CN116957630A - Power demand response system based on economic data and method thereof - Google Patents

Abstract

The application relates to the field of intelligent regulation and control, and particularly discloses an electric power demand response system and method based on economic data. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

Description

Power demand response system based on economic data and method thereof

Technical Field

The application relates to the field of intelligent regulation and control, in particular to an electric power demand response system and method based on economic data.

Background

The aim of the power demand response is to improve the energy utilization efficiency, reduce the energy waste and promote the application and development of sustainable energy under the premise of ensuring the stable operation of a power system. Through reasonable demand response measures, supply and demand balance can be realized, the pressure of a power system is reduced, and energy consumption is reduced, so that sustainable development is realized. The current method for responding to the power demand mainly comprises the step that an expert adjusts the electricity price based on the supply and demand relation, energy cost, energy policy and other factors of the power market. However, this way of adjusting the electricity price is excessively dependent on the statistics and prediction ability of the expert, and is easily affected by human subjective influences, so that the optimal adjustment effect cannot be achieved.

Thus, an optimized power demand response scheme is desired.

Disclosure of Invention

The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an electric power demand response system and method based on economic data, which intelligently code and extract the characteristics of the economic data by using artificial intelligence technology based on a deep neural network model so as to obtain more accurate classification labels for indicating that the electricity price should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

According to one aspect of the present application, there is provided an electricity demand response system based on economic data, comprising:

the data acquisition module is used for acquiring economic condition data;

a context encoding module for passing the economic situation data through a converter-based context encoder comprising an embedded layer to obtain an economic feature vector;

a feature extraction module for passing the economic feature vector through a convolutional neural network model comprising a one-dimensional convolutional kernel as a feature extractor to obtain a classified feature vector;

The optimizing module is used for carrying out sparse robustness optimization based on clustering on the classification feature vectors so as to obtain optimized classification feature vectors;

and the result generation module is used for passing the optimized classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the electricity price should be increased or decreased.

In the above-described economic data-based power demand response system, the context encoding module includes:

the embedding conversion unit is used for respectively converting each dimension data in the economic condition data into economic embedding vectors through an embedding layer to obtain a sequence of the economic embedding vectors, wherein the embedding layer performs embedded coding on each dimension data by using a learnable embedding matrix;

an encoding unit for inputting the sequence of economic embedded vectors into the converter-based context encoder to obtain the plurality of economic context semantic feature vectors;

and the cascading unit is used for cascading the plurality of economic semantic feature vectors to obtain the economic feature vectors.

In the above-described economic data-based power demand response system, the encoding unit includes:

A query vector construction subunit configured to arrange the sequence of economic embedded vectors into an input vector;

the vector conversion subunit is used for respectively converting the input vector into a query vector and a key vector through a learning embedding matrix;

a self-attention subunit, configured to calculate a product between the query vector and a transpose vector of the key vector to obtain a self-attention correlation matrix;

a normalization subunit, configured to perform normalization processing on the self-attention association matrix to obtain a normalized self-attention association matrix;

the attention calculating subunit is used for inputting the standardized self-attention association matrix into a Softmax activation function to activate so as to obtain a self-attention feature matrix;

and the attention applying subunit is used for multiplying the self-attention characteristic matrix with each economic embedding vector in the sequence of the user embedding vectors to obtain the plurality of economic semantic context characteristic vectors.

In the above-described power demand response system based on economic data, the feature extraction module is configured to:

input data are respectively subjected to forward transfer of layers by using each layer of the convolutional neural network model which is taken as a characteristic extractor and comprises a one-dimensional convolutional kernel:

Using convolution units of all layers of the convolution neural network model to carry out convolution processing on the input data based on one-dimensional convolution kernels so as to obtain a convolution feature map;

using pooling units of each layer of the convolutional neural network model to perform pooling processing of each local feature matrix along a channel dimension on the convolutional feature map so as to obtain a pooled feature map;

using an activation unit of each layer of the convolutional neural network model to perform nonlinear activation on the characteristic values of each position in the pooled characteristic map so as to obtain an activated characteristic map;

wherein the output of the last layer of the convolutional neural network model is the classification feature vector.

In the above-described economic data-based power demand response system, the optimization module includes:

the feature level expression strengthening unit is used for carrying out feature level expression strengthening on the classified feature vectors based on the Gaussian density diagram to obtain a strengthening feature matrix, wherein each strengthening feature vector in the strengthening feature matrix corresponds to a feature value of each position in the classified feature vector;

a bulldozer distance calculating unit configured to calculate bulldozer distances between each of the reinforcement feature vectors and other reinforcement feature vectors in the reinforcement feature matrix to obtain a plurality of bulldozer distances of each of the reinforcement feature vectors;

A summation unit for calculating summation values of a plurality of bulldozer distances of the respective strengthening feature vectors as robust cluster dependency feature values of the respective strengthening feature vectors;

a masking unit, configured to mask the classification feature vector based on the robust cluster dependency feature values of the respective reinforcement feature vectors;

and the optimizing unit is used for carrying out robust clustering dependency optimization on the classification characteristic vector so as to obtain an optimized classification characteristic vector.

In the above-described economic data-based power demand response system, the masking unit includes: based on the comparison between the robust cluster-dependent feature values of the respective enhanced feature vectors and a predetermined threshold, it is determined whether to zero the feature values of the corresponding positions in the classified feature vectors.

In the above-described economic data-based power demand response system, the result generation module is configured to:

processing the optimized classification feature vector with the classifier in the following classification formula to obtain the classification result;

wherein, the classification formula is: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) X, where W ₁ To W _n Is a weight matrix, B ₁ To B _n For bias vectors, X is the optimized classification feature vector, softmax represents a softmax function, and O represents the classification result.

According to another aspect of the present application, there is also provided an electricity demand response method based on economic data, including:

acquiring economic condition data;

passing the economic situation data through a converter-based context encoder comprising an embedded layer to obtain an economic feature vector;

passing the economic feature vector through a convolutional neural network model comprising a one-dimensional convolutional kernel as a feature extractor to obtain a classified feature vector;

performing sparse robustness optimization based on clustering on the classification feature vectors to obtain optimized classification feature vectors;

the optimized classification feature vector is passed through a classifier to obtain a classification result, which is used to indicate that electricity prices should be increased or decreased.

Compared with the prior art, the power demand response system and the power demand response method based on the economic data, provided by the application, use an artificial intelligence technology based on a deep neural network model to intelligently perform feature coding and extraction on the economic data so as to obtain a more accurate classification label for indicating that the electricity price should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

Drawings

The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.

FIG. 1 illustrates a block diagram of an economic data based power demand response system in accordance with an embodiment of the present application.

Fig. 2 illustrates a system architecture diagram of an economic data based power demand response system in accordance with an embodiment of the present application.

FIG. 3 illustrates a block diagram of a context encoding module in an economic data based power demand response system in accordance with an embodiment of the present application.

Fig. 4 illustrates a block diagram of an encoding unit in an economic data based power demand response system according to an embodiment of the present application.

FIG. 5 illustrates a flow chart of an economic data based power demand response method in accordance with an embodiment of the present application.

Fig. 6 illustrates a block diagram of an electronic device according to an embodiment of the application.

Detailed Description

Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

Summary of the application

As described above in the background art, the current method for responding to power demand is mainly to adjust the electricity price by an expert based on the supply and demand relationship of the power market, energy cost, energy policy, and other factors. However, this way of adjusting the electricity price is excessively dependent on the statistics and prediction ability of the expert, and is easily affected by human subjective influences, so that the optimal adjustment effect cannot be achieved. Thus, an optimized power demand response scheme is desired.

In view of the above problems, the present application proposes an electricity demand response scheme based on economic data, which uses artificial intelligence technology based on a deep neural network model to intelligently perform feature encoding and extraction on the economic data, so as to obtain more accurate classification labels for indicating that electricity prices should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

At present, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, speech signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.

In recent years, deep learning and the development of neural networks have provided new solutions and solutions for economic data-based power demand response systems.

Specifically, the economic condition data is first acquired. It should be appreciated that the economic data is multi-dimensional data. Specifically, the economic data includes economic data including power demand curves, power price data, user participation data, and cost benefit analysis data, such as power supply and demand data, power cost data, power utilization data, and the like. Such data may come from power suppliers, government agencies, users participating in surveys, and the like.

Further, the economic situation data is passed through a converter-based context encoder comprising an embedded layer to obtain an economic feature vector. The embedded layer is used to convert the input economic data into a continuous vector representation. The economic data contains various different types of information such as time series data, classification data, and the like. These different types of data can be converted into a unified vector representation by the embedding layer for subsequent feature extraction and classification tasks. The converter model is a powerful sequence-to-sequence model that can handle long-term dependencies of input sequences. It effectively captures contextual information in the input sequence through a self-attention mechanism. Thus, applying the converter to context encoding may better capture relevant features in the economic data. The context-based coding module of the converter may learn advanced representations of the input data, which may better express the complexity and diversity of the economic situation. Such advanced representations may help the feature extraction module to better extract classification features, thereby improving the predictive accuracy of the system. The use of a converter-based context coding module that includes an embedded layer may improve the system's understanding and modeling capabilities of the economic data, thereby improving the accuracy and effectiveness of the power demand prediction.

Specifically, first, the economic condition data are respectively passed through an embedding layer to convert each dimension data in the economic condition data into economic embedding vectors to obtain a sequence of economic embedding vectors, wherein the embedding layer uses a learnable embedding matrix to carry out embedded coding on each dimension data. Next, the sequence of economic embedded vectors is input to the converter-based context encoder to derive the plurality of economic context semantic feature vectors. Further, the plurality of economic semantic feature vectors are concatenated to obtain the economic feature vector.

The economic feature vector is then passed through a convolutional neural network model containing one-dimensional convolutional kernels as a feature extractor to obtain a classification feature vector. Convolutional neural networks perform well in the field of image processing, but they can also be used to process sequence data, such as economic feature vectors. The one-dimensional convolution kernel may capture local patterns and features in the sequence data and is therefore suitable for extracting classification features in the economic feature vector. By using convolutional neural networks as feature extractors, important features in economic feature vectors can be automatically learned. The filter of the convolution layer can be slid over different locations to capture features of different locations. Such local perceptibility may help the system better understand the timing and spatial relationships in economic feature vectors. By using convolutional neural networks as feature extractors, important features in economic feature vectors can be automatically learned. The filter of the convolution layer can be slid over different locations to capture features of different locations. Such local perceptibility may help the system better understand the timing and spatial relationships in economic feature vectors. By using convolutional neural networks as feature extractors, important features in economic feature vectors can be automatically learned. The filter of the convolution layer can be slid over different locations to capture features of different locations. Such local perceptibility may help the system better understand the timing and spatial relationships in economic feature vectors.

Further, the classification feature vector is passed through a classifier to obtain a classification result for indicating that the electricity price should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

In particular, in the technical scheme of the application, considering that the classification feature vector is a high-dimensional, complex and non-convex feature element set, abnormal values and noise points contained in the classification feature vector influence the classification judgment accuracy of the classification feature vector. Based on this, in the technical scheme of the application, firstly, feature level expression enhancement is performed on the classification feature vector through a gaussian density chart, which is essentially that feature values of all positions in the classification feature vector are subjected to feature level enhancement based on prior distribution information by utilizing prior distribution information of the classification feature vector, and feature levels enhancement based on prior distribution information is performed on overall feature distribution of the classification feature vector, so as to obtain an enhancement feature matrix, wherein all enhancement feature vectors in the enhancement feature matrix correspond to feature values of all positions in the classification feature vector.

Further, similarity of feature distribution between each reinforcement feature vector and other reinforcement feature vectors in the reinforcement feature matrix is expressed by bulldozer distances between each reinforcement feature vector and other reinforcement feature vectors in the reinforcement feature matrix. And calculating the summation value of the plurality of bulldozer distances of each strengthening feature vector as a robust clustering dependency feature value of each strengthening feature vector, wherein the robust clustering dependency feature value is used for representing clustering performance between each strengthening feature vector and other strengthening feature vectors in the strengthening feature matrix. Then, masking the classification feature vector based on the robust cluster dependency feature values of the respective reinforcement feature vectors to robust cluster dependency optimization of the classification feature vector to obtain an optimized classification feature vector. For example, in one specific example of the present application, it is determined whether to zero the feature value of the corresponding position in the classification feature vector based on a comparison between the robust cluster dependency feature value of the respective reinforcement feature vector and a predetermined threshold.

In this way, clustering-based sparse robustness optimization is performed on the classification feature vectors to effectively identify outliers or noise points in the classification feature vectors based on the data distribution internal characteristics of the feature sets of the classification feature vectors so as to reduce the effective dimensions of the classification feature vectors and enable the classification feature vectors to more effectively reflect the essential features and rules of data, and in this way, the accuracy of classification judgment of the classification feature vectors is improved.

Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.

Exemplary System

FIG. 1 illustrates a block diagram of an economic data based power demand response system in accordance with an embodiment of the present application. As shown in fig. 1, an economic data-based power demand response system 100 according to an embodiment of the present application includes: a data acquisition module 110 for acquiring economic condition data; a context encoding module 120 for passing the economic situation data through a converter-based context encoder comprising an embedded layer to obtain an economic feature vector; a feature extraction module 130 for passing the economic feature vector through a convolutional neural network model including a one-dimensional convolutional kernel as a feature extractor to obtain a classified feature vector; an optimizing module 140, configured to perform cluster-based sparse robustness optimization on the classification feature vector to obtain an optimized classification feature vector; and a result generation module 150, configured to pass the optimized classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate that the electricity price should be increased or decreased.

Fig. 2 illustrates a system architecture diagram of an economic data based power demand response system in accordance with an embodiment of the present application. In this system architecture, as shown in fig. 2, first, economic data is acquired. The economic situation data is then passed through a converter-based context encoder comprising an embedded layer to obtain an economic feature vector. The economic feature vector is then passed through a convolutional neural network model comprising a one-dimensional convolutional kernel as a feature extractor to obtain a classification feature vector. Further, the classification feature vector is passed through a classifier to obtain a classification result indicating that the electricity rate should be increased or decreased.

In the power demand response system 100 based on economic data, the data acquisition module 110 is configured to acquire economic data. As described above in the background art, the current method for responding to power demand is mainly to adjust the electricity price by an expert based on the supply and demand relationship of the power market, energy cost, energy policy, and other factors. However, this way of adjusting the electricity price is excessively dependent on the statistics and prediction ability of the expert, and is easily affected by human subjective influences, so that the optimal adjustment effect cannot be achieved. Thus, an optimized power demand response scheme is desired.

In view of the above problems, the present application proposes an electricity demand response scheme based on economic data, which uses artificial intelligence technology based on a deep neural network model to intelligently perform feature encoding and extraction on the economic data, so as to obtain more accurate classification labels for indicating that electricity prices should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

At present, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, speech signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.

In recent years, deep learning and the development of neural networks have provided new solutions and solutions for economic data-based power demand response systems.

Specifically, the economic condition data is first acquired. It should be appreciated that the economic data is multi-dimensional data. Specifically, the economic data includes economic data including power demand curves, power price data, user participation data, and cost benefit analysis data, such as power supply and demand data, power cost data, power utilization data, and the like. Such data may come from power suppliers, government agencies, users participating in surveys, and the like.

In the above-described economic data-based power demand response system 100, the context encoding module 120 is configured to pass the economic data through a converter-based context encoder including an embedded layer to obtain an economic feature vector. The embedded layer is used to convert the input economic data into a continuous vector representation. The economic data contains various different types of information such as time series data, classification data, and the like. These different types of data can be converted into a unified vector representation by the embedding layer for subsequent feature extraction and classification tasks. The converter model is a powerful sequence-to-sequence model that can handle long-term dependencies of input sequences. It effectively captures contextual information in the input sequence through a self-attention mechanism. Thus, applying the converter to context encoding may better capture relevant features in the economic data. The context-based coding module of the converter may learn advanced representations of the input data, which may better express the complexity and diversity of the economic situation. Such advanced representations may help the feature extraction module to better extract classification features, thereby improving the predictive accuracy of the system. The use of a converter-based context coding module that includes an embedded layer may improve the system's understanding and modeling capabilities of the economic data, thereby improving the accuracy and effectiveness of the power demand prediction.

Specifically, first, the economic condition data are respectively passed through an embedding layer to convert each dimension data in the economic condition data into economic embedding vectors to obtain a sequence of economic embedding vectors, wherein the embedding layer uses a learnable embedding matrix to carry out embedded coding on each dimension data. Next, the sequence of economic embedded vectors is input to the converter-based context encoder to derive the plurality of economic context semantic feature vectors. Further, the plurality of economic semantic feature vectors are concatenated to obtain the economic feature vector.

FIG. 3 illustrates a block diagram of a context encoding module in an economic data based power demand response system in accordance with an embodiment of the present application. As shown in fig. 3, the context encoding module 120 includes: an embedding transformation unit 121, configured to transform each dimension data in the economic situation data into an economic embedding vector by using an embedding layer to obtain a sequence of economic embedding vectors, where the embedding layer uses a learnable embedding matrix to perform embedded encoding on each dimension data; an encoding unit 122 for inputting the sequence of economic embedded vectors into the converter-based context encoder to obtain the plurality of economic context semantic feature vectors; and a cascade unit for cascading the plurality of economic semantic feature vectors to obtain the economic feature vector.

Fig. 4 illustrates a block diagram of an encoding unit in an economic data based power demand response system according to an embodiment of the present application. As shown in fig. 4, the encoding unit 122 includes: a query vector construction subunit 1221, configured to arrange the sequence of economic embedded vectors into an input vector; a vector conversion subunit 1222 for converting the input vector into a query vector and a key vector, respectively, through a learning embedding matrix; a self-attention subunit 1223, configured to calculate a product between the query vector and a transpose vector of the key vector to obtain a self-attention correlation matrix; a normalization subunit 1224, configured to perform normalization processing on the self-attention association matrix to obtain a normalized self-attention association matrix; a attention computation subunit 1225, configured to activate the normalized self-attention association matrix input Softmax activation function to obtain a self-attention feature matrix; and an attention applying subunit 1226, configured to multiply the self-attention feature matrix with each economic embedding vector in the sequence of user embedding vectors to obtain the plurality of economic semantic context feature vectors, respectively.

In the above-mentioned power demand response system 100 based on economic data, the feature extraction module 130 is configured to pass the economic feature vector through a convolutional neural network model including a one-dimensional convolutional kernel as a feature extractor to obtain a classification feature vector. Convolutional neural networks perform well in the field of image processing, but they can also be used to process sequence data, such as economic feature vectors. The one-dimensional convolution kernel may capture local patterns and features in the sequence data and is therefore suitable for extracting classification features in the economic feature vector. By using convolutional neural networks as feature extractors, important features in economic feature vectors can be automatically learned. The filter of the convolution layer can be slid over different locations to capture features of different locations. Such local perceptibility may help the system better understand the timing and spatial relationships in economic feature vectors. By using convolutional neural networks as feature extractors, important features in economic feature vectors can be automatically learned. The filter of the convolution layer can be slid over different locations to capture features of different locations. Such local perceptibility may help the system better understand the timing and spatial relationships in economic feature vectors. By using convolutional neural networks as feature extractors, important features in economic feature vectors can be automatically learned. The filter of the convolution layer can be slid over different locations to capture features of different locations. Such local perceptibility may help the system better understand the timing and spatial relationships in economic feature vectors.

Specifically, in the embodiment of the present application, the feature extraction module 130 is configured to: input data are respectively subjected to forward transfer of layers by using each layer of the convolutional neural network model which is taken as a characteristic extractor and comprises a one-dimensional convolutional kernel: using convolution units of all layers of the convolution neural network model to carry out convolution processing on the input data based on one-dimensional convolution kernels so as to obtain a convolution feature map; using pooling units of each layer of the convolutional neural network model to perform pooling processing of each local feature matrix along a channel dimension on the convolutional feature map so as to obtain a pooled feature map; using an activation unit of each layer of the convolutional neural network model to perform nonlinear activation on the characteristic values of each position in the pooled characteristic map so as to obtain an activated characteristic map; wherein the output of the last layer of the convolutional neural network model is the classification feature vector.

In the above-mentioned power demand response system 100 based on economic data, the optimizing module 140 is configured to perform sparse robustness optimization based on clustering on the classification feature vector to obtain an optimized classification feature vector. In the technical scheme of the application, the classification feature vector is considered to be a high-dimensional, complex and non-convex feature element set, and abnormal values and noise points contained in the classification feature vector influence the classification judgment accuracy of the classification feature vector. Based on this, in the technical scheme of the application, firstly, feature level expression enhancement is performed on the classification feature vector through a gaussian density chart, which is essentially that feature values of all positions in the classification feature vector are subjected to feature level enhancement based on prior distribution information by utilizing prior distribution information of the classification feature vector, and feature levels enhancement based on prior distribution information is performed on overall feature distribution of the classification feature vector, so as to obtain an enhancement feature matrix, wherein all enhancement feature vectors in the enhancement feature matrix correspond to feature values of all positions in the classification feature vector.

Further, similarity of feature distribution between each reinforcement feature vector and other reinforcement feature vectors in the reinforcement feature matrix is expressed by bulldozer distances between each reinforcement feature vector and other reinforcement feature vectors in the reinforcement feature matrix. And calculating the summation value of the plurality of bulldozer distances of each strengthening feature vector as a robust clustering dependency feature value of each strengthening feature vector, wherein the robust clustering dependency feature value is used for representing clustering performance between each strengthening feature vector and other strengthening feature vectors in the strengthening feature matrix. Then, masking the classification feature vector based on the robust cluster dependency feature values of the respective reinforcement feature vectors to robust cluster dependency optimization of the classification feature vector to obtain an optimized classification feature vector. For example, in one specific example of the present application, it is determined whether to zero the feature value of the corresponding position in the classification feature vector based on a comparison between the robust cluster dependency feature value of the respective reinforcement feature vector and a predetermined threshold.

In this way, clustering-based sparse robustness optimization is performed on the classification feature vectors to effectively identify outliers or noise points in the classification feature vectors based on the data distribution internal characteristics of the feature sets of the classification feature vectors so as to reduce the effective dimensions of the classification feature vectors and enable the classification feature vectors to more effectively reflect the essential features and rules of data, and in this way, the accuracy of classification judgment of the classification feature vectors is improved.

Specifically, in the embodiment of the present application, the optimization module 140 includes: the feature level expression strengthening unit is used for carrying out feature level expression strengthening on the classified feature vectors based on the Gaussian density diagram to obtain a strengthening feature matrix, wherein each strengthening feature vector in the strengthening feature matrix corresponds to a feature value of each position in the classified feature vector; a bulldozer distance calculating unit configured to calculate bulldozer distances between each of the reinforcement feature vectors and other reinforcement feature vectors in the reinforcement feature matrix to obtain a plurality of bulldozer distances of each of the reinforcement feature vectors; a summation unit for calculating summation values of a plurality of bulldozer distances of the respective strengthening feature vectors as robust cluster dependency feature values of the respective strengthening feature vectors; a masking unit, configured to mask the classification feature vector based on the robust cluster dependency feature values of the respective reinforcement feature vectors; and the optimizing unit is used for carrying out robust clustering dependency optimization on the classification characteristic vector so as to obtain an optimized classification characteristic vector.

Specifically, in an embodiment of the present application, the masking unit includes: based on the comparison between the robust cluster-dependent feature values of the respective enhanced feature vectors and a predetermined threshold, it is determined whether to zero the feature values of the corresponding positions in the classified feature vectors.

In the above-described economic data-based power demand response system 100, the result generation module 150 is configured to pass the optimized classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate that the electricity price should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

Specifically, in the embodiment of the present application, the result generating module 150 is configured to: processing the optimized classification feature vector with the classifier in the following classification formula to obtain the classification result; wherein, the classification formula is: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) X, where W ₁ To W _n Is a weight matrix, B ₁ To B _n Is a bias vector, X is theOptimizing the classification feature vector, softmax representing a softmax function, and O representing the classification result.

In summary, an economic data-based power demand response system according to an embodiment of the present application has been explained, which uses artificial intelligence technology based on a deep neural network model to intelligently perform feature encoding and extraction on economic data to obtain more accurate classification tags for indicating that electricity prices should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

Exemplary method

FIG. 5 illustrates a flow chart of an economic data based power demand response method in accordance with an embodiment of the present application. As shown in fig. 5, the power demand response method based on economic data according to an embodiment of the present application includes the steps of: s110, acquiring economic condition data; s120, the economic condition data is passed through a converter-based context encoder comprising an embedded layer to obtain economic feature vectors; s130, passing the economic feature vector through a convolutional neural network model which is taken as a feature extractor and comprises a one-dimensional convolutional kernel to obtain a classification feature vector; s140, performing sparse robustness optimization on the classification feature vectors based on clustering to obtain optimized classification feature vectors; and S140, passing the optimized classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the electricity price is required to be increased or decreased.

Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described economic data-based power demand response method have been described in detail in the above description of the economic data-based power demand response system with reference to fig. 1 to 4, and thus, repetitive descriptions thereof will be omitted.

As described above, the economic data-based power demand response system 100 according to the embodiment of the present application may be implemented in various terminal devices, such as an economic data-based power demand response server, etc. In one example, the economic data based power demand response system 100 according to an embodiment of the present application may be integrated into the terminal device as one software module and/or hardware module. For example, the economic data-based power demand response system 100 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the economic data based power demand response system 100 may also be one of a number of hardware modules of the terminal device.

Alternatively, in another example, the economic data-based power demand response system 100 and the terminal device may be separate devices, and the economic data-based power demand response system 100 may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information in a agreed data format.

In summary, an electric power demand response method based on economic data according to an embodiment of the present application has been explained, which uses artificial intelligence technology based on a deep neural network model to intelligently perform feature encoding and extraction on economic data to obtain more accurate classification labels for indicating that electricity prices should be increased or decreased. Therefore, an electric power demand response scheme based on the economic data is constructed to synthesize the multidimensional economic data, and further the electric power demand can be accurately predicted based on the classification result, so that the effect of reducing the resource waste is achieved.

Exemplary electronic device

Next, an electronic device according to an embodiment of the present application is described with reference to fig. 6. Fig. 6 is a block diagram of an electronic device according to an embodiment of the application. As shown in fig. 6, the electronic device 10 includes one or more processors 11 and a memory 12.

The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that can be executed by the processor 11 to perform the functions in the economic data based power demand response method and/or other desired functions of the various embodiments of the present application described above. Various contents such as economic condition data may also be stored in the computer-readable storage medium.

In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).

The input means 13 may comprise, for example, a keyboard, a mouse, etc.

The output device 14 can output various information to the outside, including that the electricity price should be increased or decreased, etc. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.

Of course, only some of the components of the electronic device 10 that are relevant to the present application are shown in fig. 6 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.

Exemplary computer program product and computer readable storage Medium

In addition to the methods and apparatus described above, embodiments of the application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform steps in the functions of the economic data based power demand response method according to the various embodiments of the application described in the "exemplary methods" section of this specification.

The computer program product may write program code for performing operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present application may also be a computer-readable storage medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform steps in the functions of the economic data based power demand response method according to the various embodiments of the present application described in the above "exemplary method" section of the present specification.

The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.

The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.

It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.

Claims (10)

1. An electrical demand response system based on economic data, comprising:

the data acquisition module is used for acquiring economic condition data;

A context encoding module for passing the economic situation data through a converter-based context encoder comprising an embedded layer to obtain an economic feature vector;

a feature extraction module for passing the economic feature vector through a convolutional neural network model comprising a one-dimensional convolutional kernel as a feature extractor to obtain a classified feature vector;

the optimizing module is used for carrying out sparse robustness optimization based on clustering on the classification feature vectors so as to obtain optimized classification feature vectors;

and the result generation module is used for passing the optimized classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the electricity price should be increased or decreased.

2. The economic data-based power demand response system of claim 1, wherein the context encoding module comprises:

the embedding conversion unit is used for respectively converting each dimension data in the economic condition data into economic embedding vectors through an embedding layer to obtain a sequence of the economic embedding vectors, wherein the embedding layer performs embedded coding on each dimension data by using a learnable embedding matrix;

An encoding unit for inputting the sequence of economic embedded vectors into the converter-based context encoder to obtain the plurality of economic context semantic feature vectors;

and the cascading unit is used for cascading the plurality of economic semantic feature vectors to obtain the economic feature vectors.

3. The economic data based power demand response system of claim 2, wherein the encoding unit comprises:

a query vector construction subunit configured to arrange the sequence of economic embedded vectors into an input vector;

the vector conversion subunit is used for respectively converting the input vector into a query vector and a key vector through a learning embedding matrix;

a self-attention subunit, configured to calculate a product between the query vector and a transpose vector of the key vector to obtain a self-attention correlation matrix;

a normalization subunit, configured to perform normalization processing on the self-attention association matrix to obtain a normalized self-attention association matrix;

the attention calculating subunit is used for inputting the standardized self-attention association matrix into a Softmax activation function to activate so as to obtain a self-attention feature matrix;

and the attention applying subunit is used for multiplying the self-attention characteristic matrix with each economic embedding vector in the sequence of the user embedding vectors to obtain the plurality of economic semantic context characteristic vectors.

4. The economic data-based power demand response system of claim 3, wherein the feature extraction module is configured to:

input data are respectively subjected to forward transfer of layers by using each layer of the convolutional neural network model which is taken as a characteristic extractor and comprises a one-dimensional convolutional kernel:

using convolution units of all layers of the convolution neural network model to carry out convolution processing on the input data based on one-dimensional convolution kernels so as to obtain a convolution feature map;

using pooling units of each layer of the convolutional neural network model to perform pooling processing of each local feature matrix along a channel dimension on the convolutional feature map so as to obtain a pooled feature map;

using an activation unit of each layer of the convolutional neural network model to perform nonlinear activation on the characteristic values of each position in the pooled characteristic map so as to obtain an activated characteristic map;

wherein the output of the last layer of the convolutional neural network model is the classification feature vector.

5. The economic data-based power demand response system of claim 4, wherein the optimization module comprises:

the feature level expression strengthening unit is used for carrying out feature level expression strengthening on the classified feature vectors based on the Gaussian density diagram to obtain a strengthening feature matrix, wherein each strengthening feature vector in the strengthening feature matrix corresponds to a feature value of each position in the classified feature vector;

A bulldozer distance calculating unit configured to calculate bulldozer distances between each of the reinforcement feature vectors and other reinforcement feature vectors in the reinforcement feature matrix to obtain a plurality of bulldozer distances of each of the reinforcement feature vectors;

a summation unit for calculating summation values of a plurality of bulldozer distances of the respective strengthening feature vectors as robust cluster dependency feature values of the respective strengthening feature vectors;

a masking unit, configured to mask the classification feature vector based on the robust cluster dependency feature values of the respective reinforcement feature vectors;

and the optimizing unit is used for carrying out robust clustering dependency optimization on the classification characteristic vector so as to obtain an optimized classification characteristic vector.

6. The economic data-based power demand response system of claim 5, wherein the masking unit comprises: based on the comparison between the robust cluster-dependent feature values of the respective enhanced feature vectors and a predetermined threshold, it is determined whether to zero the feature values of the corresponding positions in the classified feature vectors.

7. The economic data-based power demand response system of claim 6, wherein the result generation module is configured to:

Processing the optimized classification feature vector with the classifier in the following classification formula to obtain the classification result;

wherein, the classification formula is: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) X, where W ₁ To W _n Is a weight matrix, B ₁ To B _n For bias vectors, X is the optimized classification feature vector, softmax represents a softmax function, and O represents the classification result.

8. An electricity demand response method based on economic data, comprising:

acquiring economic condition data;

passing the economic situation data through a converter-based context encoder comprising an embedded layer to obtain an economic feature vector;

passing the economic feature vector through a convolutional neural network model comprising a one-dimensional convolutional kernel as a feature extractor to obtain a classified feature vector;

performing sparse robustness optimization based on clustering on the classification feature vectors to obtain optimized classification feature vectors;

the optimized classification feature vector is passed through a classifier to obtain a classification result, which is used to indicate that electricity prices should be increased or decreased.

9. The economic data-based power demand response method of claim 8, wherein passing the economic feature vector through a convolutional neural network model including a one-dimensional convolutional kernel as a feature extractor to obtain a classification feature vector comprises:

Input data are respectively subjected to forward transfer of layers by using each layer of the convolutional neural network model which is taken as a characteristic extractor and comprises a one-dimensional convolutional kernel:

using convolution units of all layers of the convolution neural network model to carry out convolution processing on the input data based on one-dimensional convolution kernels so as to obtain a convolution feature map;

using pooling units of each layer of the convolutional neural network model to perform pooling processing of each local feature matrix along a channel dimension on the convolutional feature map so as to obtain a pooled feature map;

using an activation unit of each layer of the convolutional neural network model to perform nonlinear activation on the characteristic values of each position in the pooled characteristic map so as to obtain an activated characteristic map;

wherein the output of the last layer of the convolutional neural network model is the classification feature vector.

10. The economic data-based power demand response method according to claim 9, wherein passing the optimized classification feature vector through a classifier to obtain a classification result indicating that the electricity price should be increased or decreased, comprises:

processing the optimized classification feature vector with the classifier in the following classification formula to obtain the classification result;

Wherein, the classification formula is: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) X, where W ₁ To W _n Is a weight matrix, B ₁ To B _n For bias vectors, X is the optimized classification feature vector, softmax represents a softmax function, and O represents the classification result.