CN114492212B - Dynamic capacity increasing method and device for low-voltage distribution network, terminal and storage medium - Google Patents
Dynamic capacity increasing method and device for low-voltage distribution network, terminal and storage medium Download PDFInfo
- Publication number
- CN114492212B CN114492212B CN202210392368.1A CN202210392368A CN114492212B CN 114492212 B CN114492212 B CN 114492212B CN 202210392368 A CN202210392368 A CN 202210392368A CN 114492212 B CN114492212 B CN 114492212B
- Authority
- CN
- China
- Prior art keywords
- temperature
- temperature rise
- rise model
- predicted
- output
- 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.)
- Active
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
- G06F2119/08—Thermal analysis or thermal 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
Abstract
The invention relates to the technical field of power distribution of a power grid, in particular to a dynamic capacity increasing method, a dynamic capacity increasing device, a dynamic capacity increasing terminal and a dynamic capacity increasing storage medium for a low-voltage power distribution network. According to the method provided by the embodiment of the invention, the model parameters are continuously corrected through the residual error, so that the predicted accumulated error can be reduced, and more accurate capacity indication degree can be obtained. The method adopts the model with the LSTM model structure, and the structure can selectively memorize some early output results and can also selectively forget some results, so that the method can achieve higher prediction accuracy for the power line temperature rise prediction with the hysteresis characteristic.
Description
Technical Field
The invention relates to the technical field of power distribution of a power grid, in particular to a dynamic capacity increasing method, a dynamic capacity increasing device, a dynamic capacity increasing terminal and a dynamic capacity increasing storage medium for a low-voltage power distribution network.
Background
In the past decades, power Line Rating (Dynamic Line Rating) technology has been receiving much attention and has been developed. The dynamic capacity increasing technology of the power transmission line provides a feasible alternative scheme for improving the transmission capacity of the line, deeply excavates the actual transmission potential of the power network, and improves the operation flexibility of the power system on the basis of not constructing additional lines.
The technical details related to the dynamic capacity increase of the transmission line need to be further perfected so as to ensure the safety and stability of the operation of the power system after the technology is applied. The dynamic capacity increasing technology of the power transmission line based on historical meteorological information is used for improving the transmission capacity of the line and the operation flexibility of a system, and the prior art has been researched many times, however, when the technology is applied to practice, various defects always exist. For example, the dynamic capacity-increasing technology is typically difficult to practice in laboratories due to various problems such as complex meteorological conditions, large prediction deviation of the condition of the power transmission line, complex mathematical models, large calculation amount, and the like.
Based on this, a dynamic capacity increasing method for a low-voltage distribution network needs to be developed and designed to solve various disadvantages in the prior art.
Disclosure of Invention
The embodiment of the invention provides a dynamic capacity increasing method, a dynamic capacity increasing device, a dynamic capacity increasing terminal and a dynamic capacity increasing storage medium for a low-voltage distribution network, which are used for solving the problems of complex dynamic capacity increasing model and poor application effect of the low-voltage distribution network in the prior art.
In a first aspect, an embodiment of the present invention provides a dynamic capacity increasing method for a low-voltage distribution network, including:
acquiring influence factors influencing the temperature of a power supply line and an observed temperature, wherein the observed temperature is used for representing the operating temperature of the power supply line;
inputting the influence factors into a temperature rise model to obtain predicted temperature, wherein the temperature rise model has the influence factor input, the predicted temperature output and the capacity indication output;
calculating a prediction residual, and adjusting parameters of the temperature rise model according to the prediction residual until the prediction residual is smaller than a threshold value, wherein the prediction residual is the deviation between the predicted temperature and the observed temperature, and the parameters are used for representing the weight of each transfer function in the temperature rise model;
and determining the capacity of the power supply line according to the capacity indication output by the temperature rise model.
In one possible implementation manner, obtaining influencing factors influencing the temperature of the power supply line and the observed temperature includes:
acquiring influence factors of the temperature and the observed temperature according to predetermined factor items, wherein the factor items are determined according to the following steps:
acquiring a plurality of factors, observed temperatures and actually measured temperatures of a power supply line, wherein the actually measured temperatures are the operation actually measured temperatures of the power supply line;
according to the observed temperature and the actually measured temperature, a temperature comparison table is constructed, and the comparison table is used for representing the corresponding relation between the observed temperature and the actually measured temperature;
obtaining a plurality of temperature factor influence coefficients for a plurality of factors of the power supply line according to the temperature comparison table, wherein the temperature factor influence coefficients are used for representing the correlation between the plurality of factors and the actually measured temperature;
and selecting factors with influence coefficients exceeding a threshold value as factor items.
In one possible implementation manner, the temperature rise model is constructed based on an LSTM model, and the temperature rise model includes:
the output end of the input layer and the output ends of the plurality of hidden layers are respectively connected with the input ends of the plurality of hidden layers, the input layer is provided with an input end corresponding to the input of the influence factors, the plurality of hidden layers form a full-connection network, the output ends of the plurality of hidden layers are connected with the input end of the output layer, and the output layer comprises a predicted temperature output end and a capacity indication output end, wherein the predicted temperature output end is used for outputting the predicted temperature, and the capacity indication output end is used for outputting the capacity indication.
