Load identification method and device for industrial park, equipment and medium

Abstract

The embodiment of the invention discloses a load identification method and device for an industrial park, equipment and a medium. The method comprises the following steps: acquiring power consumption data of load equipment in an industrial park; inputting the power consumption data into a load feature recognition model to obtain initial load feature data; wherein the load feature recognition model is obtained by iterating a first model through a power consumption data sample set; matching the initial load characteristic data with a set load rule base to obtain a characteristic matching result; and determining a target load feature according to the feature matching result and the initial load feature data. According to the technical scheme, the load characteristics can be comprehensively recognized according to the load characteristic recognition model and the set load rule base, the accuracy and efficiency of load recognition can be improved, and the requirement of intelligent load management can be met.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a load identification method, device, equipment and medium for an industrial park. Background Art

[0002] With the rapid development of modern social economy and the continuous growth of electricity demand, the power system is facing huge challenges. The traditional power system needs to develop in a more intelligent, digital and efficient direction to meet the growing electricity demand and diversified electricity load.

[0003] As an important part of smart grid, power load identification can provide power companies and users with accurate load characteristics and operating rules by analyzing and identifying different types of power loads, thereby achieving more accurate demand response and optimized management. However, traditional power load identification methods still have many limitations when dealing with complex and diverse load characteristics. Although existing time series analysis methods can capture the temporal characteristics of load data, it is difficult to identify the frequency characteristics and subtle differences of load types. Therefore, they rely heavily on the experience and subjective judgment of experts, lack systematic and automated identification methods, are inefficient and easily affected by human factors, and have obvious limitations when dealing with complex and diverse load characteristics, and cannot meet the needs of modern power systems for efficient, accurate and intelligent load management. Summary of the invention

[0004] The present invention provides a load identification method, device, equipment and medium for an industrial park, which can comprehensively identify load characteristics according to a load characteristic identification model and a set load rule library, and can improve the accuracy and efficiency of load identification to meet the needs of intelligent load management.

[0005] According to one aspect of the present invention, a load identification method for an industrial park is provided, comprising:

[0006] Obtain power consumption data of load equipment in the industrial park;

[0007] Inputting the power consumption data into a load characteristic recognition model to obtain initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating a first model through a power consumption data sample set;

[0008] Matching the initial load characteristic data with a set load rule library to obtain a characteristic matching result;

[0009] A target load characteristic is determined according to the characteristic matching result and the initial load characteristic data.

[0010] According to another aspect of the present invention, there is provided a load identification device for an industrial park, comprising:

[0011] A data acquisition module is used to obtain the power consumption data of load equipment in the industrial park;

[0012] An initial load characteristic data acquisition module is used to input the power consumption data into a load characteristic recognition model to obtain initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating a first model through a power consumption data sample set;

[0013] A load matching module, used for matching the initial load characteristic data with a set load rule library to obtain a characteristic matching result;

[0014] The target load characteristic determination module is used to determine the target load characteristic according to the characteristic matching result and the initial load characteristic data.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the load identification method for an industrial park described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the load identification method for an industrial park described in any embodiment of the present invention when executed.

[0020] The technical solution of the embodiment of the present invention is to obtain the power consumption data of the load equipment in the industrial park; input the power consumption data into the load characteristic recognition model to obtain the initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating the first model through the power consumption data sample set; match the initial load characteristic data with the set load rule library to obtain the characteristic matching result; determine the target load characteristic according to the characteristic matching result and the initial load characteristic data. This technical solution can comprehensively identify the load characteristics according to the load characteristic recognition model and the set load rule library, and can improve the accuracy and efficiency of load identification to meet the needs of intelligent load management.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 is a flow chart of a load identification method for an industrial park provided according to Embodiment 1 of the present invention;

[0024] Figure 2 is a flow chart of a load identification method for an industrial park provided according to Embodiment 2 of the present invention;

[0025] Figure 3 This is a schematic diagram of the structure of a load identification device for an industrial park provided according to a third embodiment of the present invention;

[0026] Figure 4 It is a schematic diagram of the structure of an electronic device provided according to Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", "target" and "initial" in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] Embodiment 1

[0030] Figure 1This is a flow chart of a method for identifying loads in an industrial park according to the first embodiment of the present invention. This embodiment is applicable to the case of identifying loads in an industrial park. The method can be executed by a load identification device for an industrial park. The load identification device for an industrial park can be implemented in the form of hardware and / or software. The load identification device for an industrial park can be configured in an electronic device with data processing capabilities. Figure 1 As shown, the method includes:

[0031] S110. Obtain power consumption data of load equipment in the industrial park.

[0032] Among them, the industrial park can be understood as a region that, according to the inherent requirements of its own economic development, uses administrative means to demarcate an area, gathers various production factors, and scientifically integrates them within a certain spatial range. Load equipment can be understood as various types of electrical equipment that consume electrical energy. The load equipment in this embodiment may refer to photovoltaic power generation equipment, air-conditioning equipment, and car charging pile equipment in the industrial park. The electricity consumption data can be understood as the electricity consumption data of various load equipment in the industrial park, which can specifically cover the use of photovoltaic power generation equipment, air-conditioning electricity, and car charging pile equipment. The electricity consumption data in this embodiment may include real-time power, voltage, current, active power, reactive power, power factor, electric energy data, and various event data. In this embodiment, real-time electricity consumption data of various load equipment in the industrial park can be obtained.

