1. The intelligent control method for the power distribution network is characterized by comprising the following steps:

acquiring a payload parameter set;

obtaining a load distribution state potential set;

inputting a situation threshold value and comparing the situation threshold value with the load distribution situation set;

and adjusting the load parameters of the power distribution network according to the comparison result.

2. The intelligent control method for the power distribution network according to claim 1, wherein the acquiring the set of load effective parameters further comprises:

detecting the load state of the power distribution network;

generating an initial load parameter;

preprocessing the initial load parameters;

a set of payload parameters is generated.

3. The intelligent control method for the power distribution network according to claim 1, wherein the obtaining of the load distribution situation set further comprises:

obtaining an offset;

inputting a set of payload parameters and the offset into a dynamic deduction model;

correcting the payload parameter set according to an offset;

and outputting the load distribution state potential set.

4. The intelligent control method for the power distribution network according to claim 1, wherein the step of adjusting the load parameters of the power distribution network according to the comparison result further comprises the following steps:

generating a load offset parameter according to the comparison result;

and performing feedback adjustment on the initial load parameters of the power distribution network according to the load deviation parameters.

5. The intelligent control method for the power distribution network according to claim 2, wherein the preprocessing the initial load parameter comprises:

performing data association analysis on the corrected initial load parameter set according to the distribution information of the power distribution network to obtain an effective load parameter set;

the distribution information comprises distribution position information of each node in the power distribution network and load correlation information among the nodes.

6. The intelligent control method for the power distribution network according to claim 3, wherein the dynamic deduction model comprises:

receiving input data;

performing analog mapping on input data to obtain mapping data;

correcting offset processing is carried out on the mapping data to obtain corrected offset data;

and carrying out batch normalization processing on the corrected bias data, and obtaining a load distribution state potential set.

7. The intelligent control method for the power distribution network according to claim 1, wherein the situation threshold comprises a first situation threshold and a second situation threshold, the first situation threshold is a lower load limit value, and the second situation threshold is an upper load limit value;

when the load distribution state potential set is smaller than a first state threshold value, generating a first comparison result, wherein the first comparison result represents that the load distribution state potential set is smaller than the first state threshold value;

and when the load distribution state potential set is larger than the second state potential threshold value, generating a second comparison result, wherein the second comparison result indicates that the load distribution state potential set is larger than the second state potential threshold value.

8. An intelligent control device for a power distribution network, wherein the method of any one of claims 1 to 7 is performed, and wherein the method comprises:

a data acquisition unit: detecting the load state of the power distribution network to obtain an initial load parameter;

a data processing unit: preprocessing the initial load parameters, and eliminating error values in the initial load parameters to obtain an effective load parameter set; the effective load parameter set is dynamically deduced to obtain a load distribution state potential set, the load distribution state potential set and a state threshold value are compared to obtain a comparison result, and the data processing unit generates a load offset parameter according to the comparison result;

a control unit: and receiving the load deviation parameters, and adjusting the load of the power distribution network by the control unit according to the load deviation parameters.

9. The intelligent control device for distribution network according to claim 8, wherein the data obtaining unit further comprises:

a sensor: and detecting the load state of the power distribution network to generate corresponding initial load parameters.

10. The intelligent control device for the power distribution network according to claim 8, wherein the data acquisition unit is in communication connection with the data processing unit, and the data processing unit is in communication connection with the control unit.

Background

The reconstruction of the power distribution network is to change the states of a section switch and a contact switch in the power distribution network so as to control the network topology structure of the power distribution network to achieve the purpose of optimizing the power distribution network, the reconstruction of the power distribution network can be generally divided into static reconstruction based on a certain time point and dynamic reconstruction based on a time interval, intermittent power sources such as photovoltaic power, wind power and the like have obvious time variation and specific certainty, and the power demand of the power distribution network becomes more complex due to the time variation of loads.

At present, most of research on dynamic loads of a power distribution network is based on constant power distribution network control parameters under different conditions, and the loads of all nodes are adjusted under a consistent daily load curve, but in actual operation, real-time values of the loads and distributed power supplies are greatly different from predicted values, so that the constant power distribution network control parameters in actual operation cannot adapt to the continuously changing power distribution network electric energy requirements.