In a possible implementation manner, the plurality of hidden layers includes a memory layer and a common transport layer, where the memory layer has a model structure of LSTM, the memory layer is located between the input layer and the common transport layer, an input end of the memory layer is connected to an output end of the input layer, and an output end of the memory layer is connected to an input end of the common transport layer.
In a possible implementation manner, the inputting the influence factors into a temperature rise model to obtain a predicted temperature includes:
acquiring a prediction period, wherein the prediction period is a period between two predictions;
acquiring a plurality of influence factors in the prediction period;
sequentially inputting a plurality of influence factors in the prediction period into the temperature rise model according to the time sequence;
and after the input of the plurality of influence factors in the prediction period is finished, acquiring the predicted temperature output by the temperature rise model as the predicted temperature.
In a possible implementation manner, the calculating a prediction residual and adjusting a parameter of the temperature rise model according to the prediction residual until the prediction residual is smaller than a threshold includes:
residual error calculation step: calculating the difference between the predicted temperature output by the temperature rise model and the observed temperature to serve as a predicted residual error;
if the predicted residual error is smaller than a threshold value, fixing the parameters of each neuron in the temperature rise model;
otherwise, updating the parameters of each neuron in the temperature rise model by using a back propagation algorithm, inputting the influence factors into the temperature rise model to obtain the predicted temperature, and skipping to the residual error calculation step.
In one possible implementation manner, the method further includes a model structure adjusting step, and the model structure adjusting step includes:
acquiring preset evaluation times;
selecting a primary prediction residual of the preset evaluation times, wherein the primary prediction residual is a prediction residual obtained by calculation before adjusting the parameters of the temperature rise model;
and if the average value of the primary prediction residuals of the preset evaluation times is greater than or equal to a threshold value, adjusting the structure of the temperature rise model in a mode of reducing the number of hidden layers.
In a second aspect, an embodiment of the present invention provides a dynamic capacity increasing device for a low-voltage distribution network, including:
the system comprises an influence factor acquisition module, a temperature measurement module and a temperature measurement module, wherein the influence factor acquisition module is used for acquiring influence factors influencing the temperature of a power supply line and an observation temperature, and the observation temperature is used for representing the operation temperature of the power supply line;
the temperature prediction acquisition module is used for inputting the influence factors into a temperature rise model to acquire predicted temperature, wherein the temperature rise model is provided with influence factor input, predicted temperature output and capacity indication output;
the iteration adjusting module is used for calculating a prediction residual error, adjusting parameters of the temperature rise model according to the prediction residual error until the prediction residual error is smaller than a threshold value, wherein the prediction residual error is the deviation of the predicted temperature and the observed temperature, and the parameters are used for representing the weight of each transfer function in the temperature rise model;
and the number of the first and second groups,
and the output module is used for determining the capacity of the power supply circuit according to the capacity indication output by the temperature rise model.
In a third aspect, an embodiment of the present invention provides a terminal, which includes a memory and a processor, where the memory stores a computer program that is executable on the processor, and the processor executes the computer program to implement the steps of the method according to the first aspect or any one of the possible implementation manners of the first aspect.
In a fourth aspect, the present invention provides a computer-readable storage medium, which stores a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any one of the possible implementations of the first aspect.
Compared with the prior art, the implementation mode of the invention has the following beneficial effects:
the embodiment of the invention discloses a dynamic capacity increasing method for a low-voltage power distribution network, which comprises the steps of firstly obtaining influence factors influencing the temperature of a power supply circuit and observed temperature, then inputting the influence factors into a temperature rise model, obtaining a predicted residual error through the output of the model and the observed temperature, and then adjusting the parameters of the temperature rise model through the residual error, so that the model can have more accurate predicted temperature output and capacity indication output. According to the method provided by the embodiment of the invention, the model parameters are continuously corrected through the residual error, so that the predicted accumulated error can be reduced, and more accurate capacity indication degree can be obtained.
The method adopts the model with the LSTM model structure, and the structure can selectively memorize some results output in the early stage (considering dynamic effect) and can also selectively forget some results (considering steady-state effect), so that the method can achieve higher prediction accuracy for the power line temperature rise prediction with hysteresis characteristic.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
Fig. 1 is a flowchart of a dynamic capacity increasing method for a low-voltage distribution network according to an embodiment of the present invention;
FIG. 2 is a topology structure diagram of a temperature rise model according to an embodiment of the present invention;
FIG. 3 is a time domain expanded view of the LSTM model provided by an embodiment of the present invention;
FIG. 4 is a block diagram of an LSTM model provided by an embodiment of the present invention at the current time;
fig. 5 is a functional block diagram of a dynamic capacity increasing device for a low-voltage distribution network according to an embodiment of the present invention;
fig. 6 is a functional block diagram of a terminal according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
In order to make the objects, technical solutions and advantages of the present invention more apparent, the following description is made with reference to the accompanying drawings.
The following is a detailed description of the embodiments of the present invention, which is implemented on the premise of the technical solution of the present invention, and the detailed implementation and the specific operation procedures are given, but the scope of the present invention is not limited to the following embodiments.
Fig. 1 is a flowchart of a dynamic capacity increasing method for a low-voltage distribution network according to an embodiment of the present invention.
As shown in fig. 1, it shows an implementation flowchart of a dynamic capacity increasing method for a low-voltage distribution network according to an embodiment of the present invention, which is detailed as follows:
in step 101, influencing factors influencing the temperature of the power supply line and an observed temperature are obtained, wherein the observed temperature is used for representing the operating temperature of the power supply line.
In some embodiments, the step 101 comprises:
acquiring influence factors of the temperature and the observed temperature according to predetermined factor items, wherein the factor items are determined according to the following steps:
acquiring a plurality of factors, observed temperatures and actually measured temperatures of a power supply line, wherein the actually measured temperatures are the operation actually measured temperatures of the power supply line;
according to the observed temperature and the actually measured temperature, a temperature comparison table is constructed and used for representing the corresponding relation between the observed temperature and the actually measured temperature;
obtaining a plurality of temperature factor influence coefficients for a plurality of factors of the power supply line according to the temperature comparison table, wherein the temperature factor influence coefficients are used for representing the correlation between the plurality of factors and the actually measured temperature;
and selecting factors with influence coefficients exceeding a threshold value as factor items.
Illustratively, the observed temperature is an observable temperature, which, as is known, is generally difficult to measure directly for the power line or the overall temperature of the power equipment, and is generally characterized by an observed value, which is generally obtained based on the temperature of a representative and easily observable point, such as the temperature of the power line connection point or the temperature of the heat generating point of the power equipment. Taking the temperature of the power line connection point as an example, the temperature of the point is usually higher than the actual overall measured temperature, that is, the actual temperature, if the observed temperature is used as the reference for dynamic capacity increase of the low-voltage distribution network, the capacity increase capacity is usually small, and at this time, a comparison table of the actual temperature and the observed temperature should be established, and the overall temperature is reflected by the observed temperature.
As the influence factors affecting the temperature of the power supply line, there may be included solar light intensity, wind speed, current flowing through the power line of the power supply line, specific heat capacity of the power line, cross-sectional shape of the power line, cross-sectional parameters, date, time, latitude, longitude in which the power line is located, and the like, but of these factors, factors that may be dominant, such as solar light intensity, wind speed, may be subordinate, or may be irrelevant to the temperature of the power supply line.
Therefore, it is necessary to screen out factors related to the power supply line from a plurality of factors, so that the complexity of the temperature rise model can be reduced, and the accuracy of the temperature rise model can be ensured.
One algorithm is to obtain the measured temperature according to a temperature comparison table and the observed temperature, then obtain the correlation coefficient between the measured temperature and a plurality of factors, and judge whether the factor is the factor influencing the power supply circuit based on the coefficient.
There are three common algorithms for correlation coefficients, for example, Pearson product difference correlation coefficient, Spearman rank correlation coefficient, and Kendall rank correlation coefficient, which are all based on statistical principles and give correlation between the two.
In step 102, the influencing factors are input into a temperature rise model, and a predicted temperature is obtained, wherein the temperature rise model has the influencing factor input, a predicted temperature output and a capacity indication output.
In some embodiments, the temperature rise model is constructed based on an LSTM model, the temperature rise model comprising:
the output end of the input layer and the output ends of the plurality of hidden layers are respectively connected with the input ends of the plurality of hidden layers, the input layer is provided with an input end corresponding to the input of the influence factors, the plurality of hidden layers form a full-connection network, the output ends of the plurality of hidden layers are connected with the input end of the output layer, and the output layer comprises a predicted temperature output end and a capacity indication output end, wherein the predicted temperature output end is used for outputting the predicted temperature, and the capacity indication output end is used for outputting the capacity indication.
In some embodiments, the plurality of hidden layers include a memory layer and a common transmission layer, wherein the memory layer has a model structure of LSTM, the memory layer is located between the input layer and the common transmission layer, an input end of the memory layer is connected to an output end of the input layer, and an output end of the memory layer is connected to an input end of the common transmission layer.
In some embodiments, the inputting the influencing factor into the temperature rise model to obtain the predicted temperature includes:
acquiring a prediction period, wherein the prediction period is a period between two predictions;
acquiring a plurality of influence factors in the prediction period;
sequentially inputting a plurality of influence factors in the prediction period into the temperature rise model according to the time sequence;
and after the input of the plurality of influence factors in the prediction period is finished, acquiring the predicted temperature output by the temperature rise model as the predicted temperature.
Exemplarily, fig. 2 shows a topology structure diagram of a temperature rise model provided by the embodiment of the present invention.
As shown in fig. 2, the temperature rise model includes an input layer, a plurality of hidden layers, and an output layer. The input layer is used for receiving input of influencing factors and comprises a plurality of input nodes, such as those in the figurex 1 、x 2 …x m The output layer is used for outputting two data which are respectively predicted temperature output and capacity indication output and comprises two nodesO 1 AndO 2 . The hidden layer is used as a key layer for realizing the embodiment of the invention and is realized based on an LSTM (Long short-term memory) model structure, wherein the first layer comprisesy 11 、y 21 …y m1 A node, the second layer comprisingy 12 、y 22 …y m2 Nodes, and so on, the nth layer includesy 1n 、y 2n …y mn And (4) nodes.
The LSTM model structure has a self-circulation Network structure, and is an improved form of an RNN (Recurrent Neural Network, RNN) model structure. Like RNN, the LSTM model structure is characterized by being output based on the result of the last output and the input of this time. This principle is just as the problem we usually think is often associated with the problem of thinking in the previous phase, for example, in expression we might say: "I love eating carrots because carrots are rich in vitamins", the latter half of which is related to the first half.
Returning to the embodiment of the present invention, the current temperature of the power line is not only determined by the current influencing factor, but also determined by the influencing factor of the previous stage and the temperature of the previous stage. For example, the temperatures of the two power lines in the previous stage are a and B, respectively, and even if the current respective factors of the two power lines are the same, the temperatures of the two power lines at this time are not the same (unless the interval time of the two stages is very long, the temperatures reach a steady state). This feature is more like hysteresis, which often causes much confusion in the construction of the model, and dynamic effects and steady-state effects should be considered in the construction.
For the structure of the LSTM model, the problem is relatively easy to solve, and the LSTM model can selectively memorize a plurality of results output earlier (considering dynamic effect) and can also selectively forget a plurality of results (considering steady effect).
In the embodiment of the invention, the hidden layer connected with the input layer has an LSTM model structure, and the hidden layer selectively sends the output result of the previous time to the next input. If the input and output of the previous time, the current time and the next time are constructed, we can obtain the time domain expansion diagram of the LSTM model as shown in fig. 3.
WhereinX t-1 、X t AndX t+1 respectively the influence factor input at the previous moment, the influence factor input at the current moment and the influence factor input at the next moment, correspondingly,h t-1 、h t andh t+1 the output of the previous moment, the output of the current moment and the output of the next moment respectively,σis the activation function sigmoid.
If the input and output at the current time are taken as the focus, we can obtain the structure diagram as shown in fig. 4.
In the figure, sigmoid and tanh are activation functions respectively, the output of the sigmoid function is between 0 and 1, and tanh is a hyperbolic tangent function, and the output of the hyperbolic tangent function is between-1 and 1.
The control system comprises three valves and a memory unit, wherein the three valves are a forgetting valve, an input valve and an output valve, the valves are controlled valves, and for the forgetting valve, the transfer function expression is as follows:
f t =sigmoid(W f ⋅[h t−1 ,X t ]+b f )，
for an input valve, the transfer function expression is:
i t =sigmoid(W i ⋅[h t−1 ,X t ]+b i )，
C t ~=tanh(W c ⋅[h t−1 ,X t ]+b c )，
for a memory cell, the transfer function expression is:
C t =f t ⋅C t−1 +i t ⋅C t ~，
for the output valve, the transfer function expression is:
o t =sigmoid(W o ⋅[h t−1 ,X t ]+b o )，
h t =o t ⋅tanh(C t )，
in the above-mentioned formula,W f 、W i 、W c andW o are all the weight values of the weight values,C t is an intermediate variable (not shown in the figure), andb f 、b i 、b c andb o all the weights and the offsets are offset, and can be modified through training to achieve the purpose of fitting.
For a generic transport layer, it may employ a generic neuron structure.
For the aspect of data input, a period is generally selected, for example, two hours are taken as a period, the influence factors collected in multiple batches within two hours are sent to the temperature rise model in batches according to the time sequence, and when the influence factors of multiple batches in the period are input, the predicted temperature at the end of the next period can be obtained.
In step 103, a prediction residual is calculated, and parameters of the temperature rise model are adjusted according to the prediction residual until the prediction residual is smaller than a threshold, wherein the prediction residual is a deviation between the predicted temperature and the observed temperature, and the parameters are used for representing weights of various transfer functions in the temperature rise model.
In some embodiments, step 103 comprises:
residual error calculation step: calculating the difference between the predicted temperature output by the temperature rise model and the observed temperature to serve as a predicted residual error;
if the predicted residual error is smaller than a threshold value, fixing the parameters of each neuron in the temperature rise model;
otherwise, updating the parameters of each neuron in the temperature rise model by using a back propagation algorithm, inputting the influence factors into the temperature rise model to obtain the predicted temperature, and skipping to the residual error calculation step.
Illustratively, for the modification aspect of the model, we use the residual error in combination with the back propagation algorithm.