[0033] S120 , inputting the power consumption data into a load characteristic recognition model to obtain initial load characteristic data.

[0034] Among them, the load characteristic recognition model is obtained by iterating the first model through the power consumption data sample set. The load characteristic recognition model can be used to extract and identify the neural network model of load characteristics. The initial load characteristic data can be understood as the load characteristic value obtained by identification and extraction based on the load characteristic recognition model. The initial load characteristic data in this embodiment may include load characteristic data of photovoltaic power generation, air conditioning power consumption and car charging piles. The load characteristic recognition model in this embodiment can be obtained by iterating the first model through the power consumption data sample set. The power consumption data sample set can be sample data obtained by performing data cleaning and data preprocessing operations based on the historical power consumption data of various types of load equipment. The first model can be understood as a first model that has not been trained with data.

[0035] In this embodiment, the acquired real-time power consumption data of the load device can be input into a trained load characteristic recognition model, so that the load characteristic recognition model outputs the corresponding initial load characteristic data after processing operations such as feature extraction and recognition.

[0036] In this embodiment, optionally, iterating the first model through the electricity consumption data sample set includes: obtaining the electricity consumption data sample set and the first model; inputting the electricity consumption data samples in the electricity consumption data sample set into the first model to obtain first load characteristic data; training the parameters of the first model according to the first load characteristic data; returning to execute the operation of inputting the electricity consumption data samples in the electricity consumption data sample set into the first model to obtain the first load characteristic data, until a load characteristic recognition model is obtained.

[0037] Among them, the power consumption data sample set can be sample data obtained by collecting detailed power consumption information of photovoltaic power generation equipment, air-conditioning equipment and car charging pile equipment in the industrial park and performing preprocessing and other operations. The power consumption data may include data such as power, voltage, current, active power, reactive power, power factor and electric energy. The first model can be an untrained convolutional neural network model (CNN). In this embodiment, a first model can be constructed, specifically including an input layer, a convolution layer, a pooling layer, a fully connected layer and an output layer. The first load characteristic data can be understood as the load characteristic value of any power consumption data sample in the power consumption data sample set input into the first model and processed and output.

[0038] In this embodiment, detailed power consumption information of photovoltaic power generation equipment, air conditioning equipment and automobile charging pile equipment in the industrial park during the historical period can be obtained and collected and preprocessed to obtain a power consumption data sample set, and a first model is constructed, and then each power consumption data sample contained in the power consumption data sample set is input into the first model to obtain the corresponding load characteristic data. In this embodiment, the specific method of training the parameters of the first model according to the first load characteristic data can be to input the power consumption data samples in each power consumption data sample set into the first model, train the model according to the different load characteristic parameters obtained, and return to execute the operation of inputting the power consumption data samples in the power consumption data sample set into the first model to obtain the first load characteristic data, and adjust the parameters in the first model through the set loss function so that the first model meets the set requirements, that is, to obtain the load characteristic recognition model.

[0039] In this embodiment, the first model can be trained using the data of the power consumption data sample set, the network parameters can be updated through back propagation, the model performance can be verified on the test set, and the model effect can be evaluated using indicators such as accuracy, recall rate and F1-score. It is also possible to adjust hyperparameters such as the convolution kernel size, the number of network layers, and the learning rate to optimize the model performance, and use the cross-validation method to further evaluate and optimize the model to ensure its generalization ability. Specifically, in this example, the feature matrix after wavelet transformation is input into the input layer. The input is an N×M matrix, where N is the number of time steps and M is the number of features. A convolution kernel of size 3×3 is applied for feature extraction, and a maximum pooling layer is used to reduce the dimension of the feature map and reduce the amount of calculation. The features output by the convolution layer are flattened and then classified through the fully connected layer. The ReLU activation function is applied after each convolution layer and the fully connected layer to improve the nonlinear representation ability of the network, the cross entropy loss function is used to evaluate the classification results, and the Adam optimization algorithm is used to accelerate network training.

[0040] Through such a setting in this embodiment, the load characteristic recognition model can be obtained by iterating the constructed first model through the acquired power consumption data sample set, thereby improving the accuracy of model load recognition.

[0041] In this embodiment, optionally, obtaining a sample set of electricity consumption data includes: collecting historical electricity consumption data of various load devices in the industrial park; preprocessing the historical electricity consumption data to obtain preprocessed historical electricity consumption data; performing wavelet packet transform on the preprocessed historical electricity consumption data to obtain decomposed load data; performing feature extraction on the decomposed load data and clustering it using a clustering algorithm to obtain clustered data; and using the clustered data of various load devices as a sample set of electricity consumption data.

[0042] Among them, the historical electricity consumption data can be the electricity consumption data generated by various load devices within the historical period. The historical period can be one month, two months or five months, and can be set according to actual needs. In this embodiment, the specific method of collecting the historical electricity consumption data of various load devices in the industrial park can be to collect detailed electricity consumption information of equipment such as photovoltaics, air conditioners and car charging piles in the park through the smart meters installed in the industrial park in the past historical period. The data types cover real-time power, voltage, current, active power, reactive power, power factor, electric energy data and various event data. The data is recorded in time series, and high-frequency sampling is used to ensure that the changing characteristics of the load are captured. It is transmitted to the central data server through the communication network for storage and management, and provides comprehensive and accurate data support for the load identification and classification of subsequent models. In this embodiment, it can be ensured that the data covers multiple time periods to enhance the comprehensiveness and representativeness of the data.