Disclosure of Invention

In view of the fact that in actual operation, the real-time numerical values of loads and distributed power supplies are different from the predicted numerical values greatly, so that constant power distribution network control parameters cannot adapt to the power demand of a power distribution network which changes continuously in actual operation, the power distribution network intelligent control method is provided.

An intelligent control method for a power distribution network comprises the following steps:

acquiring a payload parameter set; obtaining a load distribution state potential set; inputting a situation threshold value and comparing the situation threshold value with the load distribution situation set; and adjusting the load parameters of the power distribution network according to the comparison result.

Further, the acquiring the payload parameter set specifically includes the following steps:

detecting the load state of the power distribution network; generating an initial load parameter; preprocessing the initial load parameters; a set of payload parameters is generated.

Further, the obtaining of the load distribution situation set specifically further includes the following steps:

obtaining an offset; inputting a set of payload parameters and the offset into a dynamic deduction model; correcting the payload parameter set according to an offset; and outputting the load distribution state potential set.

Further, the adjusting the load parameter of the power distribution network according to the comparison result specifically includes the following steps:

generating a load offset parameter according to the comparison result; and performing feedback adjustment on the initial load parameters of the power distribution network according to the load deviation parameters.

Further, the preprocessing the initial load parameter includes:

performing data association analysis on the corrected initial load parameter set according to the distribution information of the power distribution network to obtain an effective load parameter set;

the distribution information comprises distribution position information of each node in the power distribution network and load correlation information among the nodes.

Further, the dynamic deduction model comprises:

receiving input data; performing analog mapping on input data to obtain mapping data; and correcting the mapping data to obtain corrected bias data, performing batch normalization on the corrected bias data, and obtaining a load distribution state potential set.

Further, the situation threshold includes a first situation threshold and a second situation threshold, the first situation threshold is a load lower limit, and the second situation threshold is a load upper limit;

when the load distribution state potential set is smaller than a first state threshold value, generating a first comparison result, wherein the first comparison result represents that the load distribution state potential set is smaller than the first state threshold value;

and when the load distribution state potential set is larger than the second state potential threshold value, generating a second comparison result, wherein the second comparison result indicates that the load distribution state potential set is larger than the second state potential threshold value.

An intelligent control device for a power distribution network executes an intelligent control method for the power distribution network, and comprises the following steps:

a data acquisition unit: detecting the load state of the power distribution network to obtain an initial load parameter;

a data processing unit: preprocessing the initial load parameters, and eliminating error values in the initial load parameters to obtain an effective load parameter set; the effective load parameter set is dynamically deduced to obtain a load distribution state potential set, the load distribution state potential set and a state threshold value are compared to obtain a comparison result, and the data processing unit generates a load offset parameter according to the comparison result;

a control unit: and receiving the load deviation parameters, and adjusting the load of the power distribution network by the control unit according to the load deviation parameters.

Further, the data obtaining unit further includes:

a sensor: and detecting the load state of the power distribution network to generate corresponding initial load parameters.

Further, the data acquisition unit is in communication connection with the data processing unit, and the data processing unit is in communication connection with the control unit.

According to the intelligent control method and device for the power distribution network, the effective load parameter set is obtained by preprocessing the initial load parameters, further, the load deviation parameters are generated according to the load distribution state potential set and the comparison result, the load deviation parameters are used as the adjusting parameters of the power distribution network, so that the load state of the power distribution network is sensed in time, and the corresponding deviation parameters are controlled and generated to adjust the load of each node of the power distribution network; the loads of all nodes of the power distribution network are dynamically adjusted, so that the continuously changing power demand of the power distribution network is met, and the flexibility of adjusting the power distribution network is realized.

Drawings

FIG. 1 is a process diagram of an intelligent control method for a power distribution network according to the present invention;

FIG. 2 is a detailed process diagram of step S1 according to the present invention;

FIG. 3 is a detailed process diagram of step S2 according to the present invention;

FIG. 4 is a detailed process diagram of step S4 according to the present invention;

FIG. 5 is a schematic diagram of the modules of the apparatus of the present invention.

Detailed Description

The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

The embodiments of the present disclosure are described below with specific examples, and other advantages and effects of the present disclosure will be readily apparent to those skilled in the art from the disclosure in the specification. It is to be understood that the described embodiments are merely illustrative of some, and not restrictive, of the embodiments of the disclosure. The disclosure may be embodied or carried out in various other specific embodiments, and various modifications and changes may be made in the details within the description without departing from the spirit of the disclosure. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.

The embodiment of the application provides an intelligent control method for a power distribution network, as shown in fig. 1 to 4, the intelligent control method for the power distribution network comprises the following steps:

step S1, obtaining a payload parameter set, specifically including the following steps:

step S101, detecting a load state;

in the intelligent control method for the power distribution network provided by this embodiment, the sensors are arranged at each node of the power distribution network to detect the load state of the power distribution network, so as to generate corresponding initial load parameters, each sensor uploads the corresponding initial load parameters to the server, and the processor corresponding to the server preprocesses the initial load parameters, so as to eliminate error values in the initial load parameters.

Step S102, generating an initial load parameter;

step S103, preprocessing the initial load parameters;

the preprocessing comprises the steps of carrying out data correction on the load parameters to obtain corrected data, and then carrying out data association analysis on the corrected data according to the distribution information of the power grid to obtain an effective load parameter set, wherein the data correction comprises data cleaning and data correction.

Step S104, a payload parameter set is generated.

The load of the distribution network node is in a dynamic change state, so that the initial load parameters obtained by detecting the distribution network node by a sensor at the bottom layer of a load system are easy to have errors, and in the data transmission process, because of channel transmission compiling errors, error data are introduced into the initial load parameters, therefore, invalid data or error data in the initial load parameters are removed by cleaning the initial load parameters, and data correction is carried out on the cleaned initial load parameters to correct data deviation caused by removing partial data. And the processor performs data association analysis on the corrected data according to the distribution position information and the load association information of each node in the power distribution network to obtain an effective parameter set. The effective parameter set can accurately represent the current load distribution of the whole power distribution network and the load relation among the nodes.

Step S2, obtaining a load distribution state potential set, which comprises the following steps:

step S201, obtaining an offset;

and acquiring an offset for representing the change of the load situation in a future period of time, and inputting the effective parameter set and the offset into the dynamic deduction model.

Step S202, inputting an effective load parameter set and an offset into a dynamic deduction model;

and the dynamic deduction model corrects the effective parameter set according to the offset and outputs the effective parameter set.

The input data is rapidly processed through the trained dynamic deduction model, so that the processing time can be effectively saved, and the information delay is reduced.

In a preferred embodiment of the present application, the dynamic deduction model comprises: an input layer for receiving input data; the neural network layer group is connected with the input layer and is used for carrying out preliminary simulation mapping on the input data and obtaining mapping data; the neural network layer group at least comprises a simulated neural network layer and a correction bias neural network layer; and the output layer is connected with the neural network layer group and used for carrying out batch normalization processing on the mapping data and obtaining output data.

In some embodiments, the dynamic deduction model comprises: an input layer for receiving input data; the neural network layer group is connected with the input layer and is used for carrying out preliminary simulation mapping on the input data and obtaining mapping data; the neural network layer group at least comprises a simulated neural network layer and a correction bias neural network layer; and the output layer is connected with the neural network layer group and used for carrying out batch normalization processing on the mapping data and obtaining output data. When the input data of the dynamic deduction model is an effective parameter set and an offset, the output data is a load distribution state potential set.

Furthermore, the input layer maps the input data to the corresponding neurons in the simulated neural network layer, and the neurons in the simulated neural network layer perform correction bias processing on the mapping data output by the neurons in the correction bias network layer to obtain correction bias data. And the output layer receives the correction bias data output by the neurons in the simulated neural network layer, and performs batch normalization processing on the correction bias data to obtain corresponding final result data.

In the neural network layer group, an analog neural network layer is connected with a correction bias neural network layer in cascade to be used as a neural network layer unit. And according to the actual data processing requirement, adding the number of the neural network layer units to form a corresponding neural network layer group. I.e. a plurality of neural network layer elements arranged in cascade to form a complete neural network layer set.

Namely, the input data of the situation distribution model is mapped layer by layer to deduce the voltage distribution parameters layer by layer, and the data generated by deduction is corrected adaptively. By means of the arrangement, deduction processing and correction processing are conducted alternately, and the situation prediction accuracy is prevented from being reduced due to the fact that abnormal data exist in input data.

In some embodiments, the expression of the input layer is:where a denotes the maximum value of the active parameter set, b denotes the minimum value of the active parameter set, and c denotes the offset of the dynamic variable. The dynamic variables and the effective parameter sets are used as the input of an input layer, and the initial simulation mapping is carried out on a neural network layer group.

It is understood that the offset amount indicating the end load power is obtained by a change state of the end load power in a plurality of consecutive unit times. And the offset is used as a calculation parameter for situation prediction, and the load parameters contained in the effective parameter set are used as a deduction reference, so that the load situation of each node of the power distribution network in a period of time in the future, namely a load distribution situation set is obtained.

Further, each neuron of the simulated neural network layer receives the data from the input layerAnd carrying out preliminary simulation mapping on the input data by the neurons of each simulation neural network layer.

The mapping method is as follows:

wherein the content of the first and second substances,in order to input the data, the data is,to model the mapping data of the output of the neurons of the neural network layer, W is the weight matrix obtained by pre-training the situation distribution model, z is the offset vector, and κ () is the activation function.

In addition, the data is input by simulating each neuron in the neural network layerPerforming preliminary simulation mapping to obtain mapping dataIn the correction bias neural network layer, each correction bias neuron performs correction bias processing on the mapping data and outputs the correction bias processing. Wherein, in the process of carrying out correction bias processing on the mapping data by the correction bias neuron, the following calculation is carried out:

wherein x modifies the input data of the bias neuron; and f (x) is output data of the correction bias neuron.

In the practical application process, the analog neural network layer and the correction bias neural network layer are arranged in an alternating arrangement mode to obtain a corresponding neural network layer group. The neural network layer group alternately carries out analog mapping and correction bias on input data so as to map and deduce a load distribution state potential set step by step. And the accurate dynamic load perception of the power distribution network is achieved through the load distribution state potential set, and the load distribution state potential set obtained according to the perception of the power distribution network is used as a reference for adjusting the load of the power distribution network.

And carrying out batch normalization processing on the mapping data through an output layer to obtain output data. In particular, the amount of the solvent to be used,

the input mapping data is x: β ═ x₁，…，x_mOutput data is y: y is_i＝BN_γ，β(x_i)

By pairing β ═ x₁，…，x_mCarry out batch processing to obtain the mean value mu of batch processing data_β. It is understood that the neural network formed by the above-described multilevel interconnection performs a plurality of arithmetic operations on the initial input data to obtain output data. And the output data is the solved load distribution state potential set.

Step S203, correcting the effective load parameter set according to the offset;

and the dynamic deduction model corrects the effective parameter set according to the offset and outputs the effective parameter set.

And step S204, outputting a load distribution state potential set.

When the input data of the dynamic deduction model is an effective parameter set and an offset, the output data is a load distribution state potential set.

Step S3, inputting a situation threshold value, and comparing the situation threshold value with a load distribution situation set;

the situation threshold comprises a first load situation threshold and a second load situation threshold; the load deflection parameters include: a load positive offset parameter, a load negative offset parameter; comparing the load distribution state potential set with the state threshold value to obtain a comparison result, wherein the comparison result comprises the following steps: if the load distribution state potential set is smaller than the first load state potential threshold value, generating a first comparison result; wherein the first comparison result is used to indicate that a load positive offset parameter is generated. The load distribution situation set comprises the load situations of all the nodes.

For example, the first load situation threshold and the second load situation threshold together define a threshold range, and the threshold range is a load range corresponding to the node. The first load situation threshold value is a lower load limit value, and the second load situation threshold value is an upper load limit value.

Step S4, adjusting the load parameters of the power distribution network according to the comparison result, which comprises the following steps:

step S401, generating a load offset parameter according to a comparison result;

and when the load distribution state potential set is smaller than the first load state potential threshold value, generating a first comparison result. And the first comparison result shows that the load distribution state potential set is smaller than the load distribution state potential set, namely the load state of the corresponding node is smaller than the lower limit value of the load, so that the load of the node is appropriately increased, and the loads of other nodes are reduced. And adjusting the loads of the corresponding node and the adjacent node through the positive load deviation parameter, thereby realizing the local power flow adjustment of the power distribution network. When the load distribution situation set indicates that the load of the node is within the threshold range, the node is taken as a dynamic point and is used for temporarily sharing the loads of other nodes

And S402, performing feedback adjustment on the initial load parameter of the power distribution network according to the load offset parameter.

And generating a load offset parameter according to the comparison result, and issuing the load offset parameter to a controller corresponding to each node of the power distribution network so as to adjust the electric energy distribution of the power distribution network, thereby performing feedback adjustment on the initial load parameter of the power distribution network.

In addition, the load situation perception unit stores a dynamic deduction model so as to perform situation perception processing on the input data. In some embodiments, the feedback adjusting unit is configured to obtain a feedback correction parameter according to the load distribution situation set and the real-time load distribution situation set, so as to correct the weight of each neuron of the neural network layer set stored in the load situation awareness unit.

The load correction unit further includes: the initial load parameter correcting unit is used for cleaning the initial load parameters to obtain corrected data; and the association analysis unit is used for performing data association analysis on the corrected data according to the distribution information of the power distribution network to obtain an effective parameter set.

It can be understood that, because the load of the distribution network node is in a dynamic change, the initial load parameters of the bottom-layer sensors, which are obtained by detecting the distribution network node, are prone to have errors, or error data is introduced in the data transmission process.

The initial load parameter correcting unit removes invalid data or error data in the initial load parameters by performing data cleaning on the initial load parameters, and corrects the deviation in the data by performing data correction on the cleaned initial load parameters. And the association analysis unit performs data association analysis on the corrected data according to the connection relation of each node in the power distribution network to obtain an effective parameter set. Wherein, the effective parameter set can effectively represent the current load distribution of the whole power distribution network.

The present application further provides a distribution network intelligent control device, as shown in fig. 5, including a data obtaining unit 501, a data processing unit 502, and a control unit 503, and executing the distribution network intelligent control method according to the foregoing embodiment, where:

the data acquisition unit 501: the system comprises a power distribution network, a load state detection module, a load parameter acquisition module and a load parameter acquisition module, wherein the power distribution network is used for detecting the load state of the power distribution network to obtain an initial load parameter;

502: preprocessing the initial load parameters, and eliminating error values in the initial load parameters to obtain an effective load parameter set; the effective load parameter set is dynamically deduced to obtain a load distribution state potential set, the load distribution state potential set and a state threshold value are compared to obtain a comparison result, and the data processing unit generates a load offset parameter according to the comparison result;

the control unit 503: and receiving the load deviation parameters, and adjusting the load of the power distribution network by the control unit according to the load deviation parameters.

Further, the data obtaining unit 501 further includes:

a sensor: and detecting the load state of the power distribution network to generate corresponding initial load parameters.

Further, the data acquisition unit 501 is connected to the data processing unit 502 in communication, and the data processing unit is connected to the control unit in communication.

In the description of the present invention, it is to be understood that the terms "intermediate", "length", "upper", "lower", "front", "rear", "vertical", "horizontal", "inner", "outer", "radial", "circumferential", and the like, indicate orientations and positional relationships that are based on the orientations and positional relationships shown in the drawings, are used for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and therefore, are not to be construed as limiting the present invention.

In the present invention, unless otherwise expressly stated or limited, the first feature may be "on" the second feature in direct contact with the second feature, or the first and second features may be in indirect contact via an intermediate. "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

In the present invention, unless otherwise expressly stated or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally formed; may be mechanically coupled, may be electrically coupled or may be in communication with each other; they may be directly connected or indirectly connected through intervening media, or they may be connected internally or in any other suitable relationship, unless expressly stated otherwise. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.

The above description is for the purpose of illustrating embodiments of the invention and is not intended to limit the invention, and it will be apparent to those skilled in the art that any modification, equivalent replacement, or improvement made without departing from the spirit and principle of the invention shall fall within the protection scope of the invention.