The residual error is the difference between the predicted temperature and the observed temperature, and if the residual error is smaller than a threshold value, the setting of each parameter of the temperature rise model is reasonable. If the residual is greater than or equal to the threshold, the parameters of the temperature rise model, such as the weights and biases of the activation functions described above, should be adjusted.
When the output deviates from the actual output, the contribution degree of different neurons to the deviation is different, and the weight and the bias of the neuron activation function are corrected according to the difference of the contribution degree, which is the basic idea of back propagation.
For example, in one back-propagation algorithm, the input excitation and response errors are multiplied to obtain a gradient of weights; this gradient is multiplied by a proportion and added to the weight after inversion. This ratio will affect the speed and effectiveness of the training process and is therefore referred to as the "training factor". The direction of the gradient indicates the direction of error propagation and therefore needs to be inverted when updating the weights, thereby reducing the weight-induced errors.
In step 104, the capacity of the power supply line is determined according to the capacity indication output by the temperature rise model.
For example, after the temperature rise model can obtain an accurate temperature prediction, a capacity indication can be output according to other constraint conditions, such as a power line temperature rise threshold, a power line transmission voltage drop threshold, and the like, and capacity expansion of the power supply line can be realized according to the capacity indication.
In addition, there is a model structure adjustment step 105, and the step 105 includes:
acquiring preset evaluation times;
selecting a primary prediction residual of the preset evaluation times, wherein the primary prediction residual is a prediction residual obtained by calculation before adjusting the parameters of the temperature rise model;
and if the average value of the initial prediction residuals of the preset evaluation times is greater than or equal to the threshold value, adjusting the structure of the temperature rise model in a mode of reducing the number of hidden layers.
Illustratively, as mentioned above, the number of layers of the hidden layer in the model should not be too large or too small, and should be a reasonable number, and too small may risk under-fitting, and too large may risk over-fitting.
In practical application, the neural network structure is adopted for fitting, the risk of over-fitting is far greater than the risk of under-fitting, and the characteristic of the neural network structure is determined. Overfitting is similar to the student taking an examination to try, and can only have a higher score for the questions done and a lower score for the newly issued questions. The willingness to make such a result is, on the one hand, the inclusion of noise (temperature-independent factors) in the influencing factors and, on the other hand, the complexity of the model construction.
Thus, based on the consideration of the model structure, one way to address the risk of overfitting is to reduce the number of hidden layers.
Whether overfitting exists or not is obtained through evaluation, in an evaluation mode, an evaluation number is set, for example, 100 times, if the residual errors of the last 100 times are extracted, and the average value of the residual errors of the 100 times is larger than a threshold value, it is indicated that the overfitting risk exists, and the number of hidden layers should be reduced.
According to the embodiment of the dynamic capacity increasing method for the low-voltage distribution network, firstly, the influence factors influencing the temperature of a power supply circuit and the observed temperature are obtained, then the influence factors are input into the temperature rise model, the predicted residual error is obtained through the output of the model and the observed temperature, and then the parameter of the temperature rise model is adjusted through the residual error, so that the model can have more accurate predicted temperature output and capacity indication output. According to the method provided by the embodiment of the invention, the model parameters are continuously corrected through the residual error, so that the predicted accumulated error can be reduced, and more accurate capacity indication degree can be obtained.
The method adopts the model with the LSTM model structure, and the structure can selectively memorize some results output in the early stage (considering dynamic effect) and can also selectively forget some results (considering steady-state effect), so that the method can achieve higher prediction accuracy for the power line temperature rise prediction with hysteresis characteristic.
It should be understood that the sequence numbers of the steps in the above embodiments do not mean the execution sequence, and the execution sequence of each process should be determined by the function and the inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
The following are apparatus embodiments of the invention, and for details not described in detail therein, reference may be made to the corresponding method embodiments described above.
Fig. 5 is a functional block diagram of a dynamic capacity increasing device for a low-voltage distribution network according to an embodiment of the present invention, and referring to fig. 5, the dynamic capacity increasing device 5 for a low-voltage distribution network includes: an influence factor obtaining module 501, a temperature prediction obtaining module 502, an iteration adjusting module 503, and an output module 504.
The influence factor acquiring module 501 is used for acquiring influence factors influencing the temperature of the power supply line and observed temperature, wherein the observed temperature is used for representing the running temperature of the power supply line;
a temperature prediction obtaining module 502, configured to input the influence factor into a temperature rise model to obtain a predicted temperature, where the temperature rise model has an influence factor input, a predicted temperature output, and a capacity indication output;
an iteration adjusting module 503, configured to calculate a prediction residual, and adjust a parameter of the temperature rise model according to the prediction residual until the prediction residual is smaller than a threshold, where the prediction residual is a deviation between the predicted temperature and the observed temperature, and the parameter is used to represent a weight of each transfer function in the temperature rise model;
and the number of the first and second groups,
and the output module 504 is configured to determine the capacity of the power supply line according to the capacity indication output by the temperature rise model.
Fig. 6 is a functional block diagram of a terminal according to an embodiment of the present invention. As shown in fig. 6, the terminal 6 of this embodiment includes: a processor 600 and a memory 601, in which memory 601 a computer program 602 is stored which is executable on the processor 600. The processor 600 executes the computer program 602 to implement the above-mentioned steps of the method and embodiment for dynamic capacity increase of the low voltage distribution network, such as the steps 101 to 104 shown in fig. 1.
Illustratively, the computer program 602 may be partitioned into one or more modules/units that are stored in the memory 601 and executed by the processor 600 to implement the present invention.
The terminal 6 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The terminal 6 may include, but is not limited to, a processor 600, a memory 601. It will be appreciated by those skilled in the art that fig. 6 is only an example of a terminal 6 and does not constitute a limitation of the terminal 6, and that it may comprise more or less components than those shown, or some components may be combined, or different components, for example the terminal may further comprise input output devices, network access devices, buses, etc.
The Processor 600 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 601 may be an internal storage unit of the terminal 6, such as a hard disk or a memory of the terminal 6. The memory 601 may also be an external storage device of the terminal 6, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital Card (SD), a Flash memory Card (Flash Card), and the like, provided on the terminal 6. Further, the memory 601 may also include both an internal storage unit and an external storage device of the terminal 6. The memory 601 is used for storing the computer programs and other programs and data required by the terminal. The memory 601 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit, and the integrated unit may be implemented in a form of hardware, or may be implemented in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the description of each embodiment is focused on, and for parts that are not described or illustrated in detail in a certain embodiment, reference may be made to the description of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal and method may be implemented in other manners. For example, the above-described apparatus/terminal embodiments are merely illustrative, and for example, the division of the modules or units is only one type of logical function division, and other division manners may exist in actual implementation, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each of the units may exist alone physically, or two or more units may be integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow in the method according to the above embodiment may be realized by a computer program instructing related hardware to complete, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the steps of the embodiments of the method for dynamically increasing capacity of the low voltage distribution network and the device for dynamically increasing capacity of the low voltage distribution network may be realized. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like.
The above embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (7)
1. A dynamic capacity increasing method for a low-voltage distribution network is characterized by comprising the following steps:
acquiring influence factors influencing the temperature of a power supply line and an observed temperature, wherein the observed temperature is used for representing the operating temperature of the power supply line;
inputting the influence factors into a temperature rise model to obtain predicted temperature, wherein the temperature rise model has the influence factor input, the predicted temperature output and the capacity indication output;
calculating a prediction residual, and adjusting parameters of the temperature rise model according to the prediction residual until the prediction residual is smaller than a threshold value, wherein the prediction residual is the deviation between the predicted temperature and the observed temperature, and the parameters are used for representing the weight of each transfer function in the temperature rise model;
determining the capacity of a power supply line according to the capacity indication output by the temperature rise model;
wherein the content of the first and second substances,
the temperature rise model is constructed based on an LSTM model, and comprises:
the output end of the input layer and the output ends of the plurality of hidden layers are respectively connected with the input ends of the plurality of hidden layers, the input layer is provided with an input end corresponding to the input of the influence factors, the plurality of hidden layers form a full-connection network, the output ends of the plurality of hidden layers are connected with the input end of the output layer, and the output layer comprises a predicted temperature output end and a capacity indication output end, wherein the predicted temperature output end is used for outputting the predicted temperature, and the capacity indication output end is used for outputting the capacity indication;
the hidden layers comprise a memory layer and a universal transmission layer, wherein the memory layer has an LSTM model structure, the memory layer is positioned between the input layer and the universal transmission layer, the input end of the memory layer is connected with the output end of the input layer, and the output end of the memory layer is connected with the input end of the universal transmission layer;
the calculating a prediction residual error, and adjusting the parameter of the temperature rise model according to the prediction residual error until the prediction residual error is smaller than a threshold value, includes:
residual error calculation step: calculating the difference between the predicted temperature output by the temperature rise model and the observed temperature to serve as a predicted residual error;
if the predicted residual error is smaller than a threshold value, fixing the parameters of each neuron in the temperature rise model;
otherwise, updating parameters of each neuron in the temperature rise model by using a back propagation algorithm, inputting the influence factors into the temperature rise model to obtain predicted temperature, and skipping to the residual error calculation step;
the dynamic capacity increasing method for the low-voltage distribution network further comprises a model structure adjusting step, wherein the model structure adjusting step comprises the following steps of:
acquiring preset evaluation times;
selecting a primary prediction residual of the preset evaluation times, wherein the primary prediction residual is a prediction residual obtained by calculation before adjusting the parameters of the temperature rise model;
and if the average value of the initial prediction residuals of the preset evaluation times is greater than or equal to the threshold value, adjusting the structure of the temperature rise model in a mode of reducing the number of hidden layers.