[0043] Among them, preprocessing can include data cleaning operations, data standardization operations, and noise removal processing operations. This embodiment can clean the historical power consumption data of various load devices in the industrial park, remove noise and abnormal values, fill in the missing parts in the data, and ensure that the data is complete and reliable.

[0044] In this embodiment, the specific method of preprocessing the historical power consumption data can be to clean the collected data, calculate the mean and standard deviation of the data, use the 3σ rule to detect outliers, and draw a box plot for visual analysis to identify outliers in the data. For the detected outliers, the deletion method is used to deal with obviously wrong values, and for smaller outliers, the mean of adjacent points is used to replace them to maintain the continuity and integrity of the data. For the missing data, the interpolation method is used to fill the missing values ​​according to the time trend of the data. Further, the data after the above cleaning is subjected to data standardization operation, and the data can be normalized to the [0,1] interval, which is suitable for the scenario where the input is a neural network, and then the data is converted to a standard normal distribution with a mean of 0 and a standard deviation of 1, which is suitable for machine learning algorithms. In this embodiment, the standardized data can also be subjected to feature analysis, extracting time domain features (such as mean, standard deviation, kurtosis and skewness), frequency domain features (such as FFT spectrum and energy distribution) and wavelet features (such as wavelet coefficients and energy features), and selecting the most representative features through correlation analysis and dimensionality reduction methods, and then storing the standardized data in a relational database or NoSQL database to facilitate subsequent data analysis.

[0045] In this embodiment, different filtering methods can be used for various types of electricity consumption data for horizontal comparison, and the best filtering method can be selected to remove the noise in the data. Specifically, the type of noise in the data is first identified, which may include high-frequency noise, low-frequency drift, and mutation noise, etc. Through data visualization and statistical analysis, the characteristics and distribution of the noise are understood in order to select a suitable denoising method. In this embodiment, the noise in the industrial park electricity consumption data may include high-frequency noise generated by equipment startup and shutdown, low-frequency drift caused by power grid fluctuations, and mutation noise caused by accidental events. By drawing data curves and calculating statistics, the characteristics and distribution of the noise are analyzed. For example, air-conditioning electricity consumption data may have high-frequency noise when the equipment is started and shut down, while photovoltaic power generation data may be affected by low-frequency fluctuations caused by weather changes. Therefore, in this embodiment, the initial state and covariance matrix can be set according to the data characteristics of air-conditioning electricity consumption and car charging piles, and the covariance matrix of appropriate process noise and measurement noise can be selected. Then the Kalman filter is applied to recursively filter the air-conditioning electricity consumption data, and the noise is gradually removed through prediction and update steps. For example, when processing car charging pile data, the Kalman filter can effectively smooth the fluctuations caused by changes in charging state. Furthermore, in this embodiment, the curves before and after the load data is filtered can also be drawn to compare the noise removal effect and ensure the smoothness and authenticity of the signal. For example, the photovoltaic power generation data curves before and after filtering are compared to verify the denoising effect. Then, the statistics such as the mean, standard deviation and autocorrelation of the data before and after filtering are calculated to evaluate the denoising effect. For example, check whether the standard deviation of the air-conditioning power consumption data is significantly reduced, indicating that the high-frequency noise has been effectively removed. Compare the effects of different denoising methods (moving average, low-pass filtering and Kalman filtering) to select the denoising method that best suits the current data characteristics. For example, by comparing the car charging pile data before and after filtering, it is better to select the Kalman filter.

[0046] Among them, Wavelet Packet Transform (WPT) is a data analysis method that can decompose data into wavelet packets of different frequencies to better describe the local characteristics of the signal. Decomposed load data can be understood as different data features obtained by decomposition through wavelet packet transform processing. Decomposed load data can include statistical characteristics, frequency domain characteristics, and time domain characteristics.

[0047] In this embodiment, the pre-processed historical electricity consumption data can be subjected to wavelet packet transformation, and the signal can be decomposed into different frequency levels to obtain decomposed load data, so as to extract more detailed frequency information. During the wavelet packet transformation, a suitable wavelet basis function can be selected and the number of decomposition layers can be determined. Afterwards, key features are extracted from the sub-signals obtained after the decomposition. In this embodiment, Daubechies wavelet can be selected as the wavelet basis function, and the number of decomposition layers is determined to be 5 layers according to the characteristics of the electricity consumption data. The standardized electricity consumption data is subjected to wavelet packet decomposition and decomposed into sub-signals of multiple frequency components. Among them, the number of decomposition layers can be set according to actual needs.

[0048] Specifically, in this embodiment, after the historical electricity consumption data is processed by wavelet packet transform, it is decomposed to obtain multiple layers of sub-signal data, and feature extraction can be performed on the decomposed load data, such as statistical features such as mean, standard deviation, and energy can be extracted from each layer of sub-signals. For example, the energy characteristics of photovoltaic power generation can reflect the change in sunshine intensity. In this embodiment, frequency domain features can be extracted by analyzing the spectrum center and spectrum width of each layer of sub-signals. For example, the spectrum width feature of air-conditioning electricity can reflect the frequency change of air-conditioning use. In this embodiment, time domain features can be extracted by analyzing the time domain waveform and peak features of each layer of sub-signals. For example, the peak feature of a car charging pile can reflect the power peak during the charging process. In this embodiment, the correlation between each feature and the load type (photovoltaic power generation, air-conditioning electricity, and car charging pile) can also be calculated by a correlation algorithm, and feature selection is performed based on the correlation to select features with high correlation to obtain an extracted feature set. For example, the energy characteristics of photovoltaic power generation and the spectrum center features of air-conditioning electricity. Alternatively, a random forest model can be used to calculate the importance of features and select features with high importance. The selected features are reduced in dimension by PCA to retain the main information and reduce the feature dimension. For example, multiple statistical features and frequency domain features are reduced in dimension to two principal components.

[0049] Among them, the clustering algorithm may include a K-means clustering algorithm. In this embodiment, the K-means algorithm can be used to perform preliminary clustering analysis on the extracted features, and similar power consumption patterns can be classified into different categories. The optimal number of clusters K is determined by the elbow method. Based on the preliminary clustering, the Gaussian mixture model (GMM) is used to further refine the clustering results. The parameters of the GMM are estimated by the expectation maximization (EM) algorithm, so as to more accurately identify complex power consumption data features.

[0050] In this embodiment, after extracting features from the decomposed load data, an extracted feature set is obtained, and the extracted feature set is clustered using a clustering algorithm to obtain clustered data. Then, the clustered data of various types of load equipment can be used as a sample set of electricity consumption data.

[0051] Specifically, in this embodiment, the K-means algorithm can be used to perform preliminary clustering on the extracted feature set, and the appropriate number of clusters can be determined by methods such as the elbow method or the silhouette coefficient, and the selected features are clustered by K-means to divide the data into K preliminary clusters. Then the Gaussian mixture model GMM is used for refined clustering. The Gaussian mixture model is a probabilistic model that assumes that the data consists of multiple Gaussian distributions (normal distributions), and each Gaussian distribution represents a cluster. GMM estimates the model parameters through the expectation maximization (EM) algorithm to maximize the log-likelihood value of the data. The Gaussian distribution formula is:

[0052]

[0053] Where μ is the mean vector and Σ is the covariance matrix.

[0054] In this embodiment, GMM is a linear combination of multiple Gaussian distributions, and the formula is: Among them, π K is the mixing coefficient of the kth Gaussian distribution, satisfying The maximum likelihood function for The formula for the log-likelihood function of GMM is: The expectation maximization (EM) algorithm is used to iteratively estimate the parameters of the GMM to maximize the log-likelihood value. The EM algorithm consists of two steps: the E step (expectation step) and the M step (maximization step). The E step (expectation step) is to calculate the data point x i The posterior probability (i.e., responsibility) of belonging to the kth Gaussian distribution is given by: The M step (maximization step) is to update the parameters of GMM to maximize the log-likelihood value. The formula for updating the hybrid system is: The formula for updating the mean vector is: The formula for updating the covariance matrix is:

[0055] In this embodiment, the GMM can be initialized first, the parameters of the GMM can be estimated by the expectation maximization (EM) algorithm, the results of the preliminary clustering can be refined, and finally the quality of the GMM clustering results can be evaluated using indicators such as log likelihood or BIC (Bayesian Information Criterion). In this embodiment, the elbow method is used to determine the number of clusters to be 3 for the feature set, namely photovoltaic power generation, air conditioning power consumption, and car charging piles. K-means clustering is performed on the selected features, and the power consumption data is preliminarily divided into 3 clusters. The results of K-means clustering are used as the initial parameters of GMM, and the preliminary clustering results give the initial mean μ k , covariance matrix∑ k The initial clustering results are refined by the expectation maximization (EM) algorithm to further distinguish subtle differences and calculate the responsibility γ of each data point. ik , and update the model parameters πk , μ k ,∑ k , the log-likelihood value and BIC were used to evaluate the quality of the GMM clustering results to ensure the clustering effect.

[0056] In this embodiment, the results of cluster analysis can be used as a preliminary classification of load patterns, and different characteristic patterns in electricity consumption data are preliminarily identified, such as photovoltaic power generation, air conditioning electricity consumption, and car charging piles. The preliminary clustering of K means provides a basis for the subsequent refined clustering of Gaussian mixture model (GMM). GMM further optimizes the clustering effect through the expectation maximization (EM) algorithm, so that complex load patterns can be divided more accurately. The final clustering results are used as feature inputs for the subsequent convolutional neural network (CNN) model to train the constructed first model. In this embodiment, the clustering results can also be divided into training sets and test sets at a ratio of 2:8, and the first CNN model is trained using the training set data, and the network parameters are updated by back propagation. The performance of the CNN model is verified on the test set, and the model effect is evaluated using indicators such as accuracy, recall, and F1-score.

[0057] In this embodiment, when constructing a convolutional neural network (CNN) model, i.e., the first model, the results of cluster analysis are used as the input of CNN to help the model more accurately extract high-level features of different load modes. In this embodiment, through such a setting, through the clustering results of K-means and GMM, CNN can further identify complex load behaviors based on the classified load modes, and then use the clustering data of various types of load equipment as a sample set of power consumption data, so that CNN can better extract features based on preliminary classification and avoid interference between different load modes, thereby improving the overall accuracy of load identification.

[0058] S130, matching the initial load characteristic data with the set load rule library to obtain a characteristic matching result.

[0059] The set load rule base may be a pre-established database, which may include different types of typical load characteristic rules and corresponding set data threshold ranges and other contents. The set load rule base in this embodiment may be a load rule base constructed based on historical load characteristic data and expert knowledge, wherein the set load rule base may specifically include typical load characteristics and operating rules of photovoltaic power generation, air conditioning power consumption, and automobile charging piles. In this embodiment, the initial load characteristic data may be verified by setting the load rule base to verify whether the initial load characteristic data is accurate, and then determine whether correction processing is required to improve the accuracy of the load characteristic data.

[0060] Specifically, in this embodiment, the typical characteristics of photovoltaic power generation, air conditioning power consumption and car charging piles can be defined based on historical load characteristic data and expert knowledge to construct a set load rule base. Exemplarily, typical characteristics may include the daytime energy fluctuation characteristics of photovoltaic power generation, the start-stop frequency characteristics of air conditioning power consumption, and the charging power characteristics of car charging piles. Summarize the operating rules of various types of loads, such as the peak period of photovoltaic power generation, the temperature correlation of air conditioning power consumption, and the charging time and frequency of car charging piles. In this embodiment, the structure of the set load rule base designed may include information such as feature description, threshold range, and typical mode for easy query and matching. In this embodiment, the load rule base may be stored in a database, such as a relational database (MySQL) or a NoSQL database (MongoDB), and a table structure may be designed to store the characteristics and rules of different load types. The feature matching results may include two matching results: the load feature data meets the threshold range defined by the set rule base and the load feature data does not meet the threshold range defined by the set rule base.

[0061] This embodiment matches the identified initial load characteristic data with the typical characteristics in the set load rule library. It can be determined whether the initial load characteristic value is within the threshold range defined by the set load characteristic rule library. If the load characteristic data meets the threshold range defined by the set rule library, it means that the initial load characteristic data is within a reasonable range and no correction intervention operation is required. If there is a deviation in the match, the initial load characteristic data can also be corrected using the threshold range in the set load rule library to ensure the accuracy of the final load characteristic identification result. In this embodiment, the initial load characteristic data is matched by using this set load rule library, and a feedback and verification mechanism is performed based on the set load rule library to ensure that the identification process is both accurate and reliable.

[0062] S140: Determine the target load characteristic according to the characteristic matching result and the initial load characteristic data.

[0063] Among them, the target load characteristics can be understood as the final load identification result corresponding to the acquired real-time power consumption data. In this embodiment, the specific method of determining the target load characteristics according to the characteristic matching results and the initial load characteristic data can be to first determine whether the initial load characteristics need to be corrected according to the characteristic matching results, and if the initial load characteristics need to be corrected, then determine the corrected load characteristic data, and perform fusion processing according to different weights based on the corrected load characteristic data, the initial load characteristic data, and the standard data in the set load rule library in the characteristic matching results to obtain the final load identification result.

[0064] The technical solution of the embodiment of the present invention obtains the power consumption data of the load equipment in the industrial park; inputs the power consumption data into the load characteristic recognition model to obtain the initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating the first model through the power consumption data sample set; the initial load characteristic data is matched with the set load rule library to obtain the characteristic matching result; and the target load characteristic is determined according to the characteristic matching result and the initial load characteristic data. This technical solution can comprehensively identify the load characteristics according to the load characteristic recognition model and the set load rule library, and can improve the accuracy and efficiency of load identification to meet the needs of intelligent load management.

[0065] Embodiment 2

[0066] Figure 2 This is a flow chart of a load identification method for an industrial park provided according to the second embodiment of the present invention. This embodiment is optimized based on the above embodiment. The specific optimization is: matching the initial load characteristic data with the set load rule library to obtain a characteristic matching result, including: determining the standard characteristic parameter range of the corresponding load characteristic according to the initial load characteristic data; wherein the standard characteristic parameter range is obtained from the set load rule library; comparing the initial load characteristic data with the standard characteristic parameter range, and determining the comparison result as the characteristic matching result. Figure 2 As shown, the method includes:

[0067] S210. Obtain power consumption data of load equipment in the industrial park.

[0068] S220, input the power consumption data into the load characteristic recognition model to obtain initial load characteristic data. The load characteristic recognition model is obtained by iterating the first model through the power consumption data sample set;

[0069] S230. Determine a standard characteristic parameter range corresponding to the load characteristic according to the initial load characteristic data.

[0070] The standard characteristic parameter range is obtained from the set load rule library. The standard characteristic parameter range may be the standard characteristic parameter range corresponding to each load characteristic parameter. In this embodiment, the initial load characteristic parameter obtained from the model identification may be determined to set the standard characteristic parameter range corresponding to the typical load characteristic in the set load rule library, so that the initial load characteristic data is matched with the corresponding standard characteristic parameter range. Exemplarily, the specific rules in the set load rule in this embodiment are shown in Table 1;

[0071] Table 1 Loading rules

[0072]

[0073]

[0074]

[0075] In addition, in the load identification process of this embodiment, the initial load characteristic data output by the load characteristic identification model needs to be matched with the typical characteristics in the set load rule base, specifically, the identification load characteristic parameters are matched with the characteristic range in the rule base, and whether the initial load characteristic data is reasonable can be detected. If there is a deviation in the matching result, the gap between the calculated characteristic value and the threshold in the rule base can also be fed back to the model for real-time optimization and adjustment, thereby improving the recognition accuracy of the model. The feedback mechanism can be implemented by adjusting the parameters of the load characteristic identification model. After receiving the deviation information, the model can dynamically adjust the weights and learning rates of the convolutional neural network (CNN) or other algorithms, or improve the recognition performance by optimizing hyperparameters. In addition, when the deviation is large, the model can be fine-tuned by retraining the model to adapt to new data changes.

[0076] S240: Compare the initial load characteristic data with the standard characteristic parameter range, and determine the comparison result as the characteristic matching result.

[0077] In this embodiment, the initial load characteristic data can be compared with the corresponding standard characteristic parameter range to determine whether the initial load characteristic data is within the standard characteristic parameter range. If the initial load characteristic data is within the standard characteristic parameter range, the feature matching result is that the initial load characteristic data is matched, which means that the initial load characteristic data identified by the model is relatively accurate; if the initial load characteristic data does not meet the standard characteristic parameter range, the feature matching result is that the initial load characteristic data is not matched, which means that the initial load characteristic data identified by the model is inaccurate and needs to be corrected based on the set load rule library.

[0078] In this embodiment, optionally, it further includes: if the initial load characteristic data does not conform to the standard characteristic parameter range, the initial load characteristic data is corrected based on the standard characteristic parameter range to obtain corrected load characteristic data.

[0079] The correction can be understood as correcting and adjusting the initial load characteristic data. The corrected load characteristic data can be the load characteristic data obtained by correcting the initial load characteristic data according to the standard characteristic parameter range in the set load rule library.

[0080] In this embodiment, when the initial load characteristic data does not conform to the standard characteristic parameter range in the set load rule library, the initial load characteristic data can be corrected according to the standard characteristic parameter range in the set load rule library. Specifically, the initial load characteristic data can be directly replaced by the standard value in the standard characteristic parameter range in the set load rule library, or the initial load characteristic data can be corrected by weighted averaging based on the initial load characteristic data and the standard value in the standard characteristic parameter range. Among them, the standard value can be a typical value selected within the standard characteristic parameter range, which can be selected according to actual needs, and this embodiment does not limit this.

[0081] In this embodiment, by such an arrangement, the initial load characteristic data can be corrected according to the standard value within the standard characteristic parameter range, thereby ensuring the accuracy of the final output load characteristic result.

[0082] In this embodiment, optionally, the initial load characteristic data is corrected based on the standard characteristic parameter range to obtain the corrected load characteristic data, including: fusing the set data within the standard characteristic parameter range with the initial load characteristic data to obtain fused characteristic data; replacing the initial load characteristic data with the fused characteristic data to obtain the corrected load characteristic data.

[0083] The setting data may be any selected standard value in the standard feature parameter range, or may be data selected from the standard feature parameter range according to actual needs. Fusion may be understood as a weighted average processing operation. The fused feature data may be understood as feature data obtained by weighted averaging the setting data and the initial feature data.

[0084] In this embodiment, the set data selected from the standard characteristic parameter range and the initial load characteristic data can be weighted averaged to obtain the weighted average characteristic data, and the weighted average characteristic data can be used to replace the initial load characteristic data to obtain the corrected load characteristic data; alternatively, in this embodiment, the set data in the standard characteristic parameter range can be used to directly replace the initial load characteristic data as the corrected load characteristic data.

[0085] Exemplarily, the model identification result is the characteristic value extracted from the initial load characteristic data. Taking the power generation as an example, if the power fluctuation of photovoltaic power generation should be between 400kW and 600kW, and the initial load characteristic data output by the model is 700kW, if the power fluctuation characteristic of photovoltaic power generation is 700kW in the model output, and the set data selected in the set load rule library is 600kW, then the weighted average of 700kW and 600kW is obtained to obtain 650kW, that is, the corrected load characteristic data is 650kW.

[0086] In this embodiment, by such an arrangement, the initial load characteristic data can be corrected by weighted averaging the setting data within the standard characteristic parameter range and the initial load characteristic data, so as to improve the accuracy of the load identification data.

[0087] S250: Determine the target load characteristic according to the characteristic matching result and the initial load characteristic data.

[0088] In this embodiment, optionally, the target load characteristic is determined based on the feature matching result and the initial load characteristic data, including: respectively determining a first weight parameter, a second weight parameter and a third weight parameter corresponding to the initial load characteristic data, the set data and the corrected load characteristic data; performing weighted summation of the initial load characteristic data, the set data and the corrected load characteristic data based on the first weight parameter, the second weight parameter and the third weight parameter to obtain the target load characteristic.

[0089] Among them, the first weight parameter is the weight parameter corresponding to the initial load characteristic data. The second weight parameter is the weight parameter corresponding to the setting data. The third weight parameter is the weight parameter corresponding to the corrected load characteristic data. Exemplarily, the first weight parameter in this embodiment can be set to 0.5, the second weight parameter can be set to 0.3, and the third weight parameter can be set to 0.2. The first weight parameter, the second weight parameter and the third weight parameter in this embodiment can be set according to actual needs, and this embodiment does not limit this.

[0090] In this embodiment, a weighted voting method is adopted, in which the initial load characteristic data obtained by the model output, the set data in the set load rule library, and the load characteristic data after feedback correction are used as voting members and assigned different weights for weighted summation to obtain the final target load characteristic identification result.

[0091] In this embodiment, the final output result is determined by combining the output result of the load feature identification model, the matching result of the set load rule library, and the corrected compliance result. Different weights can be given to the results of each data source according to its reliability, performance in a specific scenario, or evaluation based on historical data.

[0092] In this example, the first weight parameter is set to 0.5, the second weight parameter is set to 0.3, and the third weight parameter is set to 0.2. The specific method of weighted summing the initial load characteristic data, the setting data, and the revised load characteristic data based on the first weight parameter, the second weight parameter, and the third weight parameter can be to multiply the initial load characteristic data, the setting data, and the revised load characteristic data by the corresponding weights to perform a weighted summing operation. The calculation formula of the weighted score is as follows: Final score = (initial load characteristic data × first weight parameter) + (setting data × second weight parameter) + (revised load characteristic data × third weight parameter).

[0093] The final result of weighted voting in this embodiment will output the type and specific value of load identification. For example, if it is load identification of photovoltaic power generation, its characteristics will be output, such as power fluctuation, start and stop frequency, etc., and combined with weights to ensure that these outputs meet the actual situation and the standards of the rule base. In this embodiment, if the final target load characteristic output result still has a large error or deviation, the load characteristic identification model can be further optimized by retraining or parameter adjustment. The process of weighted voting can also continuously optimize the weight distribution to make future identification more accurate.

[0094] In this embodiment, the eigenvalues ​​extracted from the model recognition results are taken as an example. For example, if the power fluctuation of photovoltaic power generation should be between 400kW and 600kW, and the model output is 700kW, if the power fluctuation characteristic of photovoltaic power generation is 700kW in the model output, the suggestion given by the rule base matching is 600kW, the feedback correction value is 650kW, and the weights are 0.5, 0.3 and 0.2 respectively, then the value of the final power fluctuation is calculated as: Final power fluctuation = (700×0.5) + (600×0.3) + (650×0.2) = 350+180+130 = 660kW. The final recognition result in this embodiment will not be a single value, but a combination of multiple eigenvalues. These eigenvalues ​​together constitute the complete characteristics of the load, that is, the target load characteristics.

[0095] In this embodiment, through such a setting, multiple output results and their corresponding weight parameters can be combined to perform weighted summation to determine the final target load characteristics, thereby improving the accuracy and reliability of the target load characteristics.

[0096] The technical solution of the embodiment of the present invention is to obtain the electricity consumption data of the load equipment in the industrial park; input the electricity consumption data into the load characteristic recognition model to obtain the initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating the first model through the electricity consumption data sample set; determine the standard characteristic parameter range of the corresponding load characteristic according to the initial load characteristic data; wherein the standard characteristic parameter range is obtained from the set load rule library; compare the initial load characteristic data with the standard characteristic parameter range, and determine the comparison result as the characteristic matching result; determine the target load characteristic according to the characteristic matching result and the initial load characteristic data. This technical solution can comprehensively identify the load characteristics according to the load characteristic recognition model and the set load rule library, which can improve the accuracy and efficiency of load identification to meet the needs of intelligent load management, and can also provide strong support for smart grids and energy management to achieve the goals of energy conservation, emission reduction and optimization of electricity consumption strategies.

[0097] Embodiment 3

[0098] Figure 3 1 is a schematic diagram of the structure of a load identification device for an industrial park according to Embodiment 3 of the present invention. Figure 3 As shown, the device comprises:

[0099] The data acquisition module 310 is used to acquire the power consumption data of the load equipment in the industrial park;

[0100] The initial load characteristic data acquisition module 320 is used to input the power consumption data into the load characteristic recognition model to obtain the initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating the first model through the power consumption data sample set;

[0101] The load matching module 330 is used to match the initial load characteristic data with the set load rule library to obtain a characteristic matching result;

[0102] The target load characteristic determination module 340 is used to determine the target load characteristic according to the characteristic matching result and the initial load characteristic data.

[0103] Optionally, the load matching module 330 is specifically used to determine a standard characteristic parameter range of the corresponding load characteristic based on the initial load characteristic data; wherein the standard characteristic parameter range is obtained from a set load rule library; the initial load characteristic data is compared with the standard characteristic parameter range, and the comparison result is determined as a characteristic matching result.

[0104] Optionally, the device further comprises:

[0105] The data correction module is used to correct the initial load characteristic data based on the standard characteristic parameter range if the initial load characteristic data does not conform to the standard characteristic parameter range to obtain the corrected load characteristic data.

[0106] Optionally, a data correction module is specifically used to fuse the set data and the initial load characteristic data within the standard characteristic parameter range to obtain fused characteristic data; and replace the initial load characteristic data with the fused characteristic data to obtain corrected load characteristic data.

[0107] Optionally, the target load characteristic determination module 340 is specifically used to determine the first weight parameter, the second weight parameter and the third weight parameter corresponding to the initial load characteristic data, the set data and the revised load characteristic data, respectively; based on the first weight parameter, the second weight parameter and the third weight parameter, the initial load characteristic data, the set data and the revised load characteristic data are weighted and summed to obtain the target load characteristic.

[0108] Based on the above solution, optionally, the load identification device of the industrial park may further include the following modules to iterate the first model through the power consumption data sample set:

[0109] A sample set acquisition module, used to acquire a power consumption data sample set and a first model;

[0110] A sample input module, used for inputting the power consumption data samples in the power consumption data sample set into the first model to obtain the first load characteristic data;

[0111] A model training module, used for training parameters of a first model according to first load characteristic data;

[0112] The loop execution module is used to return to execute the operation of inputting the power consumption data samples in the power consumption data sample set into the first model to obtain the first load characteristic data until the load characteristic recognition model is obtained.

[0113] Optionally, a sample set acquisition module is specifically used to collect historical electricity consumption data of various load equipment in the industrial park; preprocess the historical electricity consumption data to obtain preprocessed historical electricity consumption data; perform wavelet packet transform on the preprocessed historical electricity consumption data to obtain decomposed load data; perform feature extraction on the decomposed load data and cluster it using a clustering algorithm to obtain clustered data; and use the clustered data of various load equipment as a sample set of electricity consumption data.

[0114] A load identification device for an industrial park provided by an embodiment of the present invention can execute a load identification method for an industrial park provided by any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0115] Embodiment 4

[0116] Figure 41 is a schematic diagram of the structure of an electronic device provided according to Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0117] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0118] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0119] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a load identification method for an industrial park.

[0120] In some embodiments, the load identification method for an industrial park may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the load identification method for the industrial park described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the load identification method for the industrial park in any other appropriate manner (e.g., by means of firmware).

[0121] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0122] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0123] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0124] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0125] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0126] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0127] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0128] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A load identification method for an industrial park, characterized in that: include: Obtain power consumption data of load equipment in the industrial park; Inputting the power consumption data into a load characteristic recognition model to obtain initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating a first model through a power consumption data sample set; Matching the initial load characteristic data with a set load rule library to obtain a characteristic matching result; A target load characteristic is determined according to the characteristic matching result and the initial load characteristic data.

2. The method according to claim 1, characterized in that The initial load characteristic data is matched with the set load rule library to obtain a characteristic matching result, including: Determining a standard characteristic parameter range corresponding to the load characteristic according to the initial load characteristic data; wherein the standard characteristic parameter range is obtained from the set load rule library; The initial load characteristic data is compared with the standard characteristic parameter range, and the comparison result is determined as a characteristic matching result.

3. The method according to claim 2, characterized in that Also includes: If the initial load characteristic data does not conform to the standard characteristic parameter range, the initial load characteristic data is corrected based on the standard characteristic parameter range to obtain corrected load characteristic data.

4. The method according to claim 3, characterized in that The initial load characteristic data is corrected based on the standard characteristic parameter range to obtain the corrected load characteristic data, including: Based on the set data in the standard characteristic parameter range and the initial load characteristic data, fusion is performed to obtain fused characteristic data; The initial load characteristic data is replaced by the fused characteristic data to obtain the corrected load characteristic data.

5. The method according to claim 3, characterized in that: Determining a target load characteristic according to the characteristic matching result and the initial load characteristic data includes: respectively determining a first weight parameter, a second weight parameter and a third weight parameter corresponding to the initial load characteristic data, the setting data and the modified load characteristic data; The initial load characteristic data, the set data and the corrected load characteristic data are weighted and summed based on the first weight parameter, the second weight parameter and the third weight parameter to obtain a target load characteristic.

6. The method according to claim 1, characterized in that Iterating the first model through the electricity consumption data sample set includes: Obtaining a sample set of electricity consumption data and a first model; Inputting the power consumption data samples in the power consumption data sample set into the first model to obtain first load characteristic data; training parameters of the first model according to the first load characteristic data; Return to execute the operation of inputting the power consumption data samples in the power consumption data sample set into the first model to obtain the first load characteristic data, until a load characteristic recognition model is obtained.

7. The method according to claim 6, characterized in that Get a sample set of electricity consumption data, including: Collect historical electricity consumption data of various load equipment in the industrial park; Preprocessing the historical electricity consumption data to obtain preprocessed historical electricity consumption data; Performing wavelet packet transformation on the pre-processed historical power consumption data to obtain decomposed load data; Extract features from the decomposed load data and cluster them using a clustering algorithm to obtain cluster data; The clustering data of various load devices are used as the electricity consumption data sample set.

8. A load identification device for an industrial park, characterized in that: include: A data acquisition module is used to obtain the power consumption data of load equipment in the industrial park; An initial load characteristic data acquisition module is used to input the power consumption data into a load characteristic recognition model to obtain initial load characteristic data; wherein the load characteristic recognition model is obtained by iterating a first model through a power consumption data sample set; A load matching module, used for matching the initial load characteristic data with a set load rule library to obtain a characteristic matching result; The target load characteristic determination module is used to determine the target load characteristic according to the characteristic matching result and the initial load characteristic data.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the load identification method for an industrial park described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the load identification method for an industrial park according to any one of claims 1 to 7 when executed.

Images