A load comparison decision-making method based on load composition

Abstract

The present invention discloses an electricity load comparison decision method based on load composition, which includes the following steps: first, presetting a parameter set and a preset load threshold; performing aggregation processing on electricity load data to obtain an electricity load data set after aggregation processing; performing image acquisition on the electricity load data set to obtain an electricity load image; preprocessing the electricity load image; setting a load unit sample interval in the electricity load image; classifying the target load units; judging the electricity load state, and by using the preset parameter set and load threshold, combined with the image preprocessing step, the analysis accuracy and reliability of the electricity load image can be significantly improved; by using algorithms such as image enhancement and high-frequency noise removal, the visual effect and analysis efficiency of the electricity load image are improved, providing more scientific and efficient technical support for electricity load evaluation, effectively reducing the time-consuming cost of manual evaluation, and improving evaluation efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of power planning and monitoring, and in particular to a load composition-based power load comparison and decision-making method. Background Art

[0002] With the continuous development and intelligentization of power systems, load management has become a key link in ensuring stable operation and improving energy efficiency. Load management involves collecting, analyzing, and evaluating power consumption data to identify and predict high-load areas, allowing appropriate optimization and management measures to be taken.

[0003] Traditional electricity load management relies primarily on manual monitoring and empirical judgment, a time-consuming and labor-intensive approach prone to misjudgment. In recent years, with the development of smart meters and sensor technology, real-time collection of electricity load data has become a major trend. However, accurately identifying and assessing load status from large amounts of data remains a challenge. Existing electricity load analysis methods often lack effective aggregation and processing steps, resulting in noise and outliers that affect the accuracy of analysis results. Their excessive reliance on manual judgment and lack of data-based evaluation criteria hinder the reduction of evaluation errors. Summary of the Invention

[0004] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0005] In view of the problems existing in the above-mentioned existing load composition-based power load comparison decision-making method, the present invention is proposed.

[0006] Therefore, the object of the present invention is to provide a method for comparing and deciding electricity loads based on load composition.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a load composition-based power load comparison decision method, comprising the following steps:

[0008] First, a preset parameter set and a preset load threshold;

[0009] The power load data is aggregated and processed to obtain an aggregated power load data set;

[0010] Performing image acquisition on the power load data set to obtain a power load image;

[0011] Preprocessing the power load image, including background noise removal, load intensity correction, and load unit overlap detection and separation;

[0012] Setting a load unit sample interval in the power load image, and identifying a target load unit within the load unit sample interval according to a preset parameter set;

[0013] classifying the target load units and calculating the ratio;

[0014] Comparing the ratio with a preset load threshold to determine the power load status; and

[0015] Dynamically adjusting the parameter set and load threshold according to actual analysis results to optimize the accuracy of subsequent power load analysis;

[0016] The parameter set includes load power parameters, load type parameters and load intensity parameters. The values ​​of the parameter set are set according to historical data, regional data and industry standards. The load threshold is set according to historical data, regional data and industry standards and can be used to judge the power load status.

[0017] As a preferred solution of the load composition-based power load comparison decision method of the present invention, the value of the parameter set is set according to at least one of the following: a parameter range obtained from historical data; a parameter range in industry standards; a parameter range in published academic literature;

[0018] The load threshold is set based on at least one of the following: a high load rate load threshold of actual historical data; an experience setting by power load management experts; or a load threshold in published academic literature.

[0019] As a preferred solution of the load composition-based power load comparison decision method described in the present invention, the preprocessing step also includes using an image enhancement algorithm to improve the contrast of the power load image; and applying a filtering algorithm to remove high-frequency noise in the image.

[0020] As a preferred solution of the load composition-based electricity load comparison decision method described in the present invention, the step of classifying the target load units includes classifying the target load units to distinguish between high-load units and low-load units; and calculating the proportion of high-load units to target load units based on the classification results.

