Photovoltaic user adjustable potential assessment method and system, electronic equipment and storage medium

Abstract

The invention discloses a photovoltaic user adjustable potential assessment method and system, an electronic device and a storage medium, the method classifies power consumption behaviors of photovoltaic users, takes differentiated power consumption demands of different photovoltaic users into consideration, can accurately assess the load adjustment capability of different types of surplus power Internet surfing users, and can accurately assess the adjustable potential of the photovoltaic users. According to the method, the adjustable potential of the surplus power Internet surfing user is researched for the first time, theoretical support is provided for guiding the surplus power Internet surfing user to voluntarily adjust the power consumption behavior according to the operation requirement of the power system and achieve peak load shifting and local consumption, the flexibility of the power system is improved, and safe and stable operation of the power system can be effectively guaranteed; and the renewable energy power consumption is greatly promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of power dispatching, and in particular to a photovoltaic user adjustable potential assessment method and system, electronic equipment, and a computer-readable storage medium. Background Art

[0002] With the widespread access to renewable energy and the development of the electricity market, the power system is facing the challenges of load fluctuations and imbalance between supply and demand, especially during peak hours. The load pressure on the power grid increases. How to improve the flexibility of power grid dispatching and reduce the system operation cost has become a key issue. Demand response, as an effective solution, can adjust the user's electricity consumption behavior and transfer part of the load to the off-peak period, which can not only relieve the pressure on the power grid, but also help improve the utilization efficiency of renewable energy. Therefore, studying the adjustable potential of user electricity load can not only provide more accurate decision-making basis for power grid dispatching, but also optimize the operation of the power market, reduce electricity demand during peak hours, and reduce dependence on traditional fossil energy. This research will help promote the development of smart grids and green and low-carbon, improve the stability and reliability of the power system, and promote the transformation of the energy structure.

[0003] However, the electricity demand of different photovoltaic users has personalized characteristics. For example, the electricity consumption pattern of household users is affected by multiple factors such as season, weather, and living habits. The load regulation flexibility of industrial users varies due to different equipment types and production processes. The existing adjustable potential assessment method does not take into account such personalized differences and cannot accurately assess the load regulation capabilities of different photovoltaic user groups. In addition, the existing technology does not take into account the adjustable potential of users with surplus power on the grid, and cannot effectively guide them to adjust their electricity consumption behavior according to the operation requirements of the power system to achieve peak shaving and valley filling and local consumption. In addition, the regulation capacity of a single photovoltaic user may not be sufficient to meet the needs of large-scale photovoltaic self-consumption, and it is necessary to consider the coordinated regulation of user groups in a larger range. Summary of the invention

[0004] The present invention provides a method and system for evaluating the adjustable potential of photovoltaic users, an electronic device, and a computer-readable storage medium, which can accurately evaluate the load regulation capabilities of different types of surplus power grid-connected users, provide theoretical support for guiding surplus power grid-connected users to voluntarily adjust their electricity usage behavior according to the needs of power system operation, achieve peak shaving and valley filling, and local consumption, improve the flexibility of the power system, effectively ensure the safe and stable operation of the power system, and vigorously promote the consumption of renewable energy power.

[0005] According to one aspect of the present invention, a method for evaluating photovoltaic user adjustable potential is provided, comprising the following contents: Classify and analyze the load data of photovoltaic users, classify the electricity consumption behaviors of photovoltaic users into multiple typical electricity consumption behaviors, and extract the load characteristic curve of each typical electricity consumption behavior; Construct a load power decomposition model, and use the load power decomposition model to analyze the load characteristic curve of each typical power consumption behavior to obtain the adjustable load power ratio of each typical power consumption behavior; The adjustable potential of users with surplus power on-grid is evaluated based on the adjustable load power ratio of each typical power consumption behavior.

[0006] Furthermore, the process of evaluating the adjustable potential of the surplus power grid-connected user based on the adjustable load power proportion of each typical power consumption behavior includes the following: Determine the load adjustable time interval and load acceptance boundary of users with surplus power on the Internet; The adjustable potential of each typical electricity consumption behavior is calculated based on the adjustable load power ratio of each typical electricity consumption behavior; Obtain the calculated daily load curve of users with surplus power on the grid, calculate the membership between it and each typical power consumption behavior, and construct a membership matrix; The adjustable potential of surplus power grid users is determined based on the load acceptance boundary of surplus power grid users, the adjustable potential of each typical power consumption behavior and the membership matrix.

[0007] Furthermore, the adjustable potential of each typical electricity consumption behavior is determined based on the following formula: ; ; ; ;

[0008] in, Indicates j The adjustable potential of typical electricity consumption behavior, Indicates j The adjustable potential of typical electricity consumption behaviors during photovoltaic power generation period, Indicates j The adjustable potential boundary of typical electricity consumption behavior during photovoltaic power generation period, Indicates j The adjustable potential of typical electricity consumption behavior during the period when photovoltaic power generation is not in progress. Indicates the photovoltaic power generation period, Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The total electricity consumption, Indicates photovoltaic power generation period The total photovoltaic power generation Indicates jTypical electricity consumption behaviors during photovoltaic power generation period The proportion of adjustable load power, Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The predicted electricity consumption, Indicates the period when photovoltaic power generation is not in progress. Indicates j The proportion of adjustable load power of typical electricity consumption behaviors during the period when photovoltaic power generation is not in progress. Indicates j The predicted electricity consumption of typical electricity consumption behaviors during the period when photovoltaic power generation is not in progress.

[0009] Furthermore, the adjustable potential of the surplus power grid users is determined based on the following formula:

[0010] in, Indicates i The adjustable potential of each surplus power user, E 0 means i The load acceptance boundary of the surplus power grid users, n represents the number of typical power consumption behaviors, E j Indicates j The adjustable potential of typical electricity consumption behavior, d i,j Indicates i The calculated daily load curve of the surplus power grid user is j The membership degree between the typical electricity consumption behaviors.

[0011] Furthermore, the following contents are also included: The adjustable potential of the low-voltage substation is evaluated based on the power generation and consumption of the substation and the proportion of adjustable load electricity.

[0012] Furthermore, the load power decomposition model is constructed based on a denoising autoencoder and a bidirectional temporal convolutional network. The encoder of the denoising autoencoder includes a one-dimensional convolutional layer, multiple bidirectional temporal convolutional residual blocks and a contraction convolution block arranged in sequence along the input direction, wherein the contraction convolution block is composed of two one-dimensional convolutional layers. The encoder is connected to the decoder through a one-dimensional convolutional layer. The decoder includes three extended convolutional blocks, each of which includes a one-dimensional transposed convolutional layer and two one-dimensional convolutional layers arranged in sequence along the output direction.

[0013] Furthermore, typical electricity consumption behaviors include work type, home type, commercial type and production type.

[0014] In addition, the present invention also provides a photovoltaic user adjustable potential assessment system, comprising: The electricity consumption behavior classification module is used to classify and analyze the load data of photovoltaic users, classify the electricity consumption behavior of photovoltaic users into multiple typical electricity consumption behaviors, and extract the load characteristic curve of each typical electricity consumption behavior; The load power decomposition module is used to construct a load power decomposition model, and use the load power decomposition model to analyze the load characteristic curve of each typical power consumption behavior to obtain the adjustable load power ratio of each typical power consumption behavior; The adjustable potential evaluation module is used to evaluate the adjustable potential of users who access the grid with surplus electricity based on the proportion of adjustable load electricity of each typical electricity consumption behavior.

[0015] In addition, the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0016] In addition, the present invention also provides a computer-readable storage medium for storing a computer program for evaluating photovoltaic user adjustable potential, wherein the computer program executes the steps of the above-mentioned method when running on a computer.

[0017] The present invention has the following beneficial effects: The photovoltaic user adjustable potential assessment method of the present invention first classifies photovoltaic users into a variety of typical electricity consumption behaviors, and extracts the load characteristic curve of each typical electricity consumption behavior, providing a data basis for subsequent load power decomposition and adjustable potential assessment, and then uses the load power decomposition model to analyze the load characteristic curve of each typical electricity consumption behavior, and obtains the adjustable load power ratio of each typical electricity consumption behavior, and the adjustable potential of the surplus power grid-connected users can be evaluated according to the adjustable load power ratio of each typical electricity consumption behavior. Since the electricity consumption behavior of photovoltaic users is classified, the differentiated electricity demand of different photovoltaic users is taken into account, so that the load regulation capacity of different types of surplus power grid-connected users can be accurately evaluated, and the adjustable potential of surplus power grid-connected users is studied for the first time, which provides theoretical support for guiding surplus power grid-connected users to voluntarily adjust their electricity consumption behavior according to the needs of power system operation, realize peak shaving and valley filling, and local consumption, improves the flexibility of the power system, can effectively ensure the safe and stable operation of the power system, and vigorously promote the consumption of renewable energy power.

[0018] In addition, the photovoltaic user adjustable potential assessment system of the present invention also has the above advantages.

[0019] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 It is a flow chart of a photovoltaic user adjustable potential assessment method according to a preferred embodiment of the present application; Figure 2 The load characteristic curves of typical electricity consumption behavior at work and typical electricity consumption behavior at home in the preferred embodiment of the present application; Figure 3 is a load characteristic curve of a typical commercial electricity consumption behavior in a preferred embodiment of the present application; Figure 4 is a load characteristic curve of a typical power consumption behavior of a production type in a preferred embodiment of the present application; Figure 5 It is a schematic diagram of the network architecture of the existing DAE; Figure 6 This is a schematic diagram of the network architecture of the load power decomposition model of the preferred embodiment of the present application; Figure 7 yes Figure 1 Schematic diagram of the sub-process of step S3; Figure 8 It is another flow chart of the photovoltaic user adjustable potential evaluation method of the preferred embodiment of the present application; Fig. 9 yes Figure 8 Schematic diagram of the sub-process of step S4; Fig.10 It is a schematic diagram of adjustable load power proportion of typical power consumption behavior of working people calculated in the preferred embodiment of the present application; Fig.11 This is a schematic diagram of a calculated daily load curve of a surplus power user in a preferred embodiment of the present application; Fig.12 It is a schematic diagram of the adjustable time interval and acceptable boundary of a certain low-pressure area determined in the preferred embodiment of the present application; Fig.13 It is a schematic diagram of the module structure of a photovoltaic user adjustable potential assessment system according to another embodiment of the present application. DETAILED DESCRIPTION

[0021] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0022] Reference Figure 1 The preferred embodiment of the present application provides a photovoltaic user adjustable potential assessment method, comprising the following contents: Step S1: classify and analyze the load data of photovoltaic users, classify the electricity consumption behaviors of photovoltaic users into multiple typical electricity consumption behaviors, and extract the load characteristic curve of each typical electricity consumption behavior; Step S2: construct a load power decomposition model, and use the load power decomposition model to analyze the load characteristic curve of each typical power consumption behavior to obtain the adjustable load power ratio of each typical power consumption behavior; Step S3: Evaluate the adjustable potential of users with surplus power access to the grid based on the adjustable load power ratio of each typical power consumption behavior.

[0023] It can be understood that the photovoltaic user adjustable potential assessment method of this embodiment first classifies photovoltaic users into a variety of typical electricity consumption behaviors, and extracts the load characteristic curve of each typical electricity consumption behavior, so as to provide a data basis for the subsequent load power decomposition and adjustable potential assessment, and then uses the load power decomposition model to analyze the load characteristic curve of each typical electricity consumption behavior, and obtains the adjustable load power proportion of each typical electricity consumption behavior, and the adjustable potential of the surplus power grid-connected users can be evaluated according to the adjustable load power proportion of each typical electricity consumption behavior. Since the electricity consumption behavior of photovoltaic users is classified, the differentiated electricity demand of different photovoltaic users is taken into account, so that the load regulation capacity of different types of surplus power grid-connected users can be accurately evaluated, and the adjustable potential of surplus power grid-connected users is studied for the first time, which provides theoretical support for guiding surplus power grid-connected users to voluntarily adjust their electricity consumption behavior according to the needs of power system operation, realize peak shaving and valley filling, and local consumption, improves the flexibility of the power system, can effectively ensure the safe and stable operation of the power system, and vigorously promote the consumption of renewable energy power.

[0024] It can be understood that in step S1, the improved fuzzy C-means algorithm is used to classify and analyze the load data of photovoltaic users, classify the electricity consumption behaviors of photovoltaic users into multiple typical electricity consumption behaviors, and extract the load characteristic curve of each typical electricity consumption behavior.

[0025] Specifically, fuzzy C-means clustering is a typical algorithm in the field of fuzzy clustering. It has the characteristics of simple algorithm design and wide range of problem solving. It is also suitable for occasions with large data volume and high real-time requirements. Fuzzy partitioning does not classify each sample point into a certain category, but considers it as belonging to a certain category with a certain degree of membership. Among them, fuzzy C clustering can be described as:

[0026] Among them, U represents the membership matrix, P represents the cluster center, represents the membership function, which characterizes the membership relationship between the sample and the subset, and , membership matrix , Indicates i Samples in the classx k With i Electrically conductive samples p i The distance between It represents the sum of squares of errors between samples in each category and typical samples, and m represents the weighted index. The calculation formula is: hour, ;when hour, . Therefore, the cluster centers are: The clustering criterion of the fuzzy C-means algorithm is to take The minimum value of Therefore, the fuzzy C-means clustering algorithm is essentially to iteratively adjust U and P so that the objective function J m Minimum.

[0027] The fuzzy C-means clustering algorithm is sensitive to the initial cluster center. Therefore, the present invention improves it and uses the shortest distance clustering method to provide the initial cluster center for the fuzzy C-means clustering. Assume that G represents the class and G There are m elements in express, Representation Elements x i and x j The distance between Representation Class With class The distance between Use Euclidean distance: , the shortest distance method is used to perform system clustering to obtain the cluster center, and the cluster center is used as the original cluster center for the next step of fuzzy C-means clustering. The shortest distance method defines the distance between classes as the distance between the closest samples of the two classes, that is, , if a step class G K With class G L Gather into a new class, denoted as G M ,kind G M With any existing class G J The distance between them is: .

[0028] Specifically, the shortest distance clustering process includes the following steps: a) Treat each initial sample as a class, use Euclidean distance to define the distance between samples, and calculate the distance matrix of n samples D （0） , it is a symmetric matrix; b) Search D （0） The smallest element in is set to D (K,L) ,Will G K and G L Gather into a new class, denoted as G M ,Right now G M ={ G K , G L}; c) Calculate new classes G M With any class G J The distance between them is calculated by the recursive formula:

[0029] Then, the distance matrix D （0） Update, coming soon and The row and column are merged into a new row and column, and we get , the distance between the new row and the new column is given by The membership function calculation formula is calculated when , and the values ​​on the other rows and columns remain unchanged. The new distance matrix obtained is recorded as ; d) Yes Repeat steps b) and c) to get the distance matrix ; Continue in this way until all elements are merged into one category.

[0030] Optionally, the present invention classifies the electricity consumption behavior of photovoltaic users into four typical electricity consumption behaviors: office type, home type, commercial type and production type through an improved fuzzy C-means algorithm, and then integrates the load data of each type of typical electricity consumption behavior through the mean fusion method to extract the load characteristic curves of each type of typical electricity consumption behavior, laying a data foundation for studying the adjustable potential model. Specifically, assuming that the load data of a certain type of typical electricity consumption behavior user is P i (t), where i Represents the user number, t represents the time point, and the calculation formula of the load characteristic curve of this type of typical electricity consumption behavior is: , n represents the number of users with this type of typical electricity consumption behavior. For possible load fluctuations, the curve is smoothed using methods such as moving average or Gaussian smoothing to eliminate noise and highlight typical trends. The formula for moving average is: , m Indicates the smoothing window size, usually 3 to 5 time points are selected.

[0031] In addition, the present invention uses the improved fuzzy C-mean algorithm and mean fusion method to classify and analyze the electricity consumption behavior of photovoltaic users, and obtains the load characteristic curves of four typical electricity consumption behaviors: office type, home type, commercial type and production type. Figures 2 to 4 As shown in the figure. Among them, the load characteristic curve of the office type shows obvious bimodal characteristics, with peaks appearing in the morning and evening, which is highly consistent with the work and rest patterns of the office type. The load decreases rapidly after midnight and remains at a low level in the early morning. Overall, the load curve of the office type is regular and predictable, especially the load is large during peak hours. It is a key target group in the power demand side management and has a certain adjustable load potential. The load characteristic curve of the home family shows the characteristics of load fluctuations in the morning, afternoon and evening, especially the significant increase in evening load, with a peak value higher than other periods. This shows that the power consumption behavior of the home family is mainly concentrated in the daytime and nighttime family activities. This type of load is related to family work and rest and entertainment activities, and has high flexibility. Through smart home devices, time-of-use electricity prices and power demand response, the load regulation potential of the home family can be effectively tapped. The load characteristic curve of the commercial type is characterized by high loads in the noon and afternoon periods, and reaches a peak during the noon period. This feature is closely related to the business hours and customer behavior of the daily necessities industry. Since this type of load is significantly affected by business hours and customer flow, its adjustable potential is mainly concentrated in the load peak shifting of electrical equipment and the management of equipment operation during non-business hours. The production load curve shows a continuous high load state during the day, and the load is still high at night. This type of load is related to the continuity of industrial production. In particular, the existence of night load may be related to the continuous operation of equipment or production process requirements. This type of load regulation potential can be achieved by optimizing production scheduling, introducing energy storage facilities, and reducing load during non-essential periods.

[0032] It can be understood that in order to evaluate the adjustable potential of photovoltaic users, it is necessary to first understand the electrical appliances used by photovoltaic users and determine the power consumption of each load adjustable appliance, so as to provide data support for the accurate evaluation of adjustable potential. Directly decomposing the power consumption data of the target appliance from the total load data of photovoltaic users can be regarded as a denoising problem, that is, removing the power consumption other than the target appliance from the total load data. Therefore, in the step S2, the present invention constructs a load power decomposition model based on a denoising autoencoder (DenoisingAutoencoder, DAE) and a bidirectional temporal convolution network (Bidirectional Temporal Convolution, BiTCN), efficiently learns time series features, and accurately analyzes the proportion of adjustable load power of various typical electricity consumption behaviors, so as to provide data support for the accurate evaluation of adjustable potential. The load power decomposition model can convert the noisy total energy consumption s (t) Mapped to the energy consumption of the target appliance d i (t), s (t) and d i (t) can be any electrical quantity, such as active power, reactive power, apparent power and current, etc. The present invention preferably uses active power, total active power consumption s (t) is defined as all downstream loads d i (t) and unknown load e (t) The active power consumption of each appliance can be expressed as follows: ; ;

[0033] Therefore, the remaining load d l (t) plus unknown load e (t) The total consumption is s (t) is the noise removed, therefore, the load power decomposition can be solved by removing the noise.

[0034] Among them, the denoising autoencoder was originally defined as an unsupervised feature extractor. Even if the input data is partially occluded or damaged, the model can still learn meaningful representations from high-dimensional data. The core of DAE is an autoencoder architecture in which the input is encoded into a representation with fewer dimensions - a feature encoding, called a code, and the model attempts to reconstruct an output from the code that is as close to the input as possible. For power load time series, autoencoders are usually trained with load segments x(j) of fixed duration. The encoder Map it to code, then decode it Reconstruction output During training, the encoder weights θ and the decoder weights φ The optimization is performed by minimizing the mean square error (i.e., reconstruction error) of all training samples, which can be expressed as: , where L() represents the square difference calculation, for example .

[0035] The existing DAE architecture is as follows Figure 5 As shown in the figure, the encoder subnetwork maps the input sequence to the code through a one-dimensional convolutional layer and two fully connected layers. The decoder architecture generates the estimated power consumption of the target appliance after passing the generated code through two fully connected layers and the last one-dimensional convolutional layer with only one convolution kernel. All custom parameters of the architecture, such as the number of units in the fully connected layers, the number of kernels in the convolutional layers, or the kernel size, can be determined through experiments.

[0036] The present invention takes into account that the operating characteristics of electrical appliances are time-related, especially multi-state appliances. For example, a washing machine includes a pre-wash stage, a washing stage, and a drying stage. Therefore, the load power decomposition model needs to be a network model that can capture long sequence characteristics. Compared with RNN, the temporal convolutional network is more accurate in processing long time series data problems. The temporal convolutional network TCN is composed of a stack of basic network structural units (TCN residual blocks). The present invention uses bidirectional void convolution and Batch Normalization to replace the causal void convolution and Weight Normalization in the TCN residual block, which can obtain a larger receptive field and better performance. The present invention calls this structure a bidirectional temporal convolution residual block (Bi-TCN). In addition, in order to avoid using fully connected layers in the DAE structure, the present invention also replaces the fully connected layers in the Bi-TCN network with convolutional layers. Specifically, as Figure 6As shown, the encoder of the denoising autoencoder includes a one-dimensional convolutional layer, multiple bidirectional temporal convolutional residual blocks and a contraction convolution block arranged in sequence along the input direction, wherein the contraction convolution block is composed of two one-dimensional convolutional layers, the encoder is connected to the decoder through a one-dimensional convolutional layer, and the decoder includes three dilated convolutional blocks, each of which includes a one-dimensional transposed convolutional layer and two one-dimensional convolutional layers arranged in sequence along the output direction. The encoder adopts the Bi-TCN network and replaces the fully connected layer with a contraction convolution block. The contraction convolution block consists of two one-dimensional convolution layers with a stride of 2. The two one-dimensional convolution layers with a stride of 2 reduce the input dimension along the direction of the network, and finally the input vector is represented as a feature code through a one-dimensional convolution. The decoder must convert the feature code into a reconstructed estimate of the target appliance consumption. Therefore, the decoder decodes the feature code through three dilated convolution blocks. Each dilated convolution block consists of a one-dimensional transposed convolution layer and two standard one-dimensional convolution layers with a stride of 1. The two one-dimensional convolution layers are arranged after the transposed convolution block.

[0037] Therefore, the present invention uses the load power decomposition model constructed above to perform load decomposition on the active power curve of each typical power consumption behavior, and can obtain the active power of the adjustable load device of each typical power consumption behavior during operation, and then integrate it in time series to obtain the adjustable load power of the adjustable load device of each typical power consumption behavior, so as to obtain the adjustable load power proportion of each typical power consumption behavior. Among them, the adjustable load power proportion of a certain typical power consumption behavior in a certain period of time is the sum of the load power proportions of all adjustable loads of this type of typical power consumption behavior in this period of time, which can be expressed as: , ρ T Indicates a typical electricity consumption behavior in a period T =[t1,t2] adjustable load power ratio, ρ i,T Indicates i Adjustable load in time period T The proportion of load power.

[0038] It can be understood that the adjustable potential of users of surplus power mainly refers to households or small industrial and commercial users with photovoltaic power generation systems, who can flexibly adjust their electricity consumption behavior when there is excess photovoltaic power generation, postpone or advance the use of high-energy-consuming equipment, thereby promoting the local consumption of surplus power. Figure 7 As shown, the process of evaluating the adjustable potential of surplus power grid users based on the adjustable load power proportion of each typical power consumption behavior includes the following: Step S31: Determine the load adjustable time interval and load acceptable boundary of the surplus power online user; Step S32: calculating the adjustable potential of each typical power consumption behavior based on the adjustable load power proportion of each typical power consumption behavior; Step S33: obtaining the calculated daily load curve of the surplus power grid users, and calculating the membership between the curve and each typical power consumption behavior, and constructing a membership matrix; Step S34: determining the adjustable potential of the surplus power grid-connected users according to the load acceptance boundary of the surplus power grid-connected users, the adjustable potential of each typical power consumption behavior and the membership matrix.

[0039] Specifically, define the load adjustable time interval for users with surplus power T adj For all periods when the user's electricity consumption is greater than the photovoltaic power generation, it can be expressed as: , among which, for , Indicates time t The power consumption of users, Indicates time t The photovoltaic power generation power, Respectively represent the start time and end time of the adjustable time interval.

[0040] In addition, define the load acceptance boundary of surplus power users E 0 is the difference between the total photovoltaic power generation and the total power consumption of users during photovoltaic power generation, which can be expressed as: , , , express t The power generation of the photovoltaic system of the users with surplus power at any time, It indicates the total power generation of the photovoltaic power generation users during the power generation period. express t The power consumption of users with surplus electricity at any time. It indicates the total power consumption of users with surplus power on the grid during the photovoltaic power generation period. T It is a collection of photovoltaic power generation collection time nodes.

[0041] Among them, for the period when photovoltaic power generation is not in progress, the present invention defines the adjustable potential of a typical power consumption behavior in the period when photovoltaic power generation is not in progress as: the product of the adjustable load power proportion of the typical power consumption behavior in the period when photovoltaic power generation is not in progress and the predicted power consumption in the period. The adjustable potential of each typical power consumption behavior is calculated based on the following formula:

[0042] in, Indicates the period when photovoltaic power generation is not in progress. Indicates j The proportion of adjustable load power of typical electricity consumption behaviors during the period when photovoltaic power generation is not in progress. Indicates j The predicted power consumption of typical power consumption behaviors during the period when photovoltaic power generation is not in progress. Indicates j The adjustable potential of typical electricity consumption behaviors during the period when photovoltaic power generation is not in progress.

[0043] For the photovoltaic power generation period, j The adjustable potential boundary of a typical electricity consumption behavior during the photovoltaic power generation period is the difference between the user's electricity consumption and the photovoltaic power generation in all adjustable time intervals during the photovoltaic power generation period, which can be expressed as: ,in, Indicates j The adjustable potential boundary of typical electricity consumption behavior during photovoltaic power generation period, Indicates the photovoltaic power generation period, Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The total electricity consumption, Indicates photovoltaic power generation period of total photovoltaic power generation.

[0044] The present invention defines j The adjustable potential of a typical electricity consumption behavior during the photovoltaic power generation period is: the sum of the product of the adjustable load power proportion of the typical electricity consumption behavior in all adjustable periods and the predicted power consumption in the corresponding period, which can be expressed as: ,in, Indicates j The adjustable potential of typical electricity consumption behaviors during photovoltaic power generation period, Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The proportion of adjustable load power, Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The predicted electricity consumption.

[0045] Therefore, the adjustable potential for each typical electricity consumption behavior is determined based on the following formula:

[0046] in, Indicates j Adjustable potential for typical electricity consumption behaviors.

[0047] Then, the calculated daily load curve of the surplus power grid-connected users is obtained, and the membership between it and each typical power consumption behavior is calculated to construct a membership matrix. Among them, the random matrix theory can characterize complex physical systems in the form of random matrices in practice. The high-dimensional statistical characteristics provided by the random matrix theory can reflect the non-independent and non-random characteristics existing in the system. If the correlation problems of things encountered in the project are described in the form of physical quantities, then the original data source is constructed using the physical measurement data of the things to be studied, and it is converted into the form of a random matrix, which can reflect the correlation between the things to be studied by analyzing the statistical characteristics of the random matrix. Therefore, the present invention uses random matrix theory to calculate the membership between the calculated daily load curve of the surplus power grid-connected users and each typical power consumption behavior.

[0048] Specifically, the membership degree is defined as the average spectral radius difference of the augmented matrix, which can be expressed as: ,in, It represents the membership degree between the calculated daily load curve of the surplus power grid users and each typical power consumption behavior. represents the average spectral radius of the state augmentation matrix, represents the average spectral radius of the reference state augmented matrix. The specific calculation process is as follows: 1) Select appropriate original data and construct the source data matrix, which includes the power consumption data matrix D of the surplus power grid users. A And the load data matrix D of various typical electricity consumption behaviors B ; 2) Use the source data matrix to construct an augmented matrix based on random matrix theory ; 3) In the augmented matrix Introduce white noise into the matrix and construct an augmented data matrix containing white noise data ; 4) The data matrix Standardize to get the standard data matrix , construct the reference augmented matrix D ; 5) Use singular value equivalence to obtain the standard matrix and the reference augmented matrix D The singular value equivalent matrix of and ; 6) Yes and Finding Matrix Products , ; 7) Yes , To standardize; 8) Seek The average spectral radius and The average spectral radius , and find the average spectral radius difference .

[0049] Among them, the solution process of the augmented matrix average spectral radius difference belongs to the existing technology, so the detailed solution principle will not be repeated here.

[0050] After calculating the membership between the calculated daily load curve of the surplus power grid users and each typical power consumption behavior, the membership matrix can be constructed, which can be expressed as: ,in, , Indicates i The calculated daily load curve of the surplus power grid user is j The membership degree between the typical electricity consumption behaviors.

[0051] Therefore, the low pressure area i The adjustable potential of each surplus power grid user can be determined based on the following formula:

[0052] in, Indicates i The adjustable potential of each surplus power user, E 0 means i The load acceptance boundary of the surplus power grid users, n represents the number of typical power consumption behaviors, E j Indicates j Adjustable potential for typical electricity consumption behaviors.

[0053] It is understandable that, in actual situations, the adjustable potential of some users with surplus power on the grid may be low and unable to meet the local consumption demand of photovoltaic power generation. At this time, relying solely on the adjustment capacity of a single user is obviously insufficient to achieve large-scale photovoltaic self-consumption. Therefore, the present invention further expands to a larger range of load regulation - the adjustable potential at the substation level. As a power supply unit, the substation includes all users (photovoltaic users and non-photovoltaic users). When the photovoltaic surplus power cannot be completely consumed locally, by broadening the scope of demand response subjects, all types of power users can participate in demand response, and orderly guide power users with response capabilities to participate in demand response, jointly adjust the load, and achieve load balance and stability of the power grid.

[0054] Alternatively, if Figure 8 As shown, the photovoltaic user adjustable potential assessment method also includes the following contents: Step S4: Evaluate the adjustable potential of the low-voltage substation based on the power generation and consumption of the substation and the proportion of adjustable load electricity.

[0055] Among them, Fig. 9 As shown, the process of evaluating the adjustable potential of the low-voltage area based on the power generation and consumption of the area and the proportion of adjustable load power includes the following: Step S41: Determine the load adjustable time interval and load acceptable boundary of the low voltage area; Step S42: Calculate the adjustable potential of the low voltage area during the photovoltaic power generation period and the photovoltaic non-power generation period; Step S43: Based on the load acceptance boundary of the low-voltage area and the adjustable potential of the low-voltage area during the photovoltaic power generation period and the photovoltaic non-power generation period, the adjustable potential of the low-voltage area is finally determined.

[0056] Specifically, define the load adjustable time interval of the low voltage area It is the period when the total power consumption of the area is greater than the total photovoltaic power generation in the area, which can be expressed as: ,in, , Indicates time t The total power consumption of the area, Indicates time t The total photovoltaic power generation power in the area, Respectively represent the start time and end time of the adjustable time interval.

[0057] In addition, define the load acceptance boundary of the low voltage area is the difference between the photovoltaic power generation in the area and the total electricity consumption of users in the area during photovoltaic power generation, which can be expressed as: , Indicates the photovoltaic power generation of the area during the photovoltaic power generation period, Indicates the total electricity consumption of users in the area during the photovoltaic power generation period.

[0058] For the period when photovoltaic power generation is not in progress, the present invention defines the adjustable potential of the low-voltage area as: the product of the total adjustable load power ratio of all users in the area during the period when photovoltaic power generation is not in progress and the predicted power consumption during the period, which can be expressed as: ,in, It indicates the adjustable potential of low voltage area during the period when photovoltaic power generation is not in progress. Indicates the period when photovoltaic power generation is not in progress. It indicates the proportion of total adjustable load power of all users in the low-voltage area during the period when photovoltaic power generation is not in progress. It indicates the predicted electricity consumption of all users in the low-voltage area during the period when photovoltaic power generation is not in progress.

[0059] For the photovoltaic power generation period, the present invention defines the adjustable potential boundary of the low-voltage area as: the sum of the difference between the total power consumption of the area users and the power generation of the area in all adjustable time intervals during photovoltaic power generation, which can be expressed as: ,in, It represents the adjustable potential boundary of the low voltage area during the photovoltaic power generation period. Indicates the photovoltaic power generation period, Indicates low pressure area during the period The total electricity consumption, Indicates that the photovoltaic power generation in the low voltage area is of power generation.

[0060] In addition, the present invention defines the adjustable potential of the low-voltage area during the photovoltaic power generation period as: the sum of the product of the total adjustable load power proportion of all users in the area and the predicted power consumption in the corresponding period in all adjustable periods, which can be expressed as: ,in, It indicates the adjustable potential of low voltage area during photovoltaic power generation period. Indicates that all users in the low voltage area are in the period The proportion of total adjustable load power, Indicates that all users in the low voltage area are in the period The predicted electricity consumption.

[0061] Therefore, the adjustable potential of the low-voltage area can be determined based on the following formula: ,in, .

[0062] It can be understood that the present invention is further extended to the adjustable potential assessment at the substation level. By considering the coordinated adjustment of user groups in a larger range, the overall adjustable potential is improved, which helps to solve the problem of large-scale photovoltaic power generation and achieve efficient consumption of photovoltaic power generation.

[0063] In addition, the present invention also collects load data of 265 users in a low-voltage area in 2024, with a collection frequency of 15 minutes, and quantitatively evaluates the adjustable potential of each user and the adjustable potential of the low-voltage area. Specifically, the load power is first decomposed into four typical power consumption behaviors of the low-voltage area based on historical load data, and the power proportion of each adjustable load of each typical power consumption behavior is obtained, and then the adjustable load power proportion of each typical power consumption behavior is obtained. Taking the office type as an example, the calculated adjustable load power proportion is Fig.10 Then, determine the adjustable time interval and acceptable boundary, and obtain the calculated daily load curve of the surplus power grid-connected user. The calculated daily load curve of a surplus power grid-connected user is as follows: Fig.11As shown, according to the membership calculation method of typical electricity consumption behavior of users based on random matrix theory, the membership between the surplus power online user and the typical electricity consumption behavior is obtained. From the results, it can be seen that the weekday load curve of the surplus power online user has a higher membership to the working type (0.5126), indicating that the user's weekday electricity consumption behavior will largely conform to the characteristics of the working type, and the membership to the processing and production type and the residential family is lower (0.1101 and 0.2558 respectively), indicating that the daytime electricity load is less, and the electricity consumption behavior characteristics will have a lower matching degree with the processing and production type and the residential family. Finally, the adjustable potential of all users in the substation and the adjustable potential of the entire low-voltage substation are calculated, among which the adjustable time interval and acceptable boundary of the substation are as follows Fig.12 The calculation results of the adjustable potential of some users and low-voltage areas are shown in Table 1.

[0064] Table 1. Calculation results of adjustable potential for some users and low-voltage areas

[0065] The users of surplus power on the grid in this low-voltage area account for about 7.92% of the total number of photovoltaic users. Due to economic incentives, most of the users of surplus power on the grid have adjusted their electricity consumption habits and actively responded to the needs of the power grid, but the adjustable potential of the users of surplus power on the grid is low. However, the load adjustable potential of this low-voltage area is 36.22 kWh, which has a high load adjustable potential.

[0066] In addition, if Fig.13 As shown, another embodiment of the present invention further provides a photovoltaic user adjustable potential assessment system, preferably using the photovoltaic user adjustable potential assessment method as described above, including: The electricity consumption behavior classification module is used to classify and analyze the load data of photovoltaic users, classify the electricity consumption behavior of photovoltaic users into multiple typical electricity consumption behaviors, and extract the load characteristic curve of each typical electricity consumption behavior; The load power decomposition module is used to construct a load power decomposition model, and use the load power decomposition model to analyze the load characteristic curve of each typical power consumption behavior to obtain the adjustable load power ratio of each typical power consumption behavior; The adjustable potential evaluation module is used to evaluate the adjustable potential of users who access the grid with surplus electricity based on the proportion of adjustable load electricity of each typical electricity consumption behavior.

[0067] It can be understood that the photovoltaic user adjustable potential assessment system of this embodiment first classifies photovoltaic users into a variety of typical electricity consumption behaviors, and extracts the load characteristic curve of each typical electricity consumption behavior, so as to provide a data basis for the subsequent load power decomposition and adjustable potential assessment, and then uses the load power decomposition model to analyze the load characteristic curve of each typical electricity consumption behavior, and obtains the adjustable load power proportion of each typical electricity consumption behavior, and the adjustable potential of the surplus power grid-connected users can be evaluated according to the adjustable load power proportion of each typical electricity consumption behavior. Since the electricity consumption behavior of photovoltaic users is classified, the differentiated electricity demand of different photovoltaic users is taken into account, so that the load regulation capacity of different types of surplus power grid-connected users can be accurately evaluated, and the adjustable potential of surplus power grid-connected users is studied for the first time, which provides theoretical support for guiding surplus power grid-connected users to voluntarily adjust their electricity consumption behavior according to the needs of power system operation, realize peak shaving and valley filling, and local consumption, improves the flexibility of the power system, can effectively ensure the safe and stable operation of the power system, and vigorously promote the consumption of renewable energy power.

[0068] In addition, the adjustable potential evaluation module is also used to evaluate the adjustable potential of the low-voltage substation based on the power generation and consumption conditions of the substation and the proportion of adjustable load electricity.

[0069] In addition, another embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0070] In addition, another embodiment of the present invention further provides a computer-readable storage medium for storing a computer program for evaluating photovoltaic user adjustable potential, wherein the computer program executes the steps of the method described above when running on a computer.

[0071] Common computer-readable storage media include: floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, any other optical media, punch cards, paper tapes, any other physical media with patterns of holes, random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), flash-erasable programmable read-only memory (FLASH-EPROM), any other memory chip or cartridge, or any other medium that can be read by a computer. Instructions can further be transmitted or received by a transmission medium. The term transmission medium may include any tangible or intangible medium that can be used to store, encode or carry instructions for execution by a machine, and includes digital or analog communication signals or intangible media that facilitate communication of the above instructions. Transmission media include coaxial cables, copper wire, and optical fiber, including the wires of a bus used to transmit a computer data signal.

[0072] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present application may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0073] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0074] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0075] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0076] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0077] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A photovoltaic user adjustable potential assessment method, characterized in that: Includes the following: Classify and analyze the load data of photovoltaic users, classify the electricity consumption behaviors of photovoltaic users into multiple typical electricity consumption behaviors, and extract the load characteristic curve of each typical electricity consumption behavior; Construct a load power decomposition model, and use the load power decomposition model to analyze the load characteristic curve of each typical power consumption behavior to obtain the adjustable load power ratio of each typical power consumption behavior; The adjustable potential of users with surplus power on-grid is evaluated based on the adjustable load power ratio of each typical power consumption behavior.

2. The photovoltaic user adjustable potential assessment method according to claim 1, characterized in that: The process of evaluating the adjustable potential of the surplus power grid-connected user based on the adjustable load power proportion of each typical power consumption behavior includes the following: Determine the load adjustable time interval and load acceptance boundary of users with surplus power on the Internet; The adjustable potential of each typical electricity consumption behavior is calculated based on the adjustable load power ratio of each typical electricity consumption behavior; Obtain the calculated daily load curve of users with surplus power on the grid, calculate the membership between it and each typical power consumption behavior, and construct a membership matrix; The adjustable potential of surplus power grid users is determined based on the load acceptance boundary of surplus power grid users, the adjustable potential of each typical power consumption behavior and the membership matrix.

3. The photovoltaic user adjustable potential assessment method according to claim 2, characterized in that: The adjustable potential for each typical electricity consumption behavior is determined based on the following formula: in, Indicates j The adjustable potential of typical electricity consumption behavior, Indicates j The adjustable potential of typical electricity consumption behaviors during photovoltaic power generation period, Indicates j The adjustable potential boundary of typical electricity consumption behavior during photovoltaic power generation period, Indicates j The adjustable potential of typical electricity consumption behavior during the period when photovoltaic power generation is not in progress. Indicates the photovoltaic power generation period, Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The total electricity consumption, Indicates photovoltaic power generation period The total photovoltaic power generation Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The proportion of adjustable load power, Indicates j Typical electricity consumption behaviors during photovoltaic power generation period The predicted electricity consumption, Indicates the period when photovoltaic power generation is not in progress. Indicates j The proportion of adjustable load power of typical electricity consumption behaviors during the period when photovoltaic power generation is not in progress. Indicates j The predicted electricity consumption of typical electricity consumption behaviors during the period when photovoltaic power generation is not in progress.

4. The photovoltaic user adjustable potential assessment method according to claim 2, characterized in that: The adjustable potential of surplus power grid users is determined based on the following formula: in, Indicates i The adjustable potential of each surplus power user, E 0 means i The load acceptance boundary of the surplus power grid users, n represents the number of typical power consumption behaviors, E j Indicates j The adjustable potential of typical electricity consumption behavior, d i,j Indicates i The calculated daily load curve of the surplus power grid user is j The membership degree between the typical electricity consumption behaviors.

5. The photovoltaic user adjustable potential assessment method according to claim 1, characterized in that: Also included are the following: The adjustable potential of the low-voltage substation is evaluated based on the power generation and consumption of the substation and the proportion of adjustable load electricity.

6. The photovoltaic user adjustable potential assessment method according to claim 1, characterized in that: The load power decomposition model is constructed based on a denoising autoencoder and a bidirectional temporal convolutional network. The encoder of the denoising autoencoder includes a one-dimensional convolutional layer, a plurality of bidirectional temporal convolutional residual blocks and a contraction convolutional block arranged in sequence along the input direction, wherein the contraction convolutional block is composed of two one-dimensional convolutional layers. The encoder is connected to the decoder through a one-dimensional convolutional layer. The decoder includes three dilated convolutional blocks, each of which includes a one-dimensional transposed convolutional layer and two one-dimensional convolutional layers arranged in sequence along the output direction.

7. The photovoltaic user adjustable potential assessment method according to claim 1, characterized in that: Typical electricity usage behaviors include office, home, commercial and production.

8. A photovoltaic user adjustable potential assessment system, characterized in that: include: The electricity consumption behavior classification module is used to classify and analyze the load data of photovoltaic users, classify the electricity consumption behavior of photovoltaic users into multiple typical electricity consumption behaviors, and extract the load characteristic curve of each typical electricity consumption behavior; The load power decomposition module is used to construct a load power decomposition model, and use the load power decomposition model to analyze the load characteristic curve of each typical power consumption behavior to obtain the adjustable load power ratio of each typical power consumption behavior; The adjustable potential evaluation module is used to evaluate the adjustable potential of users who access the grid with surplus electricity based on the proportion of adjustable load electricity of each typical electricity consumption behavior.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and the processor is used to execute the steps of the method according to any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium for storing a computer program for evaluating photovoltaic user adjustable potential, characterized in that: When the computer program is run on a computer, the steps of the method according to any one of claims 1 to 7 are executed.

Images