2. The dynamic capacity increasing method for the low-voltage distribution network according to claim 1, wherein the acquiring of influencing factors influencing the temperature of a power supply line and the observed temperature comprises:
acquiring influence factors of the temperature and the observed temperature according to predetermined factor items, wherein the factor items are determined according to the following steps:
acquiring a plurality of factors, observed temperatures and actually measured temperatures of a power supply line, wherein the actually measured temperatures are the actually measured operating temperatures of the power supply line;
according to the observed temperature and the actually measured temperature, a temperature comparison table is constructed, and the comparison table is used for representing the corresponding relation between the observed temperature and the actually measured temperature;
obtaining a plurality of temperature factor influence coefficients for a plurality of factors of the power supply line according to the temperature comparison table, wherein the temperature factor influence coefficients are used for representing the correlation between the plurality of factors and the actually measured temperature;
and selecting factors with influence coefficients exceeding a threshold value as factor items.
3. The dynamic capacity increasing method for the low-voltage distribution network according to claim 1, wherein the step of inputting the influencing factors into a temperature rise model to obtain the predicted temperature comprises the steps of:
acquiring a prediction period, wherein the prediction period is a period between two predictions;
acquiring a plurality of influence factors in the prediction period;
sequentially inputting a plurality of influence factors in the prediction period into the temperature rise model according to the time sequence;
and after the input of the plurality of influence factors in the prediction period is finished, acquiring the predicted temperature output by the temperature rise model as the predicted temperature.
4. The dynamic capacity increasing method for the low-voltage distribution network according to claim 3, wherein the calculating of the prediction residual and the adjusting of the parameters of the temperature rise model according to the prediction residual until the prediction residual is smaller than a threshold value comprises:
residual error calculation step: calculating the difference between the predicted temperature output by the temperature rise model and the observed temperature to serve as a predicted residual error;
if the prediction residual is smaller than a threshold value, fixing parameters of each neuron in the temperature rise model;
otherwise, updating the parameters of each neuron in the temperature rise model by using a back propagation algorithm, inputting the influence factors into the temperature rise model to obtain the predicted temperature, and skipping to the residual error calculation step.
5. A dynamic capacity increasing device for a low-voltage distribution network, comprising:
the system comprises an influence factor acquisition module, a temperature measurement module and a temperature measurement module, wherein the influence factor acquisition module is used for acquiring influence factors influencing the temperature of a power supply line and an observation temperature, and the observation temperature is used for representing the operation temperature of the power supply line;
the temperature prediction acquisition module is used for inputting the influence factors into a temperature rise model to acquire predicted temperature, wherein the temperature rise model is provided with influence factor input, predicted temperature output and capacity indication output;
the iteration adjusting module is used for calculating a prediction residual error, adjusting parameters of the temperature rise model according to the prediction residual error until the prediction residual error is smaller than a threshold value, wherein the prediction residual error is the deviation of the predicted temperature and the observed temperature, and the parameters are used for representing the weight of each transfer function in the temperature rise model;
and (c) a second step of,
the output module is used for determining the capacity of the power supply circuit according to the capacity indication output by the temperature rise model;
wherein the content of the first and second substances,
the temperature rise model is constructed based on an LSTM model, and comprises:
the output end of the input layer and the output ends of the plurality of hidden layers are respectively connected with the input ends of the plurality of hidden layers, the input layer is provided with an input end corresponding to the input of the influence factors, the plurality of hidden layers form a full-connection network, the output ends of the plurality of hidden layers are connected with the input end of the output layer, and the output layer comprises a predicted temperature output end and a capacity indication output end, wherein the predicted temperature output end is used for outputting the predicted temperature, and the capacity indication output end is used for outputting the capacity indication;
the hidden layers comprise a memory layer and a universal transmission layer, wherein the memory layer has an LSTM model structure, the memory layer is positioned between the input layer and the universal transmission layer, the input end of the memory layer is connected with the output end of the input layer, and the output end of the memory layer is connected with the input end of the universal transmission layer;
the calculating a prediction residual error, and adjusting the parameter of the temperature rise model according to the prediction residual error until the prediction residual error is smaller than a threshold value, includes:
residual error calculation step: calculating the difference between the predicted temperature output by the temperature rise model and the observed temperature to serve as a predicted residual error;
if the predicted residual error is smaller than a threshold value, fixing the parameters of each neuron in the temperature rise model;
otherwise, updating parameters of each neuron in the temperature rise model by using a back propagation algorithm, inputting the influence factors into the temperature rise model to obtain predicted temperature, and skipping to the residual error calculation step;
the dynamic capacity increasing method for the low-voltage distribution network further comprises a model structure adjusting step, wherein the model structure adjusting step comprises the following steps of:
acquiring preset evaluation times;
selecting a primary prediction residual of the preset evaluation times, wherein the primary prediction residual is a prediction residual obtained by calculation before adjusting the parameters of the temperature rise model;
and if the average value of the primary prediction residuals of the preset evaluation times is greater than or equal to a threshold value, adjusting the structure of the temperature rise model in a mode of reducing the number of hidden layers.
6. A terminal comprising a memory and a processor, the memory having stored therein a computer program operable on the processor, wherein the processor, when executing the computer program, performs the steps of the method according to any of claims 1 to 4.
7. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210392368.1A CN114492212B (en) | 2022-04-15 | 2022-04-15 | Dynamic capacity increasing method and device for low-voltage distribution network, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210392368.1A CN114492212B (en) | 2022-04-15 | 2022-04-15 | Dynamic capacity increasing method and device for low-voltage distribution network, terminal and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114492212A CN114492212A (en) | 2022-05-13 |
| CN114492212B true CN114492212B (en) | 2022-07-12 |
Family
ID=81488250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210392368.1A Active CN114492212B (en) | 2022-04-15 | 2022-04-15 | Dynamic capacity increasing method and device for low-voltage distribution network, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114492212B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102004922A (en) * | 2010-12-01 | 2011-04-06 | 南京大学 | High-resolution remote sensing image plane extraction method based on skeleton characteristic |
| CN107729927A (en) * | 2017-09-30 | 2018-02-23 | 南京理工大学 | A kind of mobile phone application class method based on LSTM neutral nets |
| CN107832289A (en) * | 2017-10-12 | 2018-03-23 | 北京知道未来信息技术有限公司 | A kind of name entity recognition method based on LSTM CNN |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5933355A (en) * | 1995-05-04 | 1999-08-03 | Deb; Anjan Kumar | Object oriented expert power line ampacity system |
| CN105678439A (en) * | 2015-12-02 | 2016-06-15 | 国网山西省电力公司电力科学研究院 | Power transmission line dynamic capacity-increasing operation risk assessment method based on BP neural network |
| CN112765873B (en) * | 2020-11-30 | 2021-11-02 | 中国南方电网有限责任公司超高压输电公司广州局 | LSTM algorithm-based power transformation equipment heating prediction method |
-
2022
- 2022-04-15 CN CN202210392368.1A patent/CN114492212B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102004922A (en) * | 2010-12-01 | 2011-04-06 | 南京大学 | High-resolution remote sensing image plane extraction method based on skeleton characteristic |
| CN107729927A (en) * | 2017-09-30 | 2018-02-23 | 南京理工大学 | A kind of mobile phone application class method based on LSTM neutral nets |
| CN107832289A (en) * | 2017-10-12 | 2018-03-23 | 北京知道未来信息技术有限公司 | A kind of name entity recognition method based on LSTM CNN |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114492212A (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Aly | An intelligent hybrid model of neuro Wavelet, time series and Recurrent Kalman Filter for wind speed forecasting | |
| Chorin et al. | Discrete approach to stochastic parametrization and dimension reduction in nonlinear dynamics | |
| Celikoglu | Application of radial basis function and generalized regression neural networks in non-linear utility function specification for travel mode choice modelling | |
| Yahya et al. | Designing weather forecasting model using computational intelligence tools | |
| Wen et al. | Robust least squares support vector machine based on recursive outlier elimination | |
| Hwang et al. | A new measure for assessing the efficiency of hydrological data-driven forecasting models | |
| Shoaib et al. | Input selection of wavelet-coupled neural network models for rainfall-runoff modelling | |
| Waheeb et al. | Nonlinear autoregressive moving-average (narma) time series forecasting using neural networks | |
| Kasiviswanathan et al. | Probabilistic and ensemble simulation approaches for input uncertainty quantification of artificial neural network hydrological models | |
| Gao et al. | Neural networks for event extraction from time series: a back propagation algorithm approach | |
| Escot et al. | Estimating Lyapunov exponents on a noisy environment by global and local Jacobian indirect algorithms | |
| CN114492212B (en) | Dynamic capacity increasing method and device for low-voltage distribution network, terminal and storage medium | |
| Jin et al. | High precision variational Bayesian inference of sparse linear networks | |
| Feng et al. | Research and application of grey neural network in equipment life prediction | |
| Du et al. | A multi-output two-stage locally regularized model construction method using the extreme learning machine | |
| Roj | Estimation of the artificial neural network uncertainty used for measurand reconstruction in a sampling transducer | |
| CN112580853A (en) | Bus short-term load prediction method based on radial basis function neural network | |
| Vaccari et al. | Multivariate polynomial regression for identification of chaotic time series | |
| Zhang et al. | A refined rank set pair analysis model based on wavelet analysis for predicting temperature series | |
| Saikhu et al. | Non-Linear Spatio-Temporal Input Selection for Rainfall Forecasting Using Recurrent Neural Networks | |
| Zhang et al. | A Nonparametric Reconstruction of the Hubble Parameter H (z) Based on Radial Basis Function Neural Networks | |
| Rudenko et al. | Adaptive identification under the maximum correntropy criterion with variable center | |
| Amirkhan | Robust DEA models for performance evaluation of systems with continuous uncertain data under CRS and VRS conditions | |
| Xie et al. | The estimation of parameters for stochastic differential equations using neural networks | |
| Feizi et al. | Nonlinear autoregressive stochastic frontier model with dynamic technical inefficiency in panel data |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |