.e., a loss evaluation model), and even if the operating condition parameters and the line information are not calculated by an algorithm, the trained neural network model may also give an interval of approximately accurate results, for example, if a training sample set and a test sample set only include a plurality of loss results with active powers of 800MW, 900MW, 1000MW, 3500MW, and 5000MW under a specific operating condition that is simulated earlier, for example: in the A-B line, the loss under 2000MW active power does not need to change parameters in a time-consuming manner, but the working condition parameters and the line information of the direct current transmission project to be evaluated can be directly input into the neural network model, so that accurate loss information can be quickly obtained.
In this embodiment, the adjusted neural network model is tested by using the second direct current transmission engineering sample in the test sample set, so that a real use scene can be simulated to verify the accuracy of the current neural network model, training is continued when the training error is greater than a preset error threshold, and the accuracy and reliability of the finally obtained loss assessment model are effectively improved.
In an embodiment, the method may further comprise the steps of:
acquiring training working condition parameters and training circuit information of a plurality of direct current transmission engineering samples; determining loss information of each direct current power transmission engineering sample according to training working condition parameters, line information and a preset loss algorithm of each direct current power transmission engineering sample, and taking the loss information of each direct current power transmission engineering sample as an engineering loss label of the direct current power transmission engineering sample; and acquiring a training sample set and a test sample set according to the plurality of direct current transmission engineering samples and the engineering loss labels thereof.
The plurality of direct current transmission engineering samples are direct current transmission engineering samples with different training working condition parameters and/or line information.
In practical application, a large number (e.g. 10000) of working condition parameters and line information of the direct current transmission engineering samples can be collected and used as training working condition parameters and training line information of the corresponding direct current transmission engineering samples. When the training condition parameters and the training line information of the direct current transmission engineering samples are obtained, a preset rapid loss algorithm can be adopted, the loss information of the direct current transmission engineering samples is determined according to the training condition parameters and the training line information, and the loss information is used as an engineering loss label corresponding to the direct current transmission engineering samples. In a specific implementation, the fast loss algorithm may be an existing algorithm for calculating the loss of the direct current transmission project, and a person skilled in the art may select the algorithm according to an actual situation.
After the engineering loss label is obtained, the engineering loss label can be associated with the corresponding direct current transmission engineering sample, and a training sample set and a test sample set are obtained by associating a plurality of transmission engineering samples with the engineering loss label. Illustratively, a plurality of electrical transmission engineering samples may be randomly allocated into a test sample set and a training sample set according to a preset ratio (e.g. 2:8), for example, if there are 10000 groups of electrical transmission engineering samples, there are 8000 groups of training sample sets, and there are 2000 groups of test sample sets.
In this embodiment, a training sample set and a testing sample set may be constructed according to a plurality of dc transmission engineering samples and engineering loss labels thereof, so as to provide a basis for subsequently training a power consumption evaluation model.
In an embodiment, the obtaining of the engineering condition parameters of the dc power transmission engineering to be evaluated in S101 may include:
acquiring power information, the number of full half-bridge submodules and fundamental frequency of a converter station in a direct current transmission project to be evaluated; and determining engineering working condition parameters of the direct current transmission engineering according to the power information, the number of full half-bridge submodules and the fundamental frequency.
Wherein, the full-half bridge sub-module may include a full-bridge sub-module and a half-bridge sub-module.
In practical application, power information of a converter station in a direct current transmission project to be evaluated, such as active power and/or reactive power, the number of full half-bridge submodules in the direct current transmission project and a fundamental frequency used by the direct current transmission project can be obtained, and further engineering working condition parameters of the direct current transmission project can be determined according to the power information, the number of full half-bridge submodules and the fundamental frequency, for example, the power information, the number of full half-bridge submodules and the fundamental frequency can be used as engineering working condition parameters, or the engineering working condition parameters can be generated by combining parameters of other engineering components.
In this embodiment, engineering condition parameters of the direct current transmission project can be determined according to the power information, the number of full-half-bridge sub-modules and the fundamental frequency, so that a basis is provided for a subsequent loss evaluation module to determine loss information of the direct current transmission project under different conditions.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the application also provides a direct current transmission engineering loss evaluation device for realizing the direct current transmission engineering loss evaluation method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so that specific limitations in one or more embodiments of the dc transmission engineering loss assessment device provided below can refer to the limitations on the dc transmission engineering loss assessment method in the above description, and details are not repeated herein.
In one embodiment, as shown in fig. 4, there is provided a dc transmission engineering loss evaluation apparatus, including:
the input information acquisition module 401 is configured to acquire engineering working condition parameters and engineering line information of a direct current transmission engineering to be evaluated;
an input module 402, configured to input the engineering condition parameters and the engineering route information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
a loss evaluation result obtaining module 403, configured to obtain loss information of the dc power transmission project output by the loss evaluation model, and determine a loss evaluation result of the dc power transmission project according to the loss information.
In an exemplary embodiment, the apparatus further comprises:
the training sample set acquisition module is used for acquiring a training sample set, wherein the training sample set comprises training working condition parameters and training line information of a plurality of different first direct current transmission engineering samples and an engineering loss label of each first direct current transmission engineering sample;
the training sample input module is used for inputting training working condition parameters and training circuit information of a plurality of first direct current transmission engineering samples in the training sample set into a neural network model to be trained, and the neural network model outputs prediction loss information of the first direct current transmission engineering samples based on the training working condition parameters of each first direct current transmission engineering sample;
and the parameter adjusting module is used for adjusting the model parameters of the neural network model according to the predicted loss information and the engineering loss label until the training end condition is met, so as to obtain a trained loss evaluation model.
In an exemplary embodiment, the parameter adjusting module includes:
the loss value determining module is used for determining a loss value according to the predicted loss information of the plurality of first direct current transmission engineering samples and the engineering loss label of each first direct current transmission engineering sample;
and the training module is used for adjusting the model parameters of the neural network model according to the loss value and the back propagation algorithm until the training end condition is met, so as to obtain a trained loss evaluation model.
In an exemplary embodiment, the training module is specifically configured to:
adjusting model parameters of the neural network model according to the loss value and a back propagation algorithm to obtain an adjusted neural network model;
obtaining a test sample set; the test sample set comprises training working condition parameters and training line information of a plurality of second direct current transmission engineering samples and an engineering loss label of each second direct current transmission engineering sample;
inputting training working condition parameters and training circuit information of a plurality of second direct current power transmission engineering samples in the test sample set into the neural network model, and outputting predicted loss information of each second direct current power transmission engineering sample by the neural network model based on the training working condition parameters and the training circuit information of the second direct current power transmission engineering sample;
determining a training error according to the predicted loss information of the plurality of second direct current transmission engineering samples and the engineering loss labels of the second direct current transmission engineering samples;
and when the training error is larger than a preset error threshold value, returning to the step of obtaining the training sample set, and adjusting the parameters of the neural network model again until the training error is not larger than the preset error threshold value to obtain the trained loss evaluation model.
In an exemplary embodiment, the apparatus further comprises:
the training sample parameter acquisition module is used for acquiring training working condition parameters and training circuit information of a plurality of direct current transmission engineering samples; the plurality of direct current transmission engineering samples are direct current transmission engineering samples with different training working condition parameters and/or line information;
the label determining module is used for determining the loss information of each direct current power transmission engineering sample according to the training working condition parameters, the line information and the preset loss algorithm of each direct current power transmission engineering sample, and taking the loss information of each direct current power transmission engineering sample as an engineering loss label of the direct current power transmission engineering sample;
and the sample set construction module is used for acquiring a training sample set and a test sample set according to the plurality of direct current transmission engineering samples and the engineering loss labels thereof.
In an exemplary embodiment, the input information obtaining module 401 is specifically configured to:
acquiring power information of a converter station, the number of full half-bridge submodules and fundamental frequency in the direct current transmission project to be evaluated;
and determining engineering working condition parameters of the direct current transmission engineering according to the power information, the number of full half-bridge sub-modules and the fundamental frequency.
All or part of each module in the direct current transmission engineering loss evaluation device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 5. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to realize a direct current transmission engineering loss assessment method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 5 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring engineering working condition parameters and engineering line information of a direct current transmission project to be evaluated;
inputting the engineering working condition parameters and the engineering line information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
and obtaining the loss information of the direct current transmission project output by the loss evaluation model, and determining the loss evaluation result of the direct current transmission project according to the loss information.
In one embodiment, the steps in the other embodiments described above are also implemented when the computer program is executed by a processor.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring engineering working condition parameters and engineering line information of a direct current transmission project to be evaluated;
inputting the engineering working condition parameters and the engineering line information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
and obtaining the loss information of the direct current transmission project output by the loss evaluation model, and determining the loss evaluation result of the direct current transmission project according to the loss information.
In one embodiment, the computer program when executed by the processor also performs the steps in the other embodiments described above.
In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, performs the steps of:
acquiring engineering working condition parameters and engineering line information of a direct current transmission project to be evaluated;
inputting the engineering working condition parameters and the engineering line information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
and obtaining the loss information of the direct current transmission project output by the loss evaluation model, and determining the loss evaluation result of the direct current transmission project according to the loss information.
In one embodiment, the computer program when executed by the processor also performs the steps in the other embodiments described above.
It should be noted that, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims (10)

1. A loss assessment method for a direct current transmission project is characterized by comprising the following steps:
acquiring engineering working condition parameters and engineering line information of a direct current transmission project to be evaluated;
inputting the engineering working condition parameters and the engineering line information into a pre-trained loss assessment model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
and obtaining the loss information of the direct current transmission project output by the loss evaluation model, and determining the loss evaluation result of the direct current transmission project according to the loss information.
2. The method of claim 1, further comprising:
acquiring a training sample set, wherein the training sample set comprises training working condition parameters and training line information of a plurality of different first direct current transmission engineering samples, and an engineering loss label of each first direct current transmission engineering sample;
inputting training working condition parameters and training circuit information of a plurality of first direct current power transmission engineering samples in the training sample set into a neural network model to be trained, and outputting predicted loss information of each first direct current power transmission engineering sample by the neural network model based on the training working condition parameters of the first direct current power transmission engineering sample;
and adjusting the model parameters of the neural network model according to the predicted loss information and the engineering loss label until the training end condition is met, so as to obtain a trained loss evaluation model.
3. The method of claim 2, wherein the adjusting model parameters of the neural network model according to the predicted loss information and the engineering loss labels until a training end condition is met to obtain a trained loss assessment model comprises:
determining a loss value according to the predicted loss information of the plurality of first direct current transmission engineering samples and the engineering loss label of each first direct current transmission engineering sample;
and adjusting the model parameters of the neural network model according to the loss value and a back propagation algorithm until the training end condition is met, so as to obtain a trained loss evaluation model.
4. The method according to claim 3, wherein the adjusting the model parameters of the neural network model to be trained according to the loss values and the back propagation algorithm until a training end condition is met to obtain a trained loss evaluation model comprises:
adjusting model parameters of the neural network model according to the loss value and a back propagation algorithm to obtain an adjusted neural network model;
obtaining a test sample set; the test sample set comprises training working condition parameters and training line information of a plurality of second direct current transmission engineering samples and an engineering loss label of each second direct current transmission engineering sample;
inputting training working condition parameters and training line information of a plurality of second direct current power transmission engineering samples in the test sample set into the neural network model, and outputting predicted loss information of each second direct current power transmission engineering sample by the neural network model based on the training working condition parameters and the training line information of the second direct current power transmission engineering sample;
determining training errors according to the predicted loss information of the plurality of second direct current transmission engineering samples and the engineering loss labels of the second direct current transmission engineering samples;
and when the training error is larger than a preset error threshold value, returning to the step of obtaining the training sample set, and adjusting the parameters of the neural network model again until the training error is not larger than the preset error threshold value to obtain the trained loss evaluation model.
5. The method of claim 4, further comprising:
acquiring training working condition parameters and training circuit information of a plurality of direct current transmission engineering samples; the plurality of direct current transmission engineering samples are direct current transmission engineering samples with different training working condition parameters and/or line information;
determining the loss information of each direct current transmission engineering sample according to the training working condition parameters, the line information and a preset loss algorithm of each direct current transmission engineering sample, and taking the loss information of each direct current transmission engineering sample as an engineering loss label of the direct current transmission engineering sample;
and acquiring a training sample set and a test sample set according to the plurality of direct current transmission engineering samples and the engineering loss labels thereof.
6. The method according to claim 1, wherein the obtaining of the engineering condition parameters of the direct current transmission engineering to be evaluated comprises:
acquiring power information of a converter station, the number of full half-bridge submodules and fundamental frequency in the direct current transmission project to be evaluated;
and determining engineering working condition parameters of the direct current transmission engineering according to the power information, the number of full half-bridge sub-modules and the fundamental frequency.
7. A dc transmission project loss assessment apparatus, said apparatus comprising:
the input information acquisition module is used for acquiring engineering working condition parameters and engineering line information of the direct current transmission engineering to be evaluated;
the input module is used for inputting the engineering working condition parameters and the engineering line information into a pre-trained loss evaluation model; the loss evaluation model is used for determining loss information of the direct current transmission project based on the project working condition parameters and the project line information; the loss evaluation model is obtained by training a neural network model based on training working condition parameters, training circuit information and engineering loss labels of a plurality of different direct-current transmission engineering samples;
and the loss evaluation result acquisition module is used for acquiring the loss information of the direct current transmission project output by the loss evaluation model and determining the loss evaluation result of the direct current transmission project according to the loss information.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
CN202210527974.XA 2022-05-16 2022-05-16 Method, device and equipment for evaluating loss of direct current transmission project and storage medium Pending CN114861542A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202210527974.XA CN114861542A (en) 2022-05-16 2022-05-16 Method, device and equipment for evaluating loss of direct current transmission project and storage medium
PCT/CN2022/131882 WO2023221426A1 (en) 2022-05-16 2022-11-15 Direct-current transmission project loss evaluation method and apparatus, device, and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210527974.XA CN114861542A (en) 2022-05-16 2022-05-16 Method, device and equipment for evaluating loss of direct current transmission project and storage medium

Publications (1)

Publication Number Publication Date
CN114861542A true CN114861542A (en) 2022-08-05

Family

ID=82636958

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210527974.XA Pending CN114861542A (en) 2022-05-16 2022-05-16 Method, device and equipment for evaluating loss of direct current transmission project and storage medium

Country Status (2)

Country Link
CN (1) CN114861542A (en)
WO (1) WO2023221426A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115238140A (en) * 2022-09-20 2022-10-25 江苏永鼎股份有限公司 Loss assessment method and system for high-voltage power transmission
CN116307946A (en) * 2023-05-19 2023-06-23 中科航迈数控软件(深圳)有限公司 Processing environment-based adaptive energy consumption assessment method, device, equipment and medium
WO2023221426A1 (en) * 2022-05-16 2023-11-23 中国南方电网有限责任公司超高压输电公司检修试验中心 Direct-current transmission project loss evaluation method and apparatus, device, and storage medium

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117434440B (en) * 2023-12-20 2024-05-07 浙江万胜智能科技股份有限公司 Automatic detection method and system for measuring switch

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102856896B (en) * 2012-08-08 2015-06-24 中国电力科学研究院 On-line analytical method for direct-current transmission loss
CN106018999A (en) * 2016-05-13 2016-10-12 国网浙江省电力公司电力科学研究院 Energy efficiency measuring method for flexible DC transmission system
US11900031B2 (en) * 2019-08-15 2024-02-13 State Grid Smart Research Institute Co., Ltd. Systems and methods of composite load modeling for electric power systems
CN110956330A (en) * 2019-12-02 2020-04-03 中国电力科学研究院有限公司 Method and system for predicting line loss of power transmission line based on multidimensional influence quantity
CN114050591B (en) * 2021-11-09 2024-01-30 福州大学 Method for reducing loss of power transmission engineering by optimizing voltage of offshore wind farm booster station
CN114861542A (en) * 2022-05-16 2022-08-05 中国南方电网有限责任公司超高压输电公司检修试验中心 Method, device and equipment for evaluating loss of direct current transmission project and storage medium

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023221426A1 (en) * 2022-05-16 2023-11-23 中国南方电网有限责任公司超高压输电公司检修试验中心 Direct-current transmission project loss evaluation method and apparatus, device, and storage medium
CN115238140A (en) * 2022-09-20 2022-10-25 江苏永鼎股份有限公司 Loss assessment method and system for high-voltage power transmission
CN116307946A (en) * 2023-05-19 2023-06-23 中科航迈数控软件(深圳)有限公司 Processing environment-based adaptive energy consumption assessment method, device, equipment and medium
CN116307946B (en) * 2023-05-19 2023-08-18 中科航迈数控软件(深圳)有限公司 Processing environment-based adaptive energy consumption assessment method, device, equipment and medium

Also Published As

Publication number Publication date
WO2023221426A1 (en) 2023-11-23

Similar Documents

Publication Publication Date Title
CN114861542A (en) Method, device and equipment for evaluating loss of direct current transmission project and storage medium
Liu et al. Reinforcement learning-based cell selection in sparse mobile crowdsensing
CN105095961A (en) Mixing system with artificial neural network and impulsive neural network
CN113887845B (en) Extreme event prediction method, device, equipment and storage medium
CN116011682B (en) Meteorological data prediction method and device, storage medium and electronic device
CN110009181A (en) Distribution network transform measure and mistake load figureofmerit relevance method for digging and device
CN116187548A (en) Photovoltaic power generation power prediction method and device, storage medium and electronic device
Zhao et al. Gaussian process-aided transfer learning for probabilistic load forecasting against anomalous events
CN117252436B (en) Method and system for land utilization change ecological risk partition
Shi et al. Network reliability analysis through survival signature and machine learning techniques
de Jongh et al. Physics-informed geometric deep learning for inference tasks in power systems
CN112800674A (en) Power system element reliability parameter calculation method based on extreme learning machine
CN117289097A (en) Power equipment partial discharge detection method, model training method, device and equipment
CN103763123A (en) Method and device for evaluating health condition of network
CN116011681A (en) Meteorological data prediction method and device, storage medium and electronic device
CN114267422B (en) Method and system for predicting surface water quality parameters, computer equipment and storage medium
CN114238106A (en) Test time prediction method and device, electronic device and storage medium
CN116400430B (en) Meteorological data prediction method and device, storage medium and electronic device
Urbanovská et al. Semantically Layered Representation for Planning Problems and Its Usage for Heuristic Computation Using Cellular Simultaneous Recurrent Neural Networks.
CN118070688B (en) Construction method and construction device of power grid load model, storage medium and electronic device
CN113392536B (en) Power communication network simulation evaluation method, device, computer equipment and medium
CN116306952B (en) Molecular property prediction method and device, storage medium and electronic device
CN108173284A (en) Active power distribution network method for estimating state and system
CN114820182B (en) Quantum searching method and device for coordination pairs in financial transaction data
CN113556266B (en) Traffic matrix prediction method taking traffic engineering as center

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination