Zone area distributed photovoltaic load non-intrusive identification method and device based on empirical value supervision and computer equipment

Abstract

The invention relates to a transformer area distributed photovoltaic load non-intrusive identification method and device based on empirical value supervision, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: obtaining a photovoltaic transformer area total table power # imgabs0 # and a transformer area illumination intensity # imgabs1 #, and constructing an optimization model of non-intrusive distributed photovoltaic load decomposition: # imgabs2 # and # imgabs3 # are used for constraining the difference between the photovoltaic transformer area total table power and a predicted value, and # imgabs4 # is used for constraining the sum of first-order differences corresponding to a basic load, a distributed photovoltaic load and a large infrequently-used load respectively; the # imgabs5 # is used for constraining the difference between the predicted value and the reference value of the distributed photovoltaic load generating capacity; # imgabs6 is used for constraining the difference between a basic load predicted value and a typical load value; the # imgabs7 # is used for restraining the starting times of the large-scale infrequently-used load; and by solving the optimization model, decomposing a target distributed photovoltaic load prediction value in the total table power of the photovoltaic transformer area. By adopting the method, the distributed photovoltaic load decomposition efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for non-invasive identification of distributed photovoltaic loads in a substation based on empirical value supervision. Background Art

[0002] With the development of power technology, the installed capacity of distributed photovoltaic power generation has grown rapidly in recent years. Since the power flow of distributed photovoltaic power is opposite to that of traditional loads, its output is manifested as a negative load at the distribution network level, which makes it difficult for grid managers to identify the difference between the increase in distributed photovoltaic output and the decrease in traditional load of the distribution network. Therefore, high penetration of distributed photovoltaic power can make the net load characteristics of the distribution network very complicated, which in turn affects the net load prediction, relay protection, demand response and other issues of the distribution network.

[0003] In order to reduce the adverse effects of distributed power grid connection, it is necessary to study the separation method of distributed photovoltaic power to further improve the stability and reliability of the power system. In related technologies, deep learning methods are usually used. However, deep learning models usually require a large amount of labeled data to train. In non-invasive load decomposition, this may mean that separate high-resolution data collection is required for each load, which is difficult to achieve in practical applications and is not conducive to the efficient realization of distributed photovoltaic load decomposition.

[0004] Therefore, there is a problem in the related art that the decomposition efficiency of distributed photovoltaic loads is low. Summary of the invention

[0005] Based on this, it is necessary to provide a non-invasive identification method, device, computer equipment, computer-readable storage medium and computer program product for distributed photovoltaic loads in a substation based on empirical value supervision, which can improve the decomposition efficiency of distributed photovoltaic loads in response to the above technical problems.

[0006] In a first aspect, the present application provides a non-intrusive identification method for distributed photovoltaic loads in a substation based on experience value supervision, comprising:

[0007] Get the total power of the photovoltaic area , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition:

[0008] ;

[0009] in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a customized hyperparameter;

[0010] in, It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the base load, the first-order difference of the distributed photovoltaic load and the first-order difference of the large infrequent use load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; Used to constrain the number of starts of large infrequently used loads within period T;

[0011] The reference value of the distributed photovoltaic load power generation is determined according to the mutual information between the illumination intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without illumination and a typical load factor; the total load dictionary of a typical day without illumination is composed of the total load time series of the typical day without illumination;

[0012] By solving the optimization model, the distributed photovoltaic load is decomposed from the total meter power of the photovoltaic area, and the target distributed photovoltaic load prediction value is obtained.

[0013] In one embodiment, the constraints of the optimization model include:

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] ;

[0021] in, Indicates the total load forecast value of the photovoltaic area; Represents the predicted value of the basic load; Table Displays the predicted value of distributed photovoltaic load; Indicates the forecast value of large infrequently used load; represents the first difference of the base load; represents the first-order difference of the distributed photovoltaic load; represents the first difference of large infrequently used loads; and Represents a custom hyperparameter; represents the residual term of the base load prediction; is a binary vector indicating whether the start of a large infrequently used load is detected at time t. If a large infrequently used load is started at time t, then =1, otherwise =0; represents the predicted value of the distributed photovoltaic load power generation, Indicates the reference value of the photovoltaic load power generation.

[0022] In one embodiment, the constraints of the optimization model also include:

[0023] ;

[0024] in, Indicates the threshold for detecting whether a large infrequently used load is activated.

[0025] In one embodiment, the method further comprises:

[0026] Obtain historical power consumption data for large, infrequently used equipment;

[0027] Analyzing the historical power consumption data to identify typical usage patterns and corresponding power consumption levels of the large infrequently used devices;

[0028] Determine a threshold for detecting whether a large infrequently used load is activated based on a typical usage pattern and a corresponding power consumption level of the large infrequently used device. .

[0029] In one embodiment, the residual term of the base load forecast The formula is as follows:

[0030] ;

[0031] ;

[0032] in, A dictionary representing the total load of N typical days without sunlight; Indicates typical load factor; Indicates the typical load value.

[0033] In one embodiment, the photovoltaic load power generation reference value The formula is as follows:

[0034] ;

[0035] in, is the rectified linear unit function, which is used to set all negative values ​​to 0 while keeping positive values ​​unchanged; express The light intensity of the area under time; , It represents the coefficient between the adjustment of the distributed photovoltaic load power generation and the light intensity of the area; , The KL divergence between the solar intensity in the area and the non-distributed photovoltaic load is obtained by minimizing:

[0036] ;

[0037] in, It represents the joint probability distribution of the solar intensity in the area and the non-distributed photovoltaic load; represents the probability distribution of light intensity; Represents the probability distribution of non-distributed PV load.

[0038] In a second aspect, the present application also provides a non-intrusive identification device for distributed photovoltaic loads in a substation based on experience value supervision, including:

[0039] Acquisition module, used to obtain the total meter power of the photovoltaic area , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition:

[0040] ;

[0041] in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a customized hyperparameter;

[0042] in, It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the base load, the first-order difference of the distributed photovoltaic load and the first-order difference of the large infrequent use load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; Used to constrain the number of starts of large infrequently used loads within period T;

[0043] The reference value of the distributed photovoltaic load power generation is determined according to the mutual information between the illumination intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without illumination and a typical load factor; the total load dictionary of a typical day without illumination is composed of the total load time series of the typical day without illumination;

[0044] The decomposition module is used to decompose the distributed photovoltaic load from the total meter power of the photovoltaic area by solving the optimization model to obtain a target distributed photovoltaic load prediction value.

[0045] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0047] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0048] The above-mentioned non-intrusive identification method, device, computer equipment, computer readable storage medium and computer program product based on experience value supervision in the distributed photovoltaic load of the photovoltaic area are obtained by obtaining the total meter power of the photovoltaic area. , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition:

[0049] ;

[0050] in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a custom hyperparameter; It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the base load, the first-order difference of the distributed photovoltaic load and the first-order difference of the large infrequent use load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; It is used to constrain the number of starts of large infrequently used loads within a period T; wherein, the reference value of distributed photovoltaic load power generation is determined according to the mutual information between the light intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without sunlight and the typical load coefficient; the total load dictionary of a typical day without sunlight is composed of the total load time series of a typical day without sunlight; by solving the optimization model, the distributed photovoltaic load is decomposed from the total meter power of the photovoltaic substation to obtain the target distributed photovoltaic load prediction value.

[0051] In this way, by decomposing the total load into distributed photovoltaic load, large infrequent load and basic load, the total load reconstruction error considering the slope is minimized, as well as three loss functions related to distributed photovoltaic load, large infrequent load and basic load, and two forms of empirical values ​​are introduced: one is to estimate the distributed photovoltaic load using mutual information, and the other is to use the basic load dictionary, that is, the total load dictionary and typical load coefficient of a typical day without sunlight, to limit the estimation error of the distributed photovoltaic load, and finally the decomposed distributed photovoltaic load is obtained in the total meter power of the photovoltaic station area. It is realized that the decomposed distributed photovoltaic load can be obtained by solving the optimization problem. Compared with the deep learning method, it does not need to rely on a large amount of labeled data or sub-metering data. While decomposing the distributed photovoltaic load more accurately, it can improve the decomposition efficiency of the distributed photovoltaic load. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1A schematic diagram of a flow chart of a non-intrusive identification method of distributed photovoltaic loads in a substation based on experience value supervision in an embodiment;

[0054] Figure 2 A flowchart of a step of determining a threshold value for detecting whether a large infrequently used load is activated in one embodiment;

[0055] Figure 3 A schematic diagram of a flow chart of a non-intrusive identification method of distributed photovoltaic loads in a substation based on experience value supervision in another embodiment;

[0056] Figure 4 It is a structural block diagram of a non-intrusive identification device for distributed photovoltaic loads in a substation based on experience value supervision in one embodiment;

[0057] FIG. 5 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0060] In one embodiment, Figure 1 As shown, a non-intrusive identification method for distributed photovoltaic loads in a substation based on experience value supervision is provided. This embodiment uses the method applied to a computer device as an example. It can be understood that the computer device can be a terminal, a server, or a system including a terminal and a server. In this embodiment, the method includes the following steps:

[0061] Step S110, obtaining the total meter power of the photovoltaic area , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition:

[0062] ;

[0063] Among them, the total power of photovoltaic area Includes base load, large infrequent load and distributed photovoltaic load.

[0064] in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a custom hyperparameter.

[0065] in, It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the base load, the first-order difference of the distributed photovoltaic load and the first-order difference of the large infrequent use load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; Used to constrain the number of starts of large infrequently used loads within a period T.

[0066] Among them, the reference value of distributed photovoltaic load power generation is determined according to the mutual information between the light intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without sunlight and the typical load coefficient; the total load dictionary of a typical day without sunlight is composed of the total load time series of a typical day without sunlight.

[0067] Among them, the constraints of the optimization model include:

[0068] ;

[0069] ;

[0070] ;

[0071] ;

[0072] ;

[0073] ;

[0074] ;

[0075] ;

[0076] in, Indicates the total load forecast value of the photovoltaic area; Represents the predicted value of the basic load; Table Displays the predicted value of distributed photovoltaic load; Indicates the forecast value of large infrequent load. , , and , obtained by solving the above optimization model.

[0077] in, represents the first difference of the base load; represents the first-order difference of the distributed photovoltaic load; represents the first difference of large infrequently used loads; and Represents a custom hyperparameter; represents the residual term of the base load prediction; is a binary vector indicating whether the start of a large infrequently used load is detected at time t. If a large infrequently used load is started at time t, then =1, otherwise =0; represents the predicted value of the distributed photovoltaic load power generation, represents the reference value of photovoltaic load power generation; wherein, Indicates the threshold for detecting whether a large infrequently used load is activated.

[0078] In some embodiments, Figure 2 As shown, the method also includes the following steps:

[0079] Step S210, obtaining historical power consumption data of large infrequently used devices.

[0080] In the specific implementation, in order to determine the threshold for detecting whether a large infrequently used load is started, The computer device may first obtain historical power consumption data of large infrequently used devices, such as timestamps, power consumption values, etc.

[0081] Step S220 , analyzing the historical power consumption data to identify typical usage patterns and corresponding power consumption levels of large infrequently used devices.

[0082] In a specific implementation, the computer device can analyze the historical power consumption data, for example, perform feature extraction: calculate at least one of the average, maximum, and minimum power consumption according to time (such as hours, days, and weeks), and count the usage frequency of large infrequently used devices in different time periods; based on the extracted feature data, use cluster analysis (such as K-means clustering) or time series analysis (such as pattern recognition algorithm) to identify the typical usage mode of the device, such as identifying the typical mode of power consumption. At the same time, the normal working state of the device in different scenarios can be identified, such as the peak usage time and the valley time of the day; and after identifying the typical usage mode, the computer device can further determine the corresponding power consumption level of each typical usage mode, for example: calculate the average power under each typical usage mode through historical data; record the percentiles (such as the 95th percentile) of the power consumption corresponding to different typical usage modes to cope with high usage demand situations.

[0083] Step S230, determining a threshold for detecting whether a large infrequently used load is activated based on a typical usage pattern and a corresponding power consumption level of the large infrequently used device .

[0084] Furthermore, the computer device may determine a threshold value for detecting whether the large infrequently used load is activated based on the typical usage pattern and corresponding power consumption level of the large infrequently used device. ,For example:

[0085] Average + K times standard deviation: For example, the threshold can be set to the average power of a typical usage pattern plus K times (such as 2 times, the specific value is set according to actual needs) the standard deviation to detect significant load increases.

[0086] In this way, by analyzing the historical power consumption data of large infrequently used equipment, the typical usage pattern of the equipment and the corresponding power consumption level can be accurately identified, so as to set a more reasonable power consumption threshold to detect whether the large infrequently used load is started.

[0087] Furthermore, the residual term of the base load forecast The formula is as follows:

[0088] ;

[0089] ;

[0090] in, A dictionary representing the total load of N typical days without sunlight; represents the typical load factor, obtained by solving the optimization model; Indicates typical load values.

[0091] Furthermore, the reference value of photovoltaic load power generation The formula is as follows:

[0092] ;

[0093] in, is the Rectified Linear Unit function, which is used to set all negative values ​​to 0 while keeping positive values ​​unchanged; Indicates The light intensity of the area on a typical day without sunlight; , It represents the coefficient between the adjustment of the distributed photovoltaic load power generation and the light intensity of the area; , The KL divergence between the solar intensity in the area and the non-distributed photovoltaic load is obtained by minimizing:

[0094] ;

[0095] in, It represents the joint probability distribution of the solar intensity in the area and the non-distributed photovoltaic load; represents the probability distribution of light intensity; Represents the probability distribution of non-distributed PV load.

[0096] Step S120, by solving the optimization model, the distributed photovoltaic load is decomposed from the total meter power of the photovoltaic area to obtain the target distributed photovoltaic load prediction value.

[0097] In this way, by constructing the above-mentioned optimization model and the corresponding constraints, and by solving the optimization model, the distributed photovoltaic load is decomposed in the total meter power of the photovoltaic area, and the distributed photovoltaic load prediction value is obtained. The distributed photovoltaic load prediction value satisfies the above-mentioned constraints. In this way, the distributed photovoltaic load prediction value can be used as the target distributed photovoltaic load prediction value.

[0098] In the above-mentioned non-intrusive identification method of distributed photovoltaic load in the substation based on experience value supervision, the total meter power of the photovoltaic substation is obtained. , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition:

[0099] ;

[0100] in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a custom hyperparameter; It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the base load, the first-order difference of the distributed photovoltaic load and the first-order difference of the large infrequent use load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; It is used to constrain the number of starts of large infrequently used loads within a period T; wherein, the reference value of distributed photovoltaic load power generation is determined according to the mutual information between the light intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without sunlight and the typical load coefficient; the total load dictionary of a typical day without sunlight is composed of the total load time series of a typical day without sunlight; by solving the optimization model, the distributed photovoltaic load is decomposed from the total meter power of the photovoltaic substation to obtain the target distributed photovoltaic load prediction value.

[0101] In this way, by decomposing the total load into distributed photovoltaic load, large infrequent load and basic load, the total load reconstruction error considering the slope is minimized, as well as three loss functions related to distributed photovoltaic load, large infrequent load and basic load, and two forms of empirical values ​​are introduced: one is to estimate the distributed photovoltaic load using mutual information, and the other is to use the basic load dictionary, that is, the total load dictionary and typical load coefficient of a typical day without sunlight, to limit the estimation error of the distributed photovoltaic load, and finally the decomposed distributed photovoltaic load is obtained in the total meter power of the photovoltaic station area. It is realized that the decomposed distributed photovoltaic load can be obtained by solving the optimization problem. Compared with the deep learning method, it does not need to rely on a large amount of labeled data or sub-metering data. While decomposing the distributed photovoltaic load more accurately, it can improve the decomposition efficiency of the distributed photovoltaic load.

[0102] In addition, the proposed method is able to process low-resolution smart meter data, which usually has a sampling rate ranging from 5 minutes to 60 minutes. In contrast, many existing methods based on hidden Markov models require a sampling rate of 1 minute or higher resolution, so the proposed scheme has stronger adaptability.

[0103] In another embodiment, Figure 3As shown, a non-intrusive identification method for distributed photovoltaic loads in a substation based on empirical value supervision is provided. The method is applied to computer equipment as an example for explanation, and includes the following steps:

[0104] Step S302, obtaining historical power consumption data of large infrequently used devices.

[0105] Step S304 , analyzing the historical power consumption data to identify typical usage patterns and corresponding power consumption levels of large infrequently used devices.

[0106] Step S306: determining a threshold for detecting whether a large infrequently used load is activated according to a typical usage pattern of the large infrequently used device and a corresponding power consumption level.

[0107] Step S308: Obtain the total power of the photovoltaic area , light intensity in the area , an optimization model for non-intrusive distributed photovoltaic load decomposition is constructed.

[0108] Step S310, by solving the optimization model, the distributed photovoltaic load is decomposed from the total meter power of the photovoltaic area to obtain the target distributed photovoltaic load prediction value.

[0109] It should be noted that the specific limitations of the above steps can refer to the specific limitations of a non-intrusive identification method for distributed photovoltaic loads in a substation based on empirical value supervision.

[0110] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0111] Based on the same inventive concept, the embodiment of the present application also provides a non-intrusive identification device for distributed photovoltaic loads in a substation based on experience value supervision for implementing the above-mentioned non-intrusive identification method for distributed photovoltaic loads in a substation based on experience value supervision. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the non-intrusive identification device for distributed photovoltaic loads in a substation based on experience value supervision provided below can be referred to the limitations of the non-intrusive identification method for distributed photovoltaic loads in a substation based on experience value supervision above, and will not be repeated here.

[0112] In an exemplary embodiment, Figure 4 As shown, a non-intrusive identification device for distributed photovoltaic loads in a substation based on experience value supervision is provided, including: an acquisition module 410 and a decomposition module 420, wherein:

[0113] Acquisition module 410, used to obtain the total meter power of the photovoltaic area , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition:

[0114] ;

[0115] in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a customized hyperparameter;

[0116] in, It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the base load, the first-order difference of the distributed photovoltaic load and the first-order difference of the large infrequent use load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; Used to constrain the number of starts of large infrequently used loads within period T;

[0117] Among them, the reference value of distributed photovoltaic load power generation is determined according to the mutual information between the light intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without light and the typical load coefficient; the total load dictionary of a typical day without light is composed of the total load time series of the typical day without light.

[0118] The decomposition module 420 is used to decompose the distributed photovoltaic load from the total meter power of the photovoltaic area by solving the optimization model to obtain a target distributed photovoltaic load prediction value.

[0119] In one embodiment, the constraints of the optimization model include:

[0120] ;

[0121] ;

[0122] ;

[0123] ;

[0124] ;

[0125] ;

[0126] ;

[0127] in, Indicates the total load forecast value of the photovoltaic area; Represents the predicted value of the basic load; Table Displays the predicted value of distributed photovoltaic load; Indicates the forecast value of large infrequently used load; represents the first difference of the base load; represents the first-order difference of the distributed photovoltaic load; represents the first difference of large infrequently used loads; and Represents a custom hyperparameter; represents the residual term of the base load prediction; is a binary vector indicating whether the start of a large infrequently used load is detected at time t. If a large infrequently used load is started at time t, then =1, otherwise =0; represents the predicted value of the distributed photovoltaic load power generation, Indicates the reference value of the photovoltaic load power generation.

[0128] In one embodiment, the constraints of the optimization model also include:

[0129] ;

[0130] in, Indicates the threshold for detecting whether a large infrequently used load is activated.

[0131] In one embodiment, the device further includes: a threshold determination module for obtaining historical power consumption data of a large infrequently used device; analyzing the historical power consumption data to identify a typical usage pattern and a corresponding power consumption level of the large infrequently used device; and determining a threshold for detecting whether a large infrequently used load is started based on the typical usage pattern and the corresponding power consumption level of the large infrequently used device. .

[0132] In one embodiment, the residual term of the base load forecast The formula is as follows:

[0133] ;

[0134] ;

[0135] in, A dictionary representing the total load of N typical days without sunlight; Indicates typical load factor; Indicates the typical load value.

[0136] In one embodiment, the photovoltaic load power generation reference value The formula is as follows:

[0137] ;

[0138] in, is the rectified linear unit function, which is used to set all negative values ​​to 0 while keeping positive values ​​unchanged; express The light intensity of the area under time; , It represents the coefficient between the adjustment of the distributed photovoltaic load power generation and the light intensity of the area; , The KL divergence between the solar intensity in the area and the non-distributed photovoltaic load is obtained by minimizing:

[0139] ;

[0140] in, It represents the joint probability distribution of the solar intensity in the area and the non-distributed photovoltaic load; represents the probability distribution of light intensity; Represents the probability distribution of non-distributed PV load.

[0141] Each module in the above-mentioned non-intrusive identification device for distributed photovoltaic loads in a substation based on experience value supervision can be implemented in whole or in part by software, hardware, and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0142] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG5. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a non-intrusive identification method of distributed photovoltaic loads in a substation area based on empirical value supervision is implemented. The display unit of the computer device is used to form a visually visible picture, and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball, or touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0143] Those skilled in the art will understand that the structure shown in FIG. 5 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0144] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0145] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0146] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0147] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0148] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0149] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0150] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A non-intrusive identification method for distributed photovoltaic loads in a substation based on empirical value supervision, characterized in that: The method comprises: Get the total power of the photovoltaic area , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition: ； in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a customized hyperparameter; in, It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the constrained base load, the first-order difference of the distributed photovoltaic load, and the first-order difference of the large infrequent load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; Used to constrain the number of starts of large infrequently used loads within period T; The reference value of the distributed photovoltaic load power generation is determined according to the mutual information between the illumination intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without illumination and a typical load factor; the total load dictionary of a typical day without illumination is composed of the total load time series of the typical day without illumination; By solving the optimization model, the distributed photovoltaic load is decomposed from the total meter power of the photovoltaic area, and the target distributed photovoltaic load prediction value is obtained.

2. The method according to claim 1, characterized in that The constraints of the optimization model include: ； ； ； ； ； ； ； in, Indicates the total load forecast value of the photovoltaic area; Represents the predicted value of the basic load; Table Displays the predicted value of distributed photovoltaic load; Indicates the forecast value of large infrequently used load; represents the first difference of the base load; represents the first-order difference of the distributed photovoltaic load; represents the first difference of large infrequently used loads; and Represents a custom hyperparameter; represents the residual term of the base load prediction; is a binary vector indicating whether the start of a large infrequently used load is detected at time t. If a large infrequently used load is started at time t, then =1, otherwise =0; represents the predicted value of the distributed photovoltaic load power generation, Indicates the reference value of the photovoltaic load power generation.

3. The method according to claim 2, characterized in that The constraints of the optimization model also include: ； in, Indicates the threshold for detecting whether a large infrequently used load is activated.

4. The method according to claim 3, characterized in that The method further comprises: Obtain historical power consumption data for large, infrequently used devices; Analyzing the historical power consumption data to identify typical usage patterns and corresponding power consumption levels of the large infrequently used devices; Determine a threshold for detecting whether a large infrequently used load is activated based on a typical usage pattern and a corresponding power consumption level of the large infrequently used device. .

5. The method according to claim 2, characterized in that: The residual term of the base load forecast The formula is as follows: ； ； in, A dictionary representing the total load of N typical days without sunlight; Indicates typical load factor; Indicates the typical load value.

6. The method according to claim 2, characterized in that The reference value of photovoltaic load power generation The formula is as follows: ； in, is the rectified linear unit function, which is used to set all negative values ​​to 0 while keeping positive values ​​unchanged; express The light intensity of the area under time; , It represents the coefficient between the adjustment of the distributed photovoltaic load power generation and the light intensity of the area; , The KL divergence between the solar intensity in the area and the non-distributed photovoltaic load is obtained by minimizing: ； in, It represents the joint probability distribution of the solar intensity in the area and the non-distributed photovoltaic load; represents the probability distribution of light intensity; Represents the probability distribution of non-distributed PV load.

7. A non-intrusive identification device for distributed photovoltaic loads in a substation based on empirical value supervision, characterized in that: The device comprises: Acquisition module, used to obtain the total meter power of the photovoltaic area , light intensity in the area , build an optimization model for non-intrusive distributed photovoltaic load decomposition: ； in, , , T represents the length of the power sequence, Characterize the load reconstruction loss function, Characterize the slope loss function, Characterize the distributed photovoltaic load estimation error loss function, Characterize the base load modeling error loss function, Characterize the loss function of large infrequently used loads, is a customized hyperparameter; in, It is used to constrain the difference between the total load forecast value of the photovoltaic area and the total meter power of the photovoltaic area. The sum of the first-order difference of the constrained base load, the first-order difference of the distributed photovoltaic load, and the first-order difference of the large infrequent load; Used to constrain the difference between the predicted value of distributed photovoltaic load power generation and the reference value of distributed photovoltaic load power generation; Used to constrain the difference between the predicted base load value and the typical load value; Used to constrain the number of starts of large infrequently used loads within period T; The reference value of the distributed photovoltaic load power generation is determined according to the mutual information between the illumination intensity of the substation and the non-distributed photovoltaic load; the typical load value is determined according to the total load dictionary of a typical day without illumination and a typical load factor; the total load dictionary of a typical day without illumination is composed of the total load time series of the typical day without illumination; The decomposition module is used to decompose the distributed photovoltaic load from the total meter power of the photovoltaic area by solving the optimization model to obtain a target distributed photovoltaic load prediction value.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images