[0021] As a preferred solution of the load composition-based electricity load comparison decision-making method described in the present invention, the dynamic adjustment step includes adjusting the parameters in the parameter set according to the deviation between the actual analysis results and the predicted results; and adjusting the preset load threshold according to the deviation between the actual analysis results and the predicted results.

[0022] As a preferred solution of the load composition-based power load comparison decision method of the present invention, it also includes:

[0023] Comparing and verifying the power load image with a known standard power load image to evaluate the accuracy of the analysis result;

[0024] Based on the results of the comparison and verification, the parameter set and load threshold are further optimized.

[0025] As a preferred solution of the load composition-based power load comparison decision method of the present invention, the specific algorithm for removing background noise in the preprocessing step is:

[0026] Calculating a local mean μ and a local standard deviation σ of the power load image, where μ represents an average value of pixels in a local area, and σ represents a standard deviation of pixels in the local area;

[0027] For each pixel , calculate its noise level ;

[0028]

[0029] in, Represents the pixel value of the image at position (x, y);

[0030] like , then the pixel is marked as a noise point, where k is the preset noise load threshold, which can control the sensitivity of noise detection;

[0031] Then, the median filter is used to smooth the noise points to reduce the impact of noise on the image.

[0032] As a preferred solution of the load composition-based power load comparison decision method of the present invention, wherein: in the preprocessing step, the specific algorithm for load intensity correction is:

[0033] Calculate the global average negative intensity of the power load image ,in, Represents the average value of all pixel values ​​in the image;

[0034] For each pixel , calculate the corrected load intensity ;

[0035]

[0036] in, Represents the original pixel value of the image at position (x, y), T IIndicates the preset target load intensity, which is used to standardize the load intensity.

[0037] As a preferred solution of the load composition-based power load comparison decision method of the present invention, in which: in the preprocessing step, the specific algorithm for load unit overlap detection and separation includes:

[0038] Use edge detection algorithm to detect the edge of the load cell and generate edge image;

[0039] For each load cell region detected, calculate its area A and perimeter P, where A represents the area of ​​the load cell region and P represents the perimeter of the load cell region;

[0040] like , it is determined that the load unit area may overlap, where C is the preset overlapping load threshold;

[0041] Overlapping load cells were separated using morphological operations to ensure the independence of each load cell region.

[0042] As a preferred embodiment of the load composition-based power load comparison decision method of the present invention, the image enhancement algorithm improves the contrast of the power load image, including through histogram equalization or adaptive contrast enhancement method;

[0043] The filtering algorithm is applied to remove high-frequency noise in the image, and a low-pass filter or a bilateral filter may be used.

[0044] The beneficial effects of the present invention are as follows: by presetting parameter sets and load thresholds, combined with image preprocessing steps, the analysis accuracy and reliability of power load images can be significantly improved; by algorithms such as image enhancement and high-frequency noise removal, the visual effect and analysis efficiency of power load images are improved, providing more scientific and efficient technical support for power load assessment; by digitizing the evaluation conditions to make the assessment more standardized, the time-consuming cost of manual assessment is effectively reduced, and the assessment efficiency and accuracy are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0046] Figure 1 The figure is a schematic diagram of the overall flow of the power load comparison decision-making method based on load composition of the present invention.

[0047] Figure 2It is a computer device of the power load comparison decision system based on load composition of the present invention. DETAILED DESCRIPTION

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0049] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0050] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0051] Furthermore, the present invention is described in detail with reference to schematic diagrams. For ease of illustration, when describing the embodiments of the present invention, cross-sectional views illustrating device structures may be partially enlarged and not to scale. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of protection of the present invention. Furthermore, in actual production, the three-dimensional dimensions of length, width, and depth should be included.

[0052] Example 1, reference Figure 1 , provides a schematic diagram of the overall flow of the load comparison decision method based on load composition, such as Figure 1 ,A power load comparison decision method based on load composition includes the following steps:

[0053] First, a preset parameter set and a preset load threshold;

[0054] The power load data is aggregated and processed to obtain an aggregated power load data set;

[0055] Perform image acquisition on the power load data set to obtain a power load image;

[0056] Preprocessing of power load images, including background noise removal, load intensity correction, and load unit overlap detection and separation;

[0057] Setting a load unit sample interval in the power load image, and identifying a target load unit within the load unit sample interval according to a preset parameter set;

[0058] Classifying the target load units and calculating the proportion; the step of classifying the target load units includes classifying the target load units to distinguish high-load units from low-load units; and calculating the proportion of high-load units to target load units based on the classification results;

[0059] Comparing the ratio with a preset load threshold to determine the power load status; and,

[0060] Dynamically adjust parameter sets and load thresholds based on actual analysis results to optimize the accuracy of subsequent power load analysis;

[0061] The parameter set includes load power parameters, load type parameters and load intensity parameters. The values ​​of the parameter set are set according to historical data, regional data and industry standards. The load threshold is set according to historical data, regional data and industry standards, and can be used to judge the power load status. The value of the parameter set is set according to at least one of the following: the parameter range obtained from historical data; the parameter range in industry standards; the parameter range in published academic literature; the setting of the load threshold is based on at least one of the following: the high load rate load threshold of actual historical data; the experience setting of power load management experts; the load threshold in published academic literature; Among them, in the field of power load management Industry standards usually include quality standards for electricity load images, standards for load unit identification and classification, and standards for aggregation and processing. For example, the quality standard for electricity load images requires clear images with moderate contrast and no obvious noise and artifacts. The standard for load unit identification and classification requires accurate classification based on the shape, size, and aggregation and processing characteristics of the load units. References include but are not limited to local historical statistical data, "Technical Specifications for Electricity Load Management", "Technical Guidelines for Electricity Load Analysis" and other industry standard documents and data. Such documents specify in detail the specific methods and standards for the acquisition, processing, analysis, and identification and classification of electricity load images.

[0062] Furthermore, the preprocessing step also includes using an image enhancement algorithm to improve the contrast of the power load image; applying a filtering algorithm to remove high-frequency noise in the image. Since the high-frequency noise here is different from the background noise type, different methods are generally used to process it; and also includes comparing and verifying the power load image with a known standard power load image to evaluate the accuracy of the analysis results; based on the results of the comparison and verification, further optimizing the parameter set and load threshold.

[0063] Furthermore, the dynamic adjustment step includes adjusting various parameters in the parameter set based on the deviation between the actual analysis results and the predicted results; and adjusting the preset load threshold based on the deviation between the actual analysis results and the predicted results. Here, the load threshold is dynamically adjusted by calculating statistical indicators such as the mean and standard deviation of historical evaluation data. For example, if the average value of the proportion of high-load units in the historical data is 30% and the standard deviation is 5%, the load threshold can be set between 25% and 35%, and the mean or median within the sliding window can be calculated in real time and used as the current load threshold. For example, a sliding window of size 10 can be set, and the mean of the proportion of high-load units within the window can be calculated in real time and used as the current load threshold. Furthermore, future trends can be predicted through time series analysis or machine learning models, and the load threshold can be dynamically adjusted accordingly. The above are specific methods for dynamic adjustment, and the basis for dynamic adjustment is: (1) adjusting various parameters in the parameter set and the preset load threshold based on the deviation between the actual analysis results and the predicted results. If the actual analysis results deviate significantly from the predicted results, the current parameter set and load threshold may not be accurate enough and need to be adjusted. (2) Feedback: The user provides accuracy feedback to the processing platform after evaluating the accuracy of the scale value through the client. Based on this user feedback, the parameter set and load threshold are adjusted to improve the accuracy of subsequent power load analysis. (3) Comparative Verification Results: The power load image is compared and verified with a known standard power load image. Based on the results of the comparative verification, the parameter set and load threshold are further optimized. If the comparative verification results show a significant difference between the current analysis results and the standard power load image, the parameter set and load threshold need to be adjusted.

[0064] In particular, in the preprocessing step, the specific algorithm for background noise removal is:

[0065] Calculate the local mean μ and local standard deviation σ of the power load image, where μ represents the average value of pixels in the local area and σ represents the standard deviation of pixels in the local area;

[0066] For each pixel , calculate its noise level ;

[0067]

[0068] in, Represents the pixel value of the image at position (x, y);

[0069] like , then the pixel is marked as a noise point, where k is the preset noise load threshold, which can control the sensitivity of noise detection;

[0070] Then, the median filter is used to smooth the noise points to reduce the impact of noise on the image.

[0071] In the preprocessing step, the specific algorithm for load intensity correction is:

[0072] Calculate the global average load intensity of the power load image ,in, Indicates the average value of the pixel values ​​in the image;

[0073] For each pixel , calculate the corrected load intensity ;

[0074]

[0075] in, Represents the original pixel value of the image at position (x, y), T I Indicates the preset target load intensity, which is used to standardize the load intensity;

[0076] In the preprocessing step, the specific algorithm for load cell overlap detection and separation includes:

[0077] Use edge detection algorithm to detect the edge of the load cell and generate edge image;

[0078] For each load cell region detected, calculate its area A and perimeter P, where A represents the area of ​​the load cell region and P represents the perimeter of the load cell region;

[0079] like , it is determined that the load unit area may overlap, where C is the preset overlapping load threshold;

[0080] Overlapping load cells were separated using morphological operations to ensure the independence of each load cell region;

[0081] Image enhancement algorithms improve the contrast of the power load image, including through histogram equalization or adaptive contrast enhancement methods;

[0082] Apply filtering algorithms to remove high-frequency noise in the image. Low-pass filters and bilateral filters can be used.

[0083] Operation process: Preset parameter sets and loads. Parameter set section: Load power parameters: Set the load cell size range, such as a load cell sample diameter between 5 and 20 mm. Load type parameters: Set the load cell sample shape characteristics, such as a load cell circularity between 0.7 and 1.0 mm. Load intensity parameters: Set the load cell load intensity range. Load cell type parameters: Set the load cell type characteristics, such as the relative size ratio of the load cell core and load cell mass.

[0084] Load threshold part: high load unit proportional load threshold; set the high load unit proportional load threshold to target load unit, the value is 30%;

[0085] Preprocess the power load data, including data cleaning, outlier processing and data normalization, to obtain the preprocessed power load data set;

[0086] Use power load data graphical acquisition and monitoring equipment to collect high-resolution images of the power load data set to obtain a power load image;

[0087] Background noise removal, load intensity correction, load unit overlap detection and separation, image enhancement and high-frequency noise removal;

[0088] Perform load unit identification and classification: set the load unit sample interval in the pre-processed power load image;

[0089] identifying a target load cell within a load cell sample interval based on a preset parameter set;

[0090] Classify the target load units and distinguish between high-load units and low-load units;

[0091] Identify high-load units: the load intensity is higher than the preset load threshold.

[0092] Identify low-load units: the load intensity is lower than the preset load threshold.

[0093] Calculate the ratio of high load units to target load units.

[0094] Then the power load state is judged: the calculated high-load unit ratio is compared with the preset load threshold (30%). If the high-load unit ratio is higher than the load threshold, it is judged as a high-load state; otherwise, it is judged as a low-load state.

[0095] Dynamic adjustment: Adjust the parameters in the parameter set and the preset load threshold based on the deviation between the actual analysis results and the predicted results. If the actual analysis results deviate significantly from the predicted results, it means that the current parameter set and load threshold may not be accurate enough and need to be adjusted.

[0096] Comparative verification: The power load image is compared with a known standard power load image to evaluate the accuracy of the analysis results. Based on the results of the comparative verification, the parameter set and load threshold are further optimized to ensure the reliability and accuracy of the analysis results.

[0097] Example 2. This example differs from the first example in that: the method also includes the following contents: data acquisition and preprocessing: collecting power load data, including but not limited to real-time monitoring of parameters such as current, voltage, and power; preprocessing the collected data, including data cleaning, outlier processing, data normalization and other steps to ensure the accuracy and reliability of the data.

[0098] Load feature extraction: Extract load features such as power factor, load rate, peak load, etc. from the preprocessed data. These features will be used for subsequent load status assessment.

[0099] Load status assessment: Based on the extracted load characteristics and preset thresholds, the power load status is assessed. For example, if the power factor is below a certain threshold, it may indicate a power factor deficiency; if the load factor exceeds a certain threshold, it may indicate that the system is overloaded.

[0100] Uncertain load state judgment rules: When the load characteristic value is between the low and high load thresholds, the uncertainty of the load state is further evaluated according to the fuzzy load state judgment rules. For example, if the load rate is between normal and overload, further analysis is required to determine whether there is local overload or nonspecific load.

[0101] Results feedback and dynamic adjustment: Based on the assessment results, preset thresholds and load signature extraction methods are dynamically adjusted to optimize the subsequent load status assessment process. For example, if a certain area is frequently misjudged, the relevant thresholds can be adjusted or the load signature extraction method can be improved.

[0102] Example 3, reference Figure 2 This embodiment differs from the above embodiment in that the present invention further provides a load composition-based power load comparison decision system, comprising:

[0103] Parameter preset module: used to preset parameter sets, which include load power parameters, load type parameters and load intensity parameters. The values ​​of the parameter sets are set according to historical data and industry standards.

[0104] Threshold preset module: used to preset thresholds. The thresholds are set according to historical data and industry standards and are used to judge the power load status.

[0105] Data acquisition module: used to collect power load data, including but not limited to real-time monitoring of parameters such as current, voltage, and power, to obtain a power load data set.

[0106] Aggregation processing module: used to preprocess the electricity load data set, including background noise removal, load intensity correction, and load unit overlap detection and separation.

[0107] Load feature extraction module: used to extract load features such as power factor, load rate, peak load, etc. from preprocessed power load data.

[0108] Load state assessment module: used to assess the power load state based on the extracted load characteristics and preset thresholds, and determine whether there is a high load state or a low load state.

[0109] Dynamic adjustment module: used to dynamically adjust parameter sets and thresholds based on actual power load diagnosis results to optimize the accuracy of subsequent power load analysis.

[0110] Result feedback module: used to feed back the evaluation results to the user and further optimize the parameter set and threshold based on the user's feedback information.

[0111] Comparison and verification module: used to compare and verify the power load data with known standard power load data to evaluate the accuracy of the analysis results.

[0112] Also included is a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program;

[0113] Also included is a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when the computer program is executed by a processor;

[0114] Furthermore, if the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0115] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0116] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, and then editing, interpreting, or processing in another suitable manner as necessary, and then storing it in a computer memory.

[0117] It is important to note that the construction and arrangement of the present application, as illustrated in various exemplary embodiments, are illustrative only. Although only a few embodiments are described in detail in this disclosure, those reading this disclosure will readily appreciate that numerous modifications are possible (e.g., variations in the size, dimensions, structure, shape, and proportions of various components, as well as parameter values ​​(e.g., temperature, pressure, etc.), mounting arrangements, use of materials, color, orientation, etc.) without materially departing from the novel teachings and advantages of the subject matter described herein. For example, components shown as integrally formed may be constructed from multiple parts or components, the positions of components may be inverted or otherwise altered, and the nature, number, or position of discrete components may be modified or changed. All such modifications are therefore intended to be encompassed within the scope of this invention. The order or sequence of any process or method steps may be altered or resequenced according to alternative embodiments. In the claims, any "means-plus-function" clause is intended to cover structures described herein that perform the recited function, and not only structural equivalents but also equivalent structures. Other substitutions, modifications, changes, and omissions may be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of this invention. Therefore, the invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0118] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment may not be described (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those that are not relevant to implementing the invention).

[0119] It will be appreciated that in the development of any actual embodiment, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but will, for those of ordinary skill having the benefit of this disclosure, be a routine undertaking of design, fabrication, and production without undue experimentation.

[0120] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A load composition-based power load comparison decision method, characterized by: The following steps are included: First, a preset parameter set and a preset load threshold; The power load data is aggregated and processed to obtain a power load data set; Performing image acquisition on the power load data set to obtain a power load image; Preprocessing the power load image, including background noise removal, load intensity correction, and load unit overlap detection and separation; Setting a load unit sample interval in the power load image, and identifying a target load unit within the load unit sample interval according to a preset parameter set; classifying the target load units and calculating the ratio; Comparing the ratio with a preset load threshold to determine the power load status; as well as, Dynamically adjusting the parameter set and load threshold according to actual analysis results to optimize the accuracy of subsequent power load analysis; Comparative verification: comparing the power load image with a known standard power load image to evaluate the accuracy of the analysis results; The parameter set includes a load power parameter, a load type parameter, and a load intensity parameter. The values ​​of the parameter set are set according to historical data, regional data, and industry standards. The load threshold is set according to historical data, regional data, and industry standards and can be used to determine the power load status. The value of the parameter set is set according to at least one of the following: a parameter range obtained from historical data; a parameter range in industry standards; a parameter range in published academic literature; The load threshold is set based on at least one of the following: a high load rate load threshold of actual historical data; an experience setting of power load management experts; a load threshold in published academic literature; The pre-processing step further comprises, Use image enhancement algorithms to improve the contrast of power load images; Apply filtering algorithms to remove high-frequency noise from images; The step of classifying the target load unit comprises: Classify the target load units and distinguish between high-load units and low-load units; The proportion of high load units to target load units is calculated based on the classification results.

2. The load composition-based power load comparison decision method according to claim 1, characterized in that: The dynamic adjustment step includes: Adjust the parameters in the parameter set according to the deviation between the actual analysis results and the predicted results; Adjust the preset load threshold based on the deviation between the actual analysis results and the predicted results.

3. The load composition-based power load comparison decision method according to claim 1 or 2, characterized in that: Also includes, Comparing and verifying the power load image with a known standard power load image to evaluate the accuracy of the analysis result; Based on the results of the comparison and verification, the parameter set and load threshold are further optimized.

4. The load composition-based power load comparison and decision-making method according to claim 3, characterized in that: In the pre-processing step, the specific algorithm for removing background noise is: Calculating a local mean μ and a local standard deviation σ of the power load image, where μ represents an average value of pixels in a local area, and σ represents a standard deviation of pixels in the local area; For each pixel I(x, y), calculate its noise level N(x, y); N(x, y) = |I(x, y - μ)| Where I(x, y) represents the pixel value of the image at position (x, y); If N(x, y)>k·σ, the pixel is marked as a noise point, where k is the preset noise load threshold, which can control the sensitivity of noise detection; Then, the median filter is used to smooth the noise points to reduce the impact of noise on the image.

5. The load composition-based power load comparison and decision-making method according to claim 1, wherein: In the pre-processing step, the specific algorithm for load intensity correction is: Calculate the global average load intensity of the power load image in, Represents the average value of all pixel values ​​in the image; For each pixel I(x, y), calculate its corrected load intensity I′(x, y); Where I(x, y) represents the original pixel value of the image at position (x, y), T I Indicates the preset target load intensity, which is used to standardize the load intensity.

6. The load composition-based power load comparison decision method according to claim 5, characterized in that: In the pre-processing step, the specific algorithm for detecting and separating the overlapped load cells includes: Use edge detection algorithm to detect the edge of the load cell and generate edge image; For each load cell region detected, calculate its area A and perimeter P, where A represents the area of ​​the load cell region and P represents the perimeter of the load cell region; like It is determined that the load unit area may overlap, where C is the preset overlapping load threshold; Overlapping load cells were separated using morphological operations to ensure the independence of each load cell region.

7. The load composition-based power load comparison decision method according to claim 3, characterized in that: The image enhancement algorithm improves the contrast of the power load image, including through histogram equalization or adaptive contrast enhancement method; The filtering algorithm is applied to remove high-frequency noise in the image, using a low-pass filter and a bilateral filter.

Images