Power demand side target user identification method and device based on multi-dimensional factor analysis

Abstract

The application relates to a power demand side target user identification method and device based on multi-dimensional factor analysis. The method performs field classification according to a historical user power consumption parameter sequence, determines a power consumption parameter sequence contained in a predicted field category, and further obtains a parameter mean sequence and a parameter offset sequence of the predicted field category. The parameter offset sequence is adjusted to a corrected offset sequence according to a prediction probability. A reference threshold sequence is obtained according to the parameter mean sequence and the corrected offset sequence. A target difference sequence is determined according to a target parameter sequence corresponding to a target user and the reference threshold sequence under the same field category. The target user target level is obtained according to the target difference sequence. The historical user power consumption parameter sequence can be adjusted to flexibly adapt to various scenes, the accuracy of target user identification is prevented from being reduced due to dynamic adjustment of the user and the power consumption parameter sequence, and the accuracy of target user identification is further improved based on field category classification.

Description

Technical Field

[0001] This invention relates to the field of power technology, and in particular to a method and device for identifying target users on the power demand side based on multidimensional factor analysis. Background Technology

[0002] With the acceleration of electricity market liberalization, demand response, as a flexible means of regulating electricity resources, is becoming increasingly important. Traditional power supply mainly relies on adjustments on the generation side to balance load. Now, demand-side resources are incorporated into a unified allocation system, prompting power supply companies to actively seek out users who can participate in response efficiently in order to cope with the complex and ever-changing market electricity supply and demand, achieve multiple goals such as peak shaving and valley filling, and alleviate congestion. This, in turn, creates a demand for technologies that accurately screen high-quality potential customers.

[0003] In existing technologies, some methods propose using machine learning and artificial intelligence models to identify target users on the electricity demand side, i.e., high-quality demand response users. However, since users' electricity consumption behavior and load characteristics are dynamically changing, existing methods are often built based on fixed parameters and rules, which cannot be adaptively adjusted in a timely manner according to changes in user conditions. This leads to significant deviations between the analysis results and the actual situation. In addition, there are huge differences in demand response potential among users in different fields (such as industry, commerce, and residential) and different types of electricity consumption (such as power consumption and lighting consumption). Existing methods often do not fully consider these differences, reducing the accuracy of the analysis.

[0004] Therefore, improving the accuracy of identifying target users on the demand side of electricity has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a method and device for identifying target users on the electricity demand side based on multidimensional factor analysis, in order to solve the problem.

[0006] Firstly, a method for identifying target users on the electricity demand side based on multidimensional factor analysis is provided, the method comprising: Obtain N historical users and the electricity consumption parameter sequence for each historical user; For any sequence of electricity consumption parameters, input the current sequence of electricity consumption parameters into the trained domain classification model to obtain the predicted domain category and the predicted probability to which the current sequence of electricity consumption parameters belongs; For any given forecasting domain category, based on all electricity consumption parameter sequences belonging to the current forecasting domain category, determine the mean parameter sequence and parameter offset sequence for the current forecasting domain category. Based on the prediction probabilities corresponding to all electricity parameter sequences belonging to the current prediction domain category, determine the reference probability of the current prediction domain category, and based on the reference probability of the current prediction domain category, determine the correction parameters corresponding to the parameter offset sequences of the current prediction domain category. Based on the parameter offset sequence of the current predicted domain category and the correction parameters corresponding to the parameter offset sequence of the current predicted domain category, determine the correction offset sequence of the current predicted domain category; Based on the parameter mean sequence and the corrected offset sequence of the current predicted domain category, determine the reference threshold sequence of the current predicted domain category; Obtain the target user and the target parameter sequence corresponding to the target user; The target parameter sequence is input into the trained domain classification model to obtain the target domain category of the target parameter sequence; Based on the target parameter sequence and the reference threshold sequence of the predicted domain category corresponding to the target domain category, determine the target difference sequence corresponding to the target parameter sequence; Based on the target difference sequence and a preset first mapping function, the target evaluation value corresponding to the target difference sequence is determined; Based on the target evaluation value and several preset evaluation value intervals, the evaluation level corresponding to the evaluation value interval to which the target evaluation value belongs is determined as the target level of the target user.

[0007] Secondly, a power demand-side target user identification device based on multidimensional factor analysis is provided, the device comprising: The first parameter acquisition module is used to acquire N historical users and the electricity consumption parameter sequence corresponding to each historical user; The first classification module is used to input the current electricity consumption parameter sequence into the trained domain classification model for any given electricity consumption parameter sequence, and obtain the predicted domain category and prediction probability to which the current electricity consumption parameter sequence belongs; The offset calculation module is used to determine the mean parameter sequence and the parameter offset sequence of the current prediction field category based on all power consumption parameter sequences belonging to the current prediction field category for any prediction field category. The correction parameter calculation module is used to determine the reference probability of the current prediction domain category based on the prediction probabilities of all power consumption parameter sequences belonging to the current prediction domain category, and to determine the correction parameters corresponding to the parameter offset sequences of the current prediction domain category based on the reference probability of the current prediction domain category. The offset correction module is used to determine the corrected offset sequence of the current predicted neighborhood category based on the parameter offset sequence of the current predicted neighborhood category and the correction parameters corresponding to the parameter offset sequence of the current predicted neighborhood category. The threshold calculation module is used to determine the reference threshold sequence for the current predicted domain category based on the parameter mean sequence and the correction offset sequence of the current predicted domain category. The second parameter acquisition module is used to acquire the target user and the target parameter sequence corresponding to the target user; The second classification module is used to input the target parameter sequence into the trained domain classification model to obtain the target domain category of the target parameter sequence; The difference calculation module is used to determine the target difference sequence corresponding to the target parameter sequence based on the target parameter sequence and the reference threshold sequence of the predicted domain category corresponding to the target domain category; The sequence evaluation module is used to determine the target evaluation value corresponding to the target difference sequence based on the target difference sequence and a preset first mapping function; The rating module is used to determine the rating level corresponding to the rating value interval to which the target rating value belongs, based on the target rating value and several preset rating value intervals, as the target level of the target user.

[0008] Thirdly, embodiments of the present invention provide a computer device, the computer device including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the electricity demand-side target user identification method as described in the first aspect.

[0009] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the electricity demand-side target user identification method as described in the first aspect.

[0010] The advantages of this invention compared to the prior art are: Based on historical user electricity consumption parameter sequences, the system classifies regions and determines the electricity consumption parameter sequences included in each predicted region category. This yields the mean parameter sequence and parameter offset sequence for each predicted region category. The corresponding parameter offset sequence is then adjusted according to the predicted probability during region classification to obtain a corrected offset sequence. A reference threshold sequence is obtained based on the mean parameter sequence and the corrected offset sequence. A target difference sequence is determined based on the target parameter sequence and the reference threshold sequence for the target user within the same region category. Evaluation is then performed based on the target difference sequence to obtain the target user's target level. This system allows for flexible adaptation to various scenarios by adjusting historical user electricity consumption parameter sequences, avoiding a decrease in the accuracy of target user identification due to dynamic adjustments to users and electricity consumption parameter sequences. Furthermore, classifying target users based on region categories further improves the accuracy of target user identification. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of an application environment for a method for identifying target users on the power demand side based on multidimensional factor analysis, provided in Embodiment 1 of the present invention. Figure 2 This is a flowchart illustrating a method for identifying target users on the power demand side based on multidimensional factor analysis, provided in Embodiment 1 of the present invention. Figure 3 This is a schematic diagram of the structure of a power demand-side target user identification device based on multidimensional factor analysis provided in Embodiment 2 of the present invention; Figure 4 This is a schematic diagram of the structure of a computer device for identifying target users on the power demand side based on multidimensional factor analysis, as provided in Embodiment 3 of the present invention. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0014] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0015] It should also be understood that the term “and / or” as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0016] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."

[0017] Furthermore, in the description of this invention and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0018] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of the invention include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0019] It should be understood that the sequence number of each step in the following embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0020] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0021] The present invention provides a method for identifying target users on the electricity demand side based on multidimensional factor analysis, which can be applied to applications such as... Figure 1 In this application environment, the server and client communicate with each other. Clients include, but are not limited to, handheld computers, desktop computers, laptops, ultra-mobile personal computers (UMPCs), netbooks, cloud terminal devices, and personal digital assistants (PDAs). The server can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0022] See Figure 2This is a flowchart illustrating a method for identifying target users on the electricity demand side based on multidimensional factor analysis, as provided in Embodiment 1 of the present invention. The above-described method for identifying target users on the electricity demand side can be applied to... Figure 1 The server receives information about historical users and their corresponding electricity consumption parameter sequences, as well as target users and their corresponding target parameter sequences. The server is equipped with a pre-trained domain classification model. It determines the target user's target level and sends it to the client to help the client identify whether the target user is a high-quality user in terms of demand response. Figure 2 As shown, the method for identifying target users on the electricity demand side may include the following steps: Step S201: Obtain N historical users and the electricity consumption parameter sequence corresponding to each historical user.

[0023] Among them, historical users refer to electricity demand-side users whose electricity consumption parameter sequences have been collected in the past. The electricity consumption parameter sequences may include the electricity load information, demand response information, etc. of historical users.

[0024] Optionally, the power consumption parameter sequence includes several power consumption parameters, which at least include load fluctuation evaluation value, power consumption evaluation value, response speed evaluation value, and response quantity evaluation value.

[0025] Among them, the load fluctuation evaluation value can be characterized by the electricity load of the corresponding historical users within a preset time period, the electricity consumption evaluation value can be characterized by the electricity consumption of the corresponding historical users within a preset time period, the response speed evaluation value can be characterized by the time point of the most recent response to the load adjustment demand of the historical users, and the response amount evaluation value can be characterized by the adjustment ratio of the most recent response to the load adjustment demand of the historical users.

[0026] Specifically, the aforementioned preset time period can be adjusted according to the implementer's needs to adapt to user identification in different scenarios. For example, the preset time period can be one day, one week, or one month, so that the user identification method provided in this embodiment can be quickly applied when the user's electricity consumption behavior and load characteristics change dynamically. By adjusting the historical user's electricity consumption parameter sequence, it can flexibly adapt to various scenarios and avoid the reduction in the accuracy of target user identification due to the dynamic adjustment of users and electricity consumption parameter sequences.

[0027] Electricity consumption parameters may also include other multi-dimensional factors such as electricity consumption regularity evaluation value, adjustable load ratio evaluation value, and power outage sensitivity evaluation value. It should be noted that the dimensions of the electricity consumption parameter sequence should be fixed. When the implementer does not need a certain electricity consumption parameter, the value of the corresponding electricity consumption parameter can be set to a specific preset value in the electricity consumption parameter sequence. In this embodiment, the preset value can be -1.

[0028] It should be noted that, for ease of subsequent processing, each power consumption parameter in this embodiment is normalized, and the condition that the closer the power consumption parameter is to 0, the better its evaluation is, and correspondingly, the closer the power consumption parameter is to 1, the worse its evaluation is. The specific normalization method can be the maximum-minimum normalization method. When the power consumption parameter is positively correlated with its evaluation, additional processing such as inverting or taking the reciprocal of the power consumption parameter can be performed to adjust the power consumption parameter and its evaluation to be negatively correlated. When the power consumption parameter and its evaluation are negatively correlated, no additional processing is required.

[0029] Step S202: For any electricity consumption parameter sequence, input the current electricity consumption parameter sequence into the trained domain classification model to obtain the predicted domain category and prediction probability to which the current electricity consumption parameter sequence belongs.

[0030] The trained domain classification model can be used to output the probability that the input electricity consumption parameter sequence belongs to each preset domain category based on the input electricity consumption parameter sequence. The maximum value of the probability that the electricity consumption parameter sequence belongs to each preset domain category is used as the prediction probability, and the preset domain category corresponding to the prediction probability is used as the prediction domain category. The preset domain categories may include categories such as industry, commerce, residential, and public affairs.

[0031] Step S203: For any prediction domain category, determine the mean parameter sequence and parameter offset sequence of the current prediction domain category based on all power consumption parameter sequences belonging to the current prediction domain category.

[0032] Since all the electricity consumption parameter sequences have the same dimension, and the electricity consumption parameters represented by the same dimension also have the same meaning, the mean parameter sequence and parameter offset sequence of the current prediction domain category can be determined based on all the electricity consumption parameter sequences belonging to the current prediction domain category.

[0033] Optionally, based on all electricity consumption parameter sequences belonging to the current forecast domain category, determine the parameter mean sequence and parameter offset sequence for the current forecast domain category, including: Calculate the mean of all electricity consumption parameter sequences belonging to the current prediction domain category to obtain the parameter mean sequence for the current prediction domain category. The parameter mean sequence includes the mean of electricity consumption parameter corresponding to each electricity consumption parameter. For any electricity consumption parameter, subtract the maximum value of the current electricity consumption parameter from the mean value of the electricity consumption parameter corresponding to the current electricity consumption parameter in all electricity consumption parameter sequences belonging to the current prediction domain category to obtain the electricity consumption parameter offset value corresponding to the current electricity consumption parameter; The parameter offset sequence for the current prediction domain category is formed by the power consumption parameter offset values ​​corresponding to each power consumption parameter.

[0034] Specifically, for any dimension of the electricity consumption parameter sequence, the mean value of the element values ​​of all electricity consumption parameter sequences belonging to the current prediction domain category in that dimension is calculated to obtain the mean value of the electricity consumption parameter corresponding to that dimension. Then, the mean values ​​of the electricity consumption parameter corresponding to all dimensions are spliced ​​together to form the parameter mean value sequence.

[0035] Specifically, since all electricity consumption parameters are adjusted to be negatively correlated with their evaluation in this embodiment, when calculating the offset value, it is only necessary to consider the difference between the maximum value of the current electricity consumption parameter in all electricity consumption parameter sequences belonging to the current prediction domain category and the average value of the current electricity consumption parameter to obtain the electricity consumption parameter offset value corresponding to the current electricity consumption parameter. This electricity consumption parameter offset value can refer to the maximum limit allowed to exceed the average value of the electricity consumption parameter under the corresponding prediction domain category. When the difference between the value of a user's electricity consumption parameter element in a certain dimension and the average value of the electricity consumption parameter exceeds the maximum limit, it can be considered that the corresponding user is unlikely to meet the general standard under the domain category in that dimension.

[0036] Step S204: Based on the prediction probabilities corresponding to all electricity parameter sequences belonging to the current prediction domain category, determine the reference probability of the current prediction domain category; based on the reference probability of the current prediction domain category, determine the correction parameters corresponding to the parameter offset sequences of the current prediction domain category.

[0037] The reference probability can characterize the predictive reliability of the current prediction domain category.

[0038] Optionally, based on the prediction probabilities corresponding to all electricity consumption parameter sequences belonging to the current prediction domain category, a reference probability for the current prediction domain category is determined. Based on the reference probability of the current prediction domain category, correction parameters corresponding to the parameter offset sequences of the current prediction domain category are determined, including: Calculate the mean of the prediction probabilities for all electricity parameter sequences belonging to the current prediction domain category, and use the mean calculation result as the reference probability for the current prediction domain category. The reference probability of the current predicted domain category is used as the correction parameter corresponding to the parameter offset sequence of the current predicted domain category.

[0039] The reference probability has a range of [0, 1]. The closer the reference probability of the current predicted domain category is to 1, the stronger the prediction reliability of the current predicted domain category. Accordingly, the parameter offset sequence of the current predicted domain category can more accurately distinguish the current predicted domain category from other predicted domain categories. Conversely, the closer the reference probability of the current predicted domain category is to 0, the weaker the prediction reliability of the current predicted domain category. Accordingly, the parameter offset sequence of the current predicted domain category is difficult to accurately distinguish the current predicted domain category from other predicted domain categories. Therefore, it is necessary to appropriately reduce the parameter offset sequence to ensure the accuracy of the current predicted domain category classification. In this embodiment, the reference probability of the current predicted domain category is directly used as the correction parameter corresponding to the parameter offset sequence of the current predicted domain category.

[0040] Step S205: Determine the corrected offset sequence of the current predicted domain category based on the parameter offset sequence of the current predicted domain category and the correction parameters corresponding to the parameter offset sequence of the current predicted domain category.

[0041] The corrected offset sequence can be used to subsequently determine the reference threshold sequence.

[0042] Optionally, based on the parameter offset sequence of the current predicted neighborhood category and the correction parameters corresponding to the parameter offset sequence of the current predicted neighborhood category, a corrected offset sequence for the current predicted neighborhood category is determined, including: Multiply the correction parameter corresponding to the parameter offset sequence of the current predicted neighborhood category with the parameter offset sequence of the current predicted neighborhood category, and use the result of the multiplication as the correction offset sequence of the current predicted neighborhood category.

[0043] The correction parameter is in numerical form, while the parameter offset sequence is in vector form. The correction offset sequence is obtained by multiplying the correction parameter by each element value in the parameter offset sequence.

[0044] Step S206: Determine the reference threshold sequence for the current predicted domain category based on the parameter mean sequence and the corrected offset sequence.

[0045] The reference threshold sequence can be used to identify target users in conjunction with the target parameter sequence.

[0046] Specifically, the mean parameter sequence and the corrected offset sequence of the current predicted domain category are added point by point, and the sum is used as the reference threshold sequence for the current predicted domain category.

[0047] Step S207: Obtain the target user and the target parameter sequence corresponding to the target user.

[0048] Among them, the target user can refer to the user to be identified, and the target parameter sequence contains the target parameter values ​​corresponding to each electricity consumption parameter.

[0049] Step S208: Input the target parameter sequence into the trained domain classification model to obtain the target domain category of the target parameter sequence.

[0050] The target domain category can refer to a preset domain category predicted based on the target parameter sequence.

[0051] Step S209: Determine the target difference sequence corresponding to the target parameter sequence based on the target parameter sequence and the reference threshold sequence of the predicted domain category of the corresponding target domain category.

[0052] The target difference sequence can be used for subsequent evaluation of target users.

[0053] Optionally, based on the target parameter sequence and the reference threshold sequence of the predicted domain category corresponding to the target domain category, the target difference sequence corresponding to the target parameter sequence is determined, including: The target parameter sequence is subtracted point by point from the reference threshold sequence of the predicted domain category of the corresponding target domain category, and the result of the subtraction is used as the target difference sequence corresponding to the target parameter sequence.

[0054] The target domain category corresponds to a preset domain category, which in turn corresponds to the predicted domain category.

[0055] Specifically, the target difference sequence can contain positive and negative elements. When the elements in the target difference sequence are positive, it means that the target user's evaluation of the corresponding dimension is lower than the minimum standard of the domain category. When the elements in the target difference sequence are negative, it means that the target user's evaluation of the corresponding dimension is higher than the minimum standard of the domain category.

[0056] Step S210: Determine the target evaluation value corresponding to the target difference sequence based on the target difference sequence and the preset first mapping function.

[0057] The target difference sequence can contain several target differences, and the first mapping function can be set as follows: ,in, Let i be the i-th target difference in the target difference sequence, where i is an integer in the range [1, I], and I is the number of target differences contained in the target difference sequence. is the mapping value corresponding to the i-th target difference in the target difference sequence, and k is the adjustment coefficient, which should be greater than 0.

[0058] The target evaluation value Y can be characterized as ,in, This could refer to the preset weights of the corresponding dimension.

[0059] Specifically, the smaller the difference of the i-th target, the larger its corresponding mapping value, indicating a better evaluation of the i-th target difference; conversely, the larger the difference of the i-th target, the smaller its corresponding mapping value, indicating a worse evaluation of the i-th target difference.

[0060] Step S211: Based on the target evaluation value and several preset evaluation value intervals, determine the evaluation level corresponding to the evaluation value interval to which the target evaluation value belongs as the target level of the target user.

[0061] The evaluation value range can be set by the implementer, but it should be set within the range of [0, 2] and there should be no overlap.

[0062] Specifically, in this embodiment, the range [0, 1] is used as the first evaluation value interval, and the range [1, 2] is used as the second evaluation value interval. That is, in this embodiment, the target level is divided into two levels, so the client side can identify the target user with the high target level as the high-quality user of demand response.

[0063] In this embodiment, domain classification is performed based on the electricity consumption parameter sequences of historical users to determine the electricity consumption parameter sequences included in each predicted domain category. This results in the parameter mean sequence and parameter offset sequence for each predicted domain category. The corresponding parameter offset sequence is then adjusted according to the predicted probability during domain classification to obtain a corrected offset sequence. A reference threshold sequence is obtained based on the parameter mean sequence and the corrected offset sequence. A target difference sequence is determined based on the target parameter sequence and the reference threshold sequence for the target user within the same domain category. Evaluation is then performed based on the target difference sequence to obtain the target user's target level. By adjusting the electricity consumption parameter sequences of historical users, this method can flexibly adapt to various scenarios, avoiding a decrease in the accuracy of target user identification due to dynamic adjustments to users and electricity consumption parameter sequences. Furthermore, classifying target users based on domain categories further improves the accuracy of target user identification.

[0064] Corresponding to the method in the above embodiments, Figure 3 This diagram illustrates a structural schematic of a power demand-side target user identification device based on multidimensional factor analysis, as provided in Embodiment 2 of the present invention. This device is applied to a server. The server receives each historical user and their corresponding electricity consumption parameter sequence, as well as target users and their corresponding target parameter sequence. A pre-trained domain classification model is deployed within the server. The server determines the target user's target level and sends it to the client to assist the client in identifying whether the target user is a high-quality user in terms of demand response. For ease of explanation, only the parts relevant to this embodiment are shown.

[0065] See Figure 3 The electricity demand-side target user identification device includes: The first parameter acquisition module 301 is used to acquire N historical users and the electricity consumption parameter sequence corresponding to each historical user; The first classification module 302 is used to input the current electricity consumption parameter sequence into the trained domain classification model for any electricity consumption parameter sequence, and obtain the predicted domain category and predicted probability to which the current electricity consumption parameter sequence belongs; The offset calculation module 303 is used to determine the mean parameter sequence and the parameter offset sequence of the current prediction field category based on all power consumption parameter sequences belonging to the current prediction field category for any prediction field category. The correction parameter calculation module 304 is used to determine the reference probability of the current prediction domain category based on the prediction probabilities of all power consumption parameter sequences belonging to the current prediction domain category, and to determine the correction parameters corresponding to the parameter offset sequence of the current prediction domain category based on the reference probability of the current prediction domain category. The offset correction module 305 is used to determine the corrected offset sequence of the current predicted neighborhood category based on the parameter offset sequence of the current predicted neighborhood category and the correction parameter corresponding to the parameter offset sequence of the current predicted neighborhood category. The threshold calculation module 306 is used to determine the reference threshold sequence for the current predicted domain category based on the parameter mean sequence and the correction offset sequence of the current predicted domain category. The second parameter acquisition module 307 is used to acquire the target user and the target parameter sequence corresponding to the target user; The second classification module 308 is used to input the target parameter sequence into the trained domain classification model to obtain the target domain category of the target parameter sequence; The difference calculation module 309 is used to determine the target difference sequence corresponding to the target parameter sequence based on the target parameter sequence and the reference threshold sequence of the predicted domain category of the corresponding target domain category; The sequence evaluation module 310 is used to determine the target evaluation value corresponding to the target difference sequence based on the target difference sequence and the preset first mapping function; The rating module 311 is used to determine the rating level corresponding to the rating value interval to which the target rating value belongs, based on the target rating value and several preset rating value intervals, as the target level of the target user.

[0066] Optionally, in the first parameter acquisition module 301 mentioned above, the power consumption parameter sequence includes several power consumption parameters, and the power consumption parameters include at least the load fluctuation evaluation value, the power consumption evaluation value, the response speed evaluation value, and the response quantity evaluation value.

[0067] Optionally, the offset calculation module 303 mentioned above includes: The mean calculation unit is used to calculate the mean of all power consumption parameter sequences belonging to the current prediction domain category, and obtain the parameter mean sequence of the current prediction domain category. The parameter mean sequence includes the power consumption parameter mean corresponding to each power consumption parameter. The offset calculation unit is used to subtract the maximum value of the current power consumption parameter in all power consumption parameter sequences belonging to the current prediction domain category from the mean value of the current power consumption parameter to obtain the offset value of the current power consumption parameter. The sequence forming unit is used to form a parameter offset sequence for the current prediction domain category from the power consumption parameter offset values ​​corresponding to each power consumption parameter.

[0068] Optionally, the above-mentioned correction parameter calculation module 304 includes: The reference probability calculation unit is used to calculate the mean of the prediction probabilities of all power consumption parameter sequences belonging to the current prediction domain category, and the mean calculation result is used as the reference probability of the current prediction domain category. The correction parameter determination unit is used to use the reference probability of the current predicted domain category as the correction parameter corresponding to the parameter offset sequence of the current predicted domain category.

[0069] Optionally, the offset correction module 305 mentioned above includes: The corrected offset sequence calculation unit is used to multiply the corrected parameter corresponding to the parameter offset sequence of the current predicted neighborhood category with the parameter offset sequence of the current predicted neighborhood category, and use the multiplication result as the corrected offset sequence of the current predicted neighborhood category.

[0070] Optionally, the threshold calculation module 306 mentioned above includes: The reference threshold sequence calculation unit is used to add the parameter mean sequence and the corrected offset sequence of the current predicted neighborhood category point by point, and use the sum as the reference threshold sequence of the current predicted neighborhood category.

[0071] Optionally, the above-mentioned difference calculation module 309 includes: The target difference sequence calculation unit is used to subtract the target parameter sequence from the reference threshold sequence of the predicted domain category of the corresponding target domain category point by point, and use the subtraction result as the target difference sequence corresponding to the target parameter sequence.

[0072] It should be noted that the information interaction and execution process between the above modules and units are based on the same concept as the method embodiments of the present invention. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0073] Figure 4This is a schematic diagram of the computer device used in a method for identifying target users on the electricity demand side based on multidimensional factor analysis, as provided in Embodiment 3 of the present invention. Figure 4 As shown, the computer device of this embodiment includes: at least one processor ( Figure 4 Only one is shown in the diagram), a memory, and a computer program stored in the memory and executable on at least one processor, which, when executed by the processor, implements the steps in any of the embodiments of the electricity demand-side target user identification method based on multidimensional factor analysis described above.

[0074] This computer device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 4 The examples of computer devices are merely examples and do not constitute a limitation on computer devices. Computer devices may include more or fewer components than shown in the illustration, or combinations of certain components, or different components, such as network interfaces, displays, and input devices.

[0075] The processor referred to can be a CPU, but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0076] Memory includes readable storage media, internal memory, etc., wherein internal memory can be the RAM of a computer device, providing an environment for the operation of the operating system and computer-readable instructions stored in the readable storage media. The readable storage media can be the hard drive of a computer device, or in other embodiments, it can be an external storage device of the computer device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, memory can include both internal storage units and external storage devices of the computer device. Memory is used to store the operating system, applications, bootloader, data, and other programs, such as program code for computer programs. Memory can also be used to temporarily store data that has been output or will be output.

[0077] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the functions described above can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the processes in the methods of the above embodiments by instructing related hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium can include at least: any entity or device capable of carrying computer program code, a recording medium, a computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0078] The present invention can implement all or part of the processes in the methods of the above embodiments, or it can be accomplished by a computer program product. When the computer program product is run on a computer device, the computer device executes the steps in the above method embodiments.

[0079] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0080] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0081] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / computer devices and methods can be implemented in other ways. For example, the apparatus / computer device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0082] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0083] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for identifying target users on the electricity demand side based on multidimensional factor analysis, characterized in that, The method includes: Obtain N historical users and the electricity consumption parameter sequence for each historical user; For any sequence of electricity consumption parameters, input the current sequence of electricity consumption parameters into the trained domain classification model to obtain the predicted domain category and the predicted probability to which the current sequence of electricity consumption parameters belongs; For any given forecasting domain category, based on all electricity consumption parameter sequences belonging to the current forecasting domain category, determine the mean parameter sequence and parameter offset sequence for the current forecasting domain category. Based on the prediction probabilities corresponding to all electricity parameter sequences belonging to the current prediction domain category, determine the reference probability of the current prediction domain category, and based on the reference probability of the current prediction domain category, determine the correction parameters corresponding to the parameter offset sequences of the current prediction domain category. Based on the parameter offset sequence of the current predicted domain category and the correction parameters corresponding to the parameter offset sequence of the current predicted domain category, determine the correction offset sequence of the current predicted domain category; Based on the parameter mean sequence and the corrected offset sequence of the current predicted domain category, determine the reference threshold sequence of the current predicted domain category; Obtain the target user and the target parameter sequence corresponding to the target user; The target parameter sequence is input into the trained domain classification model to obtain the target domain category of the target parameter sequence; Based on the target parameter sequence and the reference threshold sequence of the predicted domain category corresponding to the target domain category, determine the target difference sequence corresponding to the target parameter sequence; Based on the target difference sequence and a preset first mapping function, the target evaluation value corresponding to the target difference sequence is determined; Based on the target evaluation value and several preset evaluation value intervals, the evaluation level corresponding to the evaluation value interval to which the target evaluation value belongs is determined as the target level of the target user.

2. The method for identifying target users on the electricity demand side according to claim 1, characterized in that, The power consumption parameter sequence includes several power consumption parameters, which at least include load fluctuation evaluation value, power consumption evaluation value, response speed evaluation value, and response quantity evaluation value.

3. The method for identifying target users on the electricity demand side according to claim 2, characterized in that, The step of determining the mean parameter sequence and parameter offset sequence for the current prediction domain category based on all electricity consumption parameter sequences belonging to the current prediction domain category includes: Calculate the mean of all power consumption parameter sequences belonging to the current prediction domain category to obtain the parameter mean sequence of the current prediction domain category. The parameter mean sequence includes the power consumption parameter mean corresponding to each power consumption parameter. For any electricity consumption parameter, subtract the maximum value of the current electricity consumption parameter from the mean value of the electricity consumption parameter corresponding to the current electricity consumption parameter in all electricity consumption parameter sequences belonging to the current prediction domain category to obtain the electricity consumption parameter offset value corresponding to the current electricity consumption parameter; The parameter offset sequence for the current prediction domain category is formed by the power consumption parameter offset values ​​corresponding to each power consumption parameter.

4. The method for identifying target users on the electricity demand side according to claim 1, characterized in that, The step of determining the reference probability of the current prediction domain category based on the prediction probabilities corresponding to all electricity parameter sequences belonging to the current prediction domain category, and determining the correction parameters corresponding to the parameter offset sequences of the current prediction domain category based on the reference probability of the current prediction domain category, includes: Calculate the mean of the prediction probabilities for all electricity parameter sequences belonging to the current prediction domain category, and use the mean calculation result as the reference probability for the current prediction domain category. The reference probability of the current predicted domain category is used as the correction parameter corresponding to the parameter offset sequence of the current predicted domain category.

5. The method for identifying target users on the electricity demand side according to claim 1, characterized in that, The step of determining the corrected offset sequence for the current predicted neighborhood category based on the parameter offset sequence of the current predicted neighborhood category and the corrected parameters corresponding to the parameter offset sequence of the current predicted neighborhood category includes: Multiply the correction parameter corresponding to the parameter offset sequence of the current predicted neighborhood category with the parameter offset sequence of the current predicted neighborhood category, and use the result of the multiplication as the correction offset sequence of the current predicted neighborhood category.

6. The method for identifying target users on the electricity demand side according to claim 1, characterized in that, The step of determining the reference threshold sequence for the current predicted neighborhood category based on the parameter mean sequence and the corrected offset sequence includes: The mean sequence of parameters and the corrected offset sequence of the current predicted neighborhood category are added point by point, and the sum is used as the reference threshold sequence for the current predicted neighborhood category.

7. The method for identifying target users on the electricity demand side according to any one of claims 1 to 6, characterized in that, The step of determining the target difference sequence corresponding to the target parameter sequence based on the target parameter sequence and the reference threshold sequence of the predicted domain category corresponding to the target domain category includes: The target parameter sequence is subtracted point by point from the reference threshold sequence of the predicted domain category corresponding to the target domain category, and the result of the subtraction is used as the target difference sequence corresponding to the target parameter sequence.

8. A computer device, characterized in that, The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the electricity demand-side target user identification method as described in any one of claims 1 to 7.